import cStringIO, cgi, sys, urllib import dps.utils try: from restructuredtext import Parser except ImportError: from dps.parsers.restructuredtext import Parser # TODO: enforce model? class DumbHTMLFormatter: def __init__(self): self.out = cStringIO.StringIO() self.w = self.out.write self.section = 0 self.closers = [] def format(self, node): '''Format a node ''' for entry in node: self.formatOneTag(entry) def formatOneTag(self, tag): if tag.tagname == '#text': meth = self.format__text else: meth = getattr(self, 'format_'+tag.tagname) meth(tag) # # Root Element # # ((title, subtitle?)?, docinfo?, %structure.model;) # def format_document(self, document): ''' ((title, subtitle?)?, docinfo?, %structure.model;) ''' self.document = document self.w('\n') n = 0 # See if there's a title if document[n].tagname == 'title': title = cgi.escape(document[n][0][0].data) self.w('%s\n'%title) n += 1 if document[n].tagname == 'subtitle': title = cgi.escape(document[n][0][0].data) self.w('

%s

'%title) self.section += 1 n += 1 # Now see if there's biblio information # see if there's a field_list at the start of the document if document[n].tagname == 'docinfo': self.format_docinfo(document[n]) n += 1 self.w('\n') # now for the body l = list(document) for entry in l[n:]: self.formatOneTag(entry) self.w('\n') return self.out.getvalue() # # Title Elements # def format_title(self, node): self.w(''%self.section) if node.children: self.format(node) self.w('\n'%self.section) def format_subtitle(self, node): raise NotImplementedError, node # # Bibliographic Elements # def format_docinfo(self, node): ''' (((%bibliographic.elements;)+, abstract?) | abstract) bibliographic.elements: author | authors | organization | contact | version | revision | status | date | copyright ''' if node.children: self.format(node) def format_abstract(self, node): content = urllib.quote(node[0].data) self.w('\n'%content) def format_author(self, node): content = urllib.quote(node[0].data) self.w('\n'%content) def format_authors(self, node): ''' ((author, organization?, contact?)+) ''' self.w('\n'%content) def format_organization(self, node): content = urllib.quote(node[0].data) self.w('\n'%content) # TODO: not in DTD # def format_keywords(self, node): # content = urllib.quote(node[0].data) # self.w('\n'%content) def format_contact(self, node): addr = urllib.quote(node[0].data) self.w('\n'%content) def format_revision(self, node): addr = urllib.quote(node[0].data) self.w('\n'%content) def format_status(self, node): addr = urllib.quote(node[0].data) self.w('\n'%content) def format_date(self, node): addr = urllib.quote(node[0].data) self.w('\n'%content) def format_copyright(self, node): addr = urllib.quote(node[0].data) self.w('\n'%content) # # Structural Elements # # section # # structure.model: # ( ((%body.elements; | transition)+, (%structural.elements;)*) # | (%structural.elements;)+ ) # def format_section(self, node): self.w(''%urllib.quote(node.attributes['name'])) self.section += 1 if node.children: self.format(node) self.section -= 1 def format_transition(self, node): self.w('
') # # Body Elements # # paragraph | literal_block | block_quote | doctest_block| table # | figure | image | footnote # | bullet_list | enumerated_list | definition_list | field_list # | option_list # | note | tip | hint | warning | error | caution | danger | important # | target | substitution_definition | comment | system_message # # def format_paragraph(self, node): ''' %text.model; ''' # TODO: there are situations where the

are unnecessary self.w('

') if node.children: self.format(node) self.w('

\n') # Simple lists def format_bullet_list(self, node): ''' (list_item+) bullet CDATA ''' # TODO: handle attribute self.w('\n') def format_enumerated_list(self, node): ''' (list_item+) enumtype (arabic | loweralpha | upperalpha | lowerroman | upperroman) prefix CDATA suffix CDATA start CDATA ''' # TODO: handle attributes self.w('
    \n') if node.children: self.format(node) self.w('
\n') def format_list_item(self, node): ''' (%body.elements;)+ ''' self.w('
  • ') if node.children: self.format(node) self.w('
    • \n') # Definition List def format_definition_list(self, node): ''' (definition_list_item+) ''' self.w('
    \n') if node.children: self.format(node) self.w('
    \n') def format_definition_list_item(self, node): ''' (term, classifier?, definition) ''' self.w('
    ') if node.children: self.format(node) def format_term(self, node): ''' %text.model; ''' self.w('') if node.children:self.format(node) self.w('') def format_classifier(self, node): ''' %text.model; ''' # TODO: handle the classifier better self.w('') if node.children: self.format(node) self.w('') def format_definition(self, node): ''' (%body.elements;)+ ''' self.w('
    \n
    ') # TODO: this is way suboptimal! first = 1 for child in node.children: if child.tagname == 'paragraph' and first: # just format the contents of the para self.format(child) else: # format the whole tag self.formatOneTag(child) first = 0 self.w('
    \n') # Field List def format_field_list(self, node): ''' (field+) ''' self.w('
    ') if node.children: self.format(node) self.w('
    ') def format_field(self, node): ''' (field_name, field_argument*, field_body) ''' self.w('
    ') if node.children: self.format(node) def format_field_name(self, node): ''' (#PCDATA) ''' self.w('') if node.children:self.format(node) self.w('') def format_field_argument(self, node): ''' (#PCDATA) ''' self.w('') if node.children: self.format(node) self.w('') def format_field_body(self, node): ''' (%body.elements;)+ ''' self.w('
    \n
    ') if node.children: self.format(node) self.w('
    \n') # Option List def format_option_list(self, node): ''' (option_list_item+) ''' self.w('\n') self.w('\n') if node.children: self.format(node) self.w('
    OptionDescription
    \n') def format_option_list_item(self, node): ''' (option+, description) ''' self.w('') if node.children: self.format(node) self.w('\n') def format_option(self, node): ''' ((short_option | long_option | vms_option), option_argument?) ''' self.w('') if node.children: self.format(node) self.w('') def format_short_option(self, node): ''' (#PCDATA) ''' for option in node.children: self.w('-%s'%cgi.escape(option.data)) def format_long_option(self, node): ''' (#PCDATA) ''' for option in node.children: self.w('--%s'%cgi.escape(option.data)) def format_vms_option(self, node): ''' (#PCDATA) ''' for option in node.children: self.w('/%s'%cgi.escape(option.data)) def format_option_argument(self, node): ''' (#PCDATA) ''' for option in node.children: self.w('=%s'%cgi.escape(option.data)) def format_description(self, node): ''' (%body.elements;)+ ''' self.w('') if node.children: self.format(node) self.w('\n') # Literal Block def format_literal_block(self, node): self.w('
    ')
        if node.children: self.format(node)
        self.w('
    \n') # Block Quote def format_block_quote(self, node): # TODO: I believe this needs to be CSS'ified - blockquote is deprecated self.w('
    ') if node.children: self.format(node) self.w('
    \n') # Doctest Block def format_doctest_block(self, node): self.w('
    ')
        if node.children: self.format(node)
        self.w('
    \n') # Note, tip, hint, warning, error, caution, danger, important def format_note(self, node): ''' (%body.elements;)+ ''' self.w('') if node.children: self.format(node) self.w('') def format_tip(self, node): ''' (%body.elements;)+ ''' self.w('') if node.children: self.format(node) self.w('') def format_hint(self, node): ''' (%body.elements;)+ ''' self.w('') if node.children: self.format(node) self.w('') def format_warning(self, node): ''' (%body.elements;)+ ''' self.w('') if node.children: self.format(node) self.w('') def format_error(self, node): ''' (%body.elements;)+ ''' self.w('') if node.children: self.format(node) self.w('') def format_caution(self, node): ''' (%body.elements;)+ ''' self.w('') if node.children: self.format(node) self.w('') def format_danger(self, node): ''' (%body.elements;)+ ''' self.w('') if node.children: self.format(node) self.w('') def format_important(self, node): ''' (%body.elements;)+ ''' self.w('') if node.children: self.format(node) self.w('') # Footnote def format_footnote(self, node): ''' (label?, (%body.elements;)+) %auto.att; ''' raise NotImplementedError, node def format_label(self, node): ''' (#PCDATA) ''' for label in node.children: self.w(cgi.escape(label.data)) # Target def format_target(self, node): ''' (%text.model;) %reference.atts; %anonymous.att; ''' pass # Substitution Definition def format_substitution_definition(self, node): ''' (%text.model;) ''' raise NotImplementedError, node # Comment def format_comment(self, node): ''' (#PCDATA) %fixedspace.att; ''' # TODO: handle attrs self.w('') # Figure def format_figure(self, node): ''' (image, ((caption, legend?) | legend) ''' raise NotImplementedError, node def format_image(self, node): ''' EMPTY uri CDATA #REQUIRED alt CDATA #IMPLIED height NMTOKEN #IMPLIED width NMTOKEN #IMPLIED scale NMTOKEN #IMPLIED ''' attrs = node.attributes l = ['src="%(uri)s"'%attrs] if attrs.has_key('alt'): l.append('alt="%(alt)s"'%attrs) if attrs.has_key('alt'): l.append('alt="%(alt)s"'%attrs) if attrs.has_key('height'): l.append('height="%(height)s"'%attrs) if attrs.has_key('width'): l.append('width="%(width)s"'%attrs) # TODO: scale self.w(''%(' '.join(l))) def format_caption(self, node): ''' %text.model; ''' raise NotImplementedError, node def format_legend(self, node): ''' (%body.elements;)+ ''' raise NotImplementedError, node # System Message def format_system_message(self, node): ''' (%body.elements;)+ level NMTOKEN #IMPLIED type CDATA #IMPLIED ''' self.w(''%node.attributes['type']) if node.children: self.format(node) self.w('') # # Tables: # NOT IN DOM YET # def format_table(self, node): ''' +------------------------+------------+----------+----------+ | Header row, column 1 | Header 2 | Header 3 | Header 4 | | (header rows optional) | | | | +========================+============+==========+==========+ | body row 1, column 1 | column 2 | column 3 | column 4 | +------------------------+------------+----------+----------+ | body row 2 | Cells may span columns. | +------------------------+------------+---------------------+ | body row 3 | Cells may | - Table cells | +------------------------+ span rows. | - contain | | body row 4 | | - body elements. | +------------------------+------------+---------------------+ ''' self.w('\n') if node.children: self.format(node) self.w('
    \n') def format_tgroup(self, node): # we get the number of columns, if that's important if node.children: self.format(node) def format_colspec(self, node): # we get colwidth, but don't need it pass def format_thead(self, node): for row in node.children: self.w('') for cell in row.children: s = '' attrs = cell.attributes if attrs.has_key('morecols'): s = s + ' colspan=%d'%(attrs['morecols']+1) if attrs.has_key('morerows'): s = s + ' rowspan=%d'%(attrs['morerows']+1) self.w(''%s) if cell.children: self.format(cell) self.w('\n') self.w('\n') def format_tbody(self, node): for row in node.children: self.w('') for cell in row.children: s = '' attrs = cell.attributes if attrs.has_key('morecols'): s = s + ' colspan=%d'%(attrs['morecols']+1) if attrs.has_key('morerows'): s = s + ' rowspan=%d'%(attrs['morerows']+1) self.w(''%s) if cell.children: self.format(cell) self.w('\n') self.w('\n') # # Inline Elements # # Inline elements occur within the text contents of body elements. Some # nesting of inline elements is allowed by these definitions, with the # following caveats: # - An inline element may not contain a nested element of the same type # (e.g. may not contain another ). # - Nested inline elements may or may not be supported by individual # applications using this DTD. # - The inline elements , , and do # not support nesting. # # What that means is that all of these take (%text.model;) except: # literal (#PCDATA) # footnote_reference (#PCDATA) # # text.model: # (#PCDATA | %inline.elements;)* # def format_emphasis(self, node): ''' (%text.model;) ''' self.w('') if node.children: self.format(node) self.w('') def format_strong(self, node): ''' (%text.model;) ''' self.w('') if node.children: self.format(node) self.w('') def format_interpreted(self, node): ''' (%text.model;) type CDATA #IMPLIED ''' pass #raise NotImplementedError, node def format_literal(self, node): ''' (#PCDATA) ''' self.w('') for literal in node.children: self.w(cgi.escape(literal.data)) self.w('') def format_reference(self, node): ''' (%text.model;) %reference.atts; %anonymous.att; ''' attrs = node.attributes doc = self.document ok = 1 print node if attrs.has_key('refuri'): self.w(''%attrs['refuri']) elif doc.explicit_targets.has_key(attrs['refname']): # an external reference has been defined ref = doc.explicit_targets[attrs['refname']] if ref.attributes.has_key('refuri'): self.w(''%ref.attributes['refuri']) else: self.w(''%attrs['refname']) elif doc.implicit_targets.has_key(attrs['refname']): # internal reference name = attrs['refname'] self.w(''%urllib.quote(name)) else: ok = 0 self.w('target "%s" ' 'undefined'%attrs['refname']) if node.children: self.format(node) if ok: self.w('') def format_footnote_reference(self, node): ''' (#PCDATA) %reference.atts; %auto.att; ''' raise NotImplementedError, node def format_substitution_reference(self, node): ''' (%text.model;) %refname.att; ''' raise NotImplementedError, node def format_problematic(self, node): ''' (%text.model;) ''' raise NotImplementedError, node # # Finally, #text # def format__text(self, node): self.w(cgi.escape(node.data)) def main(filename, debug=0): parser = Parser() input = open(filename).read() document = dps.utils.newdocument() parser.parse(input, document) if debug == 1: print document.pformat() else: formatter = DumbHTMLFormatter() print formatter.format_document(document) if __name__ == '__main__': if len(sys.argv) > 2: main(sys.argv[1], debug=1) else: main(sys.argv[1])