├── .gitattributes
├── .github
├── dependabot.yml
└── workflows
│ └── test.yml
├── .gitignore
├── COPYING
├── COPYING.gpl
├── COPYING.mit
├── Documentation
├── Contributing.md
├── FAQ.md
├── converting-from-bfg-repo-cleaner.md
├── converting-from-filter-branch.md
├── examples-from-user-filed-issues.md
└── git-filter-repo.txt
├── INSTALL.md
├── Makefile
├── README.md
├── contrib
└── filter-repo-demos
│ ├── README.md
│ ├── barebones-example
│ ├── bfg-ish
│ ├── clean-ignore
│ ├── convert-svnexternals
│ ├── filter-branch-ish
│ ├── filter-lamely
│ ├── insert-beginning
│ ├── lint-history
│ └── signed-off-by
├── git-filter-repo
├── git_filter_repo.py
├── pyproject.toml
└── t
├── run_coverage
├── run_tests
├── t9390-filter-repo-basics.sh
├── t9390
├── basic
├── basic-filename
├── basic-mailmap
├── basic-message
├── basic-numbers
├── basic-replace
├── basic-ten
├── basic-twenty
├── degenerate
├── degenerate-evil-merge
├── degenerate-globme
├── degenerate-keepme
├── degenerate-keepme-noff
├── degenerate-moduleA
├── empty
├── empty-keepme
├── less-empty-keepme
├── more-empty-keepme
├── sample-mailmap
├── sample-message
├── sample-replace
├── unusual
├── unusual-filtered
└── unusual-mailmap
├── t9391-filter-repo-lib-usage.sh
├── t9391
├── commit_info.py
├── create_fast_export_output.py
├── emoji-repo
├── erroneous.py
├── file_filter.py
├── print_progress.py
├── rename-master-to-develop.py
├── splice_repos.py
├── strip-cvs-keywords.py
└── unusual.py
├── t9392-filter-repo-python-callback.sh
├── t9393-filter-repo-rerun.sh
├── t9393
├── lfs
└── simple
├── t9394-filter-repo-sanity-checks-and-bigger-repo-setup.sh
├── t9394
└── date-order
├── test-lib-functions.sh
└── test-lib.sh
/.gitattributes:
--------------------------------------------------------------------------------
1 | *.sh eol=lf
2 | *.py eol=lf
3 | /git-filter-repo eol=lf
4 | /contrib/filter-repo-demos/[a-z]* eol=lf
5 | /t/t9*/* eol=lf
6 |
--------------------------------------------------------------------------------
/.github/dependabot.yml:
--------------------------------------------------------------------------------
1 | ---
2 | version: 2
3 | updates:
4 | - package-ecosystem: "github-actions"
5 | directory: "/"
6 | schedule:
7 | interval: "monthly"
--------------------------------------------------------------------------------
/.github/workflows/test.yml:
--------------------------------------------------------------------------------
1 | name: Run tests
2 |
3 | on: [push, pull_request]
4 |
5 | jobs:
6 | run-tests:
7 | strategy:
8 | matrix:
9 | os: [ 'windows', 'ubuntu', 'macos' ]
10 | fail-fast: false
11 | runs-on: ${{ matrix.os }}-latest
12 | steps:
13 | - uses: actions/checkout@v4
14 | - name: Setup python
15 | uses: actions/setup-python@v5
16 | with:
17 | python-version: 3.x
18 | - name: test
19 | shell: bash
20 | run: |
21 | # setup-python puts `python` into the `PATH`, not `python3`, yet
22 | # `git-filter-repo` expects `python3` in the `PATH`. Let's add
23 | # a shim.
24 | printf '#!/bin/sh\n\nexec python "$@"\n' >python3 &&
25 |
26 | export PATH=$PWD:$PATH &&
27 |
28 | if ! t/run_tests -q -v -x
29 | then
30 | mkdir failed &&
31 | tar czf failed/failed.tar.gz t
32 | exit 1
33 | fi
34 | - name: upload failed tests' directories
35 | if: failure()
36 | uses: actions/upload-artifact@v4
37 | with:
38 | name: failed-${{ matrix.os }}
39 | path: failed
40 |
--------------------------------------------------------------------------------
/.gitignore:
--------------------------------------------------------------------------------
1 | /Documentation/html/
2 | /Documentation/man1/
3 | /t/test-results
4 | /t/trash directory*
5 | /__pycache__/
6 |
--------------------------------------------------------------------------------
/COPYING:
--------------------------------------------------------------------------------
1 | git-filter-repo itself and most the files in this repository (exceptions
2 | noted below) are provided under the MIT license (see COPYING.mit).
3 |
4 | The usage of the MIT license probably makes filter-repo compatible with
5 | everything, but just in case, these files can also be used under whatever
6 | open source license[1] that git.git or libgit2 use now or in the future
7 | (currently GPL[2] and GPL-with-linking-exception[3]). Further, the
8 | examples (in contrib/filter-repo-demos/ and t/t9391/) can also be used
9 | under the same license that libgit2 provides their examples under (CC0,
10 | currently[4]).
11 |
12 | Exceptions:
13 |
14 | - The test harness (t/test-lib.sh, t/test-lib-functions.sh) is a slightly
15 | modified copy of git.git's test harness (the difference being that my
16 | copy doesn't require a built version of 'git' to be present). These
17 | are thus GPL2 (see COPYING.gpl), and are individually marked as such.
18 |
19 |
20 | [1] ...as defined by the Open Source Initiative (https://opensource.org/)
21 | [2] https://git.kernel.org/pub/scm/git/git.git/tree/COPYING
22 | [3] https://github.com/libgit2/libgit2/blob/master/COPYING
23 | [4] https://github.com/libgit2/libgit2/blob/master/examples/COPYING
24 |
--------------------------------------------------------------------------------
/COPYING.gpl:
--------------------------------------------------------------------------------
1 | GNU GENERAL PUBLIC LICENSE
2 | Version 2, June 1991
3 |
4 | Copyright (C) 1989, 1991 Free Software Foundation, Inc.,
5 | 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
6 | Everyone is permitted to copy and distribute verbatim copies
7 | of this license document, but changing it is not allowed.
8 |
9 | Preamble
10 |
11 | The licenses for most software are designed to take away your
12 | freedom to share and change it. By contrast, the GNU General Public
13 | License is intended to guarantee your freedom to share and change free
14 | software--to make sure the software is free for all its users. This
15 | General Public License applies to most of the Free Software
16 | Foundation's software and to any other program whose authors commit to
17 | using it. (Some other Free Software Foundation software is covered by
18 | the GNU Lesser General Public License instead.) You can apply it to
19 | your programs, too.
20 |
21 | When we speak of free software, we are referring to freedom, not
22 | price. Our General Public Licenses are designed to make sure that you
23 | have the freedom to distribute copies of free software (and charge for
24 | this service if you wish), that you receive source code or can get it
25 | if you want it, that you can change the software or use pieces of it
26 | in new free programs; and that you know you can do these things.
27 |
28 | To protect your rights, we need to make restrictions that forbid
29 | anyone to deny you these rights or to ask you to surrender the rights.
30 | These restrictions translate to certain responsibilities for you if you
31 | distribute copies of the software, or if you modify it.
32 |
33 | For example, if you distribute copies of such a program, whether
34 | gratis or for a fee, you must give the recipients all the rights that
35 | you have. You must make sure that they, too, receive or can get the
36 | source code. And you must show them these terms so they know their
37 | rights.
38 |
39 | We protect your rights with two steps: (1) copyright the software, and
40 | (2) offer you this license which gives you legal permission to copy,
41 | distribute and/or modify the software.
42 |
43 | Also, for each author's protection and ours, we want to make certain
44 | that everyone understands that there is no warranty for this free
45 | software. If the software is modified by someone else and passed on, we
46 | want its recipients to know that what they have is not the original, so
47 | that any problems introduced by others will not reflect on the original
48 | authors' reputations.
49 |
50 | Finally, any free program is threatened constantly by software
51 | patents. We wish to avoid the danger that redistributors of a free
52 | program will individually obtain patent licenses, in effect making the
53 | program proprietary. To prevent this, we have made it clear that any
54 | patent must be licensed for everyone's free use or not licensed at all.
55 |
56 | The precise terms and conditions for copying, distribution and
57 | modification follow.
58 |
59 | GNU GENERAL PUBLIC LICENSE
60 | TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
61 |
62 | 0. This License applies to any program or other work which contains
63 | a notice placed by the copyright holder saying it may be distributed
64 | under the terms of this General Public License. The "Program", below,
65 | refers to any such program or work, and a "work based on the Program"
66 | means either the Program or any derivative work under copyright law:
67 | that is to say, a work containing the Program or a portion of it,
68 | either verbatim or with modifications and/or translated into another
69 | language. (Hereinafter, translation is included without limitation in
70 | the term "modification".) Each licensee is addressed as "you".
71 |
72 | Activities other than copying, distribution and modification are not
73 | covered by this License; they are outside its scope. The act of
74 | running the Program is not restricted, and the output from the Program
75 | is covered only if its contents constitute a work based on the
76 | Program (independent of having been made by running the Program).
77 | Whether that is true depends on what the Program does.
78 |
79 | 1. You may copy and distribute verbatim copies of the Program's
80 | source code as you receive it, in any medium, provided that you
81 | conspicuously and appropriately publish on each copy an appropriate
82 | copyright notice and disclaimer of warranty; keep intact all the
83 | notices that refer to this License and to the absence of any warranty;
84 | and give any other recipients of the Program a copy of this License
85 | along with the Program.
86 |
87 | You may charge a fee for the physical act of transferring a copy, and
88 | you may at your option offer warranty protection in exchange for a fee.
89 |
90 | 2. You may modify your copy or copies of the Program or any portion
91 | of it, thus forming a work based on the Program, and copy and
92 | distribute such modifications or work under the terms of Section 1
93 | above, provided that you also meet all of these conditions:
94 |
95 | a) You must cause the modified files to carry prominent notices
96 | stating that you changed the files and the date of any change.
97 |
98 | b) You must cause any work that you distribute or publish, that in
99 | whole or in part contains or is derived from the Program or any
100 | part thereof, to be licensed as a whole at no charge to all third
101 | parties under the terms of this License.
102 |
103 | c) If the modified program normally reads commands interactively
104 | when run, you must cause it, when started running for such
105 | interactive use in the most ordinary way, to print or display an
106 | announcement including an appropriate copyright notice and a
107 | notice that there is no warranty (or else, saying that you provide
108 | a warranty) and that users may redistribute the program under
109 | these conditions, and telling the user how to view a copy of this
110 | License. (Exception: if the Program itself is interactive but
111 | does not normally print such an announcement, your work based on
112 | the Program is not required to print an announcement.)
113 |
114 | These requirements apply to the modified work as a whole. If
115 | identifiable sections of that work are not derived from the Program,
116 | and can be reasonably considered independent and separate works in
117 | themselves, then this License, and its terms, do not apply to those
118 | sections when you distribute them as separate works. But when you
119 | distribute the same sections as part of a whole which is a work based
120 | on the Program, the distribution of the whole must be on the terms of
121 | this License, whose permissions for other licensees extend to the
122 | entire whole, and thus to each and every part regardless of who wrote it.
123 |
124 | Thus, it is not the intent of this section to claim rights or contest
125 | your rights to work written entirely by you; rather, the intent is to
126 | exercise the right to control the distribution of derivative or
127 | collective works based on the Program.
128 |
129 | In addition, mere aggregation of another work not based on the Program
130 | with the Program (or with a work based on the Program) on a volume of
131 | a storage or distribution medium does not bring the other work under
132 | the scope of this License.
133 |
134 | 3. You may copy and distribute the Program (or a work based on it,
135 | under Section 2) in object code or executable form under the terms of
136 | Sections 1 and 2 above provided that you also do one of the following:
137 |
138 | a) Accompany it with the complete corresponding machine-readable
139 | source code, which must be distributed under the terms of Sections
140 | 1 and 2 above on a medium customarily used for software interchange; or,
141 |
142 | b) Accompany it with a written offer, valid for at least three
143 | years, to give any third party, for a charge no more than your
144 | cost of physically performing source distribution, a complete
145 | machine-readable copy of the corresponding source code, to be
146 | distributed under the terms of Sections 1 and 2 above on a medium
147 | customarily used for software interchange; or,
148 |
149 | c) Accompany it with the information you received as to the offer
150 | to distribute corresponding source code. (This alternative is
151 | allowed only for noncommercial distribution and only if you
152 | received the program in object code or executable form with such
153 | an offer, in accord with Subsection b above.)
154 |
155 | The source code for a work means the preferred form of the work for
156 | making modifications to it. For an executable work, complete source
157 | code means all the source code for all modules it contains, plus any
158 | associated interface definition files, plus the scripts used to
159 | control compilation and installation of the executable. However, as a
160 | special exception, the source code distributed need not include
161 | anything that is normally distributed (in either source or binary
162 | form) with the major components (compiler, kernel, and so on) of the
163 | operating system on which the executable runs, unless that component
164 | itself accompanies the executable.
165 |
166 | If distribution of executable or object code is made by offering
167 | access to copy from a designated place, then offering equivalent
168 | access to copy the source code from the same place counts as
169 | distribution of the source code, even though third parties are not
170 | compelled to copy the source along with the object code.
171 |
172 | 4. You may not copy, modify, sublicense, or distribute the Program
173 | except as expressly provided under this License. Any attempt
174 | otherwise to copy, modify, sublicense or distribute the Program is
175 | void, and will automatically terminate your rights under this License.
176 | However, parties who have received copies, or rights, from you under
177 | this License will not have their licenses terminated so long as such
178 | parties remain in full compliance.
179 |
180 | 5. You are not required to accept this License, since you have not
181 | signed it. However, nothing else grants you permission to modify or
182 | distribute the Program or its derivative works. These actions are
183 | prohibited by law if you do not accept this License. Therefore, by
184 | modifying or distributing the Program (or any work based on the
185 | Program), you indicate your acceptance of this License to do so, and
186 | all its terms and conditions for copying, distributing or modifying
187 | the Program or works based on it.
188 |
189 | 6. Each time you redistribute the Program (or any work based on the
190 | Program), the recipient automatically receives a license from the
191 | original licensor to copy, distribute or modify the Program subject to
192 | these terms and conditions. You may not impose any further
193 | restrictions on the recipients' exercise of the rights granted herein.
194 | You are not responsible for enforcing compliance by third parties to
195 | this License.
196 |
197 | 7. If, as a consequence of a court judgment or allegation of patent
198 | infringement or for any other reason (not limited to patent issues),
199 | conditions are imposed on you (whether by court order, agreement or
200 | otherwise) that contradict the conditions of this License, they do not
201 | excuse you from the conditions of this License. If you cannot
202 | distribute so as to satisfy simultaneously your obligations under this
203 | License and any other pertinent obligations, then as a consequence you
204 | may not distribute the Program at all. For example, if a patent
205 | license would not permit royalty-free redistribution of the Program by
206 | all those who receive copies directly or indirectly through you, then
207 | the only way you could satisfy both it and this License would be to
208 | refrain entirely from distribution of the Program.
209 |
210 | If any portion of this section is held invalid or unenforceable under
211 | any particular circumstance, the balance of the section is intended to
212 | apply and the section as a whole is intended to apply in other
213 | circumstances.
214 |
215 | It is not the purpose of this section to induce you to infringe any
216 | patents or other property right claims or to contest validity of any
217 | such claims; this section has the sole purpose of protecting the
218 | integrity of the free software distribution system, which is
219 | implemented by public license practices. Many people have made
220 | generous contributions to the wide range of software distributed
221 | through that system in reliance on consistent application of that
222 | system; it is up to the author/donor to decide if he or she is willing
223 | to distribute software through any other system and a licensee cannot
224 | impose that choice.
225 |
226 | This section is intended to make thoroughly clear what is believed to
227 | be a consequence of the rest of this License.
228 |
229 | 8. If the distribution and/or use of the Program is restricted in
230 | certain countries either by patents or by copyrighted interfaces, the
231 | original copyright holder who places the Program under this License
232 | may add an explicit geographical distribution limitation excluding
233 | those countries, so that distribution is permitted only in or among
234 | countries not thus excluded. In such case, this License incorporates
235 | the limitation as if written in the body of this License.
236 |
237 | 9. The Free Software Foundation may publish revised and/or new versions
238 | of the General Public License from time to time. Such new versions will
239 | be similar in spirit to the present version, but may differ in detail to
240 | address new problems or concerns.
241 |
242 | Each version is given a distinguishing version number. If the Program
243 | specifies a version number of this License which applies to it and "any
244 | later version", you have the option of following the terms and conditions
245 | either of that version or of any later version published by the Free
246 | Software Foundation. If the Program does not specify a version number of
247 | this License, you may choose any version ever published by the Free Software
248 | Foundation.
249 |
250 | 10. If you wish to incorporate parts of the Program into other free
251 | programs whose distribution conditions are different, write to the author
252 | to ask for permission. For software which is copyrighted by the Free
253 | Software Foundation, write to the Free Software Foundation; we sometimes
254 | make exceptions for this. Our decision will be guided by the two goals
255 | of preserving the free status of all derivatives of our free software and
256 | of promoting the sharing and reuse of software generally.
257 |
258 | NO WARRANTY
259 |
260 | 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
261 | FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
262 | OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
263 | PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
264 | OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
265 | MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
266 | TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
267 | PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
268 | REPAIR OR CORRECTION.
269 |
270 | 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
271 | WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
272 | REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
273 | INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
274 | OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
275 | TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
276 | YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
277 | PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
278 | POSSIBILITY OF SUCH DAMAGES.
279 |
280 | END OF TERMS AND CONDITIONS
281 |
282 | How to Apply These Terms to Your New Programs
283 |
284 | If you develop a new program, and you want it to be of the greatest
285 | possible use to the public, the best way to achieve this is to make it
286 | free software which everyone can redistribute and change under these terms.
287 |
288 | To do so, attach the following notices to the program. It is safest
289 | to attach them to the start of each source file to most effectively
290 | convey the exclusion of warranty; and each file should have at least
291 | the "copyright" line and a pointer to where the full notice is found.
292 |
293 | <one line to give the program's name and a brief idea of what it does.>
294 | Copyright (C) <year> <name of author>
295 |
296 | This program is free software; you can redistribute it and/or modify
297 | it under the terms of the GNU General Public License as published by
298 | the Free Software Foundation; either version 2 of the License, or
299 | (at your option) any later version.
300 |
301 | This program is distributed in the hope that it will be useful,
302 | but WITHOUT ANY WARRANTY; without even the implied warranty of
303 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
304 | GNU General Public License for more details.
305 |
306 | You should have received a copy of the GNU General Public License along
307 | with this program; if not, write to the Free Software Foundation, Inc.,
308 | 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
309 |
310 | Also add information on how to contact you by electronic and paper mail.
311 |
312 | If the program is interactive, make it output a short notice like this
313 | when it starts in an interactive mode:
314 |
315 | Gnomovision version 69, Copyright (C) year name of author
316 | Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
317 | This is free software, and you are welcome to redistribute it
318 | under certain conditions; type `show c' for details.
319 |
320 | The hypothetical commands `show w' and `show c' should show the appropriate
321 | parts of the General Public License. Of course, the commands you use may
322 | be called something other than `show w' and `show c'; they could even be
323 | mouse-clicks or menu items--whatever suits your program.
324 |
325 | You should also get your employer (if you work as a programmer) or your
326 | school, if any, to sign a "copyright disclaimer" for the program, if
327 | necessary. Here is a sample; alter the names:
328 |
329 | Yoyodyne, Inc., hereby disclaims all copyright interest in the program
330 | `Gnomovision' (which makes passes at compilers) written by James Hacker.
331 |
332 | <signature of Ty Coon>, 1 April 1989
333 | Ty Coon, President of Vice
334 |
335 | This General Public License does not permit incorporating your program into
336 | proprietary programs. If your program is a subroutine library, you may
337 | consider it more useful to permit linking proprietary applications with the
338 | library. If this is what you want to do, use the GNU Lesser General
339 | Public License instead of this License.
340 |
--------------------------------------------------------------------------------
/COPYING.mit:
--------------------------------------------------------------------------------
1 | Copyright (c) 2009, 2018-2019
2 |
3 | Permission is hereby granted, free of charge, to any person obtaining a copy
4 | of this software and associated documentation files (the "Software"), to deal
5 | in the Software without restriction, including without limitation the rights
6 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
7 | copies of the Software, and to permit persons to whom the Software is
8 | furnished to do so, subject to the following conditions:
9 |
10 | The above copyright notice and this permission notice shall be included in all
11 | copies or substantial portions of the Software.
12 |
13 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
14 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
15 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
16 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
17 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
18 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
19 | SOFTWARE.
20 |
--------------------------------------------------------------------------------
/Documentation/Contributing.md:
--------------------------------------------------------------------------------
1 | Welcome to the community!
2 |
3 | Contributions need to meet the bar for inclusion in git.git. Although
4 | filter-repo is not part of the git.git repository, I want to leave the
5 | option open for it to be merged in the future. As such, any
6 | contributions need to follow the same [guidelines for contribution to
7 | git.git](https://git.kernel.org/pub/scm/git/git.git/tree/Documentation/SubmittingPatches),
8 | with a few exceptions:
9 |
10 | * While I
11 | [hate](https://public-inbox.org/git/CABPp-BG2SkH0GrRYpHLfp2Wey91ThwQoTgf9UmPa9f5Szn+v3Q@mail.gmail.com/)
12 | [GitHub](https://public-inbox.org/git/CABPp-BEcpasV4vBTm0uxQ4Vzm88MQAX-ArDG4e9QU8tEoNsZWw@mail.gmail.com/)
13 | [PRs](https://public-inbox.org/git/CABPp-BEHy8c3raHwf9aFXvXN0smf_WwCcNiYxQBwh7W6An60qQ@mail.gmail.com/)
14 | (as others point out, [it's mind-boggling in a bad way that
15 | web-based Git hosting and code review systems do such a poor
16 | job](http://nhaehnle.blogspot.com/2020/06/they-want-to-be-small-they-want-to-be.html)),
17 | git-format-patch and git-send-email can be a beast and I have not
18 | yet found time to modify Dscho's excellent
19 | [GitGitGadget](https://github.com/gitgitgadget/gitgitgadget) to
20 | work with git-filter-repo. As such:
21 | * For very short single-commit changes, feel free to open GitHub PRs.
22 | * For more involved changes, if format-patch or send-email give you
23 | too much trouble, go ahead and open a GitHub PR and just mention
24 | that email didn't work out.
25 | * If emailing patches to the git list:
26 | * Include "filter-repo" at the start of the subject,
27 | e.g. "[filter-repo PATCH] Add packaging scripts for uploading to PyPI"
28 | instead of just "[PATCH] Add packaging scripts for uploading to PyPI"
29 | * CC me instead of the git maintainer
30 | * Git's [CodingGuidlines for python
31 | code](https://github.com/git/git/blob/v2.24.0/Documentation/CodingGuidelines#L482-L494)
32 | are only partially applicable:
33 | * python3 is a hard requirement; python2 is/was EOL at the end of
34 | 2019 and should not be used. (Commit 4d0264ab723c
35 | ("filter-repo: workaround python<2.7.9 exec bug", 2019-04-30)
36 | was the last version of filter-repo that worked with python2).
37 | * You can depend on anything in python 3.6 or earlier. I may bump
38 | this minimum version over time, but do want to generally work
39 | with the python3 version found in current enterprise Linux
40 | distributions.
41 | * In filter-repo, it's not just OK to use bytestrings, you are
42 | expected to use them a lot. Using unicode strings result in
43 | lots of ugly errors since input comes from filesystem names,
44 | commit messages, file contents, etc., none of which are
45 | guaranteed to be unicode. (Plus unicode strings require lots of
46 | effort to verify, encode, and decode -- slowing the filtering
47 | process down). I tried to work with unicode strings more
48 | broadly in the code base multiple times; but it's just a bad
49 | idea to use an abstraction that doesn't fit the data.
50 | * I generally like [PEP
51 | 8](https://www.python.org/dev/peps/pep-0008/), but used
52 | two-space indents for years before learning of it and have just
53 | continued that habit. For consistency, contributions should also
54 | use two-space indents and otherwise generally follow PEP 8.
55 |
56 | There are a few extra things I would like folks to keep in mind:
57 |
58 | * Please test line coverage if you add or modify code
59 |
60 | * `make test` will run the testsuite under
61 | [coverage3](https://pypi.org/project/coverage/) (which you will
62 | need to install), and report on line coverage. Line coverage of
63 | git-filter-repo needs to remain at 100%; line coverage of
64 | contrib and test scripts can be ignored.
65 |
66 | * Please do not be intimidated by detailed feedback:
67 |
68 | * In the git community, I have been contributing for years and
69 | have had hundreds of patches accepted but I still find that even
70 | when I try to make patches perfect I am not surprised when I
71 | have to spend as much or more time fixing up patches after
72 | submitting them than I did figuring out the patches in the first
73 | place. git folks tend to do thorough reviews, which has taught
74 | me a lot, and I try to do the same for filter-repo. Plus, as
75 | noted above, I want contributions from others to be acceptable
76 | in git.git itself.
77 |
--------------------------------------------------------------------------------
/Documentation/FAQ.md:
--------------------------------------------------------------------------------
1 | # Frequently Answered Questions
2 |
3 | ## Table of Contents
4 |
5 | * [Why did `git-filter-repo` rewrite commit hashes?](#why-did-git-filter-repo-rewrite-commit-hashes)
6 | * [Why did `git-filter-repo` rewrite more commit hashes than I expected?](#why-did-git-filter-repo-rewrite-more-commit-hashes-than-i-expected)
7 | * [Why did `git-filter-repo` rewrite other branches too?](#why-did-git-filter-repo-rewrite-other-branches-too)
8 | * [How should paths be specified?](#How-should-paths-be-specified)
9 | * [Help! Can I recover or undo the filtering?](#help-can-i-recover-or-undo-the-filtering)
10 | * [Can you change `git-filter-repo` to allow future folks to recover from `--force`'d rewrites?](#can-you-change-git-filter-repo-to-allow-future-folks-to-recover-from---forced-rewrites)
11 | * [Can I use `git-filter-repo` to fix a repository with corruption?](#Can-I-use-git-filter-repo-to-fix-a-repository-with-corruption)
12 | * [What kinds of problems does `git-filter-repo` not try to solve?](#What-kinds-of-problems-does-git-filter-repo-not-try-to-solve)
13 | * [Filtering history but magically keeping the same commit IDs](#Filtering-history-but-magically-keeping-the-same-commit-IDs)
14 | * [Bidirectional development between a filtered and unfiltered repository](#Bidirectional-development-between-a-filtered-and-unfiltered-repository)
15 | * [Removing specific commits, or filtering based on the difference (a.k.a. patch or change) between commits](#Removing-specific-commits-or-filtering-based-on-the-difference-aka-patch-or-change-between-commits)
16 | * [Filtering two different clones of the same repository and getting the same new commit IDs](#Filtering-two-different-clones-of-the-same-repository-and-getting-the-same-new-commit-IDs)
17 |
18 | ## Why did `git-filter-repo` rewrite commit hashes?
19 |
20 | This is fundamental to how Git operates. In more detail...
21 |
22 | Each commit in Git is a hash of its contents. Those contents include
23 | the commit message, the author (name, email, and time authored), the
24 | committer (name, email and time committed), the toplevel tree hash,
25 | and the parent(s) of the commit. This means that if any of the commit
26 | fields change, including the tree hash or the hash of the parent(s) of
27 | the commit, then the hash for the commit will change.
28 |
29 | (The same is true for files ("blobs") and trees stored in git as well;
30 | each is a hash of its contents, so literally if anything changes, the
31 | commit hash will change.)
32 |
33 | If you attempt to write a commit (or tree or blob) object with an
34 | incorrect hash, Git will reject it as corrupt.
35 |
36 | ## Why did `git-filter-repo` rewrite more commit hashes than I expected?
37 |
38 | There are two aspects to this, or two possible underlying questions users
39 | might be asking here:
40 | * Why did commits newer than the ones I expected have their hash change?
41 | * Why did commits older than the ones I expected have their hash change?
42 |
43 | For the first question, see [why filter-repo rewrites commit
44 | hashes](#why-did-git-filter-repo-rewrite-commit-hashes), and note that
45 | if you modify some old commit, perhaps to remove a file, then obviously
46 | that commit's hash must change. Further, since that commit will have a
47 | new hash, any other commit with that commit as a parent will need to
48 | have a new hash. That will need to chain all the way to the most recent
49 | commits in history. This is fundamental to Git and there is nothing you
50 | can do to change this.
51 |
52 | For the second question, there are two causes: (1) the filter you
53 | specified applies to the older commits too, or (2) git-fast-export and
54 | git-fast-import (both of which git-filter-repo uses) canonicalize
55 | history in various ways. The second cause means that even if you have
56 | no filter, these tools sometimes change commit hashes. This can happen
57 | in any of these cases:
58 |
59 | * If you have signed commits, the signatures will be stripped
60 | * If you have commits with extended headers, the extended headers will
61 | be stripped (signed commits are actually a special case of this)
62 | * If you have commits in an encoding other than UTF-8, they will by
63 | default be re-encoded into UTF-8
64 | * If you have a commit without an author, one will be added that
65 | matches the committer.
66 | * If you have trees that are not canonical (e.g. incorrect sorting
67 | order), they will be canonicalized
68 |
69 | If this affects you and you really only want to rewrite newer commits in
70 | history, you can use the `--refs` argument to git-filter-repo to specify
71 | a range of history that you want rewritten.
72 |
73 | (For those attempting to be clever and use `--refs` for the first
74 | question: Note that if you attempt to only rewrite a few old commits,
75 | then all you'll succeed in is adding new commits that won't be part of
76 | any branch and will be subject to garbage collection. The branches will
77 | still hold on to the unrewritten versions of the commits. Thus, you
78 | have to rewrite all the way to the branch tip for the rewrite to be
79 | meaningful. Said another way, the `--refs` trick is only useful for
80 | restricting the rewrite to newer commits, never for restricting the
81 | rewrite to older commits.)
82 |
83 | ## Why did `git-filter-repo` rewrite other branches too?
84 |
85 | git-filter-repo's name is git-filter-**_repo_**. Obviously it is going
86 | to rewrite all branches by default.
87 |
88 | `git-filter-repo` can restrict its rewriting to a subset of history,
89 | such as a single branch, using the `--refs` option. However, using that
90 | comes with the risk that one branch now has a different version of some
91 | commits than other branches do; usually, when you rewrite history, you
92 | want all branches that depend on what you are rewriting to be updated.
93 |
94 | ## How should paths be specified?
95 |
96 | Arguments to `--path` should be paths as Git would report them, when run
97 | from the toplevel of the git repository (explained more below after some
98 | examples).
99 |
100 | **Good** path examples:
101 | * `README.md`
102 | * `Documentation/README.md`
103 | * `src/modules/flux/capacitor.rs`
104 |
105 | You can find examples of valid path names from your repository by
106 | running either `git diff --no-relative --name-only` or `git log
107 | --no-relative --name-only --format=""`.
108 |
109 | The following are basic rules about paths the way that Git reports and uses
110 | them:
111 | * do not use absolute paths
112 | * always treats paths as relative to the toplevel of the repository
113 | (do not add a leading slash, and do not specify paths relative to some
114 | subdirectory of the repository even if that is your current working
115 | directory)
116 | * do not use the special directories `.` or `..` anywhere in your path
117 | * do not use `\`, the Windows path separator, between directories and
118 | files; always use `/` regardless of platform.
119 |
120 | **Erroneous** path examples (do **_NOT_** use any of these styles):
121 | * `/absolute/path/to/src/modules/program.c`
122 | * `/src/modules/program.c`
123 | * `src/docs/../modules/main.java`
124 | * `scripts/config/./update.sh`
125 | * `./tests/fixtures/image.jpg`
126 | * `../src/main.rs`
127 | * `C:\absolute\path\to\src\modules\program.c`
128 | * `src\modules\program.c`
129 |
130 | ## Help! Can I recover or undo the filtering?
131 |
132 | Sure, _if_ you followed the instructions. The instructions told you to
133 | make a fresh clone before running git-filter-repo. If you did that (and
134 | didn't force push your rewritten history back over the original), you
135 | can just throw away your clone with the flubbed rewrite, and make a new
136 | clone.
137 |
138 | If you didn't make a fresh clone, and you didn't run with `--force`, you
139 | would have seen the following warning:
140 | ```
141 | Aborting: Refusing to destructively overwrite repo history since
142 | this does not look like a fresh clone.
143 | [...]
144 | Please operate on a fresh clone instead. If you want to proceed
145 | anyway, use --force.
146 | ```
147 | If you then added `--force`, well, you were warned.
148 |
149 | If you didn't make a fresh clone, and you started with `--force`, and you
150 | didn't think to read the description of the `--force` option:
151 | ```
152 | Ignore fresh clone checks and rewrite history (an irreversible
153 | operation, especially since it by default ends with an
154 | immediate pruning of reflogs and old objects).
155 | ```
156 | and you didn't read even the beginning of the manual
157 | ```
158 | git-filter-repo destructively rewrites history
159 | ```
160 | and you think it's okay to run a command with `--force` in it on
161 | something you don't have a backup of, then now is the time to reasses
162 | your life choices. `--force` should be a pretty clear warning sign.
163 | (If someone on the internet suggested `--force`, you can complain at
164 | _them_, but either way you should learn to carefully vet commands
165 | suggested by others on the internet. Sadly, even sites like Stack
166 | Overflow where someone really ought to be able to correct bad guidance
167 | still unfortunately has a fair amount of this bad advice.)
168 |
169 | See also the next question.
170 |
171 | ## Can you change `git-filter-repo` to allow future folks to recover from --force'd rewrites?
172 |
173 | This will never be supported.
174 |
175 | * Providing an alternate method to restore would require storing both
176 | the original history and the new history, meaning that those who are
177 | trying to shrink their repository size instead see it grow and have to
178 | figure out extra steps to expunge the old history to see the actual
179 | size savings. Experience with other tools showed that this was
180 | frustrating and difficult to figure out for many users.
181 |
182 | * Providing an alternate method to restore would mean that users who are
183 | trying to purge sensitive data from their repository still find the
184 | sensitive data after the rewrite because it hasn't actually been
185 | purged. In order to actually purge it, they have to take extra steps.
186 | Same as with the last bullet point, experience has shown that extra
187 | steps to purge the extra information is difficult and error-prone.
188 | This extra difficulty is particularly problematic when you're trying
189 | to expunge sensitive data.
190 |
191 | * Providing an alternate method to restore would also mean trying to
192 | figure out what should be backed up and how. The obvious choices used
193 | by previous tools only actually provided partial backups (reflogs
194 | would be ignored for example, as would uncommitted changes whether
195 | staged or not). The more you try to carefully backup everything, the
196 | more difficult the restoration from backup will be. The only backup
197 | mechanism I've found that seems reasonable, is making a separate
198 | clone. That's expensive to do automatically for the user (especially
199 | if the filtering is done via multiple invocations of the tool). Plus,
200 | it's not clear where the clone should be stored, especially to avoid
201 | the previous problems for size-reduction and sensitive-data-removal
202 | folks.
203 |
204 | * Providing an alternate method to restore would also mean providing
205 | documentation on how to restore. Past methods by other tools in the
206 | history rewriting space suggested that it was rather difficult for
207 | users to figure out. Difficult enough, in fact, that users simply
208 | didn't ever use them. They instead made a separate clone before
209 | rewriting history and if they didn't like the rewrite, then they just
210 | blew it away and made a new clone to work with. Since that was
211 | observed to be the easy restoration method, I simply enforced it with
212 | this tool, requiring users who look like they might not be operating
213 | on a fresh clone to use the --force flag.
214 |
215 | But more than all that, if there were an alternate method to restore,
216 | why would you have needed to specify the --force flag? Doesn't its
217 | existence (and the wording of its documentation) make it pretty clear on
218 | its own that there isn't going to be a way to restore?
219 |
220 | ## Can I use `git-filter-repo` to fix a repository with corruption?
221 |
222 | Some kinds of corruption can be fixed, in conjunction with `git
223 | replace`. If `git fsck` reports warnings/errors for certain objects,
224 | you can often [replace them and rewrite
225 | history](examples-from-user-filed-issues.md#Handling-repository-corruption).
226 |
227 | ## What kinds of problems does `git-filter-repo` not try to solve?
228 |
229 | This question is often asked in the form of "How do I..." or even
230 | written as a statement such as "I found a bug with `git-filter-repo`;
231 | the behavior I got was different than I expected..." But if you're
232 | trying to do one of the things below, then `git-filter-repo` is behaving
233 | as designed and either there is no solution to your problem, or you need
234 | to use a different tool to solve your problem. The following subsections
235 | address some of these common requests:
236 |
237 | ### Filtering history but magically keeping the same commit IDs
238 |
239 | This is impossible. If you modify commits, or the files contained in
240 | them, then you change their commit IDs; this is [fundamental to
241 | Git](#why-did-git-filter-repo-rewrite-commit-hashes).
242 |
243 | However, _if_ you don't need to modify commits, but just don't want to
244 | download everything, then look into one of the following:
245 | * [partial clones](https://git-scm.com/docs/partial-clone)
246 | * the ugly, retarded hack known as [shallow clones](https://git-scm.com/docs/shallow)
247 | * a massive hack like [cheap fake
248 | clones](https://github.com/newren/sequester-old-big-blobs) that at
249 | least let you put your evil overlord laugh to use
250 |
251 | ### Bidirectional development between a filtered and unfiltered repository
252 |
253 | Some folks want to extract a subset of a repository, do development work
254 | on it, then bring those changes back to the original repository, and
255 | send further changes in both directions. Such a tool can be written
256 | using fast-export and fast-import, but would need to make very different
257 | design decisions than `git-filter-repo` did. Such a tool would be
258 | capable of supporting this kind of development, but lose the ability
259 | ["to write arbitrary filters using a scripting
260 | language"](https://josh-project.github.io/josh/#concept) and other
261 | features that `git-filter-repo` has.
262 |
263 | Such a tool exists; it's called [Josh](https://github.com/josh-project/josh).
264 | Use it if this is your usecase.
265 |
266 | ### Removing specific commits, or filtering based on the difference (a.k.a. patch or change) between commits
267 |
268 | You are probably looking for `git rebase`. `git rebase` operates on the
269 | difference between commits ("diff"), allowing you to e.g. drop or modify
270 | the diff, but then runs the risk of conflicts as it attempts to apply
271 | future diffs. If you tweak one diff in the middle, since it just applies
272 | more diffs for the remaining patches, you'll still see your changes at
273 | the end.
274 |
275 | filter-repo, by contrast, uses fast-export and fast-import. Those tools
276 | treat every commit not as a diff but as a "use the same versions of most
277 | files from the parent commit, but make these five files have these exact
278 | contents". Since you don't have either the diff or ready access to the
279 | version of files from the parent commit, that makes it hard to "undo"
280 | part of the changes to some file. Further, if you attempt to drop an
281 | entire commit or tweak the contents of those new files in that commit,
282 | those changes will be reverted by the next commit in the stream that
283 | mentions that file because handling the next commit does not involve
284 | applying a diff but a "make this file have these exact contents". So,
285 | filter-repo works well for things like removing a file entirely, but if
286 | you want to make any tweaks to any files you have to make the exact same
287 | tweak over and over for every single commit that touches that file.
288 |
289 | In short, `git rebase` is the tool you want for removing specific
290 | commits or otherwise operating on the diff between commits.
291 |
292 | ### Filtering two different clones of the same repository and getting the same new commit IDs
293 |
294 | Sometimes two co-workers have a clone of the same repository and they
295 | run the same `git-filter-repo` command, and they expect to get the same
296 | new commit IDs. Often they do get the same new commit IDs, but
297 | sometimes they don't.
298 |
299 | When people get the same commit IDs, it is only by luck; not by design.
300 | There are three reasons this is unsupported and will never be reliable:
301 |
302 | * Different Git versions used could cause differences in filtering
303 |
304 | Since `git fast-export` and `git fast-import` do various
305 | canonicalizations of history, and these could change over time,
306 | having different versions of Git installed can result in differences
307 | in filtering.
308 |
309 | * Different git-filter-repo versions used could cause differences in
310 | filtering
311 |
312 | Over time, `git-filter-repo` may include new filterings by default,
313 | or fix existing filterings, or make any other number of changes. As
314 | such, having different versions of `git-filter-repo` installed can
315 | result in differences in filtering.
316 |
317 | * Different amounts of the repository cloned or differences in
318 | local-only commits can cause differences in filtering
319 |
320 | If the clones weren't made at the same time, one clone may have more
321 | commits than the other. Also, both may have made local commits the
322 | other doesn't have. These additional commits could cause history to
323 | be traversed in a different order, and filtering rules are allowed
324 | to have order-dependent rules for how they filter. Further,
325 | filtering rules are allowed to depend upon what history exists in
326 | your clone. As one example, filter-repo's default to update commit
327 | messages which refer to other commits by abbreviated hash, may be
328 | unable to find these other commits in your clone but find them in
329 | your coworkers' clone. Relatedly, filter-repo's update of
330 | abbreviated hashes in commit messages only works for commits that
331 | have already been filtered, and thus depends on the order in which
332 | fast-export traverses the history.
333 |
334 | `git-filter-repo` is designed as a _one_-shot history rewriting tool.
335 | Once you have filtered one clone of the repository, you should not be
336 | using it to filter other clones. All other clones of the repository
337 | should either be discarded and recloned, or [have all their history
338 | rebased on top of the rewritten
339 | history](https://htmlpreview.github.io/?https://github.com/newren/git-filter-repo/blob/docs/html/git-filter-repo.html#_make_sure_other_copies_are_cleaned_up_clones_of_colleagues).
340 |
341 | <!--
342 | ## How do I see what was removed?
343 |
344 | Run `git rev-list --objects --all` in both a separate fresh clone from
345 | before the rewrite and in the repo where the rewrite was done. Then
346 | find the objects that exist in the old but not the new.
347 |
348 | -->
349 |
--------------------------------------------------------------------------------
/Documentation/converting-from-bfg-repo-cleaner.md:
--------------------------------------------------------------------------------
1 | # Cheat Sheet: Converting from BFG Repo Cleaner
2 |
3 | This document is aimed at folks who are familiar with BFG Repo Cleaner
4 | and want to learn how to convert over to using filter-repo.
5 |
6 | ## Table of Contents
7 |
8 | * [Half-hearted conversions](#half-hearted-conversions)
9 | * [Intention of "equivalent" commands](#intention-of-equivalent-commands)
10 | * [Basic Differences](#basic-differences)
11 | * [Cheat Sheet: Conversion of Examples from BFG](#cheat-sheet-conversion-of-examples-from-bfg)
12 |
13 | ## Half-hearted conversions
14 |
15 | You can switch most any BFG command to use filter-repo under the
16 | covers by just replacing the `java -jar bfg.jar` part of the command
17 | with [`bfg-ish`](../contrib/filter-repo-demos/bfg-ish).
18 |
19 | bfg-ish is a reasonable tool, and provides a number of bug fixes and
20 | features on top of bfg, but most of my focus is naturally on
21 | filter-repo which has a number of capabilities lacking in bfg-ish.
22 |
23 | ## Intention of "equivalent" commands
24 |
25 | BFG and filter-repo have a few differences, highlighted in the Basic
26 | Differences section below, that make it hard to get commands that
27 | behave identically. Rather than focusing on matching BFG output as
28 | exactly as possible, I treat the BFG examples as idiomatic ways to
29 | solve a certain type of problem with BFG, and express how one would
30 | idiomatically solve the same problem in filter-repo. Sometimes that
31 | means the results are not identical, but they are largely the same in
32 | each case.
33 |
34 | ## Basic Differences
35 |
36 | BFG operates directly on tree objects, which have no notion of their
37 | leading path. Thus, it has no way of differentiating between
38 | 'README.md' at the toplevel versus in some subdirectory. You simply
39 | operate on the basename of files and directories. This precludes
40 | doing things like renaming files and directories or other bigger
41 | restructures. By directly operating on trees, it also runs into
42 | problems with loose vs. packed objects, loose vs. packed refs, not
43 | understanding replace refs or grafts, and not understanding the index
44 | and working tree as another data source.
45 |
46 | With `git filter-repo`, you are essentially given an editing tool to
47 | operate on the [fast-export](https://git-scm.com/docs/git-fast-export)
48 | serialization of a repo, which operates on filenames including their
49 | full paths from the toplevel of the repo. Directories are not
50 | separately specified, so any directory-related filtering is done by
51 | checking the leading path of each file. Further, you aren't limited
52 | to the pre-defined filtering types, python callbacks which operate on
53 | the data structures from the fast-export stream can be provided to do
54 | just about anything you want. By leveraging fast-export and
55 | fast-import, filter-repo gains automatic handling of objects and refs
56 | whether they are packed or not, automatic handling of replace refs and
57 | grafts, and future features that may appear. It also tries hard to
58 | provide a full rewrite solution, so it takes care of additional
59 | important concerns such as updating the index and working tree and
60 | running an automatic gc for the user afterwards.
61 |
62 | The "protection" and "privacy" defaults in BFG are something I
63 | fundamentally disagreed with for a variety of reasons; see the
64 | comments at the top of the
65 | [bfg-ish](../contrib/filter-repo-demos/bfg-ish) script if you want
66 | details. The bfg-ish script implemented these protection and privacy
67 | options since it was designed to act like BFG, but still flipped the
68 | default to the opposite of what BFG chose. I left the "protection"
69 | and "non-private" features out of filter-repo entirely. This means a
70 | number of things with filter-repo:
71 | * any filters you specify will also be applied to HEAD, so that you
72 | don't have a weird disconnect from your history transformations
73 | only being applied to most commits
74 | * `[formerly OLDHASH]` references are not munged into commit
75 | messages; the replace refs that filter-repo adds are a much
76 | cleaner way of looking up commits by old commit hashes.
77 | * `Former-commit-id:` footers are not added to commit messages; the
78 | replace refs that filter-repo adds are a much cleaner way of
79 | looking up commits by old commit hashes.
80 | * History is not littered with `<filename>.REMOVED.git-id` files.
81 |
82 | BFG expects you to specify the repository to rewrite as its final
83 | argument, whereas filter-repo expects you to cd into the repo and then
84 | run filter-repo.
85 |
86 | ## Cheat Sheet: Conversion of Examples from BFG
87 |
88 | ### Stripping big blobs
89 |
90 | ```shell
91 | java -jar bfg.jar --strip-blobs-bigger-than 100M some-big-repo.git
92 | ```
93 |
94 | becomes
95 |
96 | ```shell
97 | git filter-repo --strip-blobs-bigger-than 100M
98 | ```
99 |
100 | ### Deleting files
101 |
102 | ```shell
103 | java -jar bfg.jar --delete-files id_{dsa,rsa} my-repo.git
104 | ```
105 |
106 | becomes
107 |
108 | ```shell
109 | git filter-repo --use-base-name --path id_dsa --path id_rsa --invert-paths
110 | ```
111 |
112 | ### Removing sensitive content
113 |
114 | ```shell
115 | java -jar bfg.jar --replace-text passwords.txt my-repo.git
116 | ```
117 |
118 | becomes
119 |
120 | ```shell
121 | git filter-repo --replace-text passwords.txt
122 | ```
123 |
124 | The `--replace-text` was a really clever idea that the BFG came up
125 | with and I just implemented mostly as-is within filter-repo. Sadly,
126 | BFG didn't document the format of files passed to --replace text very
127 | well, but I added more detail in the filter-repo documentation.
128 |
129 | There is one small but important difference between the two tools: if
130 | you use both "regex:" and "==>" on a single line to specify a regex
131 | search and replace, then filter-repo will use "\1", "\2", "\3",
132 | etc. for replacement strings whereas BFG used "$1", "$2", "$3", etc.
133 | The reason for this difference is simply that python used backslashes
134 | in its regex format while scala used dollar signs, and both tools
135 | wanted to just pass along the strings unmodified to the underlying
136 | language. (Since bfg-ish attempts to emulate the BFG, it accepts
137 | "$1", "$2" and so forth and translates them to "\1", "\2", etc. so
138 | that filter-repo/python will understand it.)
139 |
140 | ### Removing files and folders with a certain name
141 |
142 | ```shell
143 | java -jar bfg.jar --delete-folders .git --delete-files .git --no-blob-protection my-repo.git
144 | ```
145 |
146 | becomes
147 |
148 | ```shell
149 | git filter-repo --invert-paths --path-glob '*/.git' --path .git
150 | ```
151 |
152 | Yes, that glob will handle .git directories one or more directories
153 | deep; it's a git-style glob rather than a shell-style glob. Also, the
154 | `--path .git` was added because `--path-glob '*/.git'` won't match a
155 | directory named .git in the toplevel directory since it has a '/'
156 | character in the glob expression (though I would hope the repository
157 | doesn't have a tracked .git toplevel directory in its history).
158 |
--------------------------------------------------------------------------------
/Documentation/converting-from-filter-branch.md:
--------------------------------------------------------------------------------
1 | # Cheat Sheet: Converting from filter-branch
2 |
3 | This document is aimed at folks who are familiar with filter-branch and want
4 | to learn how to convert over to using filter-repo.
5 |
6 | ## Table of Contents
7 |
8 | * [Half-hearted conversions](#half-hearted-conversions)
9 | * [Intention of "equivalent" commands](#intention-of-equivalent-commands)
10 | * [Basic Differences](#basic-differences)
11 | * [Cheat Sheet: Conversion of Examples from the filter-branch manpage](#cheat-sheet-conversion-of-examples-from-the-filter-branch-manpage)
12 | * [Cheat Sheet: Additional conversion examples](#cheat-sheet-additional-conversion-examples)
13 |
14 | ## Half-hearted conversions
15 |
16 | You can switch nearly any `git filter-branch` command to use
17 | filter-repo under the covers by just replacing the `git filter-branch`
18 | part of the command with
19 | [`filter-lamely`](../contrib/filter-repo-demos/filter-lamely). The
20 | git.git regression testsuite passes when I swap out the filter-branch
21 | script with filter-lamely, for example. (However, the filter-branch
22 | tests are not very comprehensive, so don't rely on that too much.)
23 |
24 | Doing a half-hearted conversion has nearly all of the drawbacks of
25 | filter-branch and nearly none of the benefits of filter-repo, but it
26 | will make your command run a few times faster and makes for a very
27 | simple conversion.
28 |
29 | You'll get a lot more performance, safety, and features by just
30 | switching to direct filter-repo commands.
31 |
32 | ## Intention of "equivalent" commands
33 |
34 | filter-branch and filter-repo have different defaults, as highlighted
35 | in the Basic Differences section below. As such, getting a command
36 | which behaves identically is not possible. Also, sometimes the
37 | filter-branch manpage lies, e.g. it says "suppose you want to...from
38 | all commits" and then uses a command line like "git filter-branch
39 | ... HEAD", which only operates on commits in the current branch rather
40 | than on all commits.
41 |
42 | Rather than focusing on matching filter-branch output as exactly as
43 | possible, I treat the filter-branch examples as idiomatic ways to
44 | solve a certain type of problem with filter-branch, and express how
45 | one would idiomatically solve the same problem in filter-repo.
46 | Sometimes that means the results are not identical, but they are
47 | largely the same in each case.
48 |
49 | ## Basic Differences
50 |
51 | With `git filter-branch`, you have a git repository where every single
52 | commit (within the branches or revisions you specify) is checked out
53 | and then you run one or more shell commands to transform the working
54 | copy into your desired end state.
55 |
56 | With `git filter-repo`, you are essentially given an editing tool to
57 | operate on the [fast-export](https://git-scm.com/docs/git-fast-export)
58 | serialization of a repo. That means there is an input stream of all
59 | the contents of the repository, and rather than specifying filters in
60 | the form of commands to run, you usually employ a number of common
61 | pre-defined filters that provide various ways to slice, dice, or
62 | modify the repo based on its components (such as pathnames, file
63 | content, user names or emails, etc.) That makes common operations
64 | easier, even if it's not as versatile as shell callbacks. For cases
65 | where more complexity or special casing is needed, filter-repo
66 | provides python callbacks that can operate on the data structures
67 | populated from the fast-export stream to do just about anything you
68 | want.
69 |
70 | filter-branch defaults to working on a subset of the repository, and
71 | requires you to specify a branch or branches, meaning you need to
72 | specify `-- --all` to modify all commits. filter-repo by contrast
73 | defaults to rewriting everything, and you need to specify `--refs
74 | <rev-list-args>` if you want to limit to just a certain set of
75 | branches or range of commits. (Though any `<rev-list-args>` that
76 | begin with a hyphen are not accepted by filter-repo as they look like
77 | the start of different options.)
78 |
79 | filter-repo also takes care of additional concerns automatically, like
80 | rewriting commit messages that reference old commit IDs to instead
81 | reference the rewritten commit IDs, pruning commits which do not start
82 | empty but become empty due to the specified filters, and automatically
83 | shrinking and gc'ing the repo at the end of the filtering operation.
84 |
85 | ## Cheat Sheet: Conversion of Examples from the filter-branch manpage
86 |
87 | ### Removing a file
88 |
89 | The filter-branch manual provided three different examples of removing
90 | a single file, based on different levels of ease vs. carefulness and
91 | performance:
92 |
93 | ```shell
94 | git filter-branch --tree-filter 'rm filename' HEAD
95 | ```
96 | ```shell
97 | git filter-branch --tree-filter 'rm -f filename' HEAD
98 | ```
99 | ```shell
100 | git filter-branch --index-filter 'git rm --cached --ignore-unmatch filename' HEAD
101 | ```
102 |
103 | All of these just become
104 |
105 | ```shell
106 | git filter-repo --invert-paths --path filename
107 | ```
108 |
109 | ### Extracting a subdirectory
110 |
111 | Extracting a subdirectory via
112 |
113 | ```shell
114 | git filter-branch --subdirectory-filter foodir -- --all
115 | ```
116 |
117 | is one of the easiest commands to convert; it just becomes
118 |
119 | ```shell
120 | git filter-repo --subdirectory-filter foodir
121 | ```
122 |
123 | ### Moving the whole tree into a subdirectory
124 |
125 | Keeping all files but placing them in a new subdirectory via
126 |
127 | ```shell
128 | git filter-branch --index-filter \
129 | 'git ls-files -s | sed "s-\t\"*-&newsubdir/-" |
130 | GIT_INDEX_FILE=$GIT_INDEX_FILE.new \
131 | git update-index --index-info &&
132 | mv "$GIT_INDEX_FILE.new" "$GIT_INDEX_FILE"' HEAD
133 | ```
134 |
135 | (which happens to be GNU-specific and will fail with BSD userland in
136 | very subtle ways) becomes
137 |
138 | ```shell
139 | git filter-repo --to-subdirectory-filter newsubdir
140 | ```
141 |
142 | (which works fine regardless of GNU vs BSD userland differences.)
143 |
144 | ### Re-grafting history
145 |
146 | The filter-branch manual provided one example with three different
147 | commands that could be used to achieve it, though the first of them
148 | had limited applicability (only when the repo had a single initial
149 | commit). These three examples were:
150 | ```shell
151 | git filter-branch --parent-filter 'sed "s/^\$/-p <graft-id>/"' HEAD
152 | ```
153 | ```shell
154 | git filter-branch --parent-filter \
155 | 'test $GIT_COMMIT = <commit-id> && echo "-p <graft-id>" || cat' HEAD
156 | ```
157 | ```shell
158 | git replace --graft $commit-id $graft-id
159 | git filter-branch $graft-id..HEAD
160 | ```
161 |
162 | git-replace did not exist when the original two examples were written,
163 | but it is clear that the last example is far easier to understand. As
164 | such, filter-repo just uses the same mechanism:
165 |
166 | ```shell
167 | git replace --graft $commit-id $graft-id
168 | git filter-repo --proceed
169 | ```
170 |
171 | NOTE: --proceed is needed here because filter-repo errors out if no
172 | arguments are specified (doing so is usually an error).
173 |
174 | ### Removing commits by a certain author
175 |
176 | WARNING: This is a BAD example for BOTH filter-branch and filter-repo.
177 | It does not remove the changes the user made from the repo, it just
178 | removes the commit in question while smashing the changes from it into
179 | any subsequent commits as though the subsequent authors had been
180 | responsible for those changes as well. `git rebase` is likely to be a
181 | better fit for what you really want if you are looking at this
182 | example. (See also [this explanation of the differences between
183 | rebase and
184 | filter-repo](https://github.com/newren/git-filter-repo/issues/62#issuecomment-597725502))
185 |
186 | This filter-branch example
187 |
188 | ```shell
189 | git filter-branch --commit-filter '
190 | if [ "$GIT_AUTHOR_NAME" = "Darl McBribe" ];
191 | then
192 | skip_commit "$@";
193 | else
194 | git commit-tree "$@";
195 | fi' HEAD
196 | ```
197 |
198 | becomes
199 |
200 | ```shell
201 | git filter-repo --commit-callback '
202 | if commit.author_name == b"Darl McBribe":
203 | commit.skip()
204 | '
205 | ```
206 |
207 | ### Rewriting commit messages -- removing text
208 |
209 | Removing git-svn-id: lines from commit messages via
210 |
211 | ```shell
212 | git filter-branch --msg-filter '
213 | sed -e "/^git-svn-id:/d"
214 | '
215 | ```
216 |
217 | becomes
218 |
219 | ```shell
220 | git filter-repo --message-callback '
221 | return re.sub(b"^git-svn-id:.*\n", b"", message, flags=re.MULTILINE)
222 | '
223 | ```
224 |
225 | ### Rewriting commit messages -- adding text
226 |
227 | Adding Acked-by lines to the last ten commits via
228 |
229 | ```shell
230 | git filter-branch --msg-filter '
231 | cat &&
232 | echo "Acked-by: Bugs Bunny <bunny@bugzilla.org>"
233 | ' master~10..master
234 | ```
235 |
236 | becomes
237 |
238 | ```shell
239 | git filter-repo --message-callback '
240 | return message + b"Acked-by: Bugs Bunny <bunny@bugzilla.org>\n"
241 | ' --refs master~10..master
242 | ```
243 |
244 | ### Changing author/committer(/tagger?) information
245 |
246 | ```shell
247 | git filter-branch --env-filter '
248 | if test "$GIT_AUTHOR_EMAIL" = "root@localhost"
249 | then
250 | GIT_AUTHOR_EMAIL=john@example.com
251 | fi
252 | if test "$GIT_COMMITTER_EMAIL" = "root@localhost"
253 | then
254 | GIT_COMMITTER_EMAIL=john@example.com
255 | fi
256 | ' -- --all
257 | ```
258 |
259 | becomes either
260 |
261 | ```shell
262 | # Ensure '<john@example.com> <root@localhost>' is a line in .mailmap, then:
263 | git filter-repo --use-mailmap
264 | ```
265 |
266 | or
267 |
268 | ```shell
269 | git filter-repo --email-callback '
270 | return email if email != b"root@localhost" else b"john@example.com"
271 | '
272 | ```
273 |
274 | (and as a bonus both filter-repo alternatives will fix tagger emails
275 | too, unlike the filter-branch example)
276 |
277 |
278 | ### Restricting to a range
279 |
280 | The partial examples
281 |
282 | ```shell
283 | git filter-branch ... C..H
284 | ```
285 | ```shell
286 | git filter-branch ... C..H ^D
287 | ```
288 | ```shell
289 | git filter-branch ... D..H ^C
290 | ```
291 |
292 | become
293 |
294 | ```shell
295 | git filter-repo ... --refs C..H
296 | ```
297 | ```shell
298 | git filter-repo ... --refs C..H ^D
299 | ```
300 | ```shell
301 | git filter-repo ... --refs D..H ^C
302 | ```
303 |
304 | Note that filter-branch accepts `--not` among the revision specifiers,
305 | but that appears to python to be a flag name which breaks parsing.
306 | So, instead of e.g. `--not C` as we might use with filter-branch, we
307 | can specify `^C` to filter-repo.
308 |
309 | ## Cheat Sheet: Additional conversion examples
310 |
311 | ### Running a code formatter or linter on each file with some extension
312 |
313 | Running some program on a subset of files is relatively natural in
314 | filter-branch:
315 |
316 | ```shell
317 | git filter-branch --tree-filter '
318 | git ls-files -z "*.c" \
319 | | xargs -0 -n 1 clang-format -style=file -i
320 | '
321 | ```
322 |
323 | though it has the disadvantage of running on every c file for every
324 | commit in history, even if some commits do not modify any c files. This
325 | means this kind of command can be excruciatingly slow.
326 |
327 | The same functionality is slightly more involved in filter-repo for
328 | two reasons:
329 | - fast-export and fast-import split file contents and file names into
330 | completely different data structures that aren't normally available
331 | together
332 | - to run a program on a file, you'll need to write the contents to the
333 | a file, execute the program on that file, and then read the contents
334 | of the file back in
335 |
336 | ```shell
337 | git filter-repo --file-info-callback '
338 | if not filename.endswith(b".c"):
339 | return (filename, mode, blob_id) # no changes
340 |
341 | contents = value.get_contents_by_identifier(blob_id)
342 | tmpfile = os.path.basename(filename)
343 | with open(tmpfile, "wb") as f:
344 | f.write(contents)
345 | subprocess.check_call(["clang-format", "-style=file", "-i", filename])
346 | with open(filename, "rb") as f:
347 | contents = f.read()
348 | new_blob_id = value.insert_file_with_contents(contents)
349 |
350 | return (filename, mode, new_blob_id)
351 | '
352 | ```
353 |
354 | However, one can write a script that uses filter-repo as a library to
355 | simplify this, while also gaining filter-repo's automatic handling of
356 | other concerns like rewriting commit IDs in commit messages or pruning
357 | commits that become empty. In fact, one of the [contrib
358 | demos](../contrib/filter-repo-demos),
359 | [lint-history](../contrib/filter-repo-demos/lint-history), was
360 | specifically written to make this kind of case really easy:
361 |
362 | ```shell
363 | lint-history --relevant 'return filename.endswith(b".c")' \
364 | clang-format -style=file -i
365 | ```
366 |
--------------------------------------------------------------------------------
/Documentation/examples-from-user-filed-issues.md:
--------------------------------------------------------------------------------
1 | # Examples from user-filed issues
2 |
3 | Lots of people have filed issues against git-filter-repo, and many times their
4 | issue boils down into questions of "How do I?" or "Why doesn't this work?"
5 |
6 | Below are a collection of example repository filterings in answer to their
7 | questions, which may be of interest to others.
8 |
9 | ## Table of Contents
10 |
11 | * [Adding files to root commits](#adding-files-to-root-commits)
12 | * [Purge a large list of files](#purge-a-large-list-of-files)
13 | * [Extracting a libary from a repo](#Extracting-a-libary-from-a-repo)
14 | * [Replace words in all commit messages](#Replace-words-in-all-commit-messages)
15 | * [Only keep files from two branches](#Only-keep-files-from-two-branches)
16 | * [Renormalize end-of-line characters and add a .gitattributes](#Renormalize-end-of-line-characters-and-add-a-gitattributes)
17 | * [Remove spaces at the end of lines](#Remove-spaces-at-the-end-of-lines)
18 | * [Having both exclude and include rules for filenames](#Having-both-exclude-and-include-rules-for-filenames)
19 | * [Removing paths with a certain extension](#Removing-paths-with-a-certain-extension)
20 | * [Removing a directory](#Removing-a-directory)
21 | * [Convert from NFD filenames to NFC](#Convert-from-NFD-filenames-to-NFC)
22 | * [Set the committer of the last few commits to myself](#Set-the-committer-of-the-last-few-commits-to-myself)
23 | * [Handling special characters, e.g. accents in names](#Handling-special-characters-eg-accents-in-names)
24 | * [Handling repository corruption](#Handling-repository-corruption)
25 | * [Removing all files with a backslash in them](#Removing-all-files-with-a-backslash-in-them)
26 | * [Replace a binary blob in history](#Replace-a-binary-blob-in-history)
27 | * [Remove commits older than N days](#Remove-commits-older-than-N-days)
28 | * [Replacing pngs with compressed alternative](#Replacing-pngs-with-compressed-alternative)
29 | * [Updating submodule hashes](#Updating-submodule-hashes)
30 | * [Using multi-line strings in callbacks](#Using-multi-line-strings-in-callbacks)
31 |
32 |
33 | ## Adding files to root commits
34 |
35 | <!-- https://github.com/newren/git-filter-repo/issues/21 -->
36 |
37 | Here's an example that will take `/path/to/existing/README.md` and
38 | store it as `README.md` in the repository, and take
39 | `/home/myusers/mymodule.gitignore` and store it as `src/.gitignore` in
40 | the repository:
41 |
42 | ```
43 | git filter-repo --commit-callback "if not commit.parents: commit.file_changes += [
44 | FileChange(b'M', b'README.md', b'$(git hash-object -w '/path/to/existing/README.md')', b'100644'),
45 | FileChange(b'M', b'src/.gitignore', b'$(git hash-object -w '/home/myusers/mymodule.gitignore')', b'100644')]"
46 | ```
47 |
48 | Alternatively, you could also use the [insert-beginning](../contrib/filter-repo-demos/insert-beginning) contrib script:
49 |
50 | ```
51 | mv /path/to/existing/README.md README.md
52 | mv /home/myusers/mymodule.gitignore src/.gitignore
53 | insert-beginning --file README.md
54 | insert-beginning --file src/.gitignore
55 | ```
56 |
57 | ## Purge a large list of files
58 |
59 | <!-- https://github.com/newren/git-filter-repo/issues/63 -->
60 |
61 | Stick all the files in some file (one per line),
62 | e.g. `../DELETED_FILENAMES.txt`, and then run
63 |
64 | ```
65 | git filter-repo --invert-paths --paths-from-file ../DELETED_FILENAMES.txt
66 | ```
67 |
68 | ## Extracting a libary from a repo
69 |
70 | <!-- https://github.com/newren/git-filter-repo/issues/80 -->
71 |
72 | If you want to pick out some subdirectory to keep
73 | (e.g. `src/some-filder/some-feature/`), but don't want it moved to the
74 | repository root (so that --subdirectory-filter isn't applicable) but
75 | instead want it to become some other higher level directory
76 | (e.g. `src/`):
77 |
78 | ```
79 | git filter-repo \
80 | --path src/some-folder/some-feature/ \
81 | --path-rename src/some-folder/some-feature/:src/
82 | ```
83 |
84 | ## Replace words in all commit messages
85 |
86 | <!-- https://github.com/newren/git-filter-repo/issues/83 -->
87 |
88 | Replace "stuff" in any commit message with "task".
89 |
90 | ```
91 | git filter-repo --message-callback 'return message.replace(b"stuff", b"task")'
92 | ```
93 |
94 | ## Only keep files from two branches
95 |
96 | <!-- https://github.com/newren/git-filter-repo/issues/91 -->
97 |
98 | Let's say you know that the files currently present on two branches
99 | are the only files that matter. Files that used to exist in either of
100 | these branches, or files that only exist on some other branch, should
101 | all be deleted from all versions of history. This can be accomplished
102 | by getting a list of files from each branch, combining them, sorting
103 | the list and picking out just the unique entries, then passing the
104 | result to `--paths-from-file`:
105 |
106 | ```
107 | git ls-tree -r ${BRANCH1} >../my-files
108 | git ls-tree -r ${BRANCH2} >>../my-files
109 | sort ../my-files | uniq >../my-relevant-files
110 | git filter-repo --paths-from-file ../my-relevant-files
111 | ```
112 |
113 | ## Renormalize end-of-line characters and add a .gitattributes
114 |
115 | <!-- https://github.com/newren/git-filter-repo/issues/122 -->
116 |
117 | ```
118 | contrib/filter-repo-demos/lint-history dos2unix
119 | [edit .gitattributes]
120 | contrib/filter-repo-demos/insert-beginning .gitattributes
121 | ```
122 |
123 | ## Remove spaces at the end of lines
124 |
125 | <!-- https://github.com/newren/git-filter-repo/issues/145 -->
126 |
127 | Removing all spaces at the end of lines of non-binary files, including
128 | converting CRLF to LF:
129 |
130 | ```
131 | git filter-repo --replace-text <(echo 'regex:[\r\t ]+(\n|$)==>\n')
132 | ```
133 |
134 | ## Having both exclude and include rules for filenames
135 |
136 | <!-- https://github.com/newren/git-filter-repo/issues/230 -->
137 |
138 | If you want to have rules to both include and exclude filenames, you
139 | can simply invoke `git filter-repo` multiple times. Alternatively,
140 | you can do it in one run if you dispense with `--path` arguments and
141 | instead use the more generic `--filename-callback`. For example to
142 | include all files under `src/` except for `src/README.md`:
143 |
144 | ```
145 | git filter-repo --filename-callback '
146 | if filename == b"src/README.md":
147 | return None
148 | if filename.startswith(b"src/"):
149 | return filename
150 | return None'
151 | ```
152 |
153 | ## Removing paths with a certain extension
154 |
155 | <!-- https://github.com/newren/git-filter-repo/issues/274 -->
156 |
157 | ```
158 | git filter-repo --invert-paths --path-glob '*.xsa'
159 | ```
160 |
161 | or
162 |
163 | ```
164 | git filter-repo --filename-callback '
165 | if filename.endswith(b".xsa"):
166 | return None
167 | return filename'
168 | ```
169 |
170 | ## Removing a directory
171 |
172 | <!-- https://github.com/newren/git-filter-repo/issues/278 -->
173 |
174 | ```
175 | git filter-repo --path node_modules/electron/dist/ --invert-paths
176 | ```
177 |
178 | ## Convert from NFD filenames to NFC
179 |
180 | <!-- https://github.com/newren/git-filter-repo/issues/296 -->
181 |
182 | Given that Mac does utf-8 normalization of filenames, and has
183 | historically switched which kind of normalization it does, users may
184 | have committed files with alternative normalizations to their
185 | repository. If someone wants to convert filenames in NFD form to NFC,
186 | they could run
187 |
188 | ```
189 | git filter-repo --filename-callback '
190 | try:
191 | return subprocess.check_output("iconv -f utf-8-mac -t utf-8".split(),
192 | input=filename)
193 | except:
194 | return filename
195 | '
196 | ```
197 |
198 | or instead of relying on the system iconv utility and spawning separate
199 | processes, doing it within python:
200 |
201 | ```
202 | git filter-repo --filename-callback '
203 | import unicodedata
204 | try:
205 | return bytearray(unicodedata.normalize('NFC', filename.decode('utf-8')), 'utf-8')
206 | except:
207 | return filename
208 | '
209 | ```
210 |
211 | ## Set the committer of the last few commits to myself
212 |
213 | <!-- https://github.com/newren/git-filter-repo/issues/379 -->
214 |
215 | ```
216 | git filter-repo --refs main~5..main --commit-callback '
217 | commit.commiter_name = b"My Wonderful Self"
218 | commit.committer_email = b"my@self.org"
219 | '
220 | ```
221 |
222 | ## Handling special characters, e.g. accents and umlauts in names
223 |
224 | <!-- https://github.com/newren/git-filter-repo/issues/383 -->
225 |
226 | Since characters like ë and á are multi-byte characters and python
227 | won't allow you to directly place those in a bytestring
228 | (e.g. `b"Raphaël González"` would result in a `SyntaxError: bytes can
229 | only contain ASCII literal characters` error from Python), you just
230 | need to make a normal (UTF-8) string and then convert to a bytestring
231 | to handle these. For example, changing the author name and email
232 | where the author email is currently `example@test.com`:
233 |
234 | ```
235 | git filter-repo --refs main~5..main --commit-callback '
236 | if commit.author_email = b"example@test.com":
237 | commit.author_name = "Raphaël González".encode()
238 | commit.author_email = b"rgonzalez@test.com"
239 | '
240 | ```
241 |
242 | ## Handling repository corruption
243 |
244 | <!-- https://github.com/newren/git-filter-repo/issues/420 -->
245 |
246 | First, run fsck to get a list of the corrupt objects, e.g.:
247 | ```
248 | $ git fsck --full
249 | error in commit 166f57b3fbe31257100361ecaf735f305b533b21: missingSpaceBeforeDate: invalid author/committer line - missing space before date
250 | error in tree c15680eae81cc8539af7e7de766a8a7c13bd27df: duplicateEntries: contains duplicate file entries
251 | Checking object directories: 100% (256/256), done.
252 | ```
253 |
254 | Odds are you'll only see one type of corruption, but if you see
255 | multiple, you can either do multiple filterings, or create replacement
256 | objects for all the corrupt objects (both commits and trees), and then
257 | do the filtering. Since the method for handling corrupt commits and
258 | corrupt tress is slightly different, I'll give examples below for each.
259 |
260 | ### Handling repository corruption -- commit objects
261 |
262 | Print out the corrupt object literally to a temporary file:
263 | ```
264 | $ git cat-file -p 166f57b3fbe31257100361ecaf735f305b533b21 >tmp
265 | ```
266 |
267 | Taking a look at the file would show, for example:
268 | ```
269 | $ cat tmp
270 | tree e1d871155fce791680ec899fe7869067f2b4ffd2
271 | author My Name <my@email.com>1673287380 -0800
272 | committer My Name <my@email.com> 1673287380 -0800
273 |
274 | Initial
275 | ```
276 |
277 | Edit that file to fix the error (in this case, the missing space
278 | between author email and author date). In this case, it would look
279 | like this after editing:
280 |
281 | ```
282 | tree e1d871155fce791680ec899fe7869067f2b4ffd2
283 | author My Name <my@email.com> 1673287380 -0800
284 | committer My Name <my@email.com> 1673287380 -0800
285 |
286 | Initial
287 | ```
288 |
289 | Save the updated file, then use `git replace` to make a replace reference
290 | for it.
291 | ```
292 | $ git replace -f 166f57b3fbe31257100361ecaf735f305b533b21 $(git hash-object -t commit -w tmp)
293 | ```
294 |
295 | Then remove the temporary file `tmp` and run `filter-repo` to consume
296 | the replace reference and make it permanent:
297 |
298 | ```
299 | $ rm tmp
300 | $ git filter-repo --proceed
301 | ```
302 |
303 | Note that if you have multiple corrupt objects, you need to create
304 | replacements for all of them, and then run filter-repo. Leaving any
305 | corrupt object without a replacement is likely to cause the filter-repo run
306 | to fail.
307 |
308 | ### Handling repository corruption -- tree objects
309 |
310 | <!-- GitHub customer example -->
311 |
312 | Print out the corrupt object literally to a temporary file:
313 | ```
314 | $ git cat-file -p c15680eae81cc8539af7e7de766a8a7c13bd27df >tmp
315 | ```
316 |
317 | Taking a look at the file would show, for example:
318 | ```
319 | $ cat tmp
320 | 100644 blob cd5ded43e86f80bfd384702e3f4cc7ce42de49f9 .gitignore
321 | 100644 blob 226febfcc91ec2c166a5a06834fb47c3553ec469 README.md
322 | 100644 blob e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 src
323 | 040000 tree df2b8fc99e1c1d4dbc0a854d9f72157f1d6ea078 src
324 | 040000 tree 99d732476808176bb9d73bcbfe2505e43d65cb4f t
325 | ```
326 |
327 | Edit that file to fix the error (in this case, removing either the `src`
328 | file (blob) or the `src` directory (tree)). In this case, it might look
329 | like this after editing:
330 |
331 | ```
332 | 100644 blob cd5ded43e86f80bfd384702e3f4cc7ce42de49f9 .gitignore
333 | 100644 blob 226febfcc91ec2c166a5a06834fb47c3553ec469 README.md
334 | 040000 tree df2b8fc99e1c1d4dbc0a854d9f72157f1d6ea078 src
335 | 040000 tree 99d732476808176bb9d73bcbfe2505e43d65cb4f t
336 | ```
337 |
338 | Save the updated file, then use `git mktree` to turn it into an actual
339 | tree object:
340 | ```
341 | $ git mktree <tmp
342 | ace04f50a5d13b43e94c12802d3d8a6c66a35b1d
343 | ```
344 |
345 | Now use the output of that command to create a replacement object for
346 | the original corrupt object:
347 | ```
348 | git replace -f c15680eae81cc8539af7e7de766a8a7c13bd27df ace04f50a5d13b43e94c12802d3d8a6c66a35b1d
349 | ```
350 |
351 | Then remove the temporary file `tmp` and run `filter-repo` to consume
352 | the replace reference and make it permanent:
353 |
354 | ```
355 | $ rm tmp
356 | $ git filter-repo --proceed
357 | ```
358 |
359 | As mentioned with corrupt commit objects, if you have multiple corrupt
360 | objects, as long as you create all the replacements for those objects
361 | first, you only need to run filter-repo once.
362 |
363 |
364 | ## Removing all files with a backslash in them
365 |
366 | <!-- https://github.com/newren/git-filter-repo/issues/427 -->
367 |
368 | ```
369 | git filter-repo --filename-callback 'return None if b'\\' in filename else filename'
370 | ```
371 |
372 | ## Replace a binary blob in history
373 |
374 | <!-- https://github.com/newren/git-filter-repo/issues/436 -->
375 |
376 | Let's say you committed a binary blob, perhaps an image file, with
377 | sensitive data, and never modified it. You want to replace it with
378 | the contents of some alternate file, currently found at
379 | `../alternative-file.jpg` (it can have a different filename than what
380 | is stored in the repository). Let's also say the hash of the old file
381 | was `f4ede2e944868b9a08401dafeb2b944c7166fd0a`. You can replace it
382 | with either
383 |
384 | ```
385 | git filter-repo --blob-callback '
386 | if blob.original_id == b"f4ede2e944868b9a08401dafeb2b944c7166fd0a":
387 | blob.data = open("../alternative-file.jpg", "rb").read()
388 | '
389 | ```
390 |
391 | or
392 |
393 | ```
394 | git replace -f f4ede2e944868b9a08401dafeb2b944c7166fd0a $(git hash-object -w ../alternative-file.jpg)
395 | git filter-repo --proceed
396 | ```
397 |
398 | ## Remove commits older than N days
399 |
400 | <!-- https://github.com/newren/git-filter-repo/issues/300 -->
401 |
402 | This is such a bad usecase. I'm tempted to leave it out, but it has
403 | come up multiple times, and there are people who are totally fine with
404 | changing every commit hash in their repository and throwing away
405 | history periodically. First, identify an ${OLD_COMMIT} that you want
406 | to be a new root commit, then run:
407 |
408 | ```
409 | git replace --graft ${OLD_COMMIT}
410 | git filter-repo --proceed
411 | ```
412 |
413 | (The trick here is that `git replace --graft` takes a commit to replace, and
414 | a list of new parents for the commit. Since ${OLD_COMMIT} is the final
415 | positional argument, it means the list of new parents is an empty list, i.e.
416 | we are turning it into a new root commit.)
417 |
418 | ## Replacing pngs with compressed alternative
419 |
420 | <!-- https://github.com/newren/git-filter-repo/issues/492 -->
421 |
422 | Let's say you committed thousands of pngs that were poorly compressed,
423 | but later aggressively recompressed the pngs and commited and pushed.
424 | Unfortunately, clones are slow because they still contain the poorly
425 | compressed pngs and you'd like to rewrite history to pretend that the
426 | aggressively compressed versions were used when the files were first
427 | introduced.
428 |
429 | First, take a look at the commit that aggressively recompressed the pngs:
430 |
431 | ```
432 | git log -1 --raw --no-abbrev ${COMMIT_WHERE_YOU_COMPRESSED_PNGS}
433 | ```
434 |
435 | that will show output like
436 | ```
437 | :100755 100755 edf570fde099c0705432a389b96cb86489beda09 9cce52ae0806d695956dcf662cd74b497eaa7b12 M resources/foo.png
438 | :100755 100755 644f7c55e1a88a29779dc86b9ff92f512bf9bc11 88b02e9e45c0a62db2f1751b6c065b0c2e538820 M resources/bar.png
439 | ```
440 |
441 | Use that to make a --file-info-callback to fix up the original versions:
442 | ```
443 | git filter-repo --file-info-callback '
444 | if filename == b"resources/foo.png" and blob_id == b"edf570fde099c0705432a389b96cb86489beda09":
445 | blob_id = b"9cce52ae0806d695956dcf662cd74b497eaa7b12"
446 | if filename == b"resources/bar.png" and blob_id == b"644f7c55e1a88a29779dc86b9ff92f512bf9bc11":
447 | blob_id = b"88b02e9e45c0a62db2f1751b6c065b0c2e538820"
448 | return (filename, mode, blob_id)
449 | '
450 | ```
451 |
452 | ## Updating submodule hashes
453 |
454 | <!-- https://github.com/newren/git-filter-repo/issues/537 -->
455 |
456 | Let's say you have a repo with a submodule at src/my-submodule, and
457 | that you feel the wrong commit-hashes of the submodule were commited
458 | within your project and you want them updated according to the
459 | following table:
460 | ```
461 | old new
462 | edf570fde099c0705432a389b96cb86489beda09 9cce52ae0806d695956dcf662cd74b497eaa7b12
463 | 644f7c55e1a88a29779dc86b9ff92f512bf9bc11 88b02e9e45c0a62db2f1751b6c065b0c2e538820
464 | ```
465 |
466 | You could do this as follows:
467 | ```
468 | git filter-repo --file-info-callback '
469 | if filename == b"src/my-submodule" and blob_id == b"edf570fde099c0705432a389b96cb86489beda09":
470 | blob_id = b"9cce52ae0806d695956dcf662cd74b497eaa7b12"
471 | if filename == b"src/my-submodule" and blob_id == b"644f7c55e1a88a29779dc86b9ff92f512bf9bc11":
472 | blob_id = b"88b02e9e45c0a62db2f1751b6c065b0c2e538820"
473 | return (filename, mode, blob_id)
474 | ```
475 |
476 | Yes, `blob_id` is kind of a misnomer here since the file's hash
477 | actually refers to a commit from the sub-project. But `blob_id` is
478 | the name of the parameter passed to the --file-info-callback, so that
479 | is what must be used.
480 |
481 | ## Using multi-line strings in callbacks
482 |
483 | <!-- https://lore.kernel.org/git/CABPp-BFqbiS8xsbLouNB41QTc5p0hEOy-EoV0Sjnp=xJEShkTw@mail.gmail.com/ -->
484 |
485 | Since the text for callbacks have spaces inserted at the front of every
486 | line, multi-line strings are normally munged. For example, the command
487 |
488 | ```
489 | git filter-repo --blob-callback '
490 | blob.data = bytes("""\
491 | This is the new
492 | file that I am
493 | replacing every blob
494 | with. It is great.\n""", "utf-8")
495 | '
496 | ```
497 |
498 | would result in a file with extra spaces at the front of every line:
499 | ```
500 | This is the new
501 | file that I am
502 | replacing every blob
503 | with. It is great.
504 | ```
505 |
506 | The two spaces at the beginning of every-line were inserted into every
507 | line of the callback when trying to compile it as a function.
508 | However, you can use textwrap.dedent to fix this; in fact, using it
509 | will even allow you to add more leading space so that it looks nicely
510 | indented. For example:
511 |
512 | ```
513 | git filter-repo --blob-callback '
514 | import textwrap
515 | blob.data = bytes(textwrap.dedent("""\
516 | This is the new
517 | file that I am
518 | replacing every blob
519 | with. It is great.\n"""), "utf-8")
520 | '
521 | ```
522 |
523 | That will result in a file with contents
524 | ```
525 | This is the new
526 | file that I am
527 | replacing every blob
528 | with. It is great.
529 | ```
530 |
531 | which has no leading spaces on any lines.
--------------------------------------------------------------------------------
/INSTALL.md:
--------------------------------------------------------------------------------
1 | # Table of Contents
2 |
3 | * [Pre-requisites](#pre-requisites)
4 | * [Simple Installation](#simple-installation)
5 | * [Installation via Package Manager](#installation-via-package-manager)
6 | * [Detailed installation explanation for
7 | packagers](#detailed-installation-explanation-for-packagers)
8 | * [Installation as Python Package from
9 | PyPI](#installation-as-python-package-from-pypi)
10 | * [Installation via Makefile](#installation-via-makefile)
11 | * [Notes for Windows Users](#notes-for-windows-users)
12 |
13 | # Pre-requisites
14 |
15 | Instructions on this page assume you have already installed both
16 | [Git](https://git-scm.com) and [Python](https://www.python.org/)
17 | (though the [Notes for Windows Users](#notes-for-windows-users) has
18 | some tips on Python).
19 |
20 | # Simple Installation
21 |
22 | All you need to do is download one file: the [git-filter-repo script
23 | in this repository](git-filter-repo) ([direct link to raw
24 | file](https://raw.githubusercontent.com/newren/git-filter-repo/main/git-filter-repo)),
25 | making sure to preserve its name (`git-filter-repo`, with no
26 | extension). **That's it**. You're done.
27 |
28 | Then you can run any command you want, such as
29 |
30 | $ python3 git-filter-repo --analyze
31 |
32 | If you place the git-filter-repo script in your $PATH, then you can
33 | shorten commands by replacing `python3 git-filter-repo` with `git
34 | filter-repo`; the manual assumes this but you can use the longer form.
35 |
36 | Optionally, if you also want to use some of the contrib scripts, then
37 | you need to make sure you have a `git_filter_repo.py` file which is
38 | either a link to or copy of `git-filter-repo`, and you need to place
39 | that git_filter_repo.py file in $PYTHONPATH.
40 |
41 | If you prefer an "official" installation over the manual installation
42 | explained above, the other sections may have useful tips.
43 |
44 | # Installation via Package Manager
45 |
46 | If you want to install via some [package
47 | manager](https://alternativeto.net/software/yellowdog-updater-modified/?license=opensource),
48 | you can run
49 |
50 | $ PACKAGE_TOOL install git-filter-repo
51 |
52 | The following package managers have packaged git-filter-repo:
53 |
54 | [](https://repology.org/project/git-filter-repo/versions)
55 |
56 | This list covers at least Windows (Scoop), Mac OS X (Homebrew), and
57 | Linux (most the rest). Note that I do not curate this list (and have
58 | no interest in doing so); https://repology.org tracks who packages
59 | these versions.
60 |
61 | # Detailed installation explanation for packagers
62 |
63 | filter-repo only consists of a few files that need to be installed:
64 |
65 | * git-filter-repo
66 |
67 | This is the _only_ thing needed for basic use.
68 |
69 | This can be installed in the directory pointed to by `git --exec-path`,
70 | or placed anywhere in $PATH.
71 |
72 | If your python3 executable is named "python" instead of "python3"
73 | (this particularly appears to affect a number of Windows users),
74 | then you'll also need to modify the first line of git-filter-repo
75 | to replace "python3" with "python".
76 |
77 | * git_filter_repo.py
78 |
79 | This is needed if you want to make use of one of the scripts in
80 | contrib/filter-repo-demos/, or want to write your own script making use
81 | of filter-repo as a python library.
82 |
83 | You can create this symlink to (or copy of) git-filter-repo named
84 | git_filter_repo.py and place it in your python site packages; `python
85 | -c "import site; print(site.getsitepackages())"` may help you find the
86 | appropriate location for your system. Alternatively, you can place
87 | this file anywhere within $PYTHONPATH.
88 |
89 | * git-filter-repo.1
90 |
91 | This is needed if you want `git filter-repo --help` to succeed in
92 | displaying the manpage, when help.format is "man" (the default on Linux
93 | and Mac).
94 |
95 | This can be installed in the directory pointed to by `$(git
96 | --man-path)/man1/`, or placed anywhere in $MANDIR/man1/ where $MANDIR
97 | is some entry from $MANPATH.
98 |
99 | Note that `git filter-repo -h` will show a more limited built-in set of
100 | instructions regardless of whether the manpage is installed.
101 |
102 | * git-filter-repo.html
103 |
104 | This is needed if you want `git filter-repo --help` to succeed in
105 | displaying the html version of the help, when help.format is set to
106 | "html" (the default on Windows).
107 |
108 | This can be installed in the directory pointed to by `git --html-path`.
109 |
110 | Note that `git filter-repo -h` will show a more limited built-in set of
111 | instructions regardless of whether the html version of help is
112 | installed.
113 |
114 | So, installation might look something like the following:
115 |
116 | 1. If you don't have the necessary documentation files (because you
117 | are installing from a clone of filter-repo instead of from a
118 | tarball) then you can first run:
119 |
120 | `make snag_docs`
121 |
122 | (which just copies the generated documentation files from the
123 | `docs` branch)
124 |
125 | 2. Run the following
126 |
127 | ```
128 | cp -a git-filter-repo $(git --exec-path)
129 | cp -a git-filter-repo.1 $(git --man-path)/man1 && mandb
130 | cp -a git-filter-repo.html $(git --html-path)
131 | ln -s $(git --exec-path)/git-filter-repo \
132 | $(python -c "import site; print(site.getsitepackages()[-1])")/git_filter_repo.py
133 | ```
134 |
135 | or you can use the provided Makefile, as noted below.
136 |
137 | # Installation as Python Package from PyPI
138 |
139 | `git-filter-repo` is also available as
140 | [PyPI-package](https://pypi.org/project/git-filter-repo/).
141 |
142 | Therefore, it can be installed with [pipx](https://pypa.github.io/pipx/)
143 | or [uv tool](https://docs.astral.sh/uv/concepts/tools/).
144 | Command example for pipx:
145 |
146 | `pipx install git-filter-repo`
147 |
148 | # Installation via Makefile
149 |
150 | Installing should be doable by hand, but a Makefile is provided for those
151 | that prefer it. However, usage of the Makefile really requires overriding
152 | at least a couple of the directories with sane values, e.g.
153 |
154 | $ make prefix=/usr pythondir=/usr/lib64/python3.8/site-packages install
155 |
156 | Also, the Makefile will not edit the shebang line (the first line) of
157 | git-filter-repo if your python executable is not named "python3";
158 | you'll still need to do that yourself.
159 |
160 | # Notes for Windows Users
161 |
162 | git-filter-repo can be installed with multiple tools, such as
163 | [pipx](https://pypa.github.io/pipx/) or a Windows-specific package manager
164 | like Scoop (both of which were covered above).
165 |
166 | Sadly, Windows sometimes makes things difficult. Common and historical issues:
167 |
168 | * **Non-functional Python stub**: Windows apparently ships with a
169 | [non-functional
170 | python](https://github.com/newren/git-filter-repo/issues/36#issuecomment-568933825).
171 | This can even manifest as [the app
172 | hanging](https://github.com/newren/git-filter-repo/issues/36) or
173 | [the system appearing to
174 | hang](https://github.com/newren/git-filter-repo/issues/312). Try
175 | installing
176 | [Python](https://docs.microsoft.com/en-us/windows/python/beginners)
177 | from the [Microsoft
178 | Store](https://apps.microsoft.com/store/search?publisher=Python%20Software%20Foundation)
179 | * **Modifying PATH, making the script executable**: If modifying your PATH
180 | and/or making scripts executable is difficult for you, you can skip that
181 | step by just using `python3 git-filter-repo` instead of `git filter-repo`
182 | in your commands.
183 | * **Different python executable name**: Some users don't have
184 | a `python3` executable but one named something else like `python`
185 | or `python3.8` or whatever. You may need to edit the first line
186 | of the git-filter-repo script to specify the appropriate path. Or
187 | just don't bother and instead use the long form for executing
188 | filter-repo commands. Namely, replace the `git filter-repo` part
189 | of commands with `PYTHON_EXECUTABLE git-filter-repo`. (Where
190 | `PYTHON_EXECUTABLE` is something like `python` or `python3.8` or
191 | `C:\PATH\TO\INSTALLATION\OF\python3.exe` or whatever).
192 | * **Symlink issues**: git_filter_repo.py is supposed to be a symlink to
193 | git-filter-repo, so that it appears to have identical contents.
194 | If your system messed up the symlink (usually meaning it looks like a
195 | regular file with just one line), then delete git_filter_repo.py and
196 | replace it with a copy of git-filter-repo.
197 | * **Old GitBash limitations**: older versions of GitForWindows had an
198 | unfortunate shebang length limitation (see [git-for-windows issue
199 | #3165](https://github.com/git-for-windows/git/pull/3165)). If
200 | you're affected, just use the long form for invoking filter-repo
201 | commands, i.e. replace the `git filter-repo` part of commands with
202 | `python3 git-filter-repo`.
203 |
204 | For additional historical context, see:
205 | * [#371](https://github.com/newren/git-filter-repo/issues/371#issuecomment-1267116186)
206 | * [#360](https://github.com/newren/git-filter-repo/issues/360#issuecomment-1276813596)
207 | * [#312](https://github.com/newren/git-filter-repo/issues/312)
208 | * [#307](https://github.com/newren/git-filter-repo/issues/307)
209 | * [#225](https://github.com/newren/git-filter-repo/pull/225)
210 | * [#231](https://github.com/newren/git-filter-repo/pull/231)
211 | * [#124](https://github.com/newren/git-filter-repo/issues/124)
212 | * [#36](https://github.com/newren/git-filter-repo/issues/36)
213 | * [this git mailing list
214 | thread](https://lore.kernel.org/git/nycvar.QRO.7.76.6.2004251610300.18039@tvgsbejvaqbjf.bet/)
215 |
--------------------------------------------------------------------------------
/Makefile:
--------------------------------------------------------------------------------
1 | # A bunch of installation-related paths people can override on the command line
2 | DESTDIR = /
3 | INSTALL = install
4 | prefix = $(HOME)
5 | bindir = $(prefix)/libexec/git-core
6 | localedir = $(prefix)/share/locale
7 | mandir = $(prefix)/share/man
8 | htmldir = $(prefix)/share/doc/git-doc
9 | pythondir = $(prefix)/lib64/python3.6/site-packages
10 |
11 | default: build
12 |
13 | build:
14 | @echo Nothing to do: filter-repo is a script which needs no compilation.
15 |
16 | test:
17 | time t/run_coverage
18 |
19 | # fixup_locale might matter once we actually have translations, but right now
20 | # we don't. It might not even matter then, because python has a fallback podir.
21 | fixup_locale:
22 | sed -ie s%@@LOCALEDIR@@%$(localedir)% git-filter-repo
23 |
24 | # People installing from tarball will already have man1/git-filter-repo.1 and
25 | # html/git-filter-repo.html. But let's support people installing from a git
26 | # clone too; for them, just cheat and snag a copy of the built docs that I
27 | # record in a different branch.
28 | snag_docs: Documentation/man1/git-filter-repo.1 Documentation/html/git-filter-repo.html
29 |
30 | Documentation/man1/git-filter-repo.1:
31 | mkdir -p Documentation/man1
32 | git show origin/docs:man1/git-filter-repo.1 >Documentation/man1/git-filter-repo.1
33 |
34 | Documentation/html/git-filter-repo.html:
35 | mkdir -p Documentation/html
36 | git show origin/docs:html/git-filter-repo.html >Documentation/html/git-filter-repo.html
37 |
38 | install: snag_docs #fixup_locale
39 | $(INSTALL) -Dm0755 git-filter-repo "$(DESTDIR)/$(bindir)/git-filter-repo"
40 | $(INSTALL) -dm0755 "$(DESTDIR)/$(pythondir)"
41 | ln -sf "$(bindir)/git-filter-repo" "$(DESTDIR)/$(pythondir)/git_filter_repo.py"
42 | $(INSTALL) -Dm0644 Documentation/man1/git-filter-repo.1 "$(DESTDIR)/$(mandir)/man1/git-filter-repo.1"
43 | $(INSTALL) -Dm0644 Documentation/html/git-filter-repo.html "$(DESTDIR)/$(htmldir)/git-filter-repo.html"
44 | if which mandb > /dev/null; then mandb; fi
45 |
46 |
47 | #
48 | # The remainder of the targets are meant for tasks for the maintainer; if they
49 | # don't work for you, I don't care. These tasks modify branches and upload
50 | # releases and whatnot, and presume a directory layout I have locally.
51 | #
52 | update_docs:
53 | # Set environment variables once
54 | export GIT_WORK_TREE=$(shell mktemp -d) \
55 | export GIT_INDEX_FILE=$(shell mktemp) \
56 | COMMIT=$(shell git rev-parse HEAD) \
57 | && \
58 | # Sanity check; we'll build docs in a clone of a git repo \
59 | test -d ../git && \
60 | # Sanity check; docs == origin/docs \
61 | test -z "$(git rev-parse docs origin/docs | uniq -u)" && \
62 | # Avoid spurious errors by forcing index to be well formatted, if empty \
63 | git read-tree 4b825dc642cb6eb9a060e54bf8d69288fbee4904 && # empty tree \
64 | # Symlink git-filter-repo.txt documentation into git and build it \
65 | ln -sf ../../git-filter-repo/Documentation/git-filter-repo.txt ../git/Documentation/ && \
66 | make -C ../git/Documentation -j4 man html && \
67 | # Take the built documentation and lay it out nicely \
68 | mkdir $GIT_WORK_TREE/html && \
69 | mkdir $GIT_WORK_TREE/man1 && \
70 | cp -a ../git/Documentation/*.html $GIT_WORK_TREE/html/ && \
71 | cp -a ../git/Documentation/git-filter-repo.1 $GIT_WORK_TREE/man1/ && \
72 | dos2unix $GIT_WORK_TREE/html/* && \
73 | # Add new version of the documentation as a commit, if it differs \
74 | git --work-tree $GIT_WORK_TREE add . && \
75 | git diff --quiet docs || git write-tree \
76 | | xargs git commit-tree -p docs -m "Update docs to $COMMIT" \
77 | | xargs git update-ref refs/heads/docs && \
78 | # Remove temporary files \
79 | rm -rf $GIT_WORK_TREE && \
80 | rm $GIT_INDEX_FILE && \
81 | # Push the new documentation upstream \
82 | git push origin docs && \
83 | # Notify of completion \
84 | echo && \
85 | echo === filter-repo docs branch updated ===
86 |
87 | # Call like this:
88 | # make GITHUB_COM_TOKEN=$KEY TAGNAME=v2.23.0 release
89 | release: github_release pypi_release
90 |
91 | # Call like this:
92 | # make GITHUB_COM_TOKEN=$KEY TAGNAME=v2.23.0 github_release
93 | github_release: update_docs
94 | FILEBASE=git-filter-repo-$(shell echo $(TAGNAME) | tail -c +2) \
95 | TMP_INDEX_FILE=$(shell mktemp) \
96 | COMMIT=$(shell git rev-parse HEAD) \
97 | && \
98 | test -n "$(GITHUB_COM_TOKEN)" && \
99 | test -n "$(TAGNAME)" && \
100 | test -n "$COMMIT" && \
101 | # Make sure we don't have any staged or unstaged changes \
102 | git diff --quiet --staged HEAD && git diff --quiet HEAD && \
103 | # Make sure 'jq' is installed \
104 | type -p jq && \
105 | # Tag the release, push it to GitHub \
106 | git tag -a -m "filter-repo $(TAGNAME)" $(TAGNAME) $COMMIT && \
107 | git push origin $(TAGNAME) && \
108 | # Create the tarball \
109 | GIT_INDEX_FILE=$TMP_INDEX_FILE git read-tree $COMMIT && \
110 | git ls-tree -r docs | grep filter-repo \
111 | | sed -e 's%\t%\tDocumentation/%' \
112 | | GIT_INDEX_FILE=$TMP_INDEX_FILE git update-index --index-info && \
113 | GIT_INDEX_FILE=$TMP_INDEX_FILE git write-tree \
114 | | xargs git archive --prefix=$FILEBASE/ \
115 | | xz -c >$FILEBASE.tar.xz && \
116 | rm $TMP_INDEX_FILE && \
117 | # Make GitHub mark our new tag as an official release \
118 | curl -s -H "Authorization: token $(GITHUB_COM_TOKEN)" -X POST \
119 | https://api.github.com/repos/newren/git-filter-repo/releases \
120 | --data "{ \
121 | \"tag_name\": \"$(TAGNAME)\", \
122 | \"target_commitish\": \"$COMMIT\", \
123 | \"name\": \"$(TAGNAME)\", \
124 | \"body\": \"filter-repo $(TAGNAME)\" \
125 | }" | jq -r .id >asset_id && \
126 | # Upload our tarball \
127 | cat asset_id | xargs -I ASSET_ID curl -s -H "Authorization: token $(GITHUB_COM_TOKEN)" -H "Content-Type: application/octet-stream" --data-binary @$FILEBASE.tar.xz https://uploads.github.com/repos/newren/git-filter-repo/releases/ASSET_ID/assets?name=$FILEBASE.tar.xz && \
128 | # Remove temporary file(s) \
129 | rm asset_id && \
130 | # Notify of completion \
131 | echo && \
132 | echo === filter-repo $(TAGNAME) created and uploaded to GitHub ===
133 |
134 | pypi_release: # Has an implicit dependency on github_release because...
135 | # Upload to PyPI, automatically picking tag created by github_release
136 | python3 -m venv venv
137 | venv/bin/pip install --upgrade pip
138 | venv/bin/pip install build twine
139 | venv/bin/pyproject-build
140 | # Note: Retrieve "git-filter-repo releases" token; username is 'newren'
141 | venv/bin/twine upload dist/*
142 | # Remove temporary file(s)
143 | rm -rf dist/ venv/ git_filter_repo.egg-info/
144 |
145 | # NOTE TO FUTURE SELF: If you accidentally push a bad release, you can remove
146 | # all but the git-filter-repo-$VERSION.tar.xz asset with
147 | # git push --delete origin $TAGNAME
148 | # To remove the git-filter-repo-$VERSION.tar.xz asset as well:
149 | # curl -s -H "Authorization: token $GITHUB_COM_TOKEN" -X GET \
150 | # https://api.github.com/repos/newren/git-filter-repo/releases
151 | # and look for the "id", then run
152 | # curl -s -H "Authorization: token $GITHUB_COM_TOKEN" -X DELETE \
153 | # https://api.github.com/repos/newren/git-filter-repo/releases/$ID
154 |
--------------------------------------------------------------------------------
/contrib/filter-repo-demos/README.md:
--------------------------------------------------------------------------------
1 | ## Background
2 |
3 | filter-repo is not merely a history rewriting tool, it also contains a
4 | library that can be used to write new history rewriting tools. This
5 | directory contains several examples showing the breadth of different things
6 | that could be done.
7 |
8 | ## Quick overview
9 |
10 | Command |Description
11 | -------|-----------
12 | barebones-example |Simple example with no modifications to filter-repo behavior, just showing what to import and run.
13 | insert-beginning |Add a new file (e.g. LICENSE/COPYING) to the beginning of history.
14 | signed-off-by |Add a Signed-off-by tag to a range of commits
15 | lint-history |Run some lint command on all non-binary files in history.
16 | clean-ignore |Delete files from history which match current gitignore rules.
17 | filter-lamely (or filter‑branch‑ish) |A nearly bug compatible re-implementation of filter-branch (the git testsuite passes using it instead of filter-branch), with some performance tricks to make it several times faster (though it's still glacially slow compared to filter-repo).
18 | bfg-ish |A re-implementation of most of BFG Repo Cleaner, with new features and bug fixes.
19 | convert-svnexternals |Insert Git submodules according to SVN externals.
20 |
21 | ## Purpose
22 |
23 | Please note that the point of these examples is not to provide new complete
24 | tools, but simply to demonstrate that extremely varied history rewriting
25 | tools can be created which automatically inherit lots of useful base
26 | functionality: rewriting hashes in commit messages, pruning commits that
27 | become empty, handling filenames with funny characters, non-standard
28 | encodings, handling of replace refs, etc. (Additional examples of using
29 | filter-repo as a library can also be found in [the
30 | testsuite](../../t/t9391/).) My sincerest hope is that these examples
31 | provide lots of useful functionality, but that each is missing at least one
32 | critical piece for your usecase. Go forth and extend and improve.
33 |
34 | ## Usage
35 |
36 | All the examples require a symlink to git-filter-repo in your PYTHONPATH
37 | named git_filter_repo.py in order to run; also, all have a --help flag to
38 | get a description of their usage and flags.
39 |
--------------------------------------------------------------------------------
/contrib/filter-repo-demos/barebones-example:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | This is a simple program that behaves identically to git-filter-repo. Its
5 | entire purpose is just to show what to import and run to get the normal
6 | git-filter-repo behavior, to serve as a starting point for you to figure
7 | out what you want to modify.
8 | """
9 |
10 | """
11 | Please see the
12 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
13 | near the top of git-filter-repo.
14 | """
15 |
16 | import sys
17 |
18 | try:
19 | import git_filter_repo as fr
20 | except ImportError:
21 | raise SystemExit("Error: Couldn't find git_filter_repo.py. Did you forget to make a symlink to git-filter-repo named git_filter_repo.py or did you forget to put the latter in your PYTHONPATH?")
22 |
23 | args = fr.FilteringOptions.parse_args(sys.argv[1:])
24 | if args.analyze:
25 | fr.RepoAnalyze.run(args)
26 | else:
27 | filter = fr.RepoFilter(args)
28 | filter.run()
29 |
--------------------------------------------------------------------------------
/contrib/filter-repo-demos/clean-ignore:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | This is a simple program that will delete files from history which match
5 | current gitignore rules, while also:
6 | 1) pruning commits which become empty
7 | 2) pruning merge commits which become degenerate and have no changes
8 | relative to its remaining relevant parent
9 | 3) rewriting commit hashes in commit messages to reference new commit IDs.
10 | """
11 |
12 | """
13 | Please see the
14 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
15 | near the top of git-filter-repo.
16 | """
17 |
18 | import argparse
19 | import os
20 | import subprocess
21 | import sys
22 | try:
23 | import git_filter_repo as fr
24 | except ImportError:
25 | raise SystemExit("Error: Couldn't find git_filter_repo.py. Did you forget to make a symlink to git-filter-repo named git_filter_repo.py or did you forget to put the latter in your PYTHONPATH?")
26 |
27 | class CheckIgnores:
28 | def __init__(self):
29 | self.ignored = set()
30 | self.okay = set()
31 |
32 | cmd = 'git check-ignore --stdin --verbose --non-matching --no-index -z'
33 | self.check_ignore_process = subprocess.Popen(cmd.split(),
34 | stdin=subprocess.PIPE,
35 | stdout=subprocess.PIPE)
36 |
37 | def __del__(self):
38 | if self.check_ignore_process:
39 | self.check_ignore_process.stdin.close()
40 |
41 | def get_ignored(self, filenames):
42 | ignored = set()
43 | for name in filenames:
44 | if name in self.ignored:
45 | ignored.add(name)
46 | elif name in self.okay:
47 | continue
48 | else:
49 | self.check_ignore_process.stdin.write(name+b'\0')
50 | self.check_ignore_process.stdin.flush()
51 | result = os.read(self.check_ignore_process.stdout.fileno(), 65535).rstrip(b'\0')
52 | (source, linenum, pattern, pathname) = result.split(b"\0")
53 | if name != pathname:
54 | raise SystemExit("Error: Passed {} but got {}".format(name, pathname))
55 | if not source and not linenum and not pattern:
56 | self.okay.add(name)
57 | else:
58 | if pattern[0:1] == b"!":
59 | self.okay.add(name)
60 | else:
61 | self.ignored.add(name)
62 | ignored.add(name)
63 |
64 | return ignored
65 |
66 | def skip_ignores(self, commit, metadata):
67 | filenames = [x.filename for x in commit.file_changes]
68 | bad = self.get_ignored(filenames)
69 | commit.file_changes = [x for x in commit.file_changes
70 | if x.filename not in bad]
71 |
72 |
73 | def main():
74 | checker = CheckIgnores()
75 | args = fr.FilteringOptions.parse_args(sys.argv[1:], error_on_empty = False)
76 | filter = fr.RepoFilter(args, commit_callback=checker.skip_ignores)
77 | filter.run()
78 |
79 |
80 | if __name__ == '__main__':
81 | main()
82 |
83 |
--------------------------------------------------------------------------------
/contrib/filter-repo-demos/filter-branch-ish:
--------------------------------------------------------------------------------
1 | filter-lamely
--------------------------------------------------------------------------------
/contrib/filter-repo-demos/insert-beginning:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | This is a simple program that will insert some regular file into the root
5 | commit(s) of history, e.g. adding a file named LICENSE or COPYING to the
6 | first commit. It also rewrites commit hashes in commit messages to update
7 | them based on these changes.
8 | """
9 |
10 | """
11 | Please see the
12 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
13 | near the top of git-filter-repo.
14 | """
15 |
16 | # Technically, this program could be replaced by a one-liner:
17 | # git filter-repo --commit-callback "if not commit.parents: commit.file_changes.append(FileChange(b'M', $RELATIVE_TO_PROJECT_ROOT_PATHNAME, b'$(git hash-object -w $FILENAME)', b'100644'))"
18 | # but let's do it as a full-fledged program that imports git_filter_repo
19 | # anyway...
20 |
21 | import argparse
22 | import os
23 | import subprocess
24 | try:
25 | import git_filter_repo as fr
26 | except ImportError:
27 | raise SystemExit("Error: Couldn't find git_filter_repo.py. Did you forget to make a symlink to git-filter-repo named git_filter_repo.py or did you forget to put the latter in your PYTHONPATH?")
28 |
29 | parser = argparse.ArgumentParser(
30 | description='Add a file to the root commit(s) of history')
31 | parser.add_argument('--file', type=os.fsencode,
32 | help=("Relative-path to file whose contents should be added to root commit(s)"))
33 | args = parser.parse_args()
34 | if not args.file:
35 | raise SystemExit("Error: Need to specify the --file option")
36 | if any([x == b"." or x== b".." for x in args.file.split(b"/")]):
37 | raise SystemExit(f"Error: Invalid path components in {fr.decode(args.file)}")
38 | if not os.path.isfile(args.file):
39 | raise SystemExit(f"Error: {fr.decode(args.file)} not found")
40 |
41 | fhash = subprocess.check_output(['git', 'hash-object', '-w', args.file]).strip()
42 | fmode = b'100755' if os.access(args.file, os.X_OK) else b'100644'
43 | # FIXME: I've assumed the file wasn't a directory or symlink...
44 |
45 | def fixup_commits(commit, metadata):
46 | if len(commit.parents) == 0:
47 | commit.file_changes.append(fr.FileChange(b'M', args.file, fhash, fmode))
48 | # FIXME: What if the history already had a file matching the given name,
49 | # but which didn't exist until later in history? Is the intent for the
50 | # user to keep the other version that existed when it existed, or to
51 | # overwrite the version for all of history with the specified file? I
52 | # don't know, but if it's the latter, we'd need to add an 'else' clause
53 | # like the following:
54 | #else:
55 | # commit.file_changes = [x for x in commit.file_changes
56 | # if x.filename != args.file]
57 |
58 | fr_args = fr.FilteringOptions.parse_args(['--preserve-commit-encoding',
59 | '--force',
60 | '--replace-refs', 'update-no-add'])
61 | filter = fr.RepoFilter(fr_args, commit_callback=fixup_commits)
62 | filter.run()
63 |
--------------------------------------------------------------------------------
/contrib/filter-repo-demos/lint-history:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | This is a simple program that will run a linting program on all non-binary
5 | files in history. It also rewrites commit hashes in commit messages to
6 | refer to the new commits with the rewritten files. You call it like this:
7 | lint-history my-lint-command --arg whatever --another-arg
8 | and it will repeatedly call
9 | my-lint-command --arg whatever --another-arg $TEMPORARY_FILE
10 | with $TEMPORARY_FILE having contents of some file from history.
11 |
12 | NOTE: Several people have taken and modified this script for a variety
13 | of special cases (linting python files, linting jupyter notebooks, just
14 | linting java files, etc.) and posted their modifications at
15 | https://github.com/newren/git-filter-repo/issues/45
16 | Feel free to take a look and adopt some of their ideas. Most of these
17 | modifications are probably strictly unnecessary since you could just make
18 | a lint-script that takes the filename, checks that it matches what you
19 | want, and then calls the real linter. But I guess folks don't like making
20 | an intermediate script. So I eventually added the --relevant flag for
21 | picking out certain files providing yet another way to handle it.
22 | """
23 |
24 | """
25 | Please see the
26 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
27 | near the top of git-filter-repo.
28 | """
29 |
30 | # Technically, if you are only running on all non-binary files and don't care
31 | # about filenames, then this program could be replaced by a "one-liner"; e.g.
32 | # git filter-repo --blob-callback '
33 | # if not b"\0" in blob.data[0:8192]:
34 | # filename = ".git/info/tmpfile"
35 | # with open(filename, "wb") as f:
36 | # f.write(blob.data)
37 | # subprocess.check_call(["lint_program", "--some", "arg", filename])
38 | # with open(filename, "rb") as f:
39 | # blob.data = f.read()
40 | # os.remove(filename)
41 | # '
42 | # but let's do it as a full-fledged program that imports git_filter_repo
43 | # and show how to also do it with filename handling...
44 |
45 | import argparse
46 | import os
47 | import subprocess
48 | import tempfile
49 | try:
50 | import git_filter_repo as fr
51 | except ImportError:
52 | raise SystemExit("Error: Couldn't find git_filter_repo.py. Did you forget to make a symlink to git-filter-repo named git_filter_repo.py or did you forget to put the latter in your PYTHONPATH?")
53 |
54 | example_text = '''CALLBACK
55 |
56 | When you pass --relevant 'BODY', the following style of function
57 | will be compiled and called:
58 |
59 | def is_relevant(filename):
60 | BODY
61 |
62 | Where filename is the full relative path from the toplevel of the
63 | repository.
64 |
65 | Thus, to only run on files with a ".txt" extension you would run
66 | lint-history --relevant 'return filename.endswith(b".txt")' ...
67 |
68 | EXAMPLES
69 |
70 | To run dos2unix on all non-binary files in history:
71 | lint-history dos2unix
72 |
73 | To run eslint --fix on all .js files in history:
74 | lint-history --relevant 'return filename.endswith(b".js")' eslint --fix
75 |
76 | INTERNALS
77 |
78 | Linting of files in history will be done by writing the files to a
79 | temporary directory before running the linting program; the
80 | location of this temporary directory can be controlled via the
81 | TMPDIR environment variable as per
82 | https://docs.python.org/3/library/tempfile.html#tempfile.mkdtemp.
83 | '''
84 |
85 | parser = argparse.ArgumentParser(description='Run a program (e.g. code formatter or linter) on files in history',
86 | epilog = example_text,
87 | formatter_class=argparse.RawDescriptionHelpFormatter)
88 |
89 | parser.add_argument('--relevant', metavar="FUNCTION_BODY",
90 | help=("Python code for determining whether to apply linter to a "
91 | "given filename. Implies --filenames-important. See CALLBACK "
92 | "below."))
93 | parser.add_argument('--filenames-important', action='store_true',
94 | help=("By default, contents are written to a temporary file with a "
95 | "random name. If the linting program needs to know the file "
96 | "basename to operate correctly (e.g. because it needs to know "
97 | "the file's extension), then pass this argument"))
98 | parser.add_argument('--refs', nargs='+',
99 | help=("Limit history rewriting to the specified refs. "
100 | "Implies --partial of git-filter-repo (and all its "
101 | "implications)."))
102 | parser.add_argument('command', nargs=argparse.REMAINDER,
103 | help=("Lint command to run, other than the filename at the end"))
104 | lint_args = parser.parse_args()
105 | if not lint_args.command:
106 | raise SystemExit("Error: Need to specify a lint command")
107 | if len(lint_args.command) > 1 and lint_args.command[0] == '--':
108 | lint_args.command.pop(0)
109 |
110 | tmpdir = None
111 | blobs_handled = {}
112 | cat_file_process = None
113 | def lint_with_real_filenames(commit, metadata):
114 | for change in commit.file_changes:
115 | if change.blob_id in blobs_handled:
116 | change.blob_id = blobs_handled[change.blob_id]
117 | elif change.type == b'D':
118 | continue
119 | elif not is_relevant(change.filename):
120 | continue
121 | else:
122 | # Get the old blob contents
123 | cat_file_process.stdin.write(change.blob_id + b'\n')
124 | cat_file_process.stdin.flush()
125 | objhash, objtype, objsize = cat_file_process.stdout.readline().split()
126 | contents_plus_newline = cat_file_process.stdout.read(int(objsize)+1)
127 |
128 | # Write it out to a file with the same basename
129 | filename = os.path.join(tmpdir, os.path.basename(change.filename))
130 | with open(filename, "wb") as f:
131 | f.write(contents_plus_newline[:-1])
132 |
133 | # Lint the file
134 | subprocess.check_call(lint_args.command + [filename.decode('utf-8')])
135 |
136 | # Get the new contents
137 | with open(filename, "rb") as f:
138 | blob = fr.Blob(f.read())
139 |
140 | # Insert the new file into the filter's stream, and remove the tempfile
141 | filter.insert(blob)
142 | os.remove(filename)
143 |
144 | # Record our handling of the blob and use it for this change
145 | blobs_handled[change.blob_id] = blob.id
146 | change.blob_id = blob.id
147 |
148 | def lint_non_binary_blobs(blob, metadata):
149 | if not b"\0" in blob.data[0:8192]:
150 | filename = '.git/info/tmpfile'
151 | with open(filename, "wb") as f:
152 | f.write(blob.data)
153 | subprocess.check_call(lint_args.command + [filename])
154 | with open(filename, "rb") as f:
155 | blob.data = f.read()
156 | os.remove(filename)
157 |
158 | if lint_args.filenames_important and not lint_args.relevant:
159 | lint_args.relevant = 'return True'
160 | if lint_args.relevant:
161 | body = lint_args.relevant
162 | exec('def is_relevant(filename):\n '+'\n '.join(body.splitlines()),
163 | globals())
164 | lint_args.filenames_important = True
165 | input_args = []
166 | if lint_args.refs:
167 | input_args = ["--refs",] + lint_args.refs
168 | args = fr.FilteringOptions.parse_args(input_args, error_on_empty = False)
169 | args.force = True
170 | if lint_args.filenames_important:
171 | tmpdir = tempfile.mkdtemp().encode()
172 | cat_file_process = subprocess.Popen(['git', 'cat-file', '--batch'],
173 | stdin = subprocess.PIPE,
174 | stdout = subprocess.PIPE)
175 | filter = fr.RepoFilter(args, commit_callback=lint_with_real_filenames)
176 | filter.run()
177 | cat_file_process.stdin.close()
178 | cat_file_process.wait()
179 | else:
180 | if not os.path.exists('.git/info'):
181 | os.makedirs('.git/info')
182 | filter = fr.RepoFilter(args, blob_callback=lint_non_binary_blobs)
183 | filter.run()
184 |
--------------------------------------------------------------------------------
/contrib/filter-repo-demos/signed-off-by:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | This is a simple program that will add Signed-off-by: tags to a range of
5 | commits. Example usage, to add a signed-off-by trailer to every commit that
6 | is not in next but is in any of master, develop, or maint:
7 | signed-off-by master develop maint ^next
8 | More likely called as:
9 | signed-off-by master~4..master
10 | There's no real reason to use this script since `rebase --signoff` exists;
11 | it's mostly just a demonstration of what could be done.
12 | """
13 |
14 | """
15 | Please see the
16 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
17 | near the top of git-filter-repo.
18 | """
19 |
20 | import argparse
21 | import re
22 | import subprocess
23 | try:
24 | import git_filter_repo as fr
25 | except ImportError:
26 | raise SystemExit("Error: Couldn't find git_filter_repo.py. Did you forget to make a symlink to git-filter-repo named git_filter_repo.py or did you forget to put the latter in your PYTHONPATH?")
27 |
28 | parser = argparse.ArgumentParser(
29 | description="Add 'Signed-off-by:' tags to a range of commits")
30 | parser.add_argument('rev_list_args', metavar='rev-list args',
31 | nargs=argparse.REMAINDER,
32 | help=("Range of commits (need to include ref tips) to work on"))
33 | myargs = parser.parse_args()
34 |
35 | user_name = subprocess.check_output('git config user.name'.split()).rstrip()
36 | user_email = subprocess.check_output('git config user.email'.split()).rstrip()
37 | trailer = b'Signed-off-by: %s <%s>' % (user_name, user_email)
38 |
39 | def add_signed_off_by_trailer(commit, metadata):
40 | if trailer in commit.message:
41 | return
42 |
43 | # We want to add the trailer, but we want it to be separated from any
44 | # existing paragraphs by a blank line. However, if the commit message
45 | # already ends with trailers, then we want all trailers to be on adjacent
46 | # lines.
47 | if not commit.message.endswith(b'\n'):
48 | commit.message += b'\n'
49 | lastline = commit.message.splitlines()[-1]
50 | if not re.match(b'[A-Za-z0-9-_]*: ', lastline):
51 | commit.message += b'\n'
52 | commit.message += trailer
53 |
54 | # Setting source and target to anything prevents:
55 | # * remapping origin remote tracking branches to regular branches
56 | # * deletion of the origin remote
57 | # * nuking unused refs
58 | # * nuking reflogs
59 | # * repacking
60 | # so we cheat and set source and target both to '.'
61 | args = fr.FilteringOptions.parse_args(['--force',
62 | '--refs'] + myargs.rev_list_args)
63 | args.refs = myargs.rev_list_args
64 | filter = fr.RepoFilter(args, commit_callback=add_signed_off_by_trailer)
65 | filter.run()
66 |
--------------------------------------------------------------------------------
/git_filter_repo.py:
--------------------------------------------------------------------------------
1 | git-filter-repo
--------------------------------------------------------------------------------
/pyproject.toml:
--------------------------------------------------------------------------------
1 | [project]
2 | name = "git-filter-repo"
3 | description = "Quickly rewrite git repository history"
4 | authors = [
5 | {name = "Elijah Newren", email = "newren@gmail.com"}
6 | ]
7 | readme = "README.md"
8 | classifiers = [
9 | "Development Status :: 5 - Production/Stable",
10 | "Operating System :: OS Independent",
11 | "Programming Language :: Python",
12 | "License :: OSI Approved :: MIT License",
13 | "Programming Language :: Python :: 3.6",
14 | "Programming Language :: Python :: 3.7",
15 | "Programming Language :: Python :: 3.8",
16 | "Programming Language :: Python :: 3.9",
17 | "Programming Language :: Python :: 3.10",
18 | "Programming Language :: Python :: 3.11",
19 | "Programming Language :: Python :: 3.12",
20 | "Programming Language :: Python :: 3",
21 | "Programming Language :: Python :: Implementation :: CPython",
22 | "Programming Language :: Python :: Implementation :: PyPy",
23 | ]
24 | license.text = "MIT"
25 | requires-python = ">= 3.6"
26 | dynamic = ["version"]
27 |
28 | [project.urls]
29 | Homepage = "https://github.com/newren/git-filter-repo"
30 | Issues = "https://github.com/newren/git-filter-repo/issues/"
31 | Source = "https://github.com/newren/git-filter-repo"
32 |
33 | [project.scripts]
34 | git-filter-repo = "git_filter_repo:main"
35 |
36 | [build-system]
37 | requires = ["setuptools>=61", "setuptools_scm>=8.0", "wheel"]
38 | build-backend = "setuptools.build_meta"
39 |
40 | [tool.setuptools]
41 | py-modules = ["git_filter_repo"]
42 |
43 | [tool.setuptools_scm]
44 | # This section intentionally left blank
45 |
--------------------------------------------------------------------------------
/t/run_coverage:
--------------------------------------------------------------------------------
1 | #!/bin/bash
2 |
3 | set -eu
4 |
5 | orig_dir=$(cd $(dirname $0) && pwd -P)
6 | tmpdir=$(mktemp -d)
7 |
8 | cat <<EOF >$tmpdir/.coveragerc
9 | [run]
10 | parallel=true
11 | data_file=$tmpdir/.coverage
12 | EOF
13 |
14 | cat <<EOF >$tmpdir/sitecustomize.py
15 | import coverage
16 | coverage.process_startup()
17 | EOF
18 |
19 | export COVERAGE_PROCESS_START=$tmpdir/.coveragerc
20 | export PYTHONPATH=$tmpdir:
21 |
22 | # Produce a coverage report, even if the tests fail
23 | set +e
24 | $orig_dir/run_tests
25 | exitcode=$?
26 | set -e
27 |
28 | cd $tmpdir
29 | coverage3 combine -q
30 | coverage3 html -d $orig_dir/report
31 | coverage3 report -m
32 | cd $orig_dir
33 | rm -rf $tmpdir
34 |
35 | exit $exitcode
36 |
--------------------------------------------------------------------------------
/t/run_tests:
--------------------------------------------------------------------------------
1 | #!/bin/bash
2 | set -eu
3 |
4 | cd $(dirname $0)
5 |
6 | # Put git_filter_repo.py on the front of PYTHONPATH
7 | export PYTHONPATH="$PWD/..${PYTHONPATH:+:$PYTHONPATH}"
8 |
9 | # We pretend filenames are unicode for two reasons: (1) because it exercises
10 | # more code, and (2) this setting will detect accidental use of unicode strings
11 | # for file/directory names when it should always be bytestrings.
12 | export PRETEND_UNICODE_ARGS=1
13 |
14 | export TEST_SHELL_PATH=/bin/sh
15 |
16 | failed=0
17 |
18 | for t in t[0-9]*.sh
19 | do
20 | printf '\n\n== %s ==\n' "$t"
21 | bash $t "$@" || failed=$(($failed+1))
22 | done
23 |
24 | if [ 0 -lt $failed ]
25 | then
26 | exit 1
27 | fi
28 |
--------------------------------------------------------------------------------
/t/t9390/basic:
--------------------------------------------------------------------------------
1 | feature done
2 | # Simple repo with three files, a merge where each side touches exactly one
3 | # file, and a commit at the end touching all three. Note that the original-oid
4 | # directives are very fake, but make it easy to recognize what original shas
5 | # are.
6 | blob
7 | mark :1
8 | original-oid 0000000000000000000000000000000000000001
9 | data 8
10 | initial
11 |
12 | blob
13 | mark :2
14 | original-oid 0000000000000000000000000000000000000002
15 | data 8
16 | ten-mod
17 |
18 | blob
19 | mark :3
20 | original-oid 0000000000000000000000000000000000000003
21 | data 11
22 | twenty-mod
23 |
24 | blob
25 | mark :4
26 | original-oid 0000000000000000000000000000000000000004
27 | data 6
28 | final
29 |
30 | reset refs/heads/master
31 | commit refs/heads/master
32 | mark :5
33 | original-oid 0000000000000000000000000000000000000009
34 | author Little O. Me <me@little.net> 1535228562 -0700
35 | committer Little O. Me <me@little.net> 1535228562 -0700
36 | data 8
37 | Initial
38 | M 100644 :1 filename
39 | M 100644 :1 ten
40 | M 100644 :1 twenty
41 |
42 | commit refs/heads/B
43 | mark :6
44 | original-oid 000000000000000000000000000000000000000B
45 | author Little 'ol Me <me@laptop.(none)> 1535229544 -0700
46 | committer Little 'ol Me <me@laptop.(none)> 1535229544 -0700
47 | data 11
48 | add twenty
49 | from :5
50 | M 100644 :3 twenty
51 |
52 | commit refs/heads/A
53 | mark :7
54 | original-oid 000000000000000000000000000000000000000A
55 | author Little O. Me <me@machine52.little.net> 1535229523 -0700
56 | committer Little O. Me <me@machine52.little.net> 1535229523 -0700
57 | data 8
58 | add ten
59 | from :5
60 | M 100644 :2 ten
61 |
62 | commit refs/heads/master
63 | mark :8
64 | original-oid 000000000000000000000000000000000000000C
65 | author Lit.e Me <me@fire.com> 1535229559 -0700
66 | committer Lit.e Me <me@fire.com> 1535229580 -0700
67 | data 24
68 | Merge branch 'A' into B
69 | from :6
70 | merge :7
71 | M 100644 :2 ten
72 |
73 | commit refs/heads/master
74 | mark :9
75 | original-oid 000000000000000000000000000000000000000D
76 | author Little Me <me@bigcompany.com> 1535229601 -0700
77 | committer Little Me <me@bigcompany.com> 1535229601 -0700
78 | data 9
79 | whatever
80 | from :8
81 | M 100644 :4 filename
82 | M 100644 :4 ten
83 | M 100644 :4 twenty
84 |
85 | tag v1.0
86 | from :9
87 | original-oid 000000000000000000000000000000000000000E
88 | tagger Little John <second@merry.men> 1535229618 -0700
89 | data 5
90 | v1.0
91 |
92 | reset refs/heads/master
93 | from :9
94 |
95 | done
96 |
--------------------------------------------------------------------------------
/t/t9390/basic-filename:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 8
5 | initial
6 |
7 | reset refs/heads/A
8 | commit refs/heads/A
9 | mark :2
10 | author Little O. Me <me@little.net> 1535228562 -0700
11 | committer Little O. Me <me@little.net> 1535228562 -0700
12 | data 8
13 | Initial
14 | M 100644 :1 filename
15 |
16 | blob
17 | mark :3
18 | data 6
19 | final
20 |
21 | commit refs/heads/master
22 | mark :4
23 | author Little Me <me@bigcompany.com> 1535229601 -0700
24 | committer Little Me <me@bigcompany.com> 1535229601 -0700
25 | data 9
26 | whatever
27 | from :2
28 | M 100644 :3 filename
29 |
30 | reset refs/heads/B
31 | from :2
32 |
33 | tag v1.0
34 | from :4
35 | tagger Little John <second@merry.men> 1535229618 -0700
36 | data 5
37 | v1.0
38 |
39 | done
40 |
--------------------------------------------------------------------------------
/t/t9390/basic-mailmap:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 8
5 | initial
6 |
7 | reset refs/heads/B
8 | commit refs/heads/B
9 | mark :2
10 | author Little 'ol Me <me@little.net> 1535228562 -0700
11 | committer Little 'ol Me <me@little.net> 1535228562 -0700
12 | data 8
13 | Initial
14 | M 100644 :1 filename
15 | M 100644 :1 ten
16 | M 100644 :1 twenty
17 |
18 | blob
19 | mark :3
20 | data 11
21 | twenty-mod
22 |
23 | commit refs/heads/B
24 | mark :4
25 | author Little 'ol Me <me@little.net> 1535229544 -0700
26 | committer Little 'ol Me <me@little.net> 1535229544 -0700
27 | data 11
28 | add twenty
29 | from :2
30 | M 100644 :3 twenty
31 |
32 | blob
33 | mark :5
34 | data 8
35 | ten-mod
36 |
37 | commit refs/heads/A
38 | mark :6
39 | author Little 'ol Me <me@little.net> 1535229523 -0700
40 | committer Little 'ol Me <me@little.net> 1535229523 -0700
41 | data 8
42 | add ten
43 | from :2
44 | M 100644 :5 ten
45 |
46 | commit refs/heads/master
47 | mark :7
48 | author Little 'ol Me <me@little.net> 1535229559 -0700
49 | committer Little 'ol Me <me@little.net> 1535229580 -0700
50 | data 24
51 | Merge branch 'A' into B
52 | from :4
53 | merge :6
54 | M 100644 :5 ten
55 |
56 | blob
57 | mark :8
58 | data 6
59 | final
60 |
61 | commit refs/heads/master
62 | mark :9
63 | author Little 'ol Me <me@little.net> 1535229601 -0700
64 | committer Little 'ol Me <me@little.net> 1535229601 -0700
65 | data 9
66 | whatever
67 | from :7
68 | M 100644 :8 filename
69 | M 100644 :8 ten
70 | M 100644 :8 twenty
71 |
72 | tag v1.0
73 | from :9
74 | tagger Little John <second@merry.men> 1535229618 -0700
75 | data 5
76 | v1.0
77 |
78 | done
79 |
--------------------------------------------------------------------------------
/t/t9390/basic-message:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 8
5 | initial
6 |
7 | reset refs/heads/B
8 | commit refs/heads/B
9 | mark :2
10 | author Little O. Me <me@little.net> 1535228562 -0700
11 | committer Little O. Me <me@little.net> 1535228562 -0700
12 | data 9
13 | Modified
14 | M 100644 :1 filename
15 | M 100644 :1 ten
16 | M 100644 :1 twenty
17 |
18 | blob
19 | mark :3
20 | data 11
21 | twenty-mod
22 |
23 | commit refs/heads/B
24 | mark :4
25 | author Little 'ol Me <me@laptop.(none)> 1535229544 -0700
26 | committer Little 'ol Me <me@laptop.(none)> 1535229544 -0700
27 | data 18
28 | add the number 20
29 | from :2
30 | M 100644 :3 twenty
31 |
32 | blob
33 | mark :5
34 | data 8
35 | ten-mod
36 |
37 | commit refs/heads/A
38 | mark :6
39 | author Little O. Me <me@machine52.little.net> 1535229523 -0700
40 | committer Little O. Me <me@machine52.little.net> 1535229523 -0700
41 | data 8
42 | add ten
43 | from :2
44 | M 100644 :5 ten
45 |
46 | commit refs/heads/master
47 | mark :7
48 | author Lit.e Me <me@fire.com> 1535229559 -0700
49 | committer Lit.e Me <me@fire.com> 1535229580 -0700
50 | data 24
51 | Merge branch 'A' into B
52 | from :4
53 | merge :6
54 | M 100644 :5 ten
55 |
56 | blob
57 | mark :8
58 | data 6
59 | final
60 |
61 | commit refs/heads/master
62 | mark :9
63 | author Little Me <me@bigcompany.com> 1535229601 -0700
64 | committer Little Me <me@bigcompany.com> 1535229601 -0700
65 | data 9
66 | whatever
67 | from :7
68 | M 100644 :8 filename
69 | M 100644 :8 ten
70 | M 100644 :8 twenty
71 |
72 | tag v1.0
73 | from :9
74 | tagger Little John <second@merry.men> 1535229618 -0700
75 | data 15
76 | version one :)
77 |
78 | done
79 |
--------------------------------------------------------------------------------
/t/t9390/basic-numbers:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 8
5 | initial
6 |
7 | reset refs/heads/B
8 | commit refs/heads/B
9 | mark :2
10 | author Little O. Me <me@little.net> 1535228562 -0700
11 | committer Little O. Me <me@little.net> 1535228562 -0700
12 | data 8
13 | Initial
14 | M 100644 :1 ten
15 | M 100644 :1 twenty
16 |
17 | blob
18 | mark :3
19 | data 11
20 | twenty-mod
21 |
22 | commit refs/heads/B
23 | mark :4
24 | author Little 'ol Me <me@laptop.(none)> 1535229544 -0700
25 | committer Little 'ol Me <me@laptop.(none)> 1535229544 -0700
26 | data 11
27 | add twenty
28 | from :2
29 | M 100644 :3 twenty
30 |
31 | blob
32 | mark :5
33 | data 8
34 | ten-mod
35 |
36 | commit refs/heads/A
37 | mark :6
38 | author Little O. Me <me@machine52.little.net> 1535229523 -0700
39 | committer Little O. Me <me@machine52.little.net> 1535229523 -0700
40 | data 8
41 | add ten
42 | from :2
43 | M 100644 :5 ten
44 |
45 | commit refs/heads/master
46 | mark :7
47 | author Lit.e Me <me@fire.com> 1535229559 -0700
48 | committer Lit.e Me <me@fire.com> 1535229580 -0700
49 | data 24
50 | Merge branch 'A' into B
51 | from :4
52 | merge :6
53 | M 100644 :5 ten
54 |
55 | blob
56 | mark :8
57 | data 6
58 | final
59 |
60 | commit refs/heads/master
61 | mark :9
62 | author Little Me <me@bigcompany.com> 1535229601 -0700
63 | committer Little Me <me@bigcompany.com> 1535229601 -0700
64 | data 9
65 | whatever
66 | from :7
67 | M 100644 :8 ten
68 | M 100644 :8 twenty
69 |
70 | tag v1.0
71 | from :9
72 | tagger Little John <second@merry.men> 1535229618 -0700
73 | data 5
74 | v1.0
75 |
76 | done
77 |
--------------------------------------------------------------------------------
/t/t9390/basic-replace:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 8
5 | initial
6 |
7 | reset refs/heads/B
8 | commit refs/heads/B
9 | mark :2
10 | author Little O. Me <me@little.net> 1535228562 -0700
11 | committer Little O. Me <me@little.net> 1535228562 -0700
12 | data 8
13 | Initial
14 | M 100644 :1 filename
15 | M 100644 :1 ten
16 | M 100644 :1 twenty
17 |
18 | blob
19 | mark :3
20 | data 28
21 | twenty-modified-by-gremlins
22 |
23 | commit refs/heads/B
24 | mark :4
25 | author Little 'ol Me <me@laptop.(none)> 1535229544 -0700
26 | committer Little 'ol Me <me@laptop.(none)> 1535229544 -0700
27 | data 11
28 | add twenty
29 | from :2
30 | M 100644 :3 twenty
31 |
32 | blob
33 | mark :5
34 | data 25
35 | ten-modified-by-gremlins
36 |
37 | commit refs/heads/A
38 | mark :6
39 | author Little O. Me <me@machine52.little.net> 1535229523 -0700
40 | committer Little O. Me <me@machine52.little.net> 1535229523 -0700
41 | data 8
42 | add ten
43 | from :2
44 | M 100644 :5 ten
45 |
46 | commit refs/heads/master
47 | mark :7
48 | author Lit.e Me <me@fire.com> 1535229559 -0700
49 | committer Lit.e Me <me@fire.com> 1535229580 -0700
50 | data 24
51 | Merge branch 'A' into B
52 | from :4
53 | merge :6
54 | M 100644 :5 ten
55 |
56 | blob
57 | mark :8
58 | data 6
59 | final
60 |
61 | commit refs/heads/master
62 | mark :9
63 | author Little Me <me@bigcompany.com> 1535229601 -0700
64 | committer Little Me <me@bigcompany.com> 1535229601 -0700
65 | data 9
66 | whatever
67 | from :7
68 | M 100644 :8 filename
69 | M 100644 :8 ten
70 | M 100644 :8 twenty
71 |
72 | tag v1.0
73 | from :9
74 | tagger Little John <second@merry.men> 1535229618 -0700
75 | data 5
76 | v1.0
77 |
78 | done
79 |
--------------------------------------------------------------------------------
/t/t9390/basic-ten:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 8
5 | initial
6 |
7 | reset refs/heads/B
8 | commit refs/heads/B
9 | mark :2
10 | author Little O. Me <me@little.net> 1535228562 -0700
11 | committer Little O. Me <me@little.net> 1535228562 -0700
12 | data 8
13 | Initial
14 | M 100644 :1 ten
15 |
16 | blob
17 | mark :3
18 | data 8
19 | ten-mod
20 |
21 | commit refs/heads/A
22 | mark :4
23 | author Little O. Me <me@machine52.little.net> 1535229523 -0700
24 | committer Little O. Me <me@machine52.little.net> 1535229523 -0700
25 | data 8
26 | add ten
27 | from :2
28 | M 100644 :3 ten
29 |
30 | blob
31 | mark :5
32 | data 6
33 | final
34 |
35 | commit refs/heads/master
36 | mark :6
37 | author Little Me <me@bigcompany.com> 1535229601 -0700
38 | committer Little Me <me@bigcompany.com> 1535229601 -0700
39 | data 9
40 | whatever
41 | from :4
42 | M 100644 :5 ten
43 |
44 | tag v1.0
45 | from :6
46 | tagger Little John <second@merry.men> 1535229618 -0700
47 | data 5
48 | v1.0
49 |
50 | done
51 |
--------------------------------------------------------------------------------
/t/t9390/basic-twenty:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 8
5 | initial
6 |
7 | reset refs/heads/A
8 | commit refs/heads/A
9 | mark :2
10 | author Little O. Me <me@little.net> 1535228562 -0700
11 | committer Little O. Me <me@little.net> 1535228562 -0700
12 | data 8
13 | Initial
14 | M 100644 :1 twenty
15 |
16 | blob
17 | mark :3
18 | data 11
19 | twenty-mod
20 |
21 | commit refs/heads/B
22 | mark :4
23 | author Little 'ol Me <me@laptop.(none)> 1535229544 -0700
24 | committer Little 'ol Me <me@laptop.(none)> 1535229544 -0700
25 | data 11
26 | add twenty
27 | from :2
28 | M 100644 :3 twenty
29 |
30 | blob
31 | mark :5
32 | data 6
33 | final
34 |
35 | commit refs/heads/master
36 | mark :6
37 | author Little Me <me@bigcompany.com> 1535229601 -0700
38 | committer Little Me <me@bigcompany.com> 1535229601 -0700
39 | data 9
40 | whatever
41 | from :4
42 | M 100644 :5 twenty
43 |
44 | tag v1.0
45 | from :6
46 | tagger Little John <second@merry.men> 1535229618 -0700
47 | data 5
48 | v1.0
49 |
50 | done
51 |
--------------------------------------------------------------------------------
/t/t9390/degenerate:
--------------------------------------------------------------------------------
1 | feature done
2 | # Simple repo with only three files, with a bunch of cases of dealing with
3 | # topology changes possibly causing merge commits to need to be pruned.
4 | #
5 | # As with case1, the original-oid directives are very fake, but if an error
6 | # is hit that shows one of these, it makes it really easy to know where it
7 | # came from.
8 | #
9 | # Expressed with shorthand, log history in the format
10 | # Commit Name(Parent(s)): files changed
11 | # for this case looks like the following:
12 | # W(V): moduleA/keepme
13 | # V(U,U): moduleB/nukeme
14 | # U(T): moduleA/sometimes
15 | # T(S): moduleA/keepme
16 | # S(R,R): moduleA/sometimes
17 | # R(R): moduleB/nukeme
18 | # Q(P): moduleA/keepme
19 | # P(N,M): moduleA/sometimes
20 | # O(M,N): moduleA/sometimes
21 | # N(C): moduleB/nukeme
22 | # M(L): moduleB/nukeme
23 | # L(K): moduleA/keepme
24 | # K(J): moduleB/nukeme
25 | # J(D,H): moduleA/sometimes
26 | # I(H,D): moduleA/sometimes # backwards-ish merge
27 | # H(G): moduleB/nukeme
28 | # G(F): moduleA/keepme
29 | # F(D): moduleB/nukeme
30 | # D(B,C): moduleA/sometimes
31 | # C(A): moduleB/nukeme
32 | # B(A): moduleB/nukeme
33 | # A(): moduleA/keepme
34 | #
35 | # This involved case is intended to test the following:
36 | # * Merge becoming non-merge due to both parents becoming same commit
37 | # * Two sub-cases: it has changes of its own, or it doesn't
38 | # * Merge becoming merge of commit with its own ancestor
39 | # * Two cases: and it has changes, and it doesn't have changes
40 | # * Two cases: first parent is the ancestor, second parent is the ancestor
41 | # * Merge starting as merge of commit with its own ancestor
42 | # * Two cases: has changes, doesn't have changes
43 | # * Two cases: first parent, or second parent
44 | blob
45 | mark :1
46 | original-oid 0000000000000000000000000000000000000001
47 | data 10
48 | keepme v1
49 |
50 | blob
51 | mark :2
52 | original-oid 0000000000000000000000000000000000000002
53 | data 10
54 | nukeme v1
55 |
56 | blob
57 | mark :3
58 | original-oid 0000000000000000000000000000000000000003
59 | data 10
60 | nukeme v2
61 |
62 | blob
63 | mark :4
64 | original-oid 0000000000000000000000000000000000000004
65 | data 13
66 | sometimes v1
67 |
68 | blob
69 | mark :5
70 | original-oid 0000000000000000000000000000000000000005
71 | data 10
72 | nukeme v3
73 |
74 | blob
75 | mark :6
76 | original-oid 0000000000000000000000000000000000000006
77 | data 10
78 | keepme v2
79 |
80 | blob
81 | mark :7
82 | original-oid 0000000000000000000000000000000000000007
83 | data 10
84 | nukem4 v4
85 |
86 | blob
87 | mark :8
88 | original-oid 0000000000000000000000000000000000000008
89 | data 13
90 | sometimes v2
91 |
92 | blob
93 | mark :9
94 | original-oid 0000000000000000000000000000000000000009
95 | data 13
96 | sometimes v3
97 |
98 | blob
99 | mark :10
100 | original-oid 000000000000000000000000000000000000000A
101 | data 10
102 | nukeme v4
103 |
104 | blob
105 | mark :11
106 | original-oid 000000000000000000000000000000000000000B
107 | data 10
108 | keepme v3
109 |
110 | blob
111 | mark :12
112 | original-oid 000000000000000000000000000000000000000C
113 | data 10
114 | nukeme v5
115 |
116 | blob
117 | mark :13
118 | original-oid 000000000000000000000000000000000000000D
119 | data 10
120 | nukeme v6
121 |
122 | blob
123 | mark :14
124 | original-oid 000000000000000000000000000000000000000E
125 | data 13
126 | sometimes v4
127 |
128 | blob
129 | mark :15
130 | original-oid 000000000000000000000000000000000000000F
131 | data 13
132 | sometimes v5
133 |
134 | blob
135 | mark :16
136 | original-oid 0000000000000000000000000000000000000010
137 | data 10
138 | keepme v4
139 |
140 | blob
141 | mark :17
142 | original-oid 0000000000000000000000000000000000000011
143 | data 10
144 | nukeme v7
145 |
146 | blob
147 | mark :18
148 | original-oid 0000000000000000000000000000000000000012
149 | data 13
150 | sometimes v6
151 |
152 | blob
153 | mark :19
154 | original-oid 0000000000000000000000000000000000000013
155 | data 10
156 | keepme v5
157 |
158 | blob
159 | mark :20
160 | original-oid 0000000000000000000000000000000000000014
161 | data 13
162 | sometimes v7
163 |
164 | blob
165 | mark :21
166 | original-oid 0000000000000000000000000000000000000015
167 | data 10
168 | nukeme v8
169 |
170 | blob
171 | mark :22
172 | original-oid 0000000000000000000000000000000000000016
173 | data 10
174 | keepme v6
175 |
176 | commit refs/heads/master
177 | mark :26
178 | original-oid 0000000000000000000000000000000000000020
179 | author Full Name <user@organization.tld> 2000000000 +0100
180 | committer Full Name <user@organization.tld> 2000000000 +0100
181 | data 2
182 | A
183 | M 100644 :1 moduleA/keepme
184 |
185 | commit refs/heads/master
186 | mark :27
187 | original-oid 0000000000000000000000000000000000000021
188 | author Full Name <user@organization.tld> 2000010000 +0100
189 | committer Full Name <user@organization.tld> 2000010000 +0100
190 | data 2
191 | B
192 | from :26
193 | M 100644 :2 moduleB/nukeme
194 |
195 | commit refs/heads/master
196 | mark :28
197 | original-oid 0000000000000000000000000000000000000022
198 | author Full Name <user@organization.tld> 2000020000 +0100
199 | committer Full Name <user@organization.tld> 2000020000 +0100
200 | data 2
201 | C
202 | from :26
203 | M 100644 :3 moduleB/nukeme
204 |
205 | commit refs/heads/master
206 | mark :29
207 | original-oid 0000000000000000000000000000000000000023
208 | author Full Name <user@organization.tld> 2000030000 +0100
209 | committer Full Name <user@organization.tld> 2000030000 +0100
210 | data 29
211 | D: Merge commit 'C' into 'B'
212 | from :27
213 | merge :28
214 | M 100644 :4 moduleA/sometimes
215 |
216 | commit refs/heads/master
217 | mark :30
218 | original-oid 0000000000000000000000000000000000000024
219 | author Full Name <user@organization.tld> 2000040000 +0100
220 | committer Full Name <user@organization.tld> 2000040000 +0100
221 | data 2
222 | F
223 | from :29
224 | M 100644 :5 moduleB/nukeme
225 |
226 | commit refs/heads/master
227 | mark :31
228 | original-oid 0000000000000000000000000000000000000025
229 | author Full Name <user@organization.tld> 2000050000 +0100
230 | committer Full Name <user@organization.tld> 2000050000 +0100
231 | data 2
232 | G
233 | from :30
234 | M 100644 :6 moduleA/keepme
235 |
236 | commit refs/heads/master
237 | mark :32
238 | original-oid 0000000000000000000000000000000000000026
239 | author Full Name <user@organization.tld> 2000060000 +0100
240 | committer Full Name <user@organization.tld> 2000060000 +0100
241 | data 2
242 | H
243 | from :31
244 | M 100644 :7 moduleB/nukeme
245 |
246 | commit refs/heads/branchI
247 | mark :33
248 | original-oid 0000000000000000000000000000000000000027
249 | author Full Name <user@organization.tld> 2000070000 +0100
250 | committer Full Name <user@organization.tld> 2000070000 +0100
251 | data 29
252 | I: Merge commit 'D' into 'H'
253 | from :32
254 | merge :29
255 | M 100644 :8 moduleA/sometimes
256 |
257 | commit refs/heads/master
258 | mark :34
259 | original-oid 0000000000000000000000000000000000000028
260 | author Full Name <user@organization.tld> 2000080000 +0100
261 | committer Full Name <user@organization.tld> 2000080000 +0100
262 | data 29
263 | J: Merge commit 'H' into 'D'
264 | from :29
265 | merge :32
266 | M 100644 :9 moduleA/sometimes
267 |
268 | commit refs/heads/master
269 | mark :35
270 | original-oid 0000000000000000000000000000000000000029
271 | author Full Name <user@organization.tld> 2000090000 +0100
272 | committer Full Name <user@organization.tld> 2000090000 +0100
273 | data 2
274 | K
275 | from :34
276 | M 100644 :10 moduleB/nukeme
277 |
278 | commit refs/heads/master
279 | mark :36
280 | original-oid 000000000000000000000000000000000000002A
281 | author Full Name <user@organization.tld> 2000092000 +0100
282 | committer Full Name <user@organization.tld> 2000092000 +0100
283 | data 2
284 | L
285 | from :35
286 | M 100644 :11 moduleA/keepme
287 |
288 | commit refs/heads/master
289 | mark :37
290 | original-oid 000000000000000000000000000000000000002B
291 | author Full Name <user@organization.tld> 2000094000 +0100
292 | committer Full Name <user@organization.tld> 2000094000 +0100
293 | data 2
294 | M
295 | from :36
296 | M 100644 :12 moduleB/nukeme
297 |
298 | commit refs/heads/master
299 | mark :38
300 | original-oid 000000000000000000000000000000000000002C
301 | author Full Name <user@organization.tld> 2000096000 +0100
302 | committer Full Name <user@organization.tld> 2000096000 +0100
303 | data 2
304 | N
305 | from :28
306 | M 100644 :13 moduleB/nukeme
307 |
308 | commit refs/heads/branchO
309 | mark :39
310 | original-oid 000000000000000000000000000000000000002D
311 | author Full Name <user@organization.tld> 2000098000 +0100
312 | committer Full Name <user@organization.tld> 2000098000 +0100
313 | data 29
314 | O: Merge commit 'N' into 'M'
315 | from :37
316 | merge :38
317 | D moduleA/sometimes
318 |
319 | commit refs/heads/master
320 | mark :40
321 | original-oid 000000000000000000000000000000000000002E
322 | author Full Name <user@organization.tld> 2000099000 +0100
323 | committer Full Name <user@organization.tld> 2000099000 +0100
324 | data 29
325 | P: Merge commit 'M' into 'N'
326 | from :38
327 | merge :37
328 | M 100644 :15 moduleA/sometimes
329 |
330 | commit refs/heads/master
331 | mark :41
332 | original-oid 0000000000000000000000000000000000000030
333 | author Full Name <user@organization.tld> 3000000000 +0100
334 | committer Full Name <user@organization.tld> 3000000000 +0100
335 | data 2
336 | Q
337 | from :40
338 | M 100644 :16 moduleA/keepme
339 |
340 | commit refs/heads/master
341 | mark :42
342 | original-oid 0000000000000000000000000000000000000031
343 | author Full Name <user@organization.tld> 3000010000 +0100
344 | committer Full Name <user@organization.tld> 3000010000 +0100
345 | data 2
346 | R
347 | from :41
348 | M 100644 :17 moduleB/nukeme
349 |
350 | commit refs/heads/master
351 | mark :43
352 | original-oid 0000000000000000000000000000000000000032
353 | author Full Name <user@organization.tld> 3000020000 +0100
354 | committer Full Name <user@organization.tld> 3000020000 +0100
355 | data 29
356 | S: Merge commit 'R' into 'R'
357 | from :42
358 | merge :42
359 | M 100644 :18 moduleA/sometimes
360 |
361 | commit refs/heads/master
362 | mark :44
363 | original-oid 0000000000000000000000000000000000000033
364 | author Full Name <user@organization.tld> 3000030000 +0100
365 | committer Full Name <user@organization.tld> 3000030000 +0100
366 | data 2
367 | T
368 | from :43
369 | M 100644 :19 moduleA/keepme
370 |
371 | commit refs/heads/master
372 | mark :45
373 | original-oid 0000000000000000000000000000000000000034
374 | author Full Name <user@organization.tld> 3000040000 +0100
375 | committer Full Name <user@organization.tld> 3000040000 +0100
376 | data 2
377 | U
378 | from :44
379 | M 100644 :20 moduleA/sometimes
380 |
381 | commit refs/heads/master
382 | mark :46
383 | original-oid 0000000000000000000000000000000000000035
384 | author Full Name <user@organization.tld> 3000050000 +0100
385 | committer Full Name <user@organization.tld> 3000050000 +0100
386 | data 29
387 | V: Merge commit 'U' into 'U'
388 | from :45
389 | merge :45
390 | M 100644 :21 moduleB/nukeme
391 |
392 | commit refs/heads/master
393 | mark :47
394 | original-oid 0000000000000000000000000000000000000036
395 | author Full Name <user@organization.tld> 3000060000 +0100
396 | committer Full Name <user@organization.tld> 3000060000 +0100
397 | data 2
398 | W
399 | from :46
400 | M 100644 :22 moduleA/keepme
401 |
402 | done
403 |
--------------------------------------------------------------------------------
/t/t9390/degenerate-evil-merge:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 0
5 |
6 | reset refs/heads/master
7 | commit refs/heads/master
8 | mark :2
9 | author Full Name <user@organization.tld> 2000000000 +0100
10 | committer Full Name <user@organization.tld> 2000000000 +0100
11 | data 2
12 | A
13 | M 100644 :1 irrelevant
14 | M 100644 :1 module-of-interest/file1
15 | M 100644 :1 module-of-interest/file2
16 | M 100644 :1 module-of-interest/file3
17 | M 100644 :1 other-module/files
18 | M 100644 :1 other-module/are
19 | M 100644 :1 other-module/fun
20 |
21 | commit refs/heads/master
22 | mark :3
23 | author Full Name <user@organization.tld> 2000030000 +0100
24 | committer Full Name <user@organization.tld> 2000030000 +0100
25 | data 2
26 | B
27 | from :2
28 | D irrelevant
29 | D module-of-interest/file1
30 | D module-of-interest/file2
31 | D module-of-interest/file3
32 |
33 | blob
34 | mark :4
35 | data 8
36 | content
37 |
38 | commit refs/heads/master
39 | mark :5
40 | author Full Name <user@organization.tld> 2000040000 +0100
41 | committer Full Name <user@organization.tld> 2000040000 +0100
42 | data 2
43 | D
44 | from :3
45 | M 100644 :4 other-module/fun
46 |
47 | commit refs/heads/master
48 | mark :6
49 | author Full Name <user@organization.tld> 2000020000 +0100
50 | committer Full Name <user@organization.tld> 2000020000 +0100
51 | data 2
52 | C
53 | from :2
54 | M 100644 :4 irrelevant
55 |
56 | commit refs/heads/master
57 | mark :7
58 | author Full Name <user@organization.tld> 2000050000 +0100
59 | committer Full Name <user@organization.tld> 2000050000 +0100
60 | data 31
61 | Merge and ignore the deletions
62 | from :6
63 | merge :5
64 | M 100644 :4 irrelevant
65 | M 100644 :4 other-module/fun
66 |
67 | blob
68 | mark :8
69 | data 6
70 | final
71 |
72 | commit refs/heads/master
73 | mark :7
74 | author Full Name <user@organization.tld> 2000060000 +0100
75 | committer Full Name <user@organization.tld> 2000060000 +0100
76 | data 13
77 | Final change
78 | from :7
79 | M 100644 :8 module-of-interest/file2
80 |
81 | done
82 |
--------------------------------------------------------------------------------
/t/t9390/degenerate-globme:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 10
5 | keepme v1
6 |
7 | reset refs/heads/master
8 | commit refs/heads/master
9 | mark :2
10 | author Full Name <user@organization.tld> 2000000000 +0100
11 | committer Full Name <user@organization.tld> 2000000000 +0100
12 | data 2
13 | A
14 | M 100644 :1 moduleA/keepme
15 |
16 | blob
17 | mark :3
18 | data 10
19 | nukeme v1
20 |
21 | commit refs/heads/master
22 | mark :4
23 | author Full Name <user@organization.tld> 2000010000 +0100
24 | committer Full Name <user@organization.tld> 2000010000 +0100
25 | data 2
26 | B
27 | from :2
28 | M 100644 :3 moduleB/nukeme
29 |
30 | blob
31 | mark :5
32 | data 10
33 | nukeme v2
34 |
35 | commit refs/heads/master
36 | mark :6
37 | author Full Name <user@organization.tld> 2000020000 +0100
38 | committer Full Name <user@organization.tld> 2000020000 +0100
39 | data 2
40 | C
41 | from :2
42 | M 100644 :5 moduleB/nukeme
43 |
44 | commit refs/heads/master
45 | mark :7
46 | author Full Name <user@organization.tld> 2000030000 +0100
47 | committer Full Name <user@organization.tld> 2000030000 +0100
48 | data 29
49 | D: Merge commit 'C' into 'B'
50 | from :4
51 | merge :6
52 |
53 | blob
54 | mark :8
55 | data 10
56 | nukeme v3
57 |
58 | commit refs/heads/master
59 | mark :9
60 | author Full Name <user@organization.tld> 2000040000 +0100
61 | committer Full Name <user@organization.tld> 2000040000 +0100
62 | data 2
63 | F
64 | from :7
65 | M 100644 :8 moduleB/nukeme
66 |
67 | blob
68 | mark :10
69 | data 10
70 | keepme v2
71 |
72 | commit refs/heads/master
73 | mark :11
74 | author Full Name <user@organization.tld> 2000050000 +0100
75 | committer Full Name <user@organization.tld> 2000050000 +0100
76 | data 2
77 | G
78 | from :9
79 | M 100644 :10 moduleA/keepme
80 |
81 | blob
82 | mark :12
83 | data 10
84 | nukem4 v4
85 |
86 | commit refs/heads/master
87 | mark :13
88 | author Full Name <user@organization.tld> 2000060000 +0100
89 | committer Full Name <user@organization.tld> 2000060000 +0100
90 | data 2
91 | H
92 | from :11
93 | M 100644 :12 moduleB/nukeme
94 |
95 | commit refs/heads/branchI
96 | mark :14
97 | author Full Name <user@organization.tld> 2000070000 +0100
98 | committer Full Name <user@organization.tld> 2000070000 +0100
99 | data 29
100 | I: Merge commit 'D' into 'H'
101 | from :13
102 | merge :7
103 |
104 | commit refs/heads/master
105 | mark :15
106 | author Full Name <user@organization.tld> 2000080000 +0100
107 | committer Full Name <user@organization.tld> 2000080000 +0100
108 | data 29
109 | J: Merge commit 'H' into 'D'
110 | from :7
111 | merge :13
112 |
113 | blob
114 | mark :16
115 | data 10
116 | nukeme v4
117 |
118 | commit refs/heads/master
119 | mark :17
120 | author Full Name <user@organization.tld> 2000090000 +0100
121 | committer Full Name <user@organization.tld> 2000090000 +0100
122 | data 2
123 | K
124 | from :15
125 | M 100644 :16 moduleB/nukeme
126 |
127 | blob
128 | mark :18
129 | data 10
130 | keepme v3
131 |
132 | commit refs/heads/master
133 | mark :19
134 | author Full Name <user@organization.tld> 2000092000 +0100
135 | committer Full Name <user@organization.tld> 2000092000 +0100
136 | data 2
137 | L
138 | from :17
139 | M 100644 :18 moduleA/keepme
140 |
141 | blob
142 | mark :20
143 | data 10
144 | nukeme v5
145 |
146 | commit refs/heads/master
147 | mark :21
148 | author Full Name <user@organization.tld> 2000094000 +0100
149 | committer Full Name <user@organization.tld> 2000094000 +0100
150 | data 2
151 | M
152 | from :19
153 | M 100644 :20 moduleB/nukeme
154 |
155 | blob
156 | mark :22
157 | data 10
158 | nukeme v6
159 |
160 | commit refs/heads/master
161 | mark :23
162 | author Full Name <user@organization.tld> 2000096000 +0100
163 | committer Full Name <user@organization.tld> 2000096000 +0100
164 | data 2
165 | N
166 | from :6
167 | M 100644 :22 moduleB/nukeme
168 |
169 | commit refs/heads/branchO
170 | mark :24
171 | author Full Name <user@organization.tld> 2000098000 +0100
172 | committer Full Name <user@organization.tld> 2000098000 +0100
173 | data 29
174 | O: Merge commit 'N' into 'M'
175 | from :21
176 | merge :23
177 |
178 | commit refs/heads/master
179 | mark :25
180 | author Full Name <user@organization.tld> 2000099000 +0100
181 | committer Full Name <user@organization.tld> 2000099000 +0100
182 | data 29
183 | P: Merge commit 'M' into 'N'
184 | from :23
185 | merge :21
186 |
187 | blob
188 | mark :26
189 | data 10
190 | keepme v4
191 |
192 | commit refs/heads/master
193 | mark :27
194 | author Full Name <user@organization.tld> 3000000000 +0100
195 | committer Full Name <user@organization.tld> 3000000000 +0100
196 | data 2
197 | Q
198 | from :25
199 | M 100644 :26 moduleA/keepme
200 |
201 | blob
202 | mark :28
203 | data 10
204 | nukeme v7
205 |
206 | commit refs/heads/master
207 | mark :29
208 | author Full Name <user@organization.tld> 3000010000 +0100
209 | committer Full Name <user@organization.tld> 3000010000 +0100
210 | data 2
211 | R
212 | from :27
213 | M 100644 :28 moduleB/nukeme
214 |
215 | commit refs/heads/master
216 | mark :30
217 | author Full Name <user@organization.tld> 3000020000 +0100
218 | committer Full Name <user@organization.tld> 3000020000 +0100
219 | data 29
220 | S: Merge commit 'R' into 'R'
221 | from :29
222 | merge :29
223 |
224 | blob
225 | mark :31
226 | data 10
227 | keepme v5
228 |
229 | commit refs/heads/master
230 | mark :32
231 | author Full Name <user@organization.tld> 3000030000 +0100
232 | committer Full Name <user@organization.tld> 3000030000 +0100
233 | data 2
234 | T
235 | from :30
236 | M 100644 :31 moduleA/keepme
237 |
238 | blob
239 | mark :33
240 | data 10
241 | nukeme v8
242 |
243 | commit refs/heads/master
244 | mark :34
245 | author Full Name <user@organization.tld> 3000050000 +0100
246 | committer Full Name <user@organization.tld> 3000050000 +0100
247 | data 29
248 | V: Merge commit 'U' into 'U'
249 | from :32
250 | merge :32
251 | M 100644 :33 moduleB/nukeme
252 |
253 | blob
254 | mark :35
255 | data 10
256 | keepme v6
257 |
258 | commit refs/heads/master
259 | mark :36
260 | author Full Name <user@organization.tld> 3000060000 +0100
261 | committer Full Name <user@organization.tld> 3000060000 +0100
262 | data 2
263 | W
264 | from :34
265 | M 100644 :35 moduleA/keepme
266 |
267 | done
268 |
--------------------------------------------------------------------------------
/t/t9390/degenerate-keepme:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 10
5 | keepme v1
6 |
7 | reset refs/heads/branchO
8 | commit refs/heads/branchO
9 | mark :2
10 | author Full Name <user@organization.tld> 2000000000 +0100
11 | committer Full Name <user@organization.tld> 2000000000 +0100
12 | data 2
13 | A
14 | M 100644 :1 moduleA/keepme
15 |
16 | blob
17 | mark :3
18 | data 10
19 | keepme v2
20 |
21 | commit refs/heads/branchO
22 | mark :4
23 | author Full Name <user@organization.tld> 2000050000 +0100
24 | committer Full Name <user@organization.tld> 2000050000 +0100
25 | data 2
26 | G
27 | from :2
28 | M 100644 :3 moduleA/keepme
29 |
30 | commit refs/heads/branchI
31 | mark :5
32 | author Full Name <user@organization.tld> 2000070000 +0100
33 | committer Full Name <user@organization.tld> 2000070000 +0100
34 | data 29
35 | I: Merge commit 'D' into 'H'
36 | from :4
37 | merge :2
38 |
39 | commit refs/heads/branchO
40 | mark :6
41 | author Full Name <user@organization.tld> 2000080000 +0100
42 | committer Full Name <user@organization.tld> 2000080000 +0100
43 | data 29
44 | J: Merge commit 'H' into 'D'
45 | from :2
46 | merge :4
47 |
48 | blob
49 | mark :7
50 | data 10
51 | keepme v3
52 |
53 | commit refs/heads/branchO
54 | mark :8
55 | author Full Name <user@organization.tld> 2000092000 +0100
56 | committer Full Name <user@organization.tld> 2000092000 +0100
57 | data 2
58 | L
59 | from :6
60 | M 100644 :7 moduleA/keepme
61 |
62 | blob
63 | mark :9
64 | data 10
65 | keepme v4
66 |
67 | commit refs/heads/master
68 | mark :10
69 | author Full Name <user@organization.tld> 3000000000 +0100
70 | committer Full Name <user@organization.tld> 3000000000 +0100
71 | data 2
72 | Q
73 | from :8
74 | M 100644 :9 moduleA/keepme
75 |
76 | blob
77 | mark :11
78 | data 10
79 | keepme v5
80 |
81 | commit refs/heads/master
82 | mark :12
83 | author Full Name <user@organization.tld> 3000030000 +0100
84 | committer Full Name <user@organization.tld> 3000030000 +0100
85 | data 2
86 | T
87 | from :10
88 | M 100644 :11 moduleA/keepme
89 |
90 | blob
91 | mark :13
92 | data 10
93 | keepme v6
94 |
95 | commit refs/heads/master
96 | mark :14
97 | author Full Name <user@organization.tld> 3000060000 +0100
98 | committer Full Name <user@organization.tld> 3000060000 +0100
99 | data 2
100 | W
101 | from :12
102 | M 100644 :13 moduleA/keepme
103 |
104 | done
105 |
--------------------------------------------------------------------------------
/t/t9390/degenerate-keepme-noff:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 10
5 | keepme v1
6 |
7 | reset refs/heads/master
8 | commit refs/heads/master
9 | mark :2
10 | author Full Name <user@organization.tld> 2000000000 +0100
11 | committer Full Name <user@organization.tld> 2000000000 +0100
12 | data 2
13 | A
14 | M 100644 :1 moduleA/keepme
15 |
16 | blob
17 | mark :3
18 | data 10
19 | keepme v2
20 |
21 | commit refs/heads/branchO
22 | mark :4
23 | author Full Name <user@organization.tld> 2000050000 +0100
24 | committer Full Name <user@organization.tld> 2000050000 +0100
25 | data 2
26 | G
27 | from :2
28 | M 100644 :3 moduleA/keepme
29 |
30 | commit refs/heads/branchI
31 | mark :5
32 | author Full Name <user@organization.tld> 2000070000 +0100
33 | committer Full Name <user@organization.tld> 2000070000 +0100
34 | data 29
35 | I: Merge commit 'D' into 'H'
36 | from :4
37 | merge :2
38 |
39 | commit refs/heads/branchO
40 | mark :6
41 | author Full Name <user@organization.tld> 2000080000 +0100
42 | committer Full Name <user@organization.tld> 2000080000 +0100
43 | data 29
44 | J: Merge commit 'H' into 'D'
45 | from :2
46 | merge :4
47 |
48 | blob
49 | mark :7
50 | data 10
51 | keepme v3
52 |
53 | commit refs/heads/branchO
54 | mark :8
55 | author Full Name <user@organization.tld> 2000092000 +0100
56 | committer Full Name <user@organization.tld> 2000092000 +0100
57 | data 2
58 | L
59 | from :6
60 | M 100644 :7 moduleA/keepme
61 |
62 | commit refs/heads/master
63 | mark :9
64 | author Full Name <user@organization.tld> 2000099000 +0100
65 | committer Full Name <user@organization.tld> 2000099000 +0100
66 | data 29
67 | P: Merge commit 'M' into 'N'
68 | from :2
69 | merge :8
70 |
71 | blob
72 | mark :10
73 | data 10
74 | keepme v4
75 |
76 | commit refs/heads/master
77 | mark :11
78 | author Full Name <user@organization.tld> 3000000000 +0100
79 | committer Full Name <user@organization.tld> 3000000000 +0100
80 | data 2
81 | Q
82 | from :9
83 | M 100644 :10 moduleA/keepme
84 |
85 | blob
86 | mark :12
87 | data 10
88 | keepme v5
89 |
90 | commit refs/heads/master
91 | mark :13
92 | author Full Name <user@organization.tld> 3000030000 +0100
93 | committer Full Name <user@organization.tld> 3000030000 +0100
94 | data 2
95 | T
96 | from :11
97 | M 100644 :12 moduleA/keepme
98 |
99 | blob
100 | mark :14
101 | data 10
102 | keepme v6
103 |
104 | commit refs/heads/master
105 | mark :15
106 | author Full Name <user@organization.tld> 3000060000 +0100
107 | committer Full Name <user@organization.tld> 3000060000 +0100
108 | data 2
109 | W
110 | from :13
111 | M 100644 :14 moduleA/keepme
112 |
113 | done
114 |
--------------------------------------------------------------------------------
/t/t9390/degenerate-moduleA:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 10
5 | keepme v1
6 |
7 | reset refs/heads/master
8 | commit refs/heads/master
9 | mark :2
10 | author Full Name <user@organization.tld> 2000000000 +0100
11 | committer Full Name <user@organization.tld> 2000000000 +0100
12 | data 2
13 | A
14 | M 100644 :1 moduleA/keepme
15 |
16 | blob
17 | mark :3
18 | data 13
19 | sometimes v1
20 |
21 | commit refs/heads/master
22 | mark :4
23 | author Full Name <user@organization.tld> 2000030000 +0100
24 | committer Full Name <user@organization.tld> 2000030000 +0100
25 | data 29
26 | D: Merge commit 'C' into 'B'
27 | from :2
28 | merge :2
29 | M 100644 :3 moduleA/sometimes
30 |
31 | blob
32 | mark :5
33 | data 10
34 | keepme v2
35 |
36 | commit refs/heads/master
37 | mark :6
38 | author Full Name <user@organization.tld> 2000050000 +0100
39 | committer Full Name <user@organization.tld> 2000050000 +0100
40 | data 2
41 | G
42 | from :4
43 | M 100644 :5 moduleA/keepme
44 |
45 | blob
46 | mark :7
47 | data 13
48 | sometimes v2
49 |
50 | commit refs/heads/branchI
51 | mark :8
52 | author Full Name <user@organization.tld> 2000070000 +0100
53 | committer Full Name <user@organization.tld> 2000070000 +0100
54 | data 29
55 | I: Merge commit 'D' into 'H'
56 | from :6
57 | merge :4
58 | M 100644 :7 moduleA/sometimes
59 |
60 | blob
61 | mark :9
62 | data 13
63 | sometimes v3
64 |
65 | commit refs/heads/master
66 | mark :10
67 | author Full Name <user@organization.tld> 2000080000 +0100
68 | committer Full Name <user@organization.tld> 2000080000 +0100
69 | data 29
70 | J: Merge commit 'H' into 'D'
71 | from :4
72 | merge :6
73 | M 100644 :9 moduleA/sometimes
74 |
75 | blob
76 | mark :11
77 | data 10
78 | keepme v3
79 |
80 | commit refs/heads/master
81 | mark :12
82 | author Full Name <user@organization.tld> 2000092000 +0100
83 | committer Full Name <user@organization.tld> 2000092000 +0100
84 | data 2
85 | L
86 | from :10
87 | M 100644 :11 moduleA/keepme
88 |
89 | commit refs/heads/branchO
90 | mark :13
91 | author Full Name <user@organization.tld> 2000098000 +0100
92 | committer Full Name <user@organization.tld> 2000098000 +0100
93 | data 29
94 | O: Merge commit 'N' into 'M'
95 | from :12
96 | merge :2
97 | D moduleA/sometimes
98 |
99 | blob
100 | mark :14
101 | data 13
102 | sometimes v5
103 |
104 | commit refs/heads/master
105 | mark :15
106 | author Full Name <user@organization.tld> 2000099000 +0100
107 | committer Full Name <user@organization.tld> 2000099000 +0100
108 | data 29
109 | P: Merge commit 'M' into 'N'
110 | from :2
111 | merge :12
112 | M 100644 :14 moduleA/sometimes
113 |
114 | blob
115 | mark :16
116 | data 10
117 | keepme v4
118 |
119 | commit refs/heads/master
120 | mark :17
121 | author Full Name <user@organization.tld> 3000000000 +0100
122 | committer Full Name <user@organization.tld> 3000000000 +0100
123 | data 2
124 | Q
125 | from :15
126 | M 100644 :16 moduleA/keepme
127 |
128 | blob
129 | mark :18
130 | data 13
131 | sometimes v6
132 |
133 | commit refs/heads/master
134 | mark :19
135 | author Full Name <user@organization.tld> 3000020000 +0100
136 | committer Full Name <user@organization.tld> 3000020000 +0100
137 | data 29
138 | S: Merge commit 'R' into 'R'
139 | from :17
140 | merge :17
141 | M 100644 :18 moduleA/sometimes
142 |
143 | blob
144 | mark :20
145 | data 10
146 | keepme v5
147 |
148 | commit refs/heads/master
149 | mark :21
150 | author Full Name <user@organization.tld> 3000030000 +0100
151 | committer Full Name <user@organization.tld> 3000030000 +0100
152 | data 2
153 | T
154 | from :19
155 | M 100644 :20 moduleA/keepme
156 |
157 | blob
158 | mark :22
159 | data 13
160 | sometimes v7
161 |
162 | commit refs/heads/master
163 | mark :23
164 | author Full Name <user@organization.tld> 3000040000 +0100
165 | committer Full Name <user@organization.tld> 3000040000 +0100
166 | data 2
167 | U
168 | from :21
169 | M 100644 :22 moduleA/sometimes
170 |
171 | commit refs/heads/master
172 | mark :24
173 | author Full Name <user@organization.tld> 3000050000 +0100
174 | committer Full Name <user@organization.tld> 3000050000 +0100
175 | data 29
176 | V: Merge commit 'U' into 'U'
177 | from :23
178 | merge :23
179 |
180 | blob
181 | mark :25
182 | data 10
183 | keepme v6
184 |
185 | commit refs/heads/master
186 | mark :26
187 | author Full Name <user@organization.tld> 3000060000 +0100
188 | committer Full Name <user@organization.tld> 3000060000 +0100
189 | data 2
190 | W
191 | from :24
192 | M 100644 :25 moduleA/keepme
193 |
194 | done
195 |
--------------------------------------------------------------------------------
/t/t9390/empty:
--------------------------------------------------------------------------------
1 | feature done
2 | # Simple repo with only two files, with a whole bunch of cases dealing with
3 | # empty pruning, particularly commits that start empty.
4 | #
5 | # As with case1, the original-oid directives are very fake, but if an error
6 | # is hit that shows one of these, it makes it really easy to know where it
7 | # came from.
8 | blob
9 | mark :1
10 | original-oid 0000000000000000000000000000000000000001
11 | data 10
12 | nukeme v1
13 |
14 | blob
15 | mark :2
16 | original-oid 0000000000000000000000000000000000000002
17 | data 10
18 | keepme v1
19 |
20 | blob
21 | mark :3
22 | original-oid 0000000000000000000000000000000000000003
23 | data 10
24 | nukeme v2
25 |
26 | blob
27 | mark :4
28 | original-oid 0000000000000000000000000000000000000004
29 | data 10
30 | keepme v2
31 |
32 | commit refs/heads/master
33 | mark :5
34 | original-oid 0000000000000000000000000000000000000010
35 | author Full Name <user@organization.tld> 1000000000 +0100
36 | committer Full Name <user@organization.tld> 1000000000 +0100
37 | data 2
38 | A
39 |
40 | commit refs/heads/master
41 | mark :6
42 | original-oid 0000000000000000000000000000000000000011
43 | author Full Name <user@organization.tld> 1000010000 +0100
44 | committer Full Name <user@organization.tld> 1000010000 +0100
45 | data 2
46 | B
47 | from :5
48 |
49 | reset refs/heads/master
50 | commit refs/heads/master
51 | mark :7
52 | original-oid 0000000000000000000000000000000000000012
53 | author Full Name <user@organization.tld> 1000020000 +0100
54 | committer Full Name <user@organization.tld> 1000020000 +0100
55 | data 2
56 | C
57 | M 100644 :1 nukeme
58 |
59 | commit refs/heads/master
60 | mark :8
61 | original-oid 0000000000000000000000000000000000000013
62 | author Full Name <user@organization.tld> 1000030000 +0100
63 | committer Full Name <user@organization.tld> 1000030000 +0100
64 | data 2
65 | D
66 | from :7
67 |
68 | commit refs/heads/master
69 | mark :9
70 | original-oid 0000000000000000000000000000000000000014
71 | author Full Name <user@organization.tld> 1000040000 +0100
72 | committer Full Name <user@organization.tld> 1000040000 +0100
73 | data 29
74 | E: Merge commit 'D' into 'B'
75 | from :6
76 | merge :8
77 | M 100644 :2 keepme
78 |
79 | commit refs/heads/master
80 | mark :10
81 | original-oid 0000000000000000000000000000000000000015
82 | author Full Name <user@organization.tld> 1000050000 +0100
83 | committer Full Name <user@organization.tld> 1000050000 +0100
84 | data 29
85 | F: Merge commit 'D' into 'B'
86 | from :6
87 | merge :8
88 |
89 | commit refs/heads/master
90 | mark :11
91 | original-oid 0000000000000000000000000000000000000016
92 | author Full Name <user@organization.tld> 1000060000 +0100
93 | committer Full Name <user@organization.tld> 1000060000 +0100
94 | data 2
95 | G
96 | from :9
97 | M 100644 :3 nukeme
98 |
99 | commit refs/heads/master
100 | mark :12
101 | original-oid 0000000000000000000000000000000000000017
102 | author Full Name <user@organization.tld> 1000070000 +0100
103 | committer Full Name <user@organization.tld> 1000070000 +0100
104 | data 2
105 | H
106 | from :11
107 |
108 | commit refs/heads/master
109 | mark :13
110 | original-oid 0000000000000000000000000000000000000018
111 | author Full Name <user@organization.tld> 1000080000 +0100
112 | committer Full Name <user@organization.tld> 1000080000 +0100
113 | data 2
114 | I
115 | from :10
116 | M 100644 :4 keepme
117 |
118 | commit refs/heads/master
119 | mark :14
120 | original-oid 0000000000000000000000000000000000000019
121 | author Full Name <user@organization.tld> 1000090000 +0100
122 | committer Full Name <user@organization.tld> 1000090000 +0100
123 | data 29
124 | J: Merge commit 'I' into 'H'
125 | from :12
126 | merge :13
127 |
128 | done
129 |
--------------------------------------------------------------------------------
/t/t9390/empty-keepme:
--------------------------------------------------------------------------------
1 | feature done
2 | reset refs/heads/master
3 | commit refs/heads/master
4 | mark :1
5 | author Full Name <user@organization.tld> 1000000000 +0100
6 | committer Full Name <user@organization.tld> 1000000000 +0100
7 | data 2
8 | A
9 |
10 | commit refs/heads/master
11 | mark :2
12 | author Full Name <user@organization.tld> 1000010000 +0100
13 | committer Full Name <user@organization.tld> 1000010000 +0100
14 | data 2
15 | B
16 | from :1
17 |
18 | blob
19 | mark :3
20 | data 10
21 | keepme v1
22 |
23 | commit refs/heads/master
24 | mark :4
25 | author Full Name <user@organization.tld> 1000040000 +0100
26 | committer Full Name <user@organization.tld> 1000040000 +0100
27 | data 29
28 | E: Merge commit 'D' into 'B'
29 | from :2
30 | M 100644 :3 keepme
31 |
32 | blob
33 | mark :5
34 | data 10
35 | keepme v2
36 |
37 | commit refs/heads/master
38 | mark :6
39 | author Full Name <user@organization.tld> 1000080000 +0100
40 | committer Full Name <user@organization.tld> 1000080000 +0100
41 | data 2
42 | I
43 | from :2
44 | M 100644 :5 keepme
45 |
46 | commit refs/heads/master
47 | mark :7
48 | author Full Name <user@organization.tld> 1000090000 +0100
49 | committer Full Name <user@organization.tld> 1000090000 +0100
50 | data 29
51 | J: Merge commit 'I' into 'H'
52 | from :4
53 | merge :6
54 |
55 | done
56 |
--------------------------------------------------------------------------------
/t/t9390/less-empty-keepme:
--------------------------------------------------------------------------------
1 | feature done
2 | reset refs/heads/master
3 | commit refs/heads/master
4 | mark :1
5 | author Full Name <user@organization.tld> 1000000000 +0100
6 | committer Full Name <user@organization.tld> 1000000000 +0100
7 | data 2
8 | A
9 |
10 | commit refs/heads/master
11 | mark :2
12 | author Full Name <user@organization.tld> 1000010000 +0100
13 | committer Full Name <user@organization.tld> 1000010000 +0100
14 | data 2
15 | B
16 | from :1
17 |
18 | reset refs/heads/master
19 | commit refs/heads/master
20 | mark :3
21 | author Full Name <user@organization.tld> 1000020000 +0100
22 | committer Full Name <user@organization.tld> 1000020000 +0100
23 | data 2
24 | C
25 |
26 | commit refs/heads/master
27 | mark :4
28 | author Full Name <user@organization.tld> 1000030000 +0100
29 | committer Full Name <user@organization.tld> 1000030000 +0100
30 | data 2
31 | D
32 | from :3
33 |
34 | blob
35 | mark :5
36 | data 10
37 | keepme v1
38 |
39 | commit refs/heads/master
40 | mark :6
41 | author Full Name <user@organization.tld> 1000040000 +0100
42 | committer Full Name <user@organization.tld> 1000040000 +0100
43 | data 29
44 | E: Merge commit 'D' into 'B'
45 | from :2
46 | merge :4
47 | M 100644 :5 keepme
48 |
49 | commit refs/heads/master
50 | mark :7
51 | author Full Name <user@organization.tld> 1000060000 +0100
52 | committer Full Name <user@organization.tld> 1000060000 +0100
53 | data 2
54 | G
55 | from :6
56 |
57 | commit refs/heads/master
58 | mark :8
59 | author Full Name <user@organization.tld> 1000070000 +0100
60 | committer Full Name <user@organization.tld> 1000070000 +0100
61 | data 2
62 | H
63 | from :7
64 |
65 | commit refs/heads/master
66 | mark :9
67 | author Full Name <user@organization.tld> 1000050000 +0100
68 | committer Full Name <user@organization.tld> 1000050000 +0100
69 | data 29
70 | F: Merge commit 'D' into 'B'
71 | from :2
72 | merge :4
73 |
74 | blob
75 | mark :10
76 | data 10
77 | keepme v2
78 |
79 | commit refs/heads/master
80 | mark :11
81 | author Full Name <user@organization.tld> 1000080000 +0100
82 | committer Full Name <user@organization.tld> 1000080000 +0100
83 | data 2
84 | I
85 | from :9
86 | M 100644 :10 keepme
87 |
88 | commit refs/heads/master
89 | mark :12
90 | author Full Name <user@organization.tld> 1000090000 +0100
91 | committer Full Name <user@organization.tld> 1000090000 +0100
92 | data 29
93 | J: Merge commit 'I' into 'H'
94 | from :8
95 | merge :11
96 |
97 | done
98 |
--------------------------------------------------------------------------------
/t/t9390/more-empty-keepme:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 10
5 | keepme v1
6 |
7 | reset refs/heads/master
8 | commit refs/heads/master
9 | mark :2
10 | author Full Name <user@organization.tld> 1000040000 +0100
11 | committer Full Name <user@organization.tld> 1000040000 +0100
12 | data 29
13 | E: Merge commit 'D' into 'B'
14 | M 100644 :1 keepme
15 |
16 | blob
17 | mark :3
18 | data 10
19 | keepme v2
20 |
21 | reset refs/heads/master
22 | commit refs/heads/master
23 | mark :4
24 | author Full Name <user@organization.tld> 1000080000 +0100
25 | committer Full Name <user@organization.tld> 1000080000 +0100
26 | data 2
27 | I
28 | M 100644 :3 keepme
29 |
30 | commit refs/heads/master
31 | mark :5
32 | author Full Name <user@organization.tld> 1000090000 +0100
33 | committer Full Name <user@organization.tld> 1000090000 +0100
34 | data 29
35 | J: Merge commit 'I' into 'H'
36 | from :2
37 | merge :4
38 |
39 | done
40 |
--------------------------------------------------------------------------------
/t/t9390/sample-mailmap:
--------------------------------------------------------------------------------
1 | Little 'ol Me <me@little.net>
2 | <me@little.net> <me@laptop.(none)>
3 | # Here is a comment
4 | Little 'ol Me <me@little.net> Little O. Me
5 | Little 'ol Me <me@little.net> <me@fire.com>
6 | Little 'ol Me <me@little.net> Little Me <me@bigcompany.com>
7 | Little John <second@merry.men> little.john <>
8 |
--------------------------------------------------------------------------------
/t/t9390/sample-message:
--------------------------------------------------------------------------------
1 | Initial==>Modified
2 | regex:tw.nty==>the number 20
3 | v1.0==>version one!
4 | regex:!$==> :)
5 |
--------------------------------------------------------------------------------
/t/t9390/sample-replace:
--------------------------------------------------------------------------------
1 | mod==>modified-by-gremlins
2 |
--------------------------------------------------------------------------------
/t/t9390/unusual:
--------------------------------------------------------------------------------
1 | option git quiet
2 | feature done
3 | # Input in a format filter-repo isn't generally expected to receive (either
4 | # because we don't pass certain flags to fast-export or repos don't have the
5 | # weird features or whatever other reason), but which we want to test for
6 | # completeness.
7 | progress I am starting the import, yo.
8 |
9 | checkpoint
10 |
11 | blob
12 | mark :1
13 | original-oid 0000000000000000000000000000000000000001
14 | data 5
15 | hello
16 |
17 | commit refs/heads/master
18 | mark :2
19 | original-oid 0000000000000000000000000000000000000002
20 | committer Srinivasa Ramanujan <fellow@cambridge.org> 1535228562 +051800
21 | data 8
22 | Initial
23 | M 100644 :1 greeting
24 |
25 | reset refs/heads/develop
26 | from :2
27 |
28 | tag v1.0
29 | from :2
30 | original-oid 0000000000000000000000000000000000000003
31 | tagger little.john <> 1535229618 -0700
32 | data 4
33 | v1.0
34 |
35 | done
36 |
--------------------------------------------------------------------------------
/t/t9390/unusual-filtered:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 5
5 | hello
6 | reset refs/heads/develop
7 | commit refs/heads/develop
8 | mark :2
9 | author Srinivasa Ramanujan <fellow@cambridge.org> 1535228562 +051800
10 | committer Srinivasa Ramanujan <fellow@cambridge.org> 1535228562 +051800
11 | data 8
12 | Initial
13 | M 100644 :1 greeting
14 |
15 | reset refs/heads/master
16 | from :2
17 |
18 | tag v1.0
19 | from :2
20 | tagger little.john <> 1535229618 -0700
21 | data 4
22 | v1.0
23 | done
24 |
--------------------------------------------------------------------------------
/t/t9390/unusual-mailmap:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 5
5 | hello
6 | reset refs/heads/develop
7 | commit refs/heads/develop
8 | mark :2
9 | author Srinivasa Ramanujan <fellow@cambridge.org> 1535228562 +051800
10 | committer Srinivasa Ramanujan <fellow@cambridge.org> 1535228562 +051800
11 | data 8
12 | Initial
13 | M 100644 :1 greeting
14 |
15 | reset refs/heads/master
16 | from :2
17 |
18 | tag v1.0
19 | from :2
20 | tagger Little John <second@merry.men> 1535229618 -0700
21 | data 4
22 | v1.0
23 | done
24 |
--------------------------------------------------------------------------------
/t/t9391-filter-repo-lib-usage.sh:
--------------------------------------------------------------------------------
1 | #!/bin/bash
2 |
3 | test_description='Usage of git-filter-repo as a library'
4 | . ./test-lib.sh
5 |
6 | # for git_filter_repo.py import
7 | case "$(uname -s)" in
8 | MINGW*|MSYS)
9 | export PYTHONPATH=$(cygpath -am $TEST_DIRECTORY/..)\;$PYTHONPATH
10 | ;;
11 | *)
12 | export PYTHONPATH=$(dirname $TEST_DIRECTORY):$PYTHONPATH
13 | ;;
14 | esac
15 | # Avoid writing git_filter_repo.pyc file
16 | export PYTHONDONTWRITEBYTECODE=1
17 | export CONTRIB_DIR=$TEST_DIRECTORY/../contrib/filter-repo-demos
18 |
19 | DATA="$TEST_DIRECTORY/t9391"
20 |
21 | setup()
22 | {
23 | git init $1 &&
24 | (
25 | cd $1 &&
26 | echo hello > world &&
27 | git add world &&
28 | test_tick &&
29 | git commit -m initial &&
30 | printf "The launch code is 1-2-3-4." > secret &&
31 | git add secret &&
32 | test_tick &&
33 | git commit -m "Sssh. Dont tell no one" &&
34 | echo A file that you cant trust > file.doc &&
35 | echo there >> world &&
36 | git add file.doc world &&
37 | test_tick &&
38 | printf "Random useless changes\n\nLet us be like the marketing group. Marketing is staffed with pansies" | git commit -F - &&
39 | echo Do not use a preposition to end a setence with > advice &&
40 | git add advice &&
41 | test_tick &&
42 | GIT_AUTHOR_NAME="Copy N. Paste" git commit -m "hypocrisy is fun" &&
43 | echo Avoid cliches like the plague >> advice &&
44 | test_tick &&
45 | GIT_AUTHOR_EMAIL="foo@my.crp" git commit -m "it is still fun" advice &&
46 | echo " \$Id: A bunch of junkquot; > foobar.c &&
47 | git add foobar.c &&
48 | test_tick &&
49 | git commit -m "Brain damage"
50 | )
51 | }
52 |
53 | test_expect_success 'commit_info.py' '
54 | setup commit_info &&
55 | (
56 | cd commit_info &&
57 | $TEST_DIRECTORY/t9391/commit_info.py &&
58 | test 0e5a1029 = $(git rev-parse --short=8 --verify refs/heads/master)
59 | )
60 | '
61 |
62 | test_expect_success 'file_filter.py' '
63 | setup file_filter &&
64 | (
65 | cd file_filter &&
66 | $TEST_DIRECTORY/t9391/file_filter.py &&
67 | test ee59e2b4 = $(git rev-parse --short=8 --verify refs/heads/master)
68 | )
69 | '
70 |
71 | test_expect_success 'print_progress.py' '
72 | setup print_progress &&
73 | (
74 | cd print_progress &&
75 | MASTER=$(git rev-parse --verify master) &&
76 | $TEST_DIRECTORY/t9391/print_progress.py . new &&
77 | test $MASTER = $(git rev-parse --verify refs/heads/master)
78 | )
79 | '
80 |
81 | test_expect_success 'rename-master-to-develop.py' '
82 | setup rename_master_to_develop &&
83 | (
84 | cd rename_master_to_develop &&
85 | MASTER=$(git rev-parse --verify master) &&
86 | $TEST_DIRECTORY/t9391/rename-master-to-develop.py &&
87 | test $MASTER = $(git rev-parse --verify refs/heads/develop)
88 | )
89 | '
90 |
91 | test_expect_success 'strip-cvs-keywords.py' '
92 | setup strip_cvs_keywords &&
93 | (
94 | cd strip_cvs_keywords &&
95 | $TEST_DIRECTORY/t9391/strip-cvs-keywords.py
96 | test 2306fc7c = $(git rev-parse --short=8 --verify refs/heads/master)
97 | )
98 | '
99 |
100 | test_expect_success 'setup two extra repositories' '
101 | mkdir repo1 &&
102 | cd repo1 &&
103 | git init &&
104 | echo hello > world &&
105 | git add world &&
106 | test_tick &&
107 | git commit -m "Commit A" &&
108 | echo goodbye > world &&
109 | git add world &&
110 | test_tick &&
111 | git commit -m "Commit C" &&
112 | cd .. &&
113 | mkdir repo2 &&
114 | cd repo2 &&
115 | git init &&
116 | echo foo > bar &&
117 | git add bar &&
118 | test_tick &&
119 | git commit -m "Commit B" &&
120 | echo fooey > bar &&
121 | git add bar &&
122 | test_tick &&
123 | git commit -m "Commit D" &&
124 | cd ..
125 | '
126 |
127 | test_expect_success 'splice_repos.py' '
128 | git init splice_repos &&
129 | $TEST_DIRECTORY/t9391/splice_repos.py repo1 repo2 splice_repos &&
130 | test 4 = $(git -C splice_repos rev-list master | wc -l)
131 | '
132 |
133 | test_expect_success 'create_fast_export_output.py' '
134 | git init create_fast_export_output &&
135 | (cd create_fast_export_output &&
136 | $TEST_DIRECTORY/t9391/create_fast_export_output.py &&
137 | test e5e0569b = $(git rev-parse --short=8 --verify refs/heads/master) &&
138 | test 122ead00 = $(git rev-parse --short=8 --verify refs/heads/devel) &&
139 | test f36143f9 = $(git rev-parse --short=8 --verify refs/tags/v1.0))
140 | '
141 |
142 | test_expect_success 'unusual.py' '
143 | setup unusual &&
144 | (
145 | cd unusual &&
146 | cat $TEST_DIRECTORY/t9390/unusual | \
147 | $TEST_DIRECTORY/t9391/unusual.py >output &&
148 |
149 | grep "Decipher this: .oy ,tropmi eht gnitrats ma I" output &&
150 | grep "Found 2 blobs/commits and 4 other objects" output
151 | )
152 | '
153 |
154 | test_expect_success 'erroneous.py' '
155 | setup erroneous &&
156 | (
157 | cd erroneous &&
158 | test_must_fail $TEST_DIRECTORY/t9391/erroneous.py 2>../err &&
159 |
160 | test_i18ngrep "Error: Cannot pass a tag_callback to RepoFilter AND pass --tag-callback" ../err
161 | )
162 | '
163 |
164 | test_expect_success 'other error cases' '
165 | GIT_CEILING_DIRECTORIES=$(pwd) &&
166 | export GIT_CEILING_DIRECTORIES &&
167 | (
168 | mkdir other &&
169 | cd other &&
170 |
171 | ! python3 -c "import git_filter_repo as fr; fr.GitUtils.get_commit_count(b\".\", [\"HEAD\"])" 2>err &&
172 | test_i18ngrep "\. does not appear to be a valid git repository" err
173 | )
174 | '
175 |
176 | test_lazy_prereq DOS2UNIX '
177 | dos2unix -h
178 | test $? -ne 127
179 | '
180 |
181 | test_expect_success 'lint-history' '
182 | test_create_repo lint-history &&
183 | (
184 | cd lint-history &&
185 | git config core.autocrlf false &&
186 | echo initial >content &&
187 | git add content &&
188 | git commit -m "initial" &&
189 |
190 | printf "CRLF is stupid\r\n" >content &&
191 | git add content &&
192 | git commit -m "make a statement" &&
193 |
194 | printf "CRLF is stupid\n" >content &&
195 | git add content &&
196 | git commit -m "oops, that was embarassing" &&
197 |
198 | if test_have_prereq DOS2UNIX
199 | then
200 | $CONTRIB_DIR/lint-history --filenames-important dos2unix &&
201 | echo 2 >expect &&
202 | git rev-list --count HEAD >actual &&
203 | test_cmp expect actual
204 | fi
205 | )
206 | '
207 |
208 | test_expect_success !WINDOWS 'lint-history --refs' '
209 | test_create_repo lint-history-only-some-refs &&
210 | (
211 | cd lint-history-only-some-refs &&
212 | test_commit a somefile bad &&
213 | test_commit b notherfile baaad &&
214 | test_commit c whatever baaaaaad &&
215 | git checkout -b mybranch HEAD~1 &&
216 | test_commit d somefile baaaaaaaad &&
217 | test_commit e whatever "baaaaaaaaaad to the bone" &&
218 |
219 | cat <<-EOF >linter.sh &&
220 | #!/bin/bash
221 | cat \$1 | tr -d a >tmp
222 | mv tmp \$1
223 | EOF
224 | chmod u+x linter.sh &&
225 |
226 | PATH=$PATH:. $CONTRIB_DIR/lint-history --refs master..mybranch -- linter.sh &&
227 |
228 | echo bd >expect &&
229 | echo bd to the bone >long-expect &&
230 |
231 | # Verify master is untouched
232 | git checkout master &&
233 | ! test_cmp somefile expect &&
234 | ! test_cmp notherfile expect &&
235 | ! test_cmp whatever expect &&
236 |
237 | # Verify that files touched on the branch are tweaked
238 | git checkout mybranch &&
239 | test_cmp somefile expect &&
240 | ! test_cmp notherfile expect &&
241 | test_cmp whatever long-expect
242 |
243 | )
244 | '
245 |
246 | test_expect_success 'clean-ignore with emoji in filenames' '
247 | test_create_repo clean-ignore &&
248 | (
249 | cd clean-ignore &&
250 | git fast-import --quiet <$DATA/emoji-repo &&
251 | git reset --hard &&
252 | $CONTRIB_DIR/clean-ignore --force &&
253 | printf ".gitignore\nfilename\n" >expect &&
254 | git ls-files >actual &&
255 | test_cmp expect actual
256 | )
257 | '
258 |
259 | test_done
260 |
--------------------------------------------------------------------------------
/t/t9391/commit_info.py:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | Please see the
5 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
6 | near the top of git-filter-repo
7 | """
8 |
9 | import re
10 | import datetime
11 |
12 | import git_filter_repo as fr
13 |
14 | def change_up_them_commits(commit, metadata):
15 | # Change the commit author
16 | if commit.author_name == b"Copy N. Paste":
17 | commit.author_name = b"Ima L. Oser"
18 | commit.author_email = b"aloser@my.corp"
19 |
20 | # Fix the author email
21 | commit.author_email = re.sub(b"@my.crp", b"@my.corp", commit.author_email)
22 |
23 | # Fix the committer date (bad timezone conversion in initial import)
24 | oldtime = fr.string_to_date(commit.committer_date)
25 | newtime = oldtime + datetime.timedelta(hours=-5)
26 | commit.committer_date = fr.date_to_string(newtime)
27 |
28 | # Fix the commit message
29 | commit.message = re.sub(b"Marketing is staffed with pansies", b"",
30 | commit.message)
31 |
32 | args = fr.FilteringOptions.parse_args(['--force'])
33 | filter = fr.RepoFilter(args, commit_callback = change_up_them_commits)
34 | filter.run()
35 |
--------------------------------------------------------------------------------
/t/t9391/create_fast_export_output.py:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | Please see the
5 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
6 | near the top of git-filter-repo.
7 | """
8 |
9 | import git_filter_repo as fr
10 | from git_filter_repo import Blob, Reset, FileChange, Commit, Tag, FixedTimeZone
11 | from git_filter_repo import Progress, Checkpoint
12 |
13 | from datetime import datetime, timedelta
14 |
15 | args = fr.FilteringOptions.default_options()
16 | out = fr.RepoFilter(args)
17 | out.importer_only()
18 |
19 | world = Blob(b"Hello")
20 | out.insert(world)
21 |
22 | bar = Blob(b"foo\n")
23 | out.insert(bar)
24 |
25 | master = Reset(b"refs/heads/master")
26 | out.insert(master)
27 |
28 | changes = [FileChange(b'M', b'world', world.id, mode=b"100644"),
29 | FileChange(b'M', b'bar', bar.id, mode=b"100644")]
30 | when = datetime(year=2005, month=4, day=7,
31 | hour=15, minute=16, second=10,
32 | tzinfo=FixedTimeZone(b"-0700"))
33 | when_string = fr.date_to_string(when)
34 | commit1 = Commit(b"refs/heads/master",
35 | b"A U Thor", b"au@thor.email", when_string,
36 | b"Com M. Iter", b"comm@iter.email", when_string,
37 | b"My first commit! Wooot!\n\nLonger description",
38 | changes,
39 | parents = [])
40 | out.insert(commit1)
41 |
42 | world = Blob(b"Hello\nHi")
43 | out.insert(world)
44 | world_link = Blob(b"world")
45 | out.insert(world_link)
46 |
47 | changes = [FileChange(b'M', b'world', world.id, mode=b"100644"),
48 | FileChange(b'M', b'planet', world_link.id, mode=b"120000")]
49 | when += timedelta(days=3, hours=4, minutes=6)
50 | when_string = fr.date_to_string(when)
51 | commit2 = Commit(b"refs/heads/master",
52 | b"A U Thor", b"au@thor.email", when_string,
53 | b"Com M. Iter", b"comm@iter.email", when_string,
54 | b"Make a symlink to world called planet, modify world",
55 | changes,
56 | parents = [commit1.id])
57 | out.insert(commit2)
58 |
59 | script = Blob(b"#!/bin/sh\n\necho Hello")
60 | out.insert(script)
61 | changes = [FileChange(b'M', b'runme', script.id, mode=b"100755"),
62 | FileChange(b'D', b'bar')]
63 | when_string = b"1234567890 -0700"
64 | commit3 = Commit(b"refs/heads/master",
65 | b"A U Thor", b"au@thor.email", when_string,
66 | b"Com M. Iter", b"comm@iter.email", when_string,
67 | b"Add runme script, remove bar",
68 | changes,
69 | parents = [commit2.id])
70 | out.insert(commit3)
71 |
72 | progress = Progress(b"Done with the master branch now...")
73 | out.insert(progress)
74 | checkpoint = Checkpoint()
75 | out.insert(checkpoint)
76 |
77 | devel = Reset(b"refs/heads/devel", commit1.id)
78 | out.insert(devel)
79 |
80 | world = Blob(b"Hello\nGoodbye")
81 | out.insert(world)
82 |
83 | changes = [FileChange(b'DELETEALL'),
84 | FileChange(b'M', b'world', world.id, mode=b"100644"),
85 | FileChange(b'M', b'bar', bar.id, mode=b"100644")]
86 | when = datetime(2006, 8, 17, tzinfo=FixedTimeZone(b"+0200"))
87 | when_string = fr.date_to_string(when)
88 | commit4 = Commit(b"refs/heads/devel",
89 | b"A U Thor", b"au@thor.email", when_string,
90 | b"Com M. Iter", b"comm@iter.email", when_string,
91 | b"Modify world",
92 | changes,
93 | parents = [commit1.id])
94 | out.insert(commit4)
95 |
96 | world = Blob(b"Hello\nHi\nGoodbye")
97 | out.insert(world)
98 | when = fr.string_to_date(commit3.author_date) + timedelta(days=47)
99 | when_string = fr.date_to_string(when)
100 | # git fast-import requires file changes to be listed in terms of differences
101 | # to the first parent. Thus, despite the fact that runme and planet have
102 | # not changed and bar was not modified in the devel side, we have to list them
103 | # all anyway.
104 | changes = [FileChange(b'M', b'world', world.id, mode=b"100644"),
105 | FileChange(b'D', b'bar'),
106 | FileChange(b'M', b'runme', script.id, mode=b"100755"),
107 | FileChange(b'M', b'planet', world_link.id, mode=b"120000")]
108 |
109 | commit5 = Commit(b"refs/heads/devel",
110 | b"A U Thor", b"au@thor.email", when_string,
111 | b"Com M. Iter", b"comm@iter.email", when_string,
112 | b"Merge branch 'master'\n",
113 | changes,
114 | parents = [commit4.id, commit3.id])
115 | out.insert(commit5)
116 |
117 |
118 | mytag = Tag(b"refs/tags/v1.0", commit5.id,
119 | b"His R. Highness", b"royalty@my.kingdom", when_string,
120 | b"I bequeath to my peons this royal software")
121 | out.insert(mytag)
122 | out.finish()
123 |
--------------------------------------------------------------------------------
/t/t9391/emoji-repo:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 8
5 | initial
6 |
7 | blob
8 | mark :2
9 | data 5
10 | lock
11 |
12 | blob
13 | mark :3
14 | data 11
15 | *.bak
16 | 🔒
17 |
18 | reset refs/heads/master
19 | commit refs/heads/master
20 | mark :4
21 | author Little O. Me <me@little.net> 1535228562 -0700
22 | committer Little O. Me <me@little.net> 1535228562 -0700
23 | data 10
24 | My commit
25 | M 100644 :1 filename
26 | M 100644 :2 🔒
27 | M 100644 :3 .gitignore
28 |
29 | done
30 |
--------------------------------------------------------------------------------
/t/t9391/erroneous.py:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | Please see the
5 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
6 | near the top of git-filter-repo
7 | """
8 |
9 | import git_filter_repo as fr
10 |
11 | def handle_tag(tag):
12 | print("Tagger: "+''.join(tag.tagger_name))
13 |
14 | args = fr.FilteringOptions.parse_args(['--force', '--tag-callback', 'pass'])
15 | filter = fr.RepoFilter(args, tag_callback = handle_tag)
16 | filter.run()
17 |
--------------------------------------------------------------------------------
/t/t9391/file_filter.py:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | Please see the
5 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
6 | near the top of git-filter-repo.
7 | """
8 |
9 | import sys
10 | import git_filter_repo as fr
11 |
12 | def drop_file_by_contents(blob, metadata):
13 | bad_file_contents = b'The launch code is 1-2-3-4.'
14 | if blob.data == bad_file_contents:
15 | blob.skip()
16 |
17 | def drop_files_by_name(commit, metadata):
18 | new_file_changes = []
19 | for change in commit.file_changes:
20 | if not change.filename.endswith(b'.doc'):
21 | new_file_changes.append(change)
22 | commit.file_changes = new_file_changes
23 |
24 | sys.argv.append('--force')
25 | args = fr.FilteringOptions.parse_args(sys.argv[1:])
26 |
27 | filter = fr.RepoFilter(args,
28 | blob_callback = drop_file_by_contents,
29 | commit_callback = drop_files_by_name)
30 | filter.run()
31 |
--------------------------------------------------------------------------------
/t/t9391/print_progress.py:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | Please see the
5 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
6 | near the top of git-filter-repo
7 | """
8 |
9 | import sys
10 | import git_filter_repo as fr
11 |
12 | if len(sys.argv) != 3:
13 | raise SystemExit("Syntax:\n %s SOURCE_REPO TARGET_REPO")
14 | source_repo = sys.argv[1].encode()
15 | target_repo = sys.argv[2].encode()
16 |
17 | total_objects = fr.GitUtils.get_total_objects(source_repo) # blobs+trees
18 | total_commits = fr.GitUtils.get_commit_count(source_repo)
19 | object_count = 0
20 | commit_count = 0
21 |
22 | def print_progress():
23 | global object_count, commit_count, total_objects, total_commits
24 | print("\rRewriting commits... %d/%d (%d objects)"
25 | % (commit_count, total_commits, object_count), end='')
26 |
27 | def my_blob_callback(blob, metadata):
28 | global object_count
29 | object_count += 1
30 | print_progress()
31 |
32 | def my_commit_callback(commit, metadata):
33 | global commit_count
34 | commit_count += 1
35 | print_progress()
36 |
37 | args = fr.FilteringOptions.parse_args(['--force', '--quiet'])
38 | filter = fr.RepoFilter(args,
39 | blob_callback = my_blob_callback,
40 | commit_callback = my_commit_callback)
41 | filter.run()
42 |
--------------------------------------------------------------------------------
/t/t9391/rename-master-to-develop.py:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | Please see the
5 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
6 | near the top of git-filter-repo.
7 | """
8 |
9 | import git_filter_repo as fr
10 |
11 | def my_commit_callback(commit, metadata):
12 | if commit.branch == b"refs/heads/master":
13 | commit.branch = b"refs/heads/develop"
14 |
15 | args = fr.FilteringOptions.default_options()
16 | args.force = True
17 | filter = fr.RepoFilter(args, commit_callback = my_commit_callback)
18 | filter.run()
19 |
--------------------------------------------------------------------------------
/t/t9391/splice_repos.py:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | Please see the
5 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
6 | near the top of git-filter-repo.
7 |
8 | Also, note that splicing repos may need some special care as fast-export
9 | only shows the files that changed relative to the first parent, so there
10 | may be gotchas if you are to splice near merge commits; this example does
11 | not try to handle any such special cases.
12 | """
13 |
14 | import re
15 | import sys
16 | import git_filter_repo as fr
17 |
18 | class InterleaveRepositories:
19 | def __init__(self, repo1, repo2, output_dir):
20 | self.repo1 = repo1
21 | self.repo2 = repo2
22 | self.output_dir = output_dir
23 |
24 | self.commit_map = {}
25 | self.last_commit = None
26 |
27 | def skip_reset(self, reset, metadata):
28 | reset.skip()
29 |
30 | def hold_commit(self, commit, metadata):
31 | commit.skip(new_id = commit.id)
32 | letter = re.match(b'Commit (.)', commit.message).group(1)
33 | self.commit_map[letter] = commit
34 |
35 | def weave_commit(self, commit, metadata):
36 | letter = re.match(b'Commit (.)', commit.message).group(1)
37 | prev_letter = bytes([ord(letter)-1])
38 |
39 | # Splice in any extra commits needed
40 | if prev_letter in self.commit_map:
41 | new_commit = self.commit_map[prev_letter]
42 | new_commit.dumped = 0
43 | new_commit.parents = [self.last_commit] if self.last_commit else []
44 | # direct_insertion=True to avoid weave_commit being called recursively
45 | # on the same commit
46 | self.out.insert(new_commit, direct_insertion = True)
47 | commit.parents = [new_commit.id]
48 |
49 | # Dump our commit now
50 | self.out.insert(commit, direct_insertion = True)
51 |
52 | # Make sure that commits that depended on new_commit.id will now depend
53 | # on commit.id
54 | if prev_letter in self.commit_map:
55 | self.last_commit = commit.id
56 | fr.record_id_rename(new_commit.id, commit.id)
57 |
58 | def run(self):
59 | blob = fr.Blob(b'public gpg key contents')
60 | tag = fr.Tag(b'gpg-pubkey', blob.id,
61 | b'Ima Tagger', b'ima@tagg.er', b'1136199845 +0300',
62 | b'Very important explanation and stuff')
63 |
64 | args = fr.FilteringOptions.parse_args(['--target', self.output_dir])
65 | out = fr.RepoFilter(args)
66 | out.importer_only()
67 | self.out = out
68 |
69 | i1args = fr.FilteringOptions.parse_args(['--source', self.repo1])
70 | i1 = fr.RepoFilter(i1args,
71 | reset_callback = self.skip_reset,
72 | commit_callback = self.hold_commit)
73 | i1.set_output(out)
74 | i1.run()
75 |
76 | i2args = fr.FilteringOptions.parse_args(['--source', self.repo2])
77 | i2 = fr.RepoFilter(i2args,
78 | commit_callback = self.weave_commit)
79 | i2.set_output(out)
80 | i2.run()
81 |
82 | out.insert(blob)
83 | out.insert(tag)
84 | out.finish()
85 |
86 | splicer = InterleaveRepositories(sys.argv[1], sys.argv[2], sys.argv[3])
87 | splicer.run()
88 |
--------------------------------------------------------------------------------
/t/t9391/strip-cvs-keywords.py:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | """
4 | Please see the
5 | ***** API BACKWARD COMPATIBILITY CAVEAT *****
6 | near the top of git-filter-repo.
7 | """
8 |
9 | import re
10 | import git_filter_repo as fr
11 |
12 | def strip_cvs_keywords(blob, metadata):
13 | # FIXME: Should first check if blob is a text file to avoid ruining
14 | # binaries. Could use python.magic here, or just output blob.data to
15 | # the unix 'file' command
16 | pattern = br'\$(Id|Date|Source|Header|CVSHeader|Author|Revision):.*\#39;
17 | replacement = br'$\1#39;
18 | blob.data = re.sub(pattern, replacement, blob.data)
19 |
20 | args = fr.FilteringOptions.parse_args(['--force'])
21 | filter = fr.RepoFilter(args, blob_callback = strip_cvs_keywords)
22 | filter.run()
23 |
--------------------------------------------------------------------------------
/t/t9391/unusual.py:
--------------------------------------------------------------------------------
1 | #!/usr/bin/env python3
2 |
3 | # Please: DO NOT USE THIS AS AN EXAMPLE.
4 | #
5 | # This file is NOT for demonstration of how to use git-filter-repo as a
6 | # libary; it exists to test corner cases or otherwise unusual inputs, and
7 | # to verify some invariants that git-filter-repo currently aims to maintain
8 | # (these invariants might be different in future versions of
9 | # git-filter-repo). As such, it reaches deep into the internals and does
10 | # weird things that you should probably avoid in your usage of
11 | # git-filter-repo. Any code in this testcase is much more likely to have
12 | # API breaks than other files in t9391.
13 |
14 | import collections
15 | import os
16 | import random
17 | import io
18 | import sys
19 | import textwrap
20 |
21 | import git_filter_repo as fr
22 |
23 | total_objects = {'common': 0, 'uncommon': 0}
24 | def track_everything(obj, *_ignored):
25 | if type(obj) == fr.Blob or type(obj) == fr.Commit:
26 | total_objects['common'] += 1
27 | else:
28 | total_objects['uncommon'] += 1
29 | if type(obj) == fr.Reset:
30 | def assert_not_reached(x): raise SystemExit("should have been skipped!")
31 | obj.dump = assert_not_reached
32 | obj.skip()
33 | if hasattr(obj, 'id') and type(obj) != fr.Tag:
34 | # The creation of myblob should cause objects in stream to get their ids
35 | # increased by 1; this shouldn't be depended upon as API by external
36 | # projects, I'm just verifying an invariant of the current code.
37 | assert fr._IDS._reverse_translation[obj.id] == [obj.id - 1]
38 |
39 | def handle_progress(progress):
40 | print(b"Decipher this: "+bytes(reversed(progress.message)))
41 | track_everything(progress)
42 |
43 | def handle_checkpoint(checkpoint_object):
44 | # Flip a coin; see if we want to pass the checkpoint through.
45 | if random.randint(0,1) == 0:
46 | checkpoint_object.dump(parser._output)
47 | track_everything(checkpoint_object)
48 |
49 | mystr = b'This is the contents of the blob'
50 | compare = b"Blob:\n blob\n mark :1\n data %d\n %s" % (len(mystr), mystr)
51 | # Next line's only purpose is testing code coverage of something that helps
52 | # debugging git-filter-repo; it is NOT something external folks should depend
53 | # upon.
54 | myblob = fr.Blob(mystr)
55 | assert bytes(myblob) == compare
56 | # Everyone should be using RepoFilter objects, not FastExportParser. But for
57 | # testing purposes...
58 | parser = fr.FastExportParser(blob_callback = track_everything,
59 | reset_callback = track_everything,
60 | commit_callback = track_everything,
61 | tag_callback = track_everything,
62 | progress_callback = handle_progress,
63 | checkpoint_callback = handle_checkpoint)
64 |
65 | parser.run(input = sys.stdin.detach(),
66 | output = open(os.devnull, 'bw'))
67 | # DO NOT depend upon or use _IDS directly you external script writers. I'm
68 | # only testing here for code coverage; the capacity exists to help debug
69 | # git-filter-repo itself, not for external folks to use.
70 | assert str(fr._IDS).startswith("Current count: 5")
71 | print("Found {} blobs/commits and {} other objects"
72 | .format(total_objects['common'], total_objects['uncommon']))
73 |
74 |
75 | stream = io.BytesIO(textwrap.dedent('''
76 | blob
77 | mark :1
78 | data 5
79 | hello
80 |
81 | commit refs/heads/A
82 | mark :2
83 | author Just Me <just@here.org> 1234567890 -0200
84 | committer Just Me <just@here.org> 1234567890 -0200
85 | data 2
86 | A
87 |
88 | commit refs/heads/B
89 | mark :3
90 | author Just Me <just@here.org> 1234567890 -0200
91 | committer Just Me <just@here.org> 1234567890 -0200
92 | data 2
93 | B
94 | from :2
95 | M 100644 :1 greeting
96 |
97 | reset refs/heads/B
98 | from :3
99 |
100 | commit refs/heads/C
101 | mark :4
102 | author Just Me <just@here.org> 1234567890 -0200
103 | committer Just Me <just@here.org> 1234567890 -0200
104 | data 2
105 | C
106 | from :3
107 | M 100644 :1 salutation
108 |
109 | '''[1:]).encode())
110 |
111 | counts = collections.Counter()
112 | def look_for_reset(obj, metadata):
113 | print("Processing {}".format(obj))
114 | counts[type(obj)] += 1
115 | if type(obj) == fr.Reset:
116 | assert obj.ref == b'refs/heads/B'
117 |
118 | # Use all kinds of internals that external scripts should NOT use and which
119 | # are likely to break in the future, just to verify a few invariants...
120 | args = fr.FilteringOptions.parse_args(['--stdin', '--dry-run',
121 | '--path', 'salutation'])
122 | filter = fr.RepoFilter(args,
123 | blob_callback = look_for_reset,
124 | reset_callback = look_for_reset,
125 | commit_callback = look_for_reset,
126 | tag_callback = look_for_reset)
127 | filter._input = stream
128 | filter._setup_output()
129 | filter._sanity_checks_handled = True
130 | filter.run()
131 | assert counts == collections.Counter({fr.Blob: 1, fr.Commit: 3, fr.Reset: 1})
132 |
--------------------------------------------------------------------------------
/t/t9392-filter-repo-python-callback.sh:
--------------------------------------------------------------------------------
1 | #!/bin/bash
2 |
3 | test_description='Usage of git-filter-repo with python callbacks'
4 | . ./test-lib.sh
5 |
6 | export PATH=$(dirname $TEST_DIRECTORY):$PATH # Put git-filter-repo in PATH
7 |
8 | setup()
9 | {
10 | git init $1 &&
11 | (
12 | cd $1 &&
13 | echo hello > world &&
14 | git add world &&
15 | test_tick &&
16 | git commit -m initial &&
17 | printf "The launch code is 1-2-3-4." > secret &&
18 | git add secret &&
19 | test_tick &&
20 | git commit -m "Sssh. Dont tell no one" &&
21 | echo A file that you cant trust > file.doc &&
22 | echo there >> world &&
23 | git add file.doc world &&
24 | test_tick &&
25 | printf "Random useless changes\n\nLet us be like the marketing group. Marketing is staffed with pansies" | git commit -F - &&
26 | echo Do not use a preposition to end a setence with > advice &&
27 | git add advice &&
28 | test_tick &&
29 | GIT_AUTHOR_NAME="Copy N. Paste" git commit -m "hypocrisy is fun" &&
30 | echo Avoid cliches like the plague >> advice &&
31 | test_tick &&
32 | GIT_AUTHOR_EMAIL="foo@my.crp" git commit -m "it is still fun" advice &&
33 | echo " \$Id: A bunch of junkquot; > foobar.c &&
34 | git add foobar.c &&
35 | test_tick &&
36 | git commit -m "Brain damage" &&
37 |
38 | git tag v1.0 HEAD~3 &&
39 | git tag -a -m 'Super duper snazzy release' v2.0 HEAD~1 &&
40 | git branch testing master &&
41 |
42 | # Make it look like a fresh clone (avoid need for --force)
43 | git gc &&
44 | git remote add origin . &&
45 | git update-ref refs/remotes/origin/master refs/heads/master
46 | git update-ref refs/remotes/origin/testing refs/heads/testing
47 | )
48 | }
49 |
50 | test_expect_success '--filename-callback' '
51 | setup filename-callback &&
52 | (
53 | cd filename-callback &&
54 | git filter-repo --filename-callback "return None if filename.endswith(b\".doc\") else b\"src/\"+filename" &&
55 | git log --format=%n --name-only | sort | uniq | grep -v ^$ > f &&
56 | ! grep file.doc f &&
57 | COMPARE=$(wc -l <f) &&
58 | grep src/ f >filtered_f &&
59 | test_line_count = $COMPARE filtered_f
60 | )
61 | '
62 |
63 | test_expect_success '--file-info-callback acting like --filename-callback' '
64 | setup fileinfo-as-filename-callback &&
65 | (
66 | cd fileinfo-as-filename-callback &&
67 | git filter-repo --file-info-callback "return (None if filename.endswith(b\".doc\") else b\"src/\"+filename, mode, blob_id)" &&
68 | git log --format=%n --name-only | sort | uniq | grep -v ^$ > f &&
69 | ! grep file.doc f &&
70 | COMPARE=$(wc -l <f) &&
71 | grep src/ f >filtered_f &&
72 | test_line_count = $COMPARE filtered_f
73 | )
74 | '
75 |
76 | test_expect_success '--message-callback' '
77 | setup message-callback &&
78 | (
79 | cd message-callback &&
80 | git filter-repo --message-callback "return b\"TLDR: \"+message[0:5]" &&
81 | git log --format=%s >log-messages &&
82 | grep TLDR:...... log-messages >modified-messages &&
83 | test_line_count = 6 modified-messages
84 | )
85 | '
86 |
87 | test_expect_success '--name-callback' '
88 | setup name-callback &&
89 | (
90 | cd name-callback &&
91 | git filter-repo --name-callback "return name.replace(b\"N.\", b\"And\")" &&
92 | git log --format=%an >log-person-names &&
93 | grep Copy.And.Paste log-person-names
94 | )
95 | '
96 |
97 | test_expect_success '--email-callback' '
98 | setup email-callback &&
99 | (
100 | cd email-callback &&
101 | git filter-repo --email-callback "return email.replace(b\".com\", b\".org\")" &&
102 | git log --format=%ae%n%ce >log-emails &&
103 | ! grep .com log-emails &&
104 | grep .org log-emails
105 | )
106 | '
107 |
108 | test_expect_success '--refname-callback' '
109 | setup refname-callback &&
110 | (
111 | cd refname-callback &&
112 | git filter-repo --refname-callback "
113 | dir,path = os.path.split(refname)
114 | return dir+b\"/prefix-\"+path" &&
115 | git show-ref | grep refs/heads/prefix-master &&
116 | git show-ref | grep refs/tags/prefix-v1.0 &&
117 | git show-ref | grep refs/tags/prefix-v2.0
118 | )
119 | '
120 |
121 | test_expect_success '--refname-callback sanity check' '
122 | setup refname-sanity-check &&
123 | (
124 | cd refname-sanity-check &&
125 |
126 | test_must_fail git filter-repo --refname-callback "return re.sub(b\"tags\", b\"other-tags\", refname)" 2>../err &&
127 | test_i18ngrep "fast-import requires tags to be in refs/tags/ namespace" ../err &&
128 | rm ../err
129 | )
130 | '
131 |
132 | test_expect_success '--blob-callback' '
133 | setup blob-callback &&
134 | (
135 | cd blob-callback &&
136 | git log --format=%n --name-only | sort | uniq | grep -v ^$ > f &&
137 | test_line_count = 5 f &&
138 | rm f &&
139 | git filter-repo --blob-callback "if len(blob.data) > 25: blob.skip()" &&
140 | git log --format=%n --name-only | sort | uniq | grep -v ^$ > f &&
141 | test_line_count = 2 f
142 | )
143 | '
144 |
145 | test_expect_success '--file-info-callback acting like --blob-callback' '
146 | setup fileinfo-as-blob-callback &&
147 | (
148 | cd fileinfo-as-blob-callback &&
149 | git log --format=%n --name-only | sort | uniq | grep -v ^$ > f &&
150 | test_line_count = 5 f &&
151 | rm f &&
152 | git filter-repo --file-info-callback "
153 | size = value.get_size_by_identifier(blob_id)
154 | return (None if size > 25 else filename, mode, blob_id)" &&
155 | git log --format=%n --name-only | sort | uniq | grep -v ^$ > f &&
156 | test_line_count = 2 f
157 | )
158 | '
159 |
160 | test_expect_success '--commit-callback' '
161 | setup commit-callback &&
162 | (
163 | cd commit-callback &&
164 | git filter-repo --commit-callback "
165 | commit.committer_name = commit.author_name
166 | commit.committer_email = commit.author_email
167 | commit.committer_date = commit.author_date
168 | for change in commit.file_changes:
169 | change.mode = b\"100755\"
170 | " &&
171 | git log --format=%ae%n%ce >log-emails &&
172 | ! grep committer@example.com log-emails &&
173 | git log --raw | grep ^: >file-changes &&
174 | ! grep 100644 file-changes &&
175 | grep 100755 file-changes
176 | )
177 | '
178 |
179 | test_expect_success '--tag-callback' '
180 | setup tag-callback &&
181 | (
182 | cd tag-callback &&
183 | git filter-repo --tag-callback "
184 | tag.tagger_name = b\"Dr. \"+tag.tagger_name
185 | tag.message = b\"Awesome sauce \"+tag.message
186 | " &&
187 | git cat-file -p v2.0 | grep ^tagger.Dr\\. &&
188 | git cat-file -p v2.0 | grep ^Awesome.sauce.Super
189 | )
190 | '
191 |
192 | test_expect_success '--reset-callback' '
193 | setup reset-callback &&
194 | (
195 | cd reset-callback &&
196 | git filter-repo --reset-callback "reset.from_ref = 3" &&
197 | test $(git rev-parse testing) = $(git rev-parse master~3)
198 | )
199 | '
200 |
201 | test_expect_success 'callback has return statement sanity check' '
202 | setup callback_return_sanity &&
203 | (
204 | cd callback_return_sanity &&
205 |
206 | test_must_fail git filter-repo --filename-callback "filename + b\".txt\"" 2>../err&&
207 | test_i18ngrep "Error: --filename-callback should have a return statement" ../err &&
208 | rm ../err
209 | )
210 | '
211 |
212 | test_expect_success 'Callback read from a file' '
213 | setup name-callback-from-file &&
214 | (
215 | cd name-callback-from-file &&
216 | echo "return name.replace(b\"N.\", b\"And\")" >../name-func &&
217 | git filter-repo --name-callback ../name-func &&
218 | git log --format=%an >log-person-names &&
219 | grep Copy.And.Paste log-person-names
220 | )
221 | '
222 |
223 | test_expect_success 'Filtering a blob to make it match previous version' '
224 | test_create_repo remove_unique_bits_of_blob &&
225 | (
226 | cd remove_unique_bits_of_blob &&
227 |
228 | test_write_lines foo baz >metasyntactic_names &&
229 | git add metasyntactic_names &&
230 | git commit -m init &&
231 |
232 | test_write_lines foo bar baz >metasyntactic_names &&
233 | git add metasyntactic_names &&
234 | git commit -m second &&
235 |
236 | git filter-repo --force --blob-callback "blob.data = blob.data.replace(b\"\\nbar\", b\"\")"
237 |
238 | echo 1 >expect &&
239 | git rev-list --count HEAD >actual &&
240 | test_cmp expect actual
241 | )
242 | '
243 |
244 | test_expect_success 'tweaking just a tag' '
245 | test_create_repo tweaking_just_a_tag &&
246 | (
247 | cd tweaking_just_a_tag &&
248 |
249 | test_commit foo &&
250 | git tag -a -m "Here is a tag" mytag &&
251 |
252 | git filter-repo --force --refs mytag ^mytag^{commit} --name-callback "return name.replace(b\"Mitter\", b\"L D\")" &&
253 |
254 | git cat-file -p mytag | grep C.O.L.D
255 | )
256 | '
257 |
258 | test_expect_success '--file-info-callback messing with history' '
259 | setup messing_with_files &&
260 | (
261 | cd messing_with_files &&
262 |
263 | echo "1-2-3-4==>1-2-3-4-5" >replacement &&
264 | # Trying to count the levels of backslash escaping is not fun.
265 | echo "regex:\\\$[^\$]*\\\$==>cvs is lame" >>replacement &&
266 | git filter-repo --force --file-info-callback "
267 | size = value.get_size_by_identifier(blob_id)
268 | contents = value.get_contents_by_identifier(blob_id)
269 | if not value.is_binary(contents):
270 | contents = value.apply_replace_text(contents)
271 | if contents[-1] != 10:
272 | contents += bytes([10])
273 | blob_id = value.insert_file_with_contents(contents)
274 | newname = bytes(reversed(filename))
275 | if size == 27 and len(contents) == 27:
276 | newname = None
277 | return (newname, mode, blob_id)
278 | " --replace-text replacement &&
279 |
280 | cat <<-EOF >expect &&
281 | c.raboof
282 | dlrow
283 | ecivda
284 | terces
285 | EOF
286 |
287 | git ls-files >actual &&
288 | test_cmp expect actual &&
289 |
290 | echo "The launch code is 1-2-3-4-5." >expect &&
291 | test_cmp expect terces &&
292 |
293 | echo " cvs is lame" >expect &&
294 | test_cmp expect c.raboof
295 | )
296 | '
297 |
298 | test_expect_success '--file-info-callback and deletes and drops' '
299 | setup file_info_deletes_drops &&
300 | (
301 | cd file_info_deletes_drops &&
302 |
303 | git rm file.doc &&
304 | git commit -m "Nuke doc file" &&
305 |
306 | git filter-repo --force --file-info-callback "
307 | size = value.get_size_by_identifier(blob_id)
308 | (newname, newmode) = (filename, mode)
309 | if filename == b\"world\" and size == 12:
310 | newname = None
311 | if filename == b\"advice\" and size == 77:
312 | newmode = None
313 | return (newname, newmode, blob_id)
314 | "
315 |
316 | cat <<-EOF >expect &&
317 | foobar.c
318 | secret
319 | world
320 | EOF
321 |
322 | echo 1 >expect &&
323 | git rev-list --count HEAD -- world >actual &&
324 | test_cmp expect actual &&
325 |
326 | echo 2 >expect &&
327 | git rev-list --count HEAD -- advice >actual &&
328 | test_cmp expect actual &&
329 |
330 | echo hello >expect &&
331 | test_cmp expect world
332 | )
333 | '
334 |
335 | test_lazy_prereq UNIX2DOS '
336 | unix2dos -h
337 | test $? -ne 127
338 | '
339 |
340 | test_expect_success UNIX2DOS '--file-info-callback acting like lint-history' '
341 | setup lint_history_replacement &&
342 | (
343 | cd lint_history_replacement &&
344 | git ls-files -s | grep -v file.doc >expect &&
345 |
346 | git filter-repo --force --file-info-callback "
347 | if not filename.endswith(b\".doc\"):
348 | return (filename, mode, blob_id)
349 |
350 | if blob_id in value.data:
351 | return (filename, mode, value.data[blob_id])
352 |
353 | contents = value.get_contents_by_identifier(blob_id)
354 | tmpfile = os.path.basename(filename)
355 | with open(tmpfile, \"wb\") as f:
356 | f.write(contents)
357 | subprocess.check_call([\"unix2dos\", filename])
358 | with open(filename, \"rb\") as f:
359 | contents = f.read()
360 | new_blob_id = value.insert_file_with_contents(contents)
361 |
362 | value.data[blob_id] = new_blob_id
363 | return (filename, mode, new_blob_id)
364 | " &&
365 |
366 | git ls-files -s | grep -v file.doc >actual &&
367 | test_cmp expect actual &&
368 |
369 | printf "A file that you cant trust\r\n" >expect &&
370 | test_cmp expect file.doc
371 | )
372 | '
373 |
374 | test_done
375 |
--------------------------------------------------------------------------------
/t/t9393/lfs:
--------------------------------------------------------------------------------
1 | feature done
2 | # Simple repo with a few files, some of them lfs. Note that the lfs object
3 | # ids and the original-oid directives are very fake, but make it easy to
4 | # track things.
5 | blob
6 | mark :1
7 | original-oid 0000000000000000000000000000000000000001
8 | data 39
9 | L* filter=lfs diff=lfs merge=lfs -text
10 |
11 | blob
12 | mark :2
13 | original-oid 0000000000000000000000000000000000000002
14 | data 126
15 | version https://git-lfs.github.com/spec/v1
16 | oid sha256:aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
17 | size 1
18 |
19 | blob
20 | mark :3
21 | original-oid 0000000000000000000000000000000000000003
22 | data 126
23 | version https://git-lfs.github.com/spec/v1
24 | oid sha256:bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb
25 | size 2
26 |
27 | blob
28 | mark :4
29 | original-oid 0000000000000000000000000000000000000004
30 | data 126
31 | version https://git-lfs.github.com/spec/v1
32 | oid sha256:cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc
33 | size 3
34 |
35 | blob
36 | mark :5
37 | original-oid 0000000000000000000000000000000000000005
38 | data 126
39 | version https://git-lfs.github.com/spec/v1
40 | oid sha256:dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd
41 | size 4
42 |
43 | blob
44 | mark :6
45 | original-oid 0000000000000000000000000000000000000006
46 | data 6
47 | stuff
48 |
49 | blob
50 | mark :7
51 | original-oid 0000000000000000000000000000000000000007
52 | data 1040
53 | 0000000000000000000000000000000000000000000000000000000000000000
54 | 1111111111111111111111111111111111111111111111111111111111111111
55 | 2222222222222222222222222222222222222222222222222222222222222222
56 | 3333333333333333333333333333333333333333333333333333333333333333
57 | 4444444444444444444444444444444444444444444444444444444444444444
58 | 5555555555555555555555555555555555555555555555555555555555555555
59 | 6666666666666666666666666666666666666666666666666666666666666666
60 | 7777777777777777777777777777777777777777777777777777777777777777
61 | 8888888888888888888888888888888888888888888888888888888888888888
62 | 9999999999999999999999999999999999999999999999999999999999999999
63 | aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
64 | bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb
65 | cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc
66 | dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd
67 | eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee
68 | ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff
69 |
70 | reset refs/heads/main
71 | commit refs/heads/main
72 | mark :10
73 | original-oid 0000000000000000000000000000000000000010
74 | author Little O. Me <me@little.net> 1234567890 -0700
75 | committer Little O. Me <me@little.net> 1234567890 -0700
76 | data 13
77 | First commit
78 | M 100644 :1 .gitattributes
79 | M 100644 :4 LB
80 |
81 | commit refs/heads/main
82 | mark :11
83 | original-oid 0000000000000000000000000000000000000011
84 | author Little O. Me <me@little.net> 1234567891 -0700
85 | committer Little O. Me <me@little.net> 1234567891 -0700
86 | data 14
87 | Second commit
88 | M 100644 :2 LA
89 |
90 | commit refs/heads/main
91 | mark :12
92 | original-oid 0000000000000000000000000000000000000012
93 | author Little O. Me <me@little.net> 1234567892 -0700
94 | committer Little O. Me <me@little.net> 1234567892 -0700
95 | data 13
96 | Third commit
97 | M 100644 :6 Z
98 |
99 | commit refs/heads/main
100 | mark :13
101 | original-oid 0000000000000000000000000000000000000013
102 | author Little O. Me <me@little.net> 1234567893 -0700
103 | committer Little O. Me <me@little.net> 1234567893 -0700
104 | data 14
105 | Fourth commit
106 | M 100644 :4 LC
107 |
108 | commit refs/heads/main
109 | mark :14
110 | original-oid 0000000000000000000000000000000000000014
111 | author Little O. Me <me@little.net> 1234567894 -0700
112 | committer Little O. Me <me@little.net> 1234567894 -0700
113 | data 13
114 | Fifth commit
115 | M 100644 :3 LA
116 |
117 | commit refs/heads/main
118 | mark :15
119 | original-oid 0000000000000000000000000000000000000015
120 | author Little O. Me <me@little.net> 1234567895 -0700
121 | committer Little O. Me <me@little.net> 1234567895 -0700
122 | data 13
123 | Sixth commit
124 | M 100644 :7 Y
125 | D Z
126 |
127 | commit refs/heads/main
128 | mark :16
129 | original-oid 0000000000000000000000000000000000000016
130 | author Little O. Me <me@little.net> 1234567896 -0700
131 | committer Little O. Me <me@little.net> 1234567896 -0700
132 | data 15
133 | Seventh commit
134 | M 100644 :5 LD
135 |
136 | commit refs/heads/main
137 | mark :17
138 | original-oid 0000000000000000000000000000000000000017
139 | author Little O. Me <me@little.net> 1234567897 -0700
140 | committer Little O. Me <me@little.net> 1234567897 -0700
141 | data 14
142 | Eighth commit
143 | M 100644 :2 LA
144 |
145 | done
146 |
--------------------------------------------------------------------------------
/t/t9393/simple:
--------------------------------------------------------------------------------
1 | feature done
2 | # Simple repo with a few files, and two branches with no common history.
3 | # Note that the original-oid directives are very fake, but make it easy to
4 | # track things.
5 | blob
6 | mark :1
7 | original-oid 0000000000000000000000000000000000000001
8 | data 16
9 | file 1 contents
10 |
11 | blob
12 | mark :2
13 | original-oid 0000000000000000000000000000000000000002
14 | data 16
15 | file 2 contents
16 |
17 | blob
18 | mark :3
19 | original-oid 0000000000000000000000000000000000000003
20 | data 16
21 | file 3 contents
22 |
23 | blob
24 | mark :4
25 | original-oid 0000000000000000000000000000000000000004
26 | data 16
27 | file 4 contents
28 |
29 | reset refs/heads/orphan-me
30 | commit refs/heads/orphan-me
31 | mark :5
32 | original-oid 0000000000000000000000000000000000000009
33 | author Little O. Me <me@little.net> 1535228562 -0700
34 | committer Little O. Me <me@little.net> 1535228562 -0700
35 | data 8
36 | Initial
37 | M 100644 :1 nuke-me
38 |
39 | commit refs/heads/orphan-me
40 | mark :6
41 | original-oid 000000000000000000000000000000000000000A
42 | author Little 'ol Me <me@laptop.(none)> 1535229544 -0700
43 | committer Little 'ol Me <me@laptop.(none)> 1535229544 -0700
44 | data 9
45 | Tweak it
46 | from :5
47 | M 100644 :4 nuke-me
48 |
49 | reset refs/heads/master
50 | commit refs/heads/master
51 | mark :7
52 | original-oid 000000000000000000000000000000000000000B
53 | author Little O. Me <me@machine52.little.net> 1535229523 -0700
54 | committer Little O. Me <me@machine52.little.net> 1535229523 -0700
55 | data 15
56 | Initial commit
57 | M 100644 :1 fileA
58 |
59 | commit refs/heads/master
60 | mark :8
61 | original-oid 000000000000000000000000000000000000000C
62 | author Lit.e Me <me@fire.com> 1535229559 -0700
63 | committer Lit.e Me <me@fire.com> 1535229580 -0700
64 | data 10
65 | Add fileB
66 | from :7
67 | M 100644 :2 fileB
68 |
69 | commit refs/heads/master
70 | mark :9
71 | original-oid 000000000000000000000000000000000000000D
72 | author Little Me <me@bigcompany.com> 1535229601 -0700
73 | committer Little Me <me@bigcompany.com> 1535229601 -0700
74 | data 10
75 | Add fileC
76 | from :8
77 | M 100644 :3 fileC
78 |
79 | commit refs/heads/master
80 | mark :10
81 | original-oid 000000000000000000000000000000000000000E
82 | author Little Me <me@bigcompany.com> 1535229618 -0700
83 | committer Little Me <me@bigcompany.com> 1535229618 -0700
84 | data 10
85 | Add fileD
86 | from :9
87 | M 100644 :4 fileD
88 |
89 | tag v1.0
90 | from :10
91 | original-oid 000000000000000000000000000000000000000F
92 | tagger Little John <second@merry.men> 1535229637 -0700
93 | data 5
94 | v1.0
95 |
96 | done
97 |
--------------------------------------------------------------------------------
/t/t9394/date-order:
--------------------------------------------------------------------------------
1 | feature done
2 | blob
3 | mark :1
4 | data 8
5 | initial
6 |
7 | reset refs/heads/master
8 | commit refs/heads/master
9 | mark :2
10 | author Little O. Me <me@little.net> 1535228562 -0700
11 | committer Little O. Me <me@little.net> 1535228562 -0700
12 | data 8
13 | Initial
14 | M 100644 :1 filename
15 |
16 | commit refs/heads/master
17 | mark :3
18 | author Little Me <me@bigcompany.com> 1535229601 -0700
19 | committer Little Me <me@bigcompany.com> 1535229601 -0700
20 | data 2
21 | A
22 | from :2
23 |
24 | commit refs/heads/master
25 | mark :4
26 | author Little Me <me@bigcompany.com> 1535229602 -0700
27 | committer Little Me <me@bigcompany.com> 1535229602 -0700
28 | data 2
29 | B
30 | from :2
31 |
32 | commit refs/heads/master
33 | mark :5
34 | author Little Me <me@bigcompany.com> 1535229603 -0700
35 | committer Little Me <me@bigcompany.com> 1535229603 -0700
36 | data 2
37 | C
38 | from :3
39 |
40 | commit refs/heads/master
41 | mark :6
42 | author Little Me <me@bigcompany.com> 1535229604 -0700
43 | committer Little Me <me@bigcompany.com> 1535229604 -0700
44 | data 2
45 | D
46 | from :4
47 |
48 | commit refs/heads/master
49 | mark :7
50 | author Little Me <me@bigcompany.com> 1535229605 -0700
51 | committer Little Me <me@bigcompany.com> 1535229605 -0700
52 | data 6
53 | merge
54 | from :5
55 | merge :6
56 |
57 | done
58 |
--------------------------------------------------------------------------------