|
1 /* |
|
2 * Copyright (c) 2019, Oracle and/or its affiliates. All rights reserved. |
|
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. |
|
4 * |
|
5 * This code is free software; you can redistribute it and/or modify it |
|
6 * under the terms of the GNU General Public License version 2 only, as |
|
7 * published by the Free Software Foundation. Oracle designates this |
|
8 * particular file as subject to the "Classpath" exception as provided |
|
9 * by Oracle in the LICENSE file that accompanied this code. |
|
10 * |
|
11 * This code is distributed in the hope that it will be useful, but WITHOUT |
|
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or |
|
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
|
14 * version 2 for more details (a copy is included in the LICENSE file that |
|
15 * accompanied this code). |
|
16 * |
|
17 * You should have received a copy of the GNU General Public License version |
|
18 * 2 along with this work; if not, write to the Free Software Foundation, |
|
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. |
|
20 * |
|
21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA |
|
22 * or visit www.oracle.com if you need additional information or have any |
|
23 * questions. |
|
24 */ |
|
25 |
|
26 package build.tools.fixuppandoc; |
|
27 |
|
28 import java.io.BufferedReader; |
|
29 import java.io.BufferedWriter; |
|
30 import java.io.IOException; |
|
31 import java.io.InputStreamReader; |
|
32 import java.io.OutputStreamWriter; |
|
33 import java.io.PrintStream; |
|
34 import java.io.PrintWriter; |
|
35 import java.io.Reader; |
|
36 import java.io.Writer; |
|
37 import java.nio.file.Files; |
|
38 import java.nio.file.Path; |
|
39 import java.util.ArrayList; |
|
40 import java.util.LinkedHashMap; |
|
41 import java.util.LinkedHashSet; |
|
42 import java.util.List; |
|
43 import java.util.Locale; |
|
44 import java.util.Map; |
|
45 import java.util.Objects; |
|
46 import java.util.Set; |
|
47 import java.util.regex.Matcher; |
|
48 import java.util.regex.Pattern; |
|
49 |
|
50 /** |
|
51 * Fixup HTML generated by pandoc. |
|
52 * |
|
53 * <h2>{@code <html>}</h2> |
|
54 * |
|
55 * Replace the existing element with {@code <html lang="en">}, removing references to XML. |
|
56 * |
|
57 * <h2>{@code <main>}</h2> |
|
58 * |
|
59 * {@code <main>} is inserted if palpable content is found that is not with a |
|
60 * section such as {@code header}, {@code footer}, {@code aside}. |
|
61 * |
|
62 * {@code </main>} is inserted if {@code <main>} was inserted and a section |
|
63 * is started that should not be included in the main section. |
|
64 * |
|
65 * <h2>Tables: row headings</h2> |
|
66 * |
|
67 * For simple tables, as typically generated by _pandoc_, determine the column |
|
68 * whose contents are unique, and convert the cells in that column to be header |
|
69 * cells with {@code scope="row"}. In case of ambiguity, a column containing a |
|
70 * {@code <th>} whose contents begin with <em>name</em> is preferred. |
|
71 * When converting the cell, the {@code style} attribute will be updated to |
|
72 * specify {@code font-weight: normal}, and if there is not already an explicit |
|
73 * setting for {@code text-align}, then the style will be updated to include |
|
74 * {@code text-align:left;}. |
|
75 * |
|
76 * These rules do not apply if the table contains any cells that include |
|
77 * a setting for the {@code scope} attribute, or if the table contains |
|
78 * spanning cells or nested tables. |
|
79 * |
|
80 * <h2>{@code <meta name="generator">}</h2> |
|
81 * |
|
82 * Update the content string, to indicate it has been processed by this program. |
|
83 * |
|
84 * <h2>{@code <nav id="TOC">}</h2> |
|
85 * |
|
86 * Set attribute {@code title="Table Of Contents"} |
|
87 * |
|
88 */ |
|
89 public class Main { |
|
90 /** |
|
91 * Runs the program. |
|
92 * |
|
93 * <pre> |
|
94 * java build.tools.fixuppandoc.Main [-o output-file] [input-file] |
|
95 * </pre> |
|
96 * |
|
97 * If no input file is specified, the program will read from standard input. |
|
98 * If no output file is specified, the program will write to standard output. |
|
99 * Any error messages will be written to the standard error stream. |
|
100 * |
|
101 * @param args the command-line arguments |
|
102 */ |
|
103 public static void main(String... args) { |
|
104 try { |
|
105 new Main().run(args); |
|
106 } catch (IOException | IllegalArgumentException e) { |
|
107 System.err.println(e); |
|
108 System.exit(1); |
|
109 } catch (Throwable t) { |
|
110 t.printStackTrace(System.err); |
|
111 System.exit(1); |
|
112 } |
|
113 } |
|
114 |
|
115 private void run(String... args) throws IOException { |
|
116 Path inFile = null; |
|
117 Path outFile = null; |
|
118 |
|
119 for (int i = 0; i < args.length; i++) { |
|
120 String arg = args[i]; |
|
121 if (arg.equals("-o") && i + 1 < args.length) { |
|
122 outFile = Path.of(args[++i]); |
|
123 } else if (arg.startsWith("-")) { |
|
124 throw new IllegalArgumentException(arg); |
|
125 } else if (inFile == null) { |
|
126 inFile = Path.of(arg); |
|
127 } else { |
|
128 throw new IllegalArgumentException(arg); |
|
129 } |
|
130 } |
|
131 |
|
132 new Fixup().run(inFile, outFile); |
|
133 } |
|
134 |
|
135 /** |
|
136 * A class to read HTML, copying input to output, modifying |
|
137 * fragments as needed. |
|
138 */ |
|
139 class Fixup extends HtmlParser { |
|
140 /** The output stream. */ |
|
141 PrintWriter out; |
|
142 |
|
143 /** A stream for reporting errors. */ |
|
144 PrintStream err = System.err; |
|
145 |
|
146 /** |
|
147 * Flag to indicate when {@code <main>} is permitted around palpable content. |
|
148 * Set within {@code <body>}; disabled within elements in which {@code <main>} |
|
149 * is not permitted. |
|
150 */ |
|
151 boolean allowMain = false; |
|
152 |
|
153 /** |
|
154 * Flag to indicate that {@code <main>} is required. |
|
155 * Set on {@code <body>}; reset when {@code <main>} is either found or generated. |
|
156 */ |
|
157 boolean needMain = false; |
|
158 |
|
159 /** |
|
160 * Flag to indicate that {@code </main>} is required. |
|
161 * Set if {@code <main>} is generated. |
|
162 * Reset when a start or end element is found that requires that {@code </main>} |
|
163 * needs to be generated if necessary. |
|
164 */ |
|
165 boolean needEndMain = false; |
|
166 |
|
167 /** |
|
168 * Handler for {@code <table>} elements. |
|
169 */ |
|
170 Table table; |
|
171 |
|
172 /** |
|
173 * Run the program, copying an input file to an output file. |
|
174 * If the input file is {@code null}, input is read from the standard input. |
|
175 * If the output file is {@code null}, output is written to the standard output. |
|
176 * |
|
177 * @param inFile the input file |
|
178 * @param outFile the output file |
|
179 * @throws IOException if an IO error occurs |
|
180 */ |
|
181 void run(Path inFile, Path outFile) throws IOException { |
|
182 try (Writer out = openWriter(outFile)) { |
|
183 this.out = new PrintWriter(out); |
|
184 if (inFile != null) { |
|
185 read(inFile); |
|
186 } else { |
|
187 read(new BufferedReader(new InputStreamReader(System.in))); |
|
188 } |
|
189 } |
|
190 } |
|
191 |
|
192 /** |
|
193 * Returns a writer for a file, or for the standard output if the file is {@code null}. |
|
194 * |
|
195 * @param file the file |
|
196 * @return the writer |
|
197 * @throws IOException if an IO error occurs |
|
198 */ |
|
199 private Writer openWriter(Path file) throws IOException { |
|
200 if (file != null) { |
|
201 return Files.newBufferedWriter(file); |
|
202 } else { |
|
203 return new BufferedWriter(new OutputStreamWriter(System.out) { |
|
204 @Override |
|
205 public void close() throws IOException { |
|
206 flush(); |
|
207 } |
|
208 }); |
|
209 } |
|
210 } |
|
211 |
|
212 @Override |
|
213 protected void error(Path file, int lineNumber, String message) { |
|
214 err.print(file == null ? "<stdin>" : file); |
|
215 if (lineNumber > 0) { |
|
216 err.print(":"); |
|
217 err.print(lineNumber); |
|
218 } |
|
219 err.print(": "); |
|
220 err.println(message); |
|
221 } |
|
222 |
|
223 @Override |
|
224 protected void error(Path file, int lineNumber, Throwable t) { |
|
225 error(file, lineNumber, t.toString()); |
|
226 t.printStackTrace(err); |
|
227 } |
|
228 |
|
229 /** |
|
230 * The buffer in which input is stored until an appropriate action can be determined. |
|
231 * Using the buffer ensures that the output exactly matches the input, except where |
|
232 * it is intentionally modified. |
|
233 */ |
|
234 private StringBuilder buffer = new StringBuilder(); |
|
235 |
|
236 @Override |
|
237 public int nextChar() throws IOException { |
|
238 if (ch > 0) { |
|
239 buffer.append((char) ch); |
|
240 } |
|
241 return super.nextChar(); |
|
242 } |
|
243 |
|
244 @Override |
|
245 protected void doctype(String s) { |
|
246 flushBuffer(); |
|
247 } |
|
248 |
|
249 @Override |
|
250 protected void startElement(String name, Map<String,String> attrs, boolean selfClosing) { |
|
251 switch (name) { |
|
252 case "html": |
|
253 // replace the existing <html> fragment |
|
254 out.write("<html lang=\"en\">"); |
|
255 buffer.setLength(0); |
|
256 break; |
|
257 |
|
258 case "meta": |
|
259 // update the meta-data for the generator |
|
260 if (Objects.equals(attrs.get("name"), "generator")) { |
|
261 out.write(buffer.toString() |
|
262 .replaceAll("(content=\"[^\"]*)(\")", "$1,fixuphtml$2")); |
|
263 buffer.setLength(0); |
|
264 } |
|
265 break; |
|
266 |
|
267 case "article": |
|
268 case "aside": |
|
269 case "footer": |
|
270 case "header": |
|
271 case "nav": |
|
272 // starting one of these elements will terminate <main> if one is being |
|
273 // inserted |
|
274 if (needEndMain) { |
|
275 out.write("</main>"); |
|
276 needEndMain = false; |
|
277 } |
|
278 // <main> is not permitted within these elements |
|
279 allowMain = false; |
|
280 if (name.equals("nav") && Objects.equals(attrs.get("id"), "TOC")) { |
|
281 out.write(buffer.toString() |
|
282 .replaceAll(">$", " title=\"Table Of Contents\">")); |
|
283 buffer.setLength(0); |
|
284 } |
|
285 break; |
|
286 |
|
287 case "body": |
|
288 // within <body>, <main> is both permitted and required |
|
289 allowMain = true; |
|
290 needMain = true; |
|
291 break; |
|
292 |
|
293 case "main": |
|
294 // an explicit <main> found in the input; no need to add one |
|
295 needMain = false; |
|
296 break; |
|
297 |
|
298 case "table": |
|
299 // The entire content of a <table> is buffered, until it can be |
|
300 // determined in which column of the table contains the cells |
|
301 // that can be used to identify the row. |
|
302 if (table == null) { |
|
303 table = new Table(); |
|
304 } else { |
|
305 // tables containing nested tables are not updated |
|
306 table.simple = false; |
|
307 } |
|
308 table.nestDepth++; |
|
309 break; |
|
310 |
|
311 case "thead": |
|
312 case "tbody": |
|
313 if (table != null) { |
|
314 table.endCell(); |
|
315 } |
|
316 break; |
|
317 |
|
318 case "tr": |
|
319 if (table != null) { |
|
320 table.endCell(); |
|
321 table.nextCellColumnIndex = 0; |
|
322 } |
|
323 break; |
|
324 |
|
325 case "td": |
|
326 case "th": |
|
327 if (table != null) { |
|
328 if (attrs.containsKey("rowspan") |
|
329 || attrs.containsKey("colspan") |
|
330 || attrs.containsKey("scope")) { |
|
331 // tables containing spanning cells and tables that already |
|
332 // contain scope attributes are not updated |
|
333 table.simple = false; |
|
334 } |
|
335 table.startCell(name); |
|
336 } |
|
337 break; |
|
338 } |
|
339 |
|
340 // by default, the content is deemed to be palpable content, and so |
|
341 // insert <main> if it is permitted and one is still required, |
|
342 // while also ensuring that it does not appear before <body> |
|
343 if (allowMain && needMain && !name.equals("body")) { |
|
344 out.write("<main>"); |
|
345 needMain = false; |
|
346 needEndMain = true; |
|
347 } |
|
348 |
|
349 flushBuffer(); |
|
350 } |
|
351 |
|
352 @Override |
|
353 protected void endElement(String name) { |
|
354 switch (name) { |
|
355 case "article": |
|
356 case "aside": |
|
357 case "footer": |
|
358 case "header": |
|
359 case "nav": |
|
360 // The code does not handle nested elements of these kinds, but could. |
|
361 // So, assuming they are not nested, ending these elements implies |
|
362 // that <main> is once again permitted. |
|
363 allowMain = true; |
|
364 break; |
|
365 |
|
366 case "body": |
|
367 // The document is nearly done; insert <main> if needed |
|
368 if (needEndMain) { |
|
369 out.write("</main>"); |
|
370 needEndMain = false; |
|
371 } |
|
372 break; |
|
373 |
|
374 case "table": |
|
375 // if the table is finished, analyze it and write it out |
|
376 if (table != null) { |
|
377 if (--table.nestDepth == 0) { |
|
378 table.add(buffer.toString()); |
|
379 table.write(out); |
|
380 table = null; |
|
381 buffer.setLength(0); |
|
382 } |
|
383 } |
|
384 break; |
|
385 |
|
386 case "thead": |
|
387 case "tbody": |
|
388 case "tr": |
|
389 case "td": |
|
390 case "th": |
|
391 // ending any of these elements implicity or explicitly ends the |
|
392 // current cell |
|
393 table.endCell(); |
|
394 break; |
|
395 |
|
396 } |
|
397 flushBuffer(); |
|
398 } |
|
399 |
|
400 @Override |
|
401 protected void content(String content) { |
|
402 if (table != null) { |
|
403 table.content(content); |
|
404 } else if (allowMain && needMain && !content.isBlank()) { |
|
405 // insert <main> if required and if we have palpable content |
|
406 out.write("<main>"); |
|
407 needMain = false; |
|
408 needEndMain = true; |
|
409 } |
|
410 flushBuffer(); |
|
411 } |
|
412 |
|
413 @Override |
|
414 protected void comment(String comment) { |
|
415 flushBuffer(); |
|
416 } |
|
417 |
|
418 /** |
|
419 * Flushes the buffer, either by adding it into a table, if one is |
|
420 * in progress, or by writing it out. |
|
421 */ |
|
422 private void flushBuffer() { |
|
423 String s = buffer.toString(); |
|
424 if (table != null) { |
|
425 table.add(s); |
|
426 } else { |
|
427 out.write(s); |
|
428 } |
|
429 buffer.setLength(0); |
|
430 |
|
431 } |
|
432 } |
|
433 |
|
434 /** |
|
435 * Storage for the content of a {@code <table>} element} until we can determine |
|
436 * whether we should add {@code scope="row"} to the cells in a given column, |
|
437 * and if so, which column. |
|
438 * |
|
439 * The column with the highest number of unique entries is selected; |
|
440 * in case of ambiguity, a column whose heading begins "name" is chosen. |
|
441 * |
|
442 * Only "simple" tables are supported. Tables with any of the following |
|
443 * features are not considered "simple" and will not be modified: |
|
444 * <ul> |
|
445 * <li>Tables containing nested tables</li> |
|
446 * <li>Tables containing cells that use "rowspan" and "colspan" attributes</li> |
|
447 * <li>Tables containing cells that already use "scope" attributes</li> |
|
448 * </ul> |
|
449 */ |
|
450 class Table { |
|
451 /** |
|
452 * A fragment of HTML in this table. |
|
453 */ |
|
454 class Entry { |
|
455 /** The fragment. */ |
|
456 final String html; |
|
457 /** The column for a {@code <td>} fragment, or -1. */ |
|
458 final int column; |
|
459 |
|
460 Entry(String html, int column) { |
|
461 this.html = html; |
|
462 this.column = column; |
|
463 } |
|
464 } |
|
465 |
|
466 /** Whether or not this is a "simple" table. */ |
|
467 boolean simple = true; |
|
468 |
|
469 /** The nesting depth of the current table, within enclosing tables. */ |
|
470 int nestDepth; |
|
471 |
|
472 /** A list of the HTML fragments that make up this table. */ |
|
473 List<Entry> entries; |
|
474 |
|
475 /** The plain text contents of each column, used to determine the primary column. */ |
|
476 List<Set<String>> columnContents; |
|
477 |
|
478 /** The column index of the next cell to be found. */ |
|
479 int nextCellColumnIndex; |
|
480 |
|
481 /** A flag to mark the start of a {@code <td>} cell. */ |
|
482 boolean startTDCell; |
|
483 |
|
484 /** The column index of the current cell, or -1 if not in a cell. */ |
|
485 int currCellColumnIndex; |
|
486 |
|
487 /** The plain text contents of the current column. */ |
|
488 Set<String> currColumnContents; |
|
489 |
|
490 /** The plain text content of the current cell. */ |
|
491 StringBuilder currCellContent; |
|
492 |
|
493 /** The kind ({@code th} or {@code td}) of the current cell. */ |
|
494 String currCellKind; |
|
495 |
|
496 /** |
|
497 * The index of the column, if any, containing a heading beginning "name". |
|
498 * This column is given preferential treatment when deciding the primary column. |
|
499 */ |
|
500 int nameColumn; |
|
501 |
|
502 Table() { |
|
503 entries = new ArrayList<>(); |
|
504 columnContents = new ArrayList<>(); |
|
505 } |
|
506 |
|
507 void startCell(String name) { |
|
508 endCell(); |
|
509 startTDCell = name.equals("td"); |
|
510 currCellColumnIndex = nextCellColumnIndex++; |
|
511 currColumnContents = getColumn(currCellColumnIndex); |
|
512 currCellContent = new StringBuilder(); |
|
513 currCellKind = name; |
|
514 } |
|
515 |
|
516 void endCell() { |
|
517 if (currCellContent != null) { |
|
518 String c = currCellContent.toString().trim(); |
|
519 if (Objects.equals(currCellKind, "th") |
|
520 && c.toLowerCase(Locale.US).startsWith("name")) { |
|
521 nameColumn = currCellColumnIndex; |
|
522 } |
|
523 currColumnContents.add(c); |
|
524 currCellContent = null; |
|
525 currCellColumnIndex = -1; |
|
526 currColumnContents = null; |
|
527 } |
|
528 } |
|
529 |
|
530 void content(String content) { |
|
531 if (currCellContent != null) { |
|
532 currCellContent.append(content); |
|
533 } |
|
534 } |
|
535 |
|
536 void add(String html) { |
|
537 int index = startTDCell ? currCellColumnIndex : -1; |
|
538 entries.add(new Entry(html, index)); |
|
539 startTDCell = false; |
|
540 } |
|
541 |
|
542 void write(PrintWriter out) { |
|
543 int max = -1; |
|
544 int maxIndex = -1; |
|
545 int index = 0; |
|
546 for (Set<String> c : columnContents) { |
|
547 if (c.size() > max || c.size() == max && index == nameColumn) { |
|
548 max = c.size(); |
|
549 maxIndex = index; |
|
550 } |
|
551 index++; |
|
552 } |
|
553 boolean updateEndTd = false; |
|
554 Pattern styleAttr = Pattern.compile("(?<before>.*style=\")(?<style>[^\"]*)(?<after>\".*)"); |
|
555 for (Entry e : entries) { |
|
556 if (simple && e.column == maxIndex) { |
|
557 String attrs = e.html.substring(3, e.html.length() - 1); |
|
558 out.write("<th"); |
|
559 Matcher m = styleAttr.matcher(attrs); |
|
560 if (m.matches()) { |
|
561 out.write(m.group("before")); |
|
562 out.write("font-weight: normal; "); |
|
563 String style = m.group("style"); |
|
564 if (!style.contains("text-align")) { |
|
565 out.write("text-align: left; "); |
|
566 } |
|
567 out.write(style); |
|
568 out.write(m.group("after")); |
|
569 } else { |
|
570 out.write(" style=\"font-weight: normal; text-align:left;\" "); |
|
571 out.write(attrs); |
|
572 } |
|
573 out.write(" scope=\"row\""); |
|
574 out.write(">"); |
|
575 updateEndTd = true; |
|
576 } else if (updateEndTd && e.html.equalsIgnoreCase("</td>")) { |
|
577 out.write("</th>"); |
|
578 updateEndTd = false; |
|
579 } else { |
|
580 out.write(e.html); |
|
581 if (updateEndTd && e.html.regionMatches(true, 0, "<td", 0, 3)) { |
|
582 // a new cell has been started without explicitly closing the |
|
583 // cell that was being updated |
|
584 updateEndTd = false; |
|
585 } |
|
586 } |
|
587 } |
|
588 } |
|
589 |
|
590 private Set<String> getColumn(int index) { |
|
591 while (columnContents.size() <= index) { |
|
592 columnContents.add(new LinkedHashSet<>()); |
|
593 } |
|
594 |
|
595 return columnContents.get(index); |
|
596 } |
|
597 } |
|
598 |
|
599 /** |
|
600 * A basic HTML parser. |
|
601 * Override the protected methods as needed to get notified of significant items |
|
602 * in any file that is read. |
|
603 */ |
|
604 abstract class HtmlParser { |
|
605 |
|
606 private Path file; |
|
607 private Reader in; |
|
608 protected int ch; |
|
609 private int lineNumber; |
|
610 private boolean inScript; |
|
611 private boolean xml; |
|
612 |
|
613 /** |
|
614 * Read a file. |
|
615 * @param file the file |
|
616 */ |
|
617 void read(Path file) { |
|
618 try (Reader r = Files.newBufferedReader(file)) { |
|
619 this.file = file; |
|
620 read(r); |
|
621 } catch (IOException e) { |
|
622 error(file, -1, e); |
|
623 } |
|
624 } |
|
625 |
|
626 HtmlParser() { } |
|
627 |
|
628 /** |
|
629 * Read a stream. |
|
630 * @param r the stream |
|
631 */ |
|
632 void read(Reader r) { |
|
633 try { |
|
634 this.in = r; |
|
635 StringBuilder content = new StringBuilder(); |
|
636 |
|
637 startFile(file); |
|
638 try { |
|
639 lineNumber = 1; |
|
640 xml = false; |
|
641 nextChar(); |
|
642 |
|
643 while (ch != -1) { |
|
644 if (ch == '<') { |
|
645 content(content.toString()); |
|
646 content.setLength(0); |
|
647 html(); |
|
648 } else { |
|
649 content.append((char) ch); |
|
650 if (ch == '\n') { |
|
651 content(content.toString()); |
|
652 content.setLength(0); |
|
653 } |
|
654 nextChar(); |
|
655 } |
|
656 } |
|
657 } finally { |
|
658 endFile(); |
|
659 } |
|
660 } catch (IOException e) { |
|
661 error(file, lineNumber, e); |
|
662 } catch (Throwable t) { |
|
663 error(file, lineNumber, t); |
|
664 t.printStackTrace(System.err); |
|
665 } |
|
666 } |
|
667 |
|
668 protected int getLineNumber() { |
|
669 return lineNumber; |
|
670 } |
|
671 |
|
672 /** |
|
673 * Called when a file has been opened, before parsing begins. |
|
674 * This is always the first notification when reading a file. |
|
675 * This implementation does nothing. |
|
676 * |
|
677 * @param file the file |
|
678 */ |
|
679 protected void startFile(Path file) { } |
|
680 |
|
681 /** |
|
682 * Called when the parser has finished reading a file. |
|
683 * This is always the last notification when reading a file, |
|
684 * unless any errors occur while closing the file. |
|
685 * This implementation does nothing. |
|
686 */ |
|
687 protected void endFile() { } |
|
688 |
|
689 /** |
|
690 * Called when a doctype declaration is found, at the beginning of the file. |
|
691 * This implementation does nothing. |
|
692 * @param s the doctype declaration |
|
693 */ |
|
694 protected void doctype(String s) { } |
|
695 |
|
696 /** |
|
697 * Called when the opening tag of an HTML element is encountered. |
|
698 * This implementation does nothing. |
|
699 * @param name the name of the tag |
|
700 * @param attrs the attribute |
|
701 * @param selfClosing whether or not this is a self-closing tag |
|
702 */ |
|
703 protected void startElement(String name, Map<String,String> attrs, boolean selfClosing) { } |
|
704 |
|
705 /** |
|
706 * Called when the closing tag of an HTML tag is encountered. |
|
707 * This implementation does nothing. |
|
708 * @param name the name of the tag |
|
709 */ |
|
710 protected void endElement(String name) { } |
|
711 |
|
712 /** |
|
713 * Called for sequences of character content. |
|
714 * @param content the character content |
|
715 */ |
|
716 protected void content(String content) { } |
|
717 |
|
718 /** |
|
719 * Called for sequences of comment. |
|
720 * @param comment the comment |
|
721 */ |
|
722 protected void comment(String comment) { } |
|
723 |
|
724 /** |
|
725 * Called when an error has been encountered. |
|
726 * @param file the file being read |
|
727 * @param lineNumber the line number of line containing the error |
|
728 * @param message a description of the error |
|
729 */ |
|
730 protected abstract void error(Path file, int lineNumber, String message); |
|
731 |
|
732 /** |
|
733 * Called when an exception has been encountered. |
|
734 * @param file the file being read |
|
735 * @param lineNumber the line number of the line being read when the exception was found |
|
736 * @param t the exception |
|
737 */ |
|
738 protected abstract void error(Path file, int lineNumber, Throwable t); |
|
739 |
|
740 protected int nextChar() throws IOException { |
|
741 ch = in.read(); |
|
742 if (ch == '\n') |
|
743 lineNumber++; |
|
744 return ch; |
|
745 } |
|
746 |
|
747 /** |
|
748 * Read the start or end of an HTML tag, or an HTML comment |
|
749 * {@literal <identifier attrs> } or {@literal </identifier> } |
|
750 * @throws java.io.IOException if there is a problem reading the file |
|
751 */ |
|
752 protected void html() throws IOException { |
|
753 nextChar(); |
|
754 if (isIdentifierStart((char) ch)) { |
|
755 String name = readIdentifier().toLowerCase(Locale.US); |
|
756 Map<String,String> attrs = htmlAttrs(); |
|
757 if (attrs != null) { |
|
758 boolean selfClosing = false; |
|
759 if (ch == '/') { |
|
760 nextChar(); |
|
761 selfClosing = true; |
|
762 } |
|
763 if (ch == '>') { |
|
764 nextChar(); |
|
765 startElement(name, attrs, selfClosing); |
|
766 if (name.equals("script")) { |
|
767 inScript = true; |
|
768 } |
|
769 return; |
|
770 } |
|
771 } |
|
772 } else if (ch == '/') { |
|
773 nextChar(); |
|
774 if (isIdentifierStart((char) ch)) { |
|
775 String name = readIdentifier().toLowerCase(Locale.US); |
|
776 skipWhitespace(); |
|
777 if (ch == '>') { |
|
778 nextChar(); |
|
779 endElement(name); |
|
780 if (name.equals("script")) { |
|
781 inScript = false; |
|
782 } |
|
783 return; |
|
784 } |
|
785 } |
|
786 } else if (ch == '!') { |
|
787 nextChar(); |
|
788 if (ch == '-') { |
|
789 nextChar(); |
|
790 if (ch == '-') { |
|
791 nextChar(); |
|
792 StringBuilder comment = new StringBuilder(); |
|
793 while (ch != -1) { |
|
794 int dash = 0; |
|
795 while (ch == '-') { |
|
796 dash++; |
|
797 comment.append(ch); |
|
798 nextChar(); |
|
799 } |
|
800 // Strictly speaking, a comment should not contain "--" |
|
801 // so dash > 2 is an error, dash == 2 implies ch == '>' |
|
802 // See http://www.w3.org/TR/html-markup/syntax.html#syntax-comments |
|
803 // for more details. |
|
804 if (dash >= 2 && ch == '>') { |
|
805 comment.setLength(comment.length() - 2); |
|
806 comment(comment.toString()); |
|
807 nextChar(); |
|
808 return; |
|
809 } |
|
810 |
|
811 comment.append(ch); |
|
812 nextChar(); |
|
813 } |
|
814 } |
|
815 } else if (ch == '[') { |
|
816 nextChar(); |
|
817 if (ch == 'C') { |
|
818 nextChar(); |
|
819 if (ch == 'D') { |
|
820 nextChar(); |
|
821 if (ch == 'A') { |
|
822 nextChar(); |
|
823 if (ch == 'T') { |
|
824 nextChar(); |
|
825 if (ch == 'A') { |
|
826 nextChar(); |
|
827 if (ch == '[') { |
|
828 while (true) { |
|
829 nextChar(); |
|
830 if (ch == ']') { |
|
831 nextChar(); |
|
832 if (ch == ']') { |
|
833 nextChar(); |
|
834 if (ch == '>') { |
|
835 nextChar(); |
|
836 return; |
|
837 } |
|
838 } |
|
839 } |
|
840 } |
|
841 |
|
842 } |
|
843 } |
|
844 } |
|
845 } |
|
846 } |
|
847 } |
|
848 } else { |
|
849 StringBuilder sb = new StringBuilder(); |
|
850 while (ch != -1 && ch != '>') { |
|
851 sb.append((char) ch); |
|
852 nextChar(); |
|
853 } |
|
854 Pattern p = Pattern.compile("(?is)doctype\\s+html\\s?.*"); |
|
855 String s = sb.toString(); |
|
856 if (p.matcher(s).matches()) { |
|
857 doctype(s); |
|
858 return; |
|
859 } |
|
860 } |
|
861 } else if (ch == '?') { |
|
862 nextChar(); |
|
863 if (ch == 'x') { |
|
864 nextChar(); |
|
865 if (ch == 'm') { |
|
866 nextChar(); |
|
867 if (ch == 'l') { |
|
868 Map<String,String> attrs = htmlAttrs(); |
|
869 if (ch == '?') { |
|
870 nextChar(); |
|
871 if (ch == '>') { |
|
872 nextChar(); |
|
873 xml = true; |
|
874 return; |
|
875 } |
|
876 } |
|
877 } |
|
878 } |
|
879 |
|
880 } |
|
881 } |
|
882 |
|
883 if (!inScript) { |
|
884 error(file, lineNumber, "bad html"); |
|
885 } |
|
886 } |
|
887 |
|
888 /** |
|
889 * Read a series of HTML attributes, terminated by {@literal > }. |
|
890 * Each attribute is of the form {@literal identifier[=value] }. |
|
891 * "value" may be unquoted, single-quoted, or double-quoted. |
|
892 */ |
|
893 private Map<String,String> htmlAttrs() throws IOException { |
|
894 Map<String, String> map = new LinkedHashMap<>(); |
|
895 skipWhitespace(); |
|
896 |
|
897 while (isIdentifierStart((char) ch)) { |
|
898 String name = readAttributeName().toLowerCase(Locale.US); |
|
899 skipWhitespace(); |
|
900 String value = null; |
|
901 if (ch == '=') { |
|
902 nextChar(); |
|
903 skipWhitespace(); |
|
904 if (ch == '\'' || ch == '"') { |
|
905 char quote = (char) ch; |
|
906 nextChar(); |
|
907 StringBuilder sb = new StringBuilder(); |
|
908 while (ch != -1 && ch != quote) { |
|
909 sb.append((char) ch); |
|
910 nextChar(); |
|
911 } |
|
912 value = sb.toString() // hack to replace common entities |
|
913 .replace("<", "<") |
|
914 .replace(">", ">") |
|
915 .replace("&", "&"); |
|
916 nextChar(); |
|
917 } else { |
|
918 StringBuilder sb = new StringBuilder(); |
|
919 while (ch != -1 && !isUnquotedAttrValueTerminator((char) ch)) { |
|
920 sb.append((char) ch); |
|
921 nextChar(); |
|
922 } |
|
923 value = sb.toString(); |
|
924 } |
|
925 skipWhitespace(); |
|
926 } |
|
927 map.put(name, value); |
|
928 } |
|
929 |
|
930 return map; |
|
931 } |
|
932 |
|
933 private boolean isIdentifierStart(char ch) { |
|
934 return Character.isUnicodeIdentifierStart(ch); |
|
935 } |
|
936 |
|
937 private String readIdentifier() throws IOException { |
|
938 StringBuilder sb = new StringBuilder(); |
|
939 sb.append((char) ch); |
|
940 nextChar(); |
|
941 while (ch != -1 && Character.isUnicodeIdentifierPart(ch)) { |
|
942 sb.append((char) ch); |
|
943 nextChar(); |
|
944 } |
|
945 return sb.toString(); |
|
946 } |
|
947 |
|
948 private String readAttributeName() throws IOException { |
|
949 StringBuilder sb = new StringBuilder(); |
|
950 sb.append((char) ch); |
|
951 nextChar(); |
|
952 while (ch != -1 && Character.isUnicodeIdentifierPart(ch) |
|
953 || ch == '-' |
|
954 || (xml || sb.toString().startsWith("xml")) && ch == ':') { |
|
955 sb.append((char) ch); |
|
956 nextChar(); |
|
957 } |
|
958 return sb.toString(); |
|
959 } |
|
960 |
|
961 private boolean isWhitespace(char ch) { |
|
962 return Character.isWhitespace(ch); |
|
963 } |
|
964 |
|
965 private void skipWhitespace() throws IOException { |
|
966 while (isWhitespace((char) ch)) { |
|
967 nextChar(); |
|
968 } |
|
969 } |
|
970 |
|
971 private boolean isUnquotedAttrValueTerminator(char ch) { |
|
972 switch (ch) { |
|
973 case '\f': case '\n': case '\r': case '\t': |
|
974 case ' ': |
|
975 case '"': case '\'': case '`': |
|
976 case '=': case '<': case '>': |
|
977 return true; |
|
978 default: |
|
979 return false; |
|
980 } |
|
981 } |
|
982 } |
|
983 |
|
984 } |