sandbox/aahz/OO/OOwriter.py

   1 # Author: Aahz
   2 # Contact: aahz@pythoncraft.com
   3 # Revision:
   4 # Date: $Date$
   5 # Copyright: This module has been placed in the public domain.
   6
   7 """
   8 OpenOffice writer
   9
  10 The output is an OpenOffice.org 1.0-compatible document.
  11 """
  12
  13 __docformat__ = 'reStructuredText'
  14
  15
  16 import sys
  17 from warnings import warn
  18 import re
  19
  20 import docutils
  21 from docutils import nodes, utils, writers, languages
  22
  23 import OOtext
  24
  25
  26 section_styles = [
  27     '.ch title',
  28     '.head 1',
  29     '.head 2'
  30     ]
  31
  32 class Writer(writers.Writer):
  33
  34     supported = ('OpenOffice')
  35     """Formats this writer supports."""
  36
  37     def __init__(self):
  38         writers.Writer.__init__(self)
  39         self.output = None
  40         self.translator_class = Translator
  41
  42     def translate(self):
  43         visitor = self.translator_class(self.document)
  44         self.document.walkabout(visitor)
  45         self.output = visitor.astext()
  46
  47
  48 class Translator(nodes.NodeVisitor):
  49
  50     header = [OOtext.content_header]
  51     footer = [OOtext.content_footer]
  52
  53     start_para = '\n<text:p text:style-name="%s">\n'
  54     end_para = '\n</text:p>\n'
  55
  56     start_charstyle = '<text:span text:style-name="%s">'
  57     end_charstyle = '</text:span>'
  58
  59     line_break = '\n<text:line-break/>'
  60     re_spaces = re.compile('  +')
  61     spaces = '<text:s text:c="%d"/>'
  62
  63     re_annotation = re.compile(r'#\d+(?:, #\d+)*$')
  64
  65     def __init__(self, document):
  66         nodes.NodeVisitor.__init__(self, document)
  67         self.settings = document.settings
  68         self.body = []
  69         self.section_level = 0
  70         self.skip_para_tag = False
  71
  72     def astext(self):
  73         return ''.join(self.header + self.body + self.footer)
  74
  75     def encode(self, text):
  76         """Encode special characters in `text` & return."""
  77         # @@@ A codec to do these and all other HTML entities would be nice.
  78         text = text.replace("&", "&amp;")
  79         text = text.replace("<", "&lt;")
  80         text = text.replace('"', "&quot;")
  81         text = text.replace(">", "&gt;")
  82         return text
  83
  84     def compress_spaces(self, line):
  85         while 1:
  86             match = self.re_spaces.search(line)
  87             if match:
  88                 start, end = match.span()
  89                 numspaces = end - start
  90                 line = line[:start] + (self.spaces % numspaces) + line[end:]
  91             else:
  92                 break
  93         return line
  94
  95     def fix_annotation(self, line):
  96         match = self.re_annotation.search(line)
  97         if match:
  98             pos = match.start()
  99             line = line[:pos] + '|' + line[pos:]
 100         return line
 101
 102     def visit_Text(self, node):
 103         self.body.append(self.encode(node.astext()))
 104
 105     def depart_Text(self, node):
 106         pass
 107
 108     def visit_admonition(self, node, name):
 109         self.skip_para_tag = True
 110         self.body.append(self.start_para % '.CALLOUT')
 111
 112     def depart_admonition(self):
 113         self.body.append(self.end_para)
 114         self.skip_para_tag = False
 115
 116     def visit_attention(self, node):
 117         self.visit_admonition(node, 'attention')
 118
 119     def depart_attention(self, node):
 120         self.depart_admonition()
 121
 122     def visit_block_quote(self, node):
 123         self.skip_para_tag = True
 124         self.body.append(self.start_para % '.quotes')
 125
 126     def depart_block_quote(self, node):
 127         self.body.append(self.end_para)
 128         self.skip_para_tag = False
 129
 130     def visit_bullet_list(self, node):
 131         self.body.append('\n<text:unordered-list text:style-name=".bullet">\n')
 132
 133     def depart_bullet_list(self, node):
 134         self.body.append('</text:unordered-list>\n')
 135
 136     def visit_caption(self, node):
 137         self.body.append(self.starttag(node, 'p', '', CLASS='caption'))
 138
 139     def depart_caption(self, node):
 140         self.body.append('</p>\n')
 141
 142     def visit_caution(self, node):
 143         self.visit_admonition(node, 'caution')
 144
 145     def depart_caution(self, node):
 146         self.depart_admonition()
 147
 148     def visit_citation(self, node):
 149         self.body.append(self.starttag(node, 'table', CLASS='citation',
 150                                        frame="void", rules="none"))
 151         self.footnote_backrefs(node)
 152
 153     def depart_citation(self, node):
 154         self.body.append('</td></tr>\n'
 155                          '</tbody>\n</table>\n')
 156
 157     def visit_citation_reference(self, node):
 158         href = ''
 159         if node.has_key('refid'):
 160             href = '#' + node['refid']
 161         elif node.has_key('refname'):
 162             href = '#' + self.document.nameids[node['refname']]
 163         self.body.append(self.starttag(node, 'a', '[', href=href,
 164                                        CLASS='citation-reference'))
 165
 166     def depart_citation_reference(self, node):
 167         self.body.append(']</a>')
 168
 169     def visit_classifier(self, node):
 170         self.body.append(' <span class="classifier-delimiter">:</span> ')
 171         self.body.append(self.starttag(node, 'span', '', CLASS='classifier'))
 172
 173     def depart_classifier(self, node):
 174         self.body.append('</span>')
 175
 176     def visit_colspec(self, node):
 177         self.colspecs.append(node)
 178
 179     def depart_colspec(self, node):
 180         pass
 181
 182     def write_colspecs(self):
 183         width = 0
 184         for node in self.colspecs:
 185             width += node['colwidth']
 186         for node in self.colspecs:
 187             colwidth = int(node['colwidth'] * 100.0 / width + 0.5)
 188             self.body.append(self.emptytag(node, 'col',
 189                                            colwidth='%i%%' % colwidth))
 190         self.colspecs = []
 191
 192     def visit_comment(self, node):
 193         raise nodes.SkipNode
 194
 195     def visit_decoration(self, node):
 196         pass
 197
 198     def depart_decoration(self, node):
 199         pass
 200
 201     def visit_definition(self, node):
 202         self.body.append('</dt>\n')
 203         self.body.append(self.starttag(node, 'dd', ''))
 204         if len(node) and isinstance(node[0], nodes.paragraph):
 205             node[0].set_class('first')
 206
 207     def depart_definition(self, node):
 208         self.body.append('</dd>\n')
 209
 210     def visit_definition_list(self, node):
 211         print node.astext()
 212         self.body.append(self.starttag(node, 'dl'))
 213
 214     def depart_definition_list(self, node):
 215         self.body.append('</dl>\n')
 216
 217     def visit_definition_list_item(self, node):
 218         pass
 219
 220     def depart_definition_list_item(self, node):
 221         pass
 222
 223     def visit_description(self, node):
 224         self.body.append(self.starttag(node, 'td', ''))
 225         if len(node) and isinstance(node[0], nodes.paragraph):
 226             node[0].set_class('first')
 227
 228     def depart_description(self, node):
 229         self.body.append('</td>')
 230
 231     def visit_doctest_block(self, node):
 232         self.visit_literal_block(node)
 233
 234     def visit_document(self, node):
 235         pass
 236
 237     def depart_document(self, node):
 238         pass
 239
 240     def visit_emphasis(self, node):
 241         self.body.append(self.start_charstyle % 'italic')
 242
 243     def depart_emphasis(self, node):
 244         self.body.append(self.end_charstyle)
 245
 246     def visit_entry(self, node):
 247         if isinstance(node.parent.parent, nodes.thead):
 248             tagname = 'th'
 249         else:
 250             tagname = 'td'
 251         atts = {}
 252         if node.has_key('morerows'):
 253             atts['rowspan'] = node['morerows'] + 1
 254         if node.has_key('morecols'):
 255             atts['colspan'] = node['morecols'] + 1
 256         self.body.append(self.starttag(node, tagname, '', **atts))
 257         self.context.append('</%s>\n' % tagname.lower())
 258         if len(node) == 0:              # empty cell
 259             self.body.append('&nbsp;')
 260         elif isinstance(node[0], nodes.paragraph):
 261             node[0].set_class('first')
 262
 263     def depart_entry(self, node):
 264         self.body.append(self.context.pop())
 265
 266     def visit_enumerated_list(self, node):
 267         """
 268         The 'start' attribute does not conform to HTML 4.01's strict.dtd, but
 269         CSS1 doesn't help. CSS2 isn't widely enough supported yet to be
 270         usable.
 271         """
 272         atts = {}
 273         if node.has_key('start'):
 274             atts['start'] = node['start']
 275         if node.has_key('enumtype'):
 276             atts['class'] = node['enumtype']
 277         # @@@ To do: prefix, suffix. How? Change prefix/suffix to a
 278         # single "format" attribute? Use CSS2?
 279         old_compact_simple = self.compact_simple
 280         self.context.append((self.compact_simple, self.compact_p))
 281         self.compact_p = None
 282         self.compact_simple = (self.options.compact_lists and
 283                                (self.compact_simple
 284                                 or self.topic_class == 'contents'
 285                                 or self.check_simple_list(node)))
 286         if self.compact_simple and not old_compact_simple:
 287             atts['class'] = (atts.get('class', '') + ' simple').strip()
 288         self.body.append(self.starttag(node, 'ol', **atts))
 289
 290     def depart_enumerated_list(self, node):
 291         self.compact_simple, self.compact_p = self.context.pop()
 292         self.body.append('</ol>\n')
 293
 294     def visit_error(self, node):
 295         self.visit_admonition(node, 'error')
 296
 297     def depart_error(self, node):
 298         self.depart_admonition()
 299
 300     def visit_field(self, node):
 301         self.body.append(self.starttag(node, 'tr', '', CLASS='field'))
 302
 303     def depart_field(self, node):
 304         self.body.append('</tr>\n')
 305
 306     def visit_field_body(self, node):
 307         self.body.append(self.starttag(node, 'td', '', CLASS='field-body'))
 308         if len(node) and isinstance(node[0], nodes.paragraph):
 309             node[0].set_class('first')
 310
 311     def depart_field_body(self, node):
 312         self.body.append('</td>\n')
 313
 314     def visit_field_list(self, node):
 315         self.body.append(self.starttag(node, 'table', frame='void',
 316                                        rules='none', CLASS='field-list'))
 317         self.body.append('<col class="field-name" />\n'
 318                          '<col class="field-body" />\n'
 319                          '<tbody valign="top">\n')
 320
 321     def depart_field_list(self, node):
 322         self.body.append('</tbody>\n</table>\n')
 323
 324     def visit_field_name(self, node):
 325         atts = {}
 326         if self.in_docinfo:
 327             atts['class'] = 'docinfo-name'
 328         else:
 329             atts['class'] = 'field-name'
 330         if len(node.astext()) > 14:
 331             atts['colspan'] = 2
 332             self.context.append('</tr>\n<tr><td>&nbsp;</td>')
 333         else:
 334             self.context.append('')
 335         self.body.append(self.starttag(node, 'th', '', **atts))
 336
 337     def depart_field_name(self, node):
 338         self.body.append(':</th>')
 339         self.body.append(self.context.pop())
 340
 341     def visit_figure(self, node):
 342         self.body.append(self.start_para % '.figure')
 343
 344     def depart_figure(self, node):
 345         self.body.append(self.end_para)
 346
 347     def visit_footnote(self, node):
 348         raise nodes.SkipNode
 349
 350     def footnote_backrefs(self, node):
 351         warn("footnote backrefs not available")
 352
 353     def depart_footnote(self, node):
 354         pass
 355
 356     def visit_footnote_reference(self, node):
 357         name = node['refid']
 358         id = node['id']
 359         number = node['auto']
 360         for footnote in self.document.autofootnotes:
 361             if name == footnote['name']:
 362                 break
 363         self.body.append('<text:footnote text:id="%s">\n' % id)
 364         self.body.append('<text:footnote-citation text:string-value="%s"/>\n' % number)
 365         self.body.append('<text:footnote-body>\n')
 366         self.body.append(self.start_para % '.body')
 367         for child in footnote.children:
 368             if isinstance(child, nodes.paragraph):
 369                 self.body.append(child.astext())
 370         self.body.append(self.end_para)
 371         self.body.append('</text:footnote-body>\n')
 372         self.body.append('</text:footnote>')
 373         raise nodes.SkipNode
 374
 375     def depart_footnote_reference(self, node):
 376         pass
 377
 378     def visit_generated(self, node):
 379         pass
 380
 381     def depart_generated(self, node):
 382         pass
 383
 384     def visit_header(self, node):
 385         self.context.append(len(self.body))
 386
 387     def depart_header(self, node):
 388         start = self.context.pop()
 389         self.body_prefix.append(self.starttag(node, 'div', CLASS='header'))
 390         self.body_prefix.extend(self.body[start:])
 391         self.body_prefix.append('<hr />\n</div>\n')
 392         del self.body[start:]
 393
 394     def visit_hint(self, node):
 395         self.visit_admonition(node, 'hint')
 396
 397     def depart_hint(self, node):
 398         self.depart_admonition()
 399
 400     def visit_image(self, node):
 401         name = "Figure: %s\n" % node.attributes['uri']
 402         self.body.append(name)
 403
 404     def depart_image(self, node):
 405         pass
 406
 407     def visit_important(self, node):
 408         self.visit_admonition(node, 'important')
 409
 410     def depart_important(self, node):
 411         self.depart_admonition()
 412
 413     def visit_index_entry(self, node):
 414         index_format = '<text:alphabetical-index-mark text:string-value="%s"/>\n'
 415         self.body.append(self.start_para % '.body')
 416         entries = node.astext().split('\n')
 417         for entry in entries:
 418             self.body.append(index_format % self.encode(entry))
 419         self.body.append(self.end_para)
 420         raise nodes.SkipNode
 421
 422     def visit_interpreted(self, node):
 423         # @@@ Incomplete, pending a proper implementation on the
 424         # Parser/Reader end.
 425         #self.body.append(node['role'] + ':')
 426         self.body.append(node.astext())
 427         raise nodes.SkipNode
 428
 429     def depart_interpreted(self, node):
 430         pass
 431
 432     # Don't need footnote labels/numbers
 433     def visit_label(self, node):
 434         print "!"
 435         raise nodes.SkipNode
 436
 437     def visit_legend(self, node):
 438         self.body.append(self.starttag(node, 'div', CLASS='legend'))
 439
 440     def depart_legend(self, node):
 441         self.body.append('</div>\n')
 442
 443     def visit_line_block(self, node):
 444         self.body.append(self.start_para % '.quotes')
 445         lines = node.astext()
 446         lines = lines.split('\n')
 447         lines = self.line_break.join(lines)
 448         self.body.append(lines)
 449         self.body.append(self.end_para)
 450         raise nodes.SkipNode
 451
 452     def visit_list_item(self, node):
 453         self.body.append('<text:list-item>')
 454
 455     def depart_list_item(self, node):
 456         self.body.append('</text:list-item>\n')
 457
 458     def visit_literal(self, node):
 459         self.body.append(self.start_charstyle % 'code')
 460
 461     def depart_literal(self, node):
 462         self.body.append(self.end_charstyle)
 463
 464     def visit_literal_block(self, node):
 465         self.body.append(self.start_para % '.code first')
 466         self.body.append(self.end_para)
 467         lines = self.encode(node.astext())
 468         lines = lines.split('\n')
 469         while lines[-1] == '':
 470             lines.pop()
 471         for line in lines:
 472             self.body.append(self.start_para % '.code')
 473             line = self.fix_annotation(line)
 474             line = self.compress_spaces(line)
 475             self.body.append(line)
 476             self.body.append(self.end_para)
 477         self.body.append(self.start_para % '.code last')
 478         self.body.append(self.end_para)
 479         raise nodes.SkipNode
 480
 481     def visit_note(self, node):
 482         self.visit_admonition(node, '.note')
 483
 484     def depart_note(self, node):
 485         self.depart_admonition()
 486
 487     def visit_option(self, node):
 488         if self.context[-1]:
 489             self.body.append(', ')
 490
 491     def depart_option(self, node):
 492         self.context[-1] += 1
 493
 494     def visit_option_argument(self, node):
 495         self.body.append(node.get('delimiter', ' '))
 496         self.body.append(self.starttag(node, 'var', ''))
 497
 498     def depart_option_argument(self, node):
 499         self.body.append('</var>')
 500
 501     def visit_option_group(self, node):
 502         atts = {}
 503         if len(node.astext()) > 14:
 504             atts['colspan'] = 2
 505             self.context.append('</tr>\n<tr><td>&nbsp;</td>')
 506         else:
 507             self.context.append('')
 508         self.body.append(self.starttag(node, 'td', **atts))
 509         self.body.append('<kbd>')
 510         self.context.append(0)          # count number of options
 511
 512     def depart_option_group(self, node):
 513         self.context.pop()
 514         self.body.append('</kbd></td>\n')
 515         self.body.append(self.context.pop())
 516
 517     def visit_option_list(self, node):
 518         self.body.append(
 519               self.starttag(node, 'table', CLASS='option-list',
 520                             frame="void", rules="none"))
 521         self.body.append('<col class="option" />\n'
 522                          '<col class="description" />\n'
 523                          '<tbody valign="top">\n')
 524
 525     def depart_option_list(self, node):
 526         self.body.append('</tbody>\n</table>\n')
 527
 528     def visit_option_list_item(self, node):
 529         self.body.append(self.starttag(node, 'tr', ''))
 530
 531     def depart_option_list_item(self, node):
 532         self.body.append('</tr>\n')
 533
 534     def visit_option_string(self, node):
 535         self.body.append(self.starttag(node, 'span', '', CLASS='option'))
 536
 537     def depart_option_string(self, node):
 538         self.body.append('</span>')
 539
 540     def visit_paragraph(self, node):
 541         if not self.skip_para_tag:
 542             self.body.append(self.start_para % '.body')
 543
 544     def depart_paragraph(self, node):
 545         if not self.skip_para_tag:
 546             self.body.append(self.end_para)
 547
 548     def visit_problematic(self, node):
 549         if node.hasattr('refid'):
 550             self.body.append('<a href="#%s" name="%s">' % (node['refid'],
 551                                                            node['id']))
 552             self.context.append('</a>')
 553         else:
 554             self.context.append('')
 555         self.body.append(self.starttag(node, 'span', '', CLASS='problematic'))
 556
 557     def depart_problematic(self, node):
 558         self.body.append('</span>')
 559         self.body.append(self.context.pop())
 560
 561     def visit_raw(self, node):
 562         if node.has_key('format') and node['format'] == 'html':
 563             self.body.append(node.astext())
 564         raise nodes.SkipNode
 565
 566     def visit_reference(self, node):
 567         pass
 568
 569     def depart_reference(self, node):
 570         pass
 571
 572     def visit_row(self, node):
 573         self.body.append(self.starttag(node, 'tr', ''))
 574
 575     def depart_row(self, node):
 576         self.body.append('</tr>\n')
 577
 578     def visit_section(self, node):
 579         self.section_level += 1
 580
 581     def depart_section(self, node):
 582         self.section_level -= 1
 583
 584     def visit_strong(self, node):
 585         self.body.append('<strong>')
 586
 587     def depart_strong(self, node):
 588         self.body.append('</strong>')
 589
 590     def visit_table(self, node):
 591         self.body.append(
 592               self.starttag(node, 'table', CLASS="table",
 593                             frame='border', rules='all'))
 594
 595     def depart_table(self, node):
 596         self.body.append('</table>\n')
 597
 598     def visit_target(self, node):
 599         if not (node.has_key('refuri') or node.has_key('refid')
 600                 or node.has_key('refname')):
 601             self.body.append(self.starttag(node, 'a', '', CLASS='target'))
 602             self.context.append('</a>')
 603         else:
 604             self.context.append('')
 605
 606     def depart_target(self, node):
 607         self.body.append(self.context.pop())
 608
 609     def visit_tbody(self, node):
 610         self.write_colspecs()
 611         self.body.append(self.context.pop()) # '</colgroup>\n' or ''
 612         self.body.append(self.starttag(node, 'tbody', valign='top'))
 613
 614     def depart_tbody(self, node):
 615         self.body.append('</tbody>\n')
 616
 617     def visit_term(self, node):
 618         self.body.append(self.starttag(node, 'dt', ''))
 619
 620     def depart_term(self, node):
 621         """
 622         Leave the end tag to `self.visit_definition()`, in case there's a
 623         classifier.
 624         """
 625         pass
 626
 627     def visit_tgroup(self, node):
 628         # Mozilla needs <colgroup>:
 629         self.body.append(self.starttag(node, 'colgroup'))
 630         # Appended by thead or tbody:
 631         self.context.append('</colgroup>\n')
 632
 633     def depart_tgroup(self, node):
 634         pass
 635
 636     def visit_thead(self, node):
 637         self.write_colspecs()
 638         self.body.append(self.context.pop()) # '</colgroup>\n'
 639         # There may or may not be a <thead>; this is for <tbody> to use:
 640         self.context.append('')
 641         self.body.append(self.starttag(node, 'thead', valign='bottom'))
 642
 643     def depart_thead(self, node):
 644         self.body.append('</thead>\n')
 645
 646     def visit_tip(self, node):
 647         self.visit_admonition(node, 'tip')
 648
 649     def depart_tip(self, node):
 650         self.depart_admonition()
 651
 652     def visit_title(self, node):
 653         """Only 3 section levels are supported by this writer."""
 654         title_tag = self.start_para % section_styles[self.section_level]
 655         self.body.append(title_tag)
 656
 657     def depart_title(self, node):
 658         self.body.append(self.end_para)
 659
 660     def visit_warning(self, node):
 661         self.visit_admonition(node, 'warning')
 662
 663     def depart_warning(self, node):
 664         self.depart_admonition()
 665
 666     def visit_system_message(self, node):
 667         print node.astext()
 668
 669     def depart_system_message(self, node):
 670         pass
 671
 672     def unknown_visit(self, node):
 673         print "Failure processing at line", node.line
 674         print "Failure is", node.astext()
 675         raise NotImplementedError('visiting unimplemented node type: %s'
 676                                   % node.__class__.__name__)