87 | 1、转换为 HTML 类型
55 | *2、需要将文件中的图片下载到本地，且替换图片地址
56 | *3、处理代码块，因为 markdown 转换时没有给代码块加 ```
57 | */ 58 | protected String convertHtml(MarkDown markDown, Document document) { 59 | if ((markDown.getImagePath() != null && !"".equals(markDown.getImagePath())) && 60 | (markDown.getImageUrl() != null && !"".equals(markDown.getImageUrl()))) { 61 | 62 | // 处理图片 63 | try { 64 | handleImg(markDown, document); 65 | } catch (InterruptedException e) { 66 | log.info("图片处理异常."); 67 | e.printStackTrace(); 68 | } 69 | } else { 70 | log.info("没有图片需要下载.图片处理结束."); 71 | } 72 | 73 | return document.html(); 74 | } 75 | 76 | /** 77 | * 处理代码块，主要是在代码块中的前面增加 ````
` 即可 78 | * 79 | * @param document 80 | */ 81 | protected void handlePre(Document document) { 82 | 83 | } 84 | 85 | /** 86 | * 处理图片 87 | * 88 | * @param markDown 89 | * @param document 90 | */ 91 | private void handleImg(MarkDown markDown, Document document) throws InterruptedException { 92 | // 获取所有的 img 标签 93 | Elements elements = document.getElementsByTag("img"); 94 | 95 | if (elements != null && elements.size() > 0) { 96 | 97 | // 新建文件 98 | File pathFile = new File(markDown.getImagePath()); 99 | 100 | if (!pathFile.exists() && !pathFile.mkdirs()) { 101 | throw new RuntimeException("新建目录失败..."); 102 | } 103 | 104 | doHandleImg(elements, markDown); 105 | } 106 | } 107 | 108 | 109 | public void doHandleImg(Elements elements, MarkDown markDown) throws InterruptedException { 110 | String id = markDown.getId(); 111 | 112 | 113 | 114 | 115 | long l = System.currentTimeMillis(); 116 | CountDownLatch cdl = new CountDownLatch(elements.size()); 117 | Semaphore semaphore = new Semaphore(15); 118 | ExecutorService threadPool = ThreadPoolConfig.getThreadPool(); 119 | AtomicReferenceIt is recommended that you saveToFile this class if it is going to be reused for better performance. This class 36 | * is thread-safe, but can only process a single document concurrently.
37 | * 38 | *Usage:
39 | * 40 | *Basic usage involves instantiating this class with a specific set of options, and calling one of the 41 | * {@code convert*} methods on some form of input.
42 | * 43 | *Examples:
44 | * 45 | *
46 | * // Create a generic remark that converts to pure-Markdown spec.
47 | * Remark remark = new Remark();
48 | * String cleanedUp = remark.convertFragment(inputString);
49 | *
50 | * // Create a remark that converts to pegdown with all extensions enabled.
51 | * Remark pegdownAll = new Remark(Options.pegdownAllExtensions());
52 | * cleanedUp = pegdownAll.convert(new URL("http://www.example.com"), 15000);
53 | *
54 | * // stream the conversion
55 | * pegdownAll.withStream(System.out).convert(new URL("http://www.overzealous.com"), 15000);
56 | *
57 | *
58 | *
59 | * @author Phil DeJarnett
60 | */
61 | public class Remark {
62 | private final Cleaner cleaner;
63 | @SuppressWarnings({"FieldCanBeLocal", "UnusedDeclaration"})
64 | private final Options options;
65 | private final DocumentConverter converter;
66 | private final ReentrantLock converterLock = new ReentrantLock();
67 | private boolean cleanedHtmlEchoed = false;
68 |
69 | /**
70 | * Creates a default, pure Markdown-compatible Remark instance.
71 | */
72 | public Remark() {
73 | this(Options.markdown());
74 | }
75 |
76 | /**
77 | * Creates a Remark instance with the specified options.
78 | *
79 | * @param options Specified options to use on this instance. See the docs for the Options class for common options sets.
80 | */
81 | public Remark(Options options) {
82 | this.options = options.getCopy();
83 | Whitelist whitelist = Whitelist.basicWithImages()
84 | .addTags("div",
85 | "h1", "h2", "h3", "h4", "h5", "h6",
86 | "table", "tbody", "td", "tfoot", "th", "thead", "tr",
87 | "hr",
88 | "span", "font")
89 | .addAttributes("th", "colspan", "align", "style")
90 | .addAttributes("td", "colspan", "align", "style")
91 | .addAttributes(":all", "title", "style");
92 | if(options.preserveRelativeLinks) {
93 | whitelist.preserveRelativeLinks(true);
94 | }
95 | if(options.abbreviations) {
96 | whitelist.addTags("abbr", "acronym");
97 | }
98 | if(options.headerIds) {
99 | for(int i=1; i<=6; i++) {
100 | whitelist.addAttributes("h"+i, "id");
101 | }
102 | }
103 | for(final IgnoredHtmlElement el : options.getIgnoredHtmlElements()) {
104 | whitelist.addTags(el.getTagName());
105 | if(!el.getAttributes().isEmpty()) {
106 | whitelist.addAttributes(el.getTagName(), el.getAttributes().toArray(new String[el.getAttributes().size()]));
107 | }
108 | }
109 | cleaner = new Cleaner(whitelist);
110 |
111 | if(options.getTables().isLeftAsHtml()) {
112 | // we need to allow the table nodes to be ignored
113 | // since they are automatically ignored recursively, this is the only node we worry about.
114 | options.getIgnoredHtmlElements().add(IgnoredHtmlElement.create("table"));
115 | }
116 |
117 | converter = new DocumentConverter(options);
118 | }
119 |
120 | /**
121 | * Provides access to the DocumentConverter for customization.
122 | *
123 | * @return the configured DocumentConverter.
124 | */
125 | @SuppressWarnings({"UnusedDeclaration"})
126 | public DocumentConverter getConverter() {
127 | return converter;
128 | }
129 |
130 | /**
131 | * Returns true if the cleaned HTML document is echoed to {@code System.out}.
132 | * @return true if the cleaned HTML document is echoed
133 | */
134 | @SuppressWarnings({"UnusedDeclaration"})
135 | public boolean isCleanedHtmlEchoed() {
136 | return cleanedHtmlEchoed;
137 | }
138 |
139 | /**
140 | * To see the cleaned and processed HTML document, set this to true. It will
141 | * be rendered to {@code System.out} for debugging purposes.
142 | * @param cleanedHtmlEchoed true to echo out the cleaned HTML document
143 | */
144 | public void setCleanedHtmlEchoed(boolean cleanedHtmlEchoed) {
145 | this.cleanedHtmlEchoed = cleanedHtmlEchoed;
146 | }
147 |
148 | /**
149 | * This class is used to handle conversions that convert directly to streams.
150 | */
151 | private final class StreamRemark extends Remark {
152 | private final Remark remark;
153 | private final Writer writer;
154 | private final OutputStream os;
155 |
156 | private StreamRemark(Remark remark, Writer writer) {
157 | this.remark = remark;
158 | this.writer = writer;
159 | this.os = null;
160 | }
161 | private StreamRemark(Remark remark, OutputStream out) {
162 | this.remark = remark;
163 | this.writer = null;
164 | this.os = out;
165 | }
166 |
167 | @Override
168 | public Remark withWriter(Writer writer) {
169 | return remark.withWriter(writer);
170 | }
171 |
172 | @Override
173 | public Remark withOutputStream(OutputStream os) {
174 | return remark.withOutputStream(os);
175 | }
176 |
177 | @Override
178 | public String convert(Document doc) {
179 | return remark.processConvert(doc, writer, os);
180 | }
181 | }
182 |
183 | /**
184 | * Use this method in a chain to handle streaming the output to a Writer.
185 | * The returned class can be saved for repeated writing to the same streams.
186 | *
187 | * Note: The convert methods on the returned class will always return {@code null}.
188 | * 189 | *Note: It is up to the calling class to handle closing the writer!
190 | * 191 | *Example:
192 | * 193 | *{@code new Remark(options).withWriter(myWiter).convert(htmlText);}194 | * 195 | * @param writer Writer to receive the converted output 196 | * @return A Remark that writes to streams. 197 | */ 198 | @SuppressWarnings({"WeakerAccess"}) 199 | public synchronized Remark withWriter(Writer writer) { 200 | if(writer == null) { 201 | throw new NullPointerException("Writer cannot be null."); 202 | } 203 | return new StreamRemark(this, writer); 204 | } 205 | 206 | /** 207 | * Use this method in a chain to handle streaming the output to an OutputStream. 208 | * The returned class can be saved for repeated writing to the same streams. 209 | * 210 | *
Note: The convert methods on the returned class will always return {@code null}.
211 | * 212 | *Note: It is up to the calling class to handle closing the stream!
213 | * 214 | *Example:
215 | * 216 | *{@code new Remark(options).withOutputStream(myOut).convert(htmlText);}217 | * 218 | * @param os OutputStream to receive the converted output 219 | * @return A Remark that writes to streams. 220 | */ 221 | @SuppressWarnings({"WeakerAccess"}) 222 | public synchronized Remark withOutputStream(OutputStream os) { 223 | if(os == null) { 224 | throw new NullPointerException("OutputStream cannot be null."); 225 | } 226 | return new StreamRemark(this, os); 227 | } 228 | 229 | /** 230 | * Converts an HTML document retrieved from a URL to Markdown. 231 | * @param url URL to connect to. 232 | * @param timeoutMillis Maximum time to wait before giving up on the connection. 233 | * @return Markdown text. 234 | * @throws IOException If an error occurs while retrieving the document. 235 | * @see org.jsoup.Jsoup#parse(URL, int) 236 | */ 237 | public String convert(URL url, int timeoutMillis) throws IOException { 238 | Document doc = Jsoup.parse(url, timeoutMillis); 239 | return convert(doc); 240 | } 241 | 242 | 243 | /** 244 | * Converts an HTML file to Markdown. 245 | * @param file The file to load. 246 | * @return Markdown text. 247 | * @throws IOException If an error occurs while loading the file. 248 | * @see org.jsoup.Jsoup#parse(File, String, String) 249 | */ 250 | public String convert(File file) throws IOException { 251 | return convert(file, null); 252 | } 253 | 254 | 255 | /** 256 | * Converts an HTML file to Markdown. 257 | * @param file The file to load. 258 | * @param charset The charset of the file (if not specified and not UTF-8). Set to {@code null} to determine from {@code http-equiv} meta tag, if present, or fall back to {@code UTF-8} (which is often safe to do). 259 | * @return Markdown text. 260 | * @throws IOException If an error occurs while loading the file. 261 | * @see org.jsoup.Jsoup#parse(File, String, String) 262 | */ 263 | @SuppressWarnings({"WeakerAccess", "SameParameterValue"}) 264 | public String convert(File file, String charset) throws IOException { 265 | return convert(file, charset, ""); 266 | } 267 | 268 | 269 | /** 270 | * Converts an HTML file to Markdown. 271 | * @param file The file to load. 272 | * @param charset The charset of the file (if not specified and not UTF-8). Set to {@code null} to determine from {@code http-equiv} meta tag, if present, or fall back to {@code UTF-8} (which is often safe to do). 273 | * @param baseUri The base URI for resolving relative links. 274 | * @return Markdown text. 275 | * @throws IOException If an error occurs while loading the file. 276 | * @see org.jsoup.Jsoup#parse(File, String, String) 277 | */ 278 | public String convert(File file, String charset, String baseUri) throws IOException { 279 | Document doc = Jsoup.parse(file, charset, baseUri); 280 | return convert(doc); 281 | } 282 | 283 | 284 | /** 285 | * Converts HTML in memory to Markdown. 286 | * @param html The string to processConvert from HTML 287 | * @return Markdown text. 288 | * @see org.jsoup.Jsoup#parse(String, String) 289 | */ 290 | public String convert(String html) { 291 | return convert(html, ""); 292 | } 293 | 294 | 295 | /** 296 | * Converts HTML in memory to Markdown. 297 | * @param html The string to processConvert from HTML 298 | * @param baseUri The base URI for resolving relative links. 299 | * @return Markdown text. 300 | * @see org.jsoup.Jsoup#parse(String, String) 301 | */ 302 | @SuppressWarnings({"WeakerAccess", "SameParameterValue"}) 303 | public String convert(String html, String baseUri) { 304 | Document doc = Jsoup.parse(html, baseUri); 305 | return convert(doc); 306 | } 307 | 308 | 309 | /** 310 | * Converts an HTML body fragment to Markdown. 311 | * @param body The fragment string to processConvert from HTML 312 | * @return Markdown text. 313 | * @see org.jsoup.Jsoup#parseBodyFragment(String, String) 314 | */ 315 | @SuppressWarnings({"UnusedDeclaration"}) 316 | public String convertFragment(String body) { 317 | return convertFragment(body, ""); 318 | } 319 | 320 | 321 | /** 322 | * Converts an HTML body fragment to Markdown. 323 | * @param body The fragment string to processConvert from HTML 324 | * @param baseUri The base URI for resolving relative links. 325 | * @return Markdown text. 326 | * @see org.jsoup.Jsoup#parseBodyFragment(String, String) 327 | */ 328 | public String convertFragment(String body, String baseUri) { 329 | Document doc = Jsoup.parseBodyFragment(body, baseUri); 330 | return convert(doc); 331 | } 332 | 333 | /** 334 | * Converts an already-loaded JSoup Document to Markdown. 335 | * 336 | * @param doc Document to be processed 337 | * @return Markdown text. 338 | */ 339 | @SuppressWarnings({"WeakerAccess"}) 340 | public String convert(Document doc) { 341 | // Note: all convert methods should end up going through this method! 342 | return processConvert(doc, null, null); 343 | } 344 | 345 | /** 346 | * Handles the actual conversion 347 | * @param doc document to convert 348 | * @param writer Optional Writer for output 349 | * @param os Optional OutputStream for output 350 | * @return String result if not using an output stream, else null 351 | */ 352 | private String processConvert(Document doc, Writer writer, OutputStream os) { 353 | String cleanString = Jsoup.clean(doc.html(), "https://www.baidu.com", Whitelist.relaxed().preserveRelativeLinks(true)); 354 | Document parse = Jsoup.parse(cleanString); 355 | doc = parse; 356 | if(cleanedHtmlEchoed) { 357 | System.out.println("Cleaned and processed HTML document:"); 358 | System.out.println(doc.toString()); 359 | System.out.println(); 360 | } 361 | String result = null; 362 | converterLock.lock(); 363 | try { 364 | if(writer != null) { 365 | converter.convert(doc, writer); 366 | } else if(os != null) { 367 | converter.convert(doc, os); 368 | } else { 369 | result = converter.convert(doc); 370 | } 371 | } finally { 372 | converterLock.unlock(); 373 | } 374 | return result; 375 | } 376 | } 377 | -------------------------------------------------------------------------------- /src/main/java/com/overzealous/remark/convert/CmLine.java: -------------------------------------------------------------------------------- 1 | /* 2 | * Copyright 2011 OverZealous Creations, LLC 3 | * 4 | * Licensed under the Apache License, Version 2.0 (the "License"); 5 | * you may not use this file except in compliance with the License. 6 | * You may obtain a copy of the License at 7 | * 8 | * http://www.apache.org/licenses/LICENSE-2.0 9 | * 10 | * Unless required by applicable law or agreed to in writing, software 11 | * distributed under the License is distributed on an "AS IS" BASIS, 12 | * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 13 | * See the License for the specific language governing permissions and 14 | * limitations under the License. 15 | */ 16 | 17 | package com.overzealous.remark.convert; 18 | 19 | import com.overzealous.remark.Options; 20 | import com.overzealous.remark.util.BlockWriter; 21 | import com.overzealous.remark.util.StringUtils; 22 | import org.jsoup.nodes.Element; 23 | 24 | /** 25 | * Handles preformatted sections (pre), renders them as code blocks. 26 | * 27 | * @author Phil DeJarnett 28 | */ 29 | public class CmLine extends AbstractNodeHandler { 30 | 31 | /** 32 | * Converts a pre-formatted block of code. 33 | * Depending on the options, this may render as a block with four spaces added to the beginning, 34 | * or as a fenced code block. 35 | * 36 | * @param parent The previous node walker, in case we just want to remove an element. 37 | * @param node Node to handle 38 | * @param converter Parent converter for this object. 39 | */ 40 | @Override 41 | public void handleNode(NodeHandler parent, Element node, DocumentConverter converter) { 42 | BlockWriter out; 43 | Options.FencedCodeBlocks fenced = converter.options.getFencedCodeBlocks(); 44 | if(fenced.isEnabled()) { 45 | String fence = StringUtils.multiply(fenced.getSeparatorCharacter(), 46 | converter.options.fencedCodeBlocksWidth); 47 | out = converter.output; 48 | converter.output.startBlock(); 49 | out.println(fence); 50 | out.write(converter.cleaner.cleanCode(node)); 51 | out.println(); 52 | out.print(fence); 53 | converter.output.endBlock(); 54 | } else { 55 | converter.output.startBlock(); 56 | out = new BlockWriter(converter.output); 57 | out.print("\r\n"); 58 | out.write(converter.cleaner.cleanCode(node)); 59 | out.print("\r\n"); 60 | converter.output.endBlock(); 61 | } 62 | } 63 | } 64 | -------------------------------------------------------------------------------- /src/main/java/com/overzealous/remark/convert/Codeblock.java: -------------------------------------------------------------------------------- 1 | /* 2 | * Copyright 2011 OverZealous Creations, LLC 3 | * 4 | * Licensed under the Apache License, Version 2.0 (the "License"); 5 | * you may not use this file except in compliance with the License. 6 | * You may obtain a copy of the License at 7 | * 8 | * http://www.apache.org/licenses/LICENSE-2.0 9 | * 10 | * Unless required by applicable law or agreed to in writing, software 11 | * distributed under the License is distributed on an "AS IS" BASIS, 12 | * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 13 | * See the License for the specific language governing permissions and 14 | * limitations under the License. 15 | */ 16 | 17 | package com.overzealous.remark.convert; 18 | 19 | import com.overzealous.remark.Options; 20 | import com.overzealous.remark.util.BlockWriter; 21 | import com.overzealous.remark.util.StringUtils; 22 | import org.jsoup.nodes.Element; 23 | 24 | /** 25 | * Handles preformatted sections (pre), renders them as code blocks. 26 | * 27 | * @author Phil DeJarnett 28 | */ 29 | public class Codeblock extends AbstractNodeHandler { 30 | 31 | /** 32 | * Converts a pre-formatted block of code. 33 | * Depending on the options, this may render as a block with four spaces added to the beginning, 34 | * or as a fenced code block. 35 | * 36 | * @param parent The previous node walker, in case we just want to remove an element. 37 | * @param node Node to handle 38 | * @param converter Parent converter for this object. 39 | */ 40 | @Override 41 | public void handleNode(NodeHandler parent, Element node, DocumentConverter converter) { 42 | BlockWriter out; 43 | Options.FencedCodeBlocks fenced = converter.options.getFencedCodeBlocks(); 44 | if(fenced.isEnabled()) { 45 | String fence = StringUtils.multiply(fenced.getSeparatorCharacter(), 46 | converter.options.fencedCodeBlocksWidth); 47 | out = converter.output; 48 | converter.output.startBlock(); 49 | out.println(fence); 50 | out.write(converter.cleaner.cleanCode(node)); 51 | out.println(); 52 | out.print(fence); 53 | converter.output.endBlock(); 54 | } else { 55 | converter.output.startBlock(); 56 | out = new BlockWriter(converter.output); 57 | out.print("```java"); 58 | out.print("\r\n"); 59 | out.write(converter.cleaner.cleanCode(node)); 60 | out.print("\r\n"); 61 | out.print("```"); 62 | converter.output.endBlock(); 63 | } 64 | } 65 | } 66 | -------------------------------------------------------------------------------- /src/main/java/com/overzealous/remark/convert/Header.java: -------------------------------------------------------------------------------- 1 | /* 2 | * Copyright 2011 OverZealous Creations, LLC 3 | * 4 | * Licensed under the Apache License, Version 2.0 (the "License"); 5 | * you may not use this file except in compliance with the License. 6 | * You may obtain a copy of the License at 7 | * 8 | * http://www.apache.org/licenses/LICENSE-2.0 9 | * 10 | * Unless required by applicable law or agreed to in writing, software 11 | * distributed under the License is distributed on an "AS IS" BASIS, 12 | * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 13 | * See the License for the specific language governing permissions and 14 | * limitations under the License. 15 | */ 16 | 17 | package com.overzealous.remark.convert; 18 | 19 | import com.overzealous.remark.util.BlockWriter; 20 | import com.overzealous.remark.util.StringUtils; 21 | import org.jsoup.nodes.Element; 22 | 23 | /** 24 | * Handles header nodes (h1 through h6) 25 | * 26 | * @author Phil DeJarnett 27 | */ 28 | public class Header extends AbstractNodeHandler { 29 | 30 | /** 31 | * Renders a header node (h1..h6). If enabled, also handles the headerID attribute. 32 | * 33 | * @param parent The previous node walker, in case we just want to remove an element. 34 | * @param node Node to handle 35 | * @param converter Parent converter for this object. 36 | */ 37 | public void handleNode(NodeHandler parent, Element node, DocumentConverter converter) { 38 | int depth = Integer.parseInt(node.tagName().substring(1, 2)); 39 | BlockWriter out = converter.output; 40 | out.startBlock(); 41 | StringUtils.multiply(out, '#', depth); 42 | out.print(' '); 43 | out.print(converter.getInlineContent(this, node).replace("\n", " ")); 44 | out.print(' '); 45 | if(converter.options.headerIds && node.hasAttr("id")) { 46 | out.printf(" {#%s}", node.attr("id")); 47 | } 48 | out.endBlock(); 49 | } 50 | } 51 | -------------------------------------------------------------------------------- /src/main/java/com/overzealous/remark/convert/Image.java: -------------------------------------------------------------------------------- 1 | package com.overzealous.remark.convert; 2 | 3 | 4 | import com.overzealous.remark.util.BlockWriter; 5 | import org.apache.commons.lang3.StringUtils; 6 | import org.jsoup.nodes.Element; 7 | 8 | /** 9 | * Handles img tags. 10 | * @author Phil DeJarnett 11 | */ 12 | public class Image extends AbstractNodeHandler { 13 | 14 | @Override 15 | public void handleNode(NodeHandler parent, Element node, DocumentConverter converter) { 16 | String url = converter.cleaner.cleanUrl(node.attr("src")); 17 | if (StringUtils.isNotBlank(url)) { 18 | String alt = node.attr("alt"); 19 | converter.output.printf("", alt, url); 20 | } 21 | 22 | } 23 | } 24 | -------------------------------------------------------------------------------- /src/main/java/com/overzealous/remark/convert/InlineStyle.java: -------------------------------------------------------------------------------- 1 | /* 2 | * Copyright 2011 OverZealous Creations, LLC 3 | * 4 | * Licensed under the Apache License, Version 2.0 (the "License"); 5 | * you may not use this file except in compliance with the License. 6 | * You may obtain a copy of the License at 7 | * 8 | * http://www.apache.org/licenses/LICENSE-2.0 9 | * 10 | * Unless required by applicable law or agreed to in writing, software 11 | * distributed under the License is distributed on an "AS IS" BASIS, 12 | * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 13 | * See the License for the specific language governing permissions and 14 | * limitations under the License. 15 | */ 16 | 17 | package com.overzealous.remark.convert; 18 | 19 | import com.overzealous.remark.Options; 20 | import org.jsoup.nodes.Element; 21 | import org.jsoup.nodes.Node; 22 | import org.jsoup.nodes.TextNode; 23 | 24 | import java.util.regex.Matcher; 25 | import java.util.regex.Pattern; 26 | 27 | /** 28 | * Handles various inline styling (italics and bold), such as em, i, strong, b, span, and font tags. 29 | * @author Phil DeJarnett 30 | */ 31 | public class InlineStyle extends AbstractNodeHandler { 32 | 33 | private static final char ITALICS_WRAPPER = '*'; 34 | private static final String BOLD_WRAPPER = "**"; 35 | 36 | private static final Pattern ITALICS_PATTERN = Pattern.compile("font-style:\\s*italic", Pattern.CASE_INSENSITIVE); 37 | private static final Pattern BOLD_PATTERN = Pattern.compile("font-weight:\\s*bold", Pattern.CASE_INSENSITIVE); 38 | 39 | private static final Pattern INWORD_CHARACTER = Pattern.compile("\\w"); 40 | 41 | private static final Pattern SPACE_CONTENT_SPACE = Pattern.compile("^(\\s*+)(.*?)(\\s*)$", Pattern.DOTALL); 42 | 43 | private int italicDepth = 0; 44 | private int boldDepth = 0; 45 | 46 | /** 47 | * Renders inline styling (bold, italics) for the given tag. It handles implicit styling ({@code em}, {@code strong}) as 48 | * well as explicit styling via the {@code style} attribute. 49 | *
This object keeps track of the depth of the styling, to prevent recursive situations like this:
50 | * 51 | *{@code hello world}52 | * 53 | *
A naive method would be render the example incorrectly (the output would be {@code *hello **world*})
54 | * 55 | * @param parent The previous node walker, in case we just want to remove an element. 56 | * @param node Node to handle 57 | * @param converter Parent converter for this object. 58 | */ 59 | public void handleNode(NodeHandler parent, Element node, DocumentConverter converter) { 60 | if(checkInnerBlock(node)) { 61 | // not valid to have an inline node around block nodes, so we have to 62 | // simply ignore them. 63 | // just recurse like it's not here. 64 | converter.walkNodes(parent, node); 65 | } else { 66 | Rules rules = checkInword(node, converter); 67 | if(rules.emphasisPreserved) { 68 | checkTag(node, rules); 69 | 70 | if(rules.bold || rules.italics) { 71 | handleStyled(parent, node, converter, rules); 72 | } else { 73 | converter.walkNodes(this, node, converter.inlineNodes); 74 | } 75 | } else { // emphasis has been disabled for this section 76 | // mark as if emphasis was already processed 77 | italicDepth++; 78 | boldDepth++; 79 | converter.walkNodes(this, node, converter.inlineNodes); 80 | italicDepth--; 81 | boldDepth--; 82 | } 83 | } 84 | } 85 | 86 | @Override 87 | public void handleTextNode(TextNode node, DocumentConverter converter) { 88 | // Override to provide special handling for ignoring 89 | // leading or trailing all-space nodes. 90 | if((node.previousSibling() != null && node.nextSibling() != null) || 91 | node.text().trim().length() != 0) { 92 | super.handleTextNode(node, converter); 93 | } 94 | } 95 | 96 | /** 97 | * Minor class to hold onto the styling rules for this class. 98 | */ 99 | private class Rules { 100 | boolean emphasisPreserved = true; 101 | boolean addSpacing = false; 102 | boolean italics = false; 103 | boolean bold = false; 104 | } 105 | 106 | /** 107 | * Handles dealing with a styled node (one that has markers on either side). 108 | * 109 | *It's unique because we have to deal with leading and trailing spaces, among other issues.
110 | * 111 | * @param parent The previous node walker, in case we just want to remove an element. 112 | * @param node Node to handle 113 | * @param converter Parent converter for this object. 114 | * @param rules The styling rules that are active 115 | */ 116 | private void handleStyled(NodeHandler parent, Element node, DocumentConverter converter, Rules rules) { 117 | // prevent double styling 118 | if(rules.bold) { boldDepth++; } 119 | if(rules.italics) { italicDepth++; } 120 | String content = converter.getInlineContent(this, node, true); 121 | if(rules.bold) { boldDepth--; } 122 | if(rules.italics) { italicDepth--; } 123 | 124 | // only proceed if we have content 125 | if(content.length() > 0) { 126 | 127 | 128 | Matcher parts = SPACE_CONTENT_SPACE.matcher(content); 129 | if(parts.find()) { 130 | // write any leading space 131 | converter.output.write(parts.group(1)); 132 | 133 | // don't write the markers if the content ends up empty 134 | if(parts.group(2).length() > 0) { 135 | 136 | // write content 137 | converter.output.write(parts.group(2)); 138 | 139 | 140 | } 141 | 142 | // write any trailing space 143 | converter.output.write(parts.group(3)); 144 | 145 | } // else, something weird happened, like (1 == 0) 146 | } 147 | } 148 | 149 | /** 150 | * Check to see if there is a block-level node somewhere inside this node. 151 | * 152 | * @param node Current node 153 | * @return True is there is a block inside this node (which would be invalid HTML) 154 | */ 155 | private boolean checkInnerBlock(Element node) { 156 | boolean blockExists = false; 157 | for(final Element child : node.children()) { 158 | blockExists = child.isBlock() || checkInnerBlock(child); 159 | if(blockExists) { 160 | break; 161 | } 162 | } 163 | return blockExists; 164 | } 165 | 166 | /** 167 | * Handles the situation where InWordEmphasis needs to be manipulated. 168 | * 169 | *This isn't a terribly intelligent check - it merely looks for the 170 | * situation where a styled node is immediately followed by a 171 | * text node, and that text node starts with a word character.
172 | * 173 | * @param node The current node (should be an inline-styled node) 174 | * @param converter The current converter 175 | * @return flags for checking. 176 | */ 177 | private Rules checkInword(Element node, DocumentConverter converter) { 178 | Rules result = new Rules(); 179 | Options.InWordEmphasis iwe = converter.options.getInWordEmphasis(); 180 | if(!iwe.isEmphasisPreserved() || iwe.isAdditionalSpacingNeeded()) { 181 | // peek behind for inline styling 182 | Node n = node.previousSibling(); 183 | if(n != null && n instanceof TextNode) { 184 | TextNode tn = (TextNode)n; 185 | String text = tn.text(); 186 | if(INWORD_CHARACTER.matcher(text.substring(text.length()-1)).matches()) { 187 | result.emphasisPreserved = iwe.isEmphasisPreserved(); 188 | result.addSpacing = iwe.isAdditionalSpacingNeeded(); 189 | } 190 | } 191 | // peek ahead for inline styling 192 | n = node.nextSibling(); 193 | if(n != null && n instanceof TextNode) { 194 | TextNode tn = (TextNode)n; 195 | if(INWORD_CHARACTER.matcher(tn.text().substring(0,1)).matches()) { 196 | result.emphasisPreserved = iwe.isEmphasisPreserved(); 197 | result.addSpacing = iwe.isAdditionalSpacingNeeded(); 198 | } 199 | } 200 | } 201 | return result; 202 | } 203 | 204 | /** 205 | * Check the styling rules that may or may not apply to this tag. 206 | * @param node The node to look at 207 | * @param rules The rules object to hold the result 208 | */ 209 | private void checkTag(Element node, Rules rules) { 210 | String tn = node.tagName(); 211 | if(tn.equals("i") || tn.equals("em")) { 212 | rules.italics = (italicDepth == 0); 213 | } else if(tn.equals("b") || tn.equals("strong")) { 214 | rules.bold = (boldDepth == 0); 215 | } else { 216 | // check inline-style 217 | if(node.hasAttr("style")) { 218 | String style = node.attr("style"); 219 | if(ITALICS_PATTERN.matcher(style).find()) { 220 | rules.italics = (italicDepth == 0); 221 | } 222 | if(BOLD_PATTERN.matcher(style).find()) { 223 | rules.bold = (boldDepth == 0); 224 | } 225 | } 226 | } 227 | } 228 | 229 | /** 230 | * Render the starting styling tag as necessary. 231 | * 232 | * @param style Rules to render 233 | * @param leadingSpaces Leading spaces string (if any) 234 | * @param converter parent converter 235 | */ 236 | private void start(Rules style, String leadingSpaces, DocumentConverter converter) { 237 | if(style.addSpacing && 238 | (italicDepth == 0 || boldDepth == 0) && 239 | (leadingSpaces == null || leadingSpaces.length() == 0)) { 240 | converter.output.write(' '); 241 | } 242 | if(style.italics) { 243 | if(italicDepth == 0) { 244 | converter.output.write(ITALICS_WRAPPER); 245 | } 246 | } 247 | if(style.bold) { 248 | if(boldDepth == 0) { 249 | converter.output.write(BOLD_WRAPPER); 250 | } 251 | } 252 | } 253 | 254 | /** 255 | * Render the ending tag as necessary. 256 | * 257 | * @param style Rules to render 258 | * @param trailingSpaces Trailing spaces (if any) 259 | * @param converter parent converter 260 | */ 261 | private void end(Rules style, String trailingSpaces, DocumentConverter converter) { 262 | if(style.bold) { 263 | if(boldDepth == 0) { 264 | converter.output.write(BOLD_WRAPPER); 265 | } 266 | } 267 | if(style.italics) { 268 | if(italicDepth == 0) { 269 | converter.output.write(ITALICS_WRAPPER); 270 | } 271 | } 272 | if(style.addSpacing && 273 | (italicDepth == 0 || boldDepth == 0) && 274 | (trailingSpaces == null || trailingSpaces.length() == 0)) { 275 | converter.output.write(' '); 276 | } 277 | } 278 | } 279 | -------------------------------------------------------------------------------- /src/main/resources/application.yml: -------------------------------------------------------------------------------- 1 | server: 2 | port: 9999 3 | 4 | spring: 5 | thymeleaf: 6 | prefix: classpath:/templates/ 7 | suffix: .html 8 | mode: HTML 9 | cache: true 10 | encoding: UTF-8 11 | 12 | datasource: 13 | url: jdbc:h2:file:~/markdown;DB_CLOSE_DELAY=-1;DB_CLOSE_ON_EXIT=FALSE;mode=MySQL 14 | platform: h2 15 | driver-class-name: org.h2.Driver 16 | username: sa 17 | password: 18 | schema: classpath:schema.sql 19 | 20 | 21 | 22 | 23 | 24 | resources: 25 | chain: 26 | strategy: 27 | content: 28 | enabled: true 29 | paths: /** 30 | h2: 31 | console: 32 | enabled: true 33 | path: /db 34 | settings: 35 | web-allow-others: true 36 | jpa: 37 | hibernate: 38 | ddl-auto: update 39 | show-sql: true 40 | 41 | 42 | 43 | 44 | ## markdown存储位置配置区: 45 | markdown: 46 | filePath: ./mds/ 47 | -------------------------------------------------------------------------------- /src/main/resources/banner.txt: -------------------------------------------------------------------------------- 1 | 服务开始启动...... -------------------------------------------------------------------------------- /src/main/resources/data.sql: -------------------------------------------------------------------------------- 1 | INSERT IGNORE INTO PUBLIC.SETTING (ID, CONFIG_NAME, CONFIG_VALUE, CREATE_TIME, REMARK) VALUES (1, 'Image_Proxy_Path', 'http://127.0.0.1:9999/images', null, '图片映射地址'); 2 | INSERT IGNORE INTO PUBLIC.SETTING (ID, CONFIG_NAME, CONFIG_VALUE, CREATE_TIME, REMARK) VALUES (2, 'MD_Save_Path', './mds', null, 'MD文件保存路径'); 3 | INSERT IGNORE INTO PUBLIC.SETTING (ID, CONFIG_NAME, CONFIG_VALUE, CREATE_TIME, REMARK) VALUES (3, 'Image_Save_Path', './pics', null, '图片保存路径'); 4 | INSERT IGNORE INTO PUBLIC.SETTING (ID, CONFIG_NAME, CONFIG_VALUE, CREATE_TIME, REMARK) VALUES (4, 'User_template_id', '1', null, '自定义文件头部和尾部'); 5 | INSERT IGNORE INTO PUBLIC.SETTING (ID, CONFIG_NAME, CONFIG_VALUE, CREATE_TIME, REMARK) VALUES (36, 'Image_DEFAULT_NAME', 'pic', null, '默认图片文件名'); -------------------------------------------------------------------------------- /src/main/resources/schema.sql: -------------------------------------------------------------------------------- 1 | create table IF NOT EXISTS md( 2 | id int primary key auto_increment, 3 | title text, 4 | context longtext, 5 | pname varchar(100), 6 | save_path varchar(100), 7 | blog_url varchar(300), 8 | create_time datetime 9 | ); 10 | 11 | create table IF NOT EXISTS pic( 12 | id int primary key auto_increment, 13 | pname varchar(100), 14 | path varchar(255), 15 | create_time datetime 16 | ); 17 | create table IF NOT EXISTS setting( 18 | id int primary key auto_increment, 19 | config_name varchar(100), 20 | config_value varchar(100), 21 | create_time datetime, 22 | remark varchar(200) 23 | ); 24 | create table IF NOT EXISTS USER_TEMPLATE( 25 | id int primary key auto_increment, 26 | header longtext, 27 | bottom longtext, 28 | create_time datetime 29 | ); -------------------------------------------------------------------------------- /src/main/resources/static/css/about.ae075234.css: -------------------------------------------------------------------------------- 1 | .img-container[data-v-4b17ba9c]{width:210px;height:140px}.tr[data-v-4b17ba9c]{border-radius:6px;margin-bottom:11px;overflow:hidden}.if[data-v-eb84ce2c]{height:768px;width:1024px} -------------------------------------------------------------------------------- /src/main/resources/static/css/app.400b5e86.css: -------------------------------------------------------------------------------- 1 | #components-layout-demo-custom-trigger .trigger{font-size:18px;line-height:64px;padding:0 24px;cursor:pointer;transition:color .3s}#components-layout-demo-custom-trigger .trigger:hover{color:#1890ff}#components-layout-demo-custom-trigger .logo{height:32px;background:hsla(0,0%,100%,.2);margin:16px}.mde[data-v-0524ac7a]{height:850px} -------------------------------------------------------------------------------- /src/main/resources/static/favicon.ico: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/liangtengyu/to_markdown/2e8cfc5910b94e7d26aa22ba6fc1646f6d2be53b/src/main/resources/static/favicon.ico -------------------------------------------------------------------------------- /src/main/resources/static/fonts/fontello.068ca2b3.ttf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/liangtengyu/to_markdown/2e8cfc5910b94e7d26aa22ba6fc1646f6d2be53b/src/main/resources/static/fonts/fontello.068ca2b3.ttf -------------------------------------------------------------------------------- /src/main/resources/static/fonts/fontello.8d4a4e6f.woff2: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/liangtengyu/to_markdown/2e8cfc5910b94e7d26aa22ba6fc1646f6d2be53b/src/main/resources/static/fonts/fontello.8d4a4e6f.woff2 -------------------------------------------------------------------------------- /src/main/resources/static/fonts/fontello.a782baa8.woff: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/liangtengyu/to_markdown/2e8cfc5910b94e7d26aa22ba6fc1646f6d2be53b/src/main/resources/static/fonts/fontello.a782baa8.woff -------------------------------------------------------------------------------- /src/main/resources/static/fonts/fontello.e73a0647.eot: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/liangtengyu/to_markdown/2e8cfc5910b94e7d26aa22ba6fc1646f6d2be53b/src/main/resources/static/fonts/fontello.e73a0647.eot -------------------------------------------------------------------------------- /src/main/resources/static/img/1614755729311.82e3994d.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/liangtengyu/to_markdown/2e8cfc5910b94e7d26aa22ba6fc1646f6d2be53b/src/main/resources/static/img/1614755729311.82e3994d.jpg -------------------------------------------------------------------------------- /src/main/resources/static/img/fontello.9354499c.svg: -------------------------------------------------------------------------------- 1 | 2 | 3 | -------------------------------------------------------------------------------- /src/main/resources/static/index.html: -------------------------------------------------------------------------------- 1 |将HTTP页面 解析为Markdown格式
5 |目前支持: 微信公众号,知乎,知乎专栏,简书,知否(SegmentFault),掘金,CSDN,V2EX,博客园
6 |----------------------
7 | 8 | 9 |----------------------
17 |
19 |
20 |
21 | {{msg}}
34 |线上试用版本,为防止被恶意篡改配置,不提供此功能! 4 |
5 | 6 | 7 | 8 | 9 |