minirst: add basic HTML formatting support

mercurial/minirst.py

@@ -1,579 +1,666 b''
 # minirst.py - minimal reStructuredText parser
 #
 # Copyright 2009, 2010 Matt Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 """simplified reStructuredText parser.
 
 This parser knows just enough about reStructuredText to parse the
 Mercurial docstrings.
 
 It cheats in a major way: nested blocks are not really nested. They
 are just indented blocks that look like they are nested. This relies
 on the user to keep the right indentation for the blocks.
 
 Remember to update http://mercurial.selenic.com/wiki/HelpStyleGuide
 when adding support for new constructs.
 """
 
 import re
 import util, encoding
 from i18n import _
 
 def replace(text, substs):
     for f, t in substs:
         text = text.replace(f, t)
     return text
 
 _blockre = re.compile(r"\n(?:\s*\n)+")
 
 def findblocks(text):
     """Find continuous blocks of lines in text.
 
     Returns a list of dictionaries representing the blocks. Each block
     has an 'indent' field and a 'lines' field.
     """
     blocks = []
     for b in _blockre.split(text.lstrip('\n').rstrip()):
         lines = b.splitlines()
         if lines:
             indent = min((len(l) - len(l.lstrip())) for l in lines)
             lines = [l[indent:] for l in lines]
             blocks.append(dict(indent=indent, lines=lines))
     return blocks
 
 def findliteralblocks(blocks):
     """Finds literal blocks and adds a 'type' field to the blocks.
 
     Literal blocks are given the type 'literal', all other blocks are
     given type the 'paragraph'.
     """
     i = 0
     while i < len(blocks):
         # Searching for a block that looks like this:
         #
         # +------------------------------+
         # | paragraph                    |
         # | (ends with "::")             |
         # +------------------------------+
         #    +---------------------------+
         #    | indented literal block    |
         #    +---------------------------+
         blocks[i]['type'] = 'paragraph'
         if blocks[i]['lines'][-1].endswith('::') and i + 1 < len(blocks):
             indent = blocks[i]['indent']
             adjustment = blocks[i + 1]['indent'] - indent
 
             if blocks[i]['lines'] == ['::']:
                 # Expanded form: remove block
                 del blocks[i]
                 i -= 1
             elif blocks[i]['lines'][-1].endswith(' ::'):
                 # Partially minimized form: remove space and both
                 # colons.
                 blocks[i]['lines'][-1] = blocks[i]['lines'][-1][:-3]
             else:
                 # Fully minimized form: remove just one colon.
                 blocks[i]['lines'][-1] = blocks[i]['lines'][-1][:-1]
 
             # List items are formatted with a hanging indent. We must
             # correct for this here while we still have the original
             # information on the indentation of the subsequent literal
             # blocks available.
             m = _bulletre.match(blocks[i]['lines'][0])
             if m:
                 indent += m.end()
                 adjustment -= m.end()
 
             # Mark the following indented blocks.
             while i + 1 < len(blocks) and blocks[i + 1]['indent'] > indent:
                 blocks[i + 1]['type'] = 'literal'
                 blocks[i + 1]['indent'] -= adjustment
                 i += 1
         i += 1
     return blocks
 
 _bulletre = re.compile(r'(-|[0-9A-Za-z]+\.|\(?[0-9A-Za-z]+\)|\|) ')
 _optionre = re.compile(r'^(-([a-zA-Z0-9]), )?(--[a-z0-9-]+)'
                        r'((.*)  +)(.*)$')
 _fieldre = re.compile(r':(?![: ])([^:]*)(?<! ):[ ]+(.*)')
 _definitionre = re.compile(r'[^ ]')
 _tablere = re.compile(r'(=+\s+)*=+')
 
 def splitparagraphs(blocks):
     """Split paragraphs into lists."""
     # Tuples with (list type, item regexp, single line items?). Order
     # matters: definition lists has the least specific regexp and must
     # come last.
     listtypes = [('bullet', _bulletre, True),
                  ('option', _optionre, True),
                  ('field', _fieldre, True),
                  ('definition', _definitionre, False)]
 
     def match(lines, i, itemre, singleline):
         """Does itemre match an item at line i?
 
         A list item can be followed by an idented line or another list
         item (but only if singleline is True).
         """
         line1 = lines[i]
         line2 = i + 1 < len(lines) and lines[i + 1] or ''
         if not itemre.match(line1):
             return False
         if singleline:
             return line2 == '' or line2[0] == ' ' or itemre.match(line2)
         else:
             return line2.startswith(' ')
 
     i = 0
     while i < len(blocks):
         if blocks[i]['type'] == 'paragraph':
             lines = blocks[i]['lines']
             for type, itemre, singleline in listtypes:
                 if match(lines, 0, itemre, singleline):
                     items = []
                     for j, line in enumerate(lines):
                         if match(lines, j, itemre, singleline):
                             items.append(dict(type=type, lines=[],
                                               indent=blocks[i]['indent']))
                         items[-1]['lines'].append(line)
                     blocks[i:i + 1] = items
                     break
         i += 1
     return blocks
 
 _fieldwidth = 12
 
 def updatefieldlists(blocks):
     """Find key and maximum key width for field lists."""
     i = 0
     while i < len(blocks):
         if blocks[i]['type'] != 'field':
             i += 1
             continue
 
         keywidth = 0
         j = i
         while j < len(blocks) and blocks[j]['type'] == 'field':
             m = _fieldre.match(blocks[j]['lines'][0])
             key, rest = m.groups()
             blocks[j]['lines'][0] = rest
             blocks[j]['key'] = key
             keywidth = max(keywidth, len(key))
             j += 1
 
         for block in blocks[i:j]:
             block['keywidth'] = keywidth
         i = j + 1
 
     return blocks
 
 def updateoptionlists(blocks):
     i = 0
     while i < len(blocks):
         if blocks[i]['type'] != 'option':
             i += 1
             continue
 
         optstrwidth = 0
         j = i
         while j < len(blocks) and blocks[j]['type'] == 'option':
             m = _optionre.match(blocks[j]['lines'][0])
 
             shortoption = m.group(2)
             group3 = m.group(3)
             longoption = group3[2:].strip()
             desc = m.group(6).strip()
             longoptionarg = m.group(5).strip()
             blocks[j]['lines'][0] = desc
 
             noshortop = ''
             if not shortoption:
                 noshortop = '   '
 
             opt = "%s%s" %   (shortoption and "-%s " % shortoption or '',
                             ("%s--%s %s") % (noshortop, longoption,
                                              longoptionarg))
             opt = opt.rstrip()
             blocks[j]['optstr'] = opt
             optstrwidth = max(optstrwidth, encoding.colwidth(opt))
             j += 1
 
         for block in blocks[i:j]:
             block['optstrwidth'] = optstrwidth
         i = j + 1
     return blocks
 
 def prunecontainers(blocks, keep):
     """Prune unwanted containers.
 
     The blocks must have a 'type' field, i.e., they should have been
     run through findliteralblocks first.
     """
     pruned = []
     i = 0
     while i + 1 < len(blocks):
         # Searching for a block that looks like this:
         #
         # +-------+---------------------------+
         # | ".. container ::" type            |
         # +---+                               |
         #     | blocks                        |
         #     +-------------------------------+
         if (blocks[i]['type'] == 'paragraph' and
             blocks[i]['lines'][0].startswith('.. container::')):
             indent = blocks[i]['indent']
             adjustment = blocks[i + 1]['indent'] - indent
             containertype = blocks[i]['lines'][0][15:]
             prune = containertype not in keep
             if prune:
                 pruned.append(containertype)
 
             # Always delete "..container:: type" block
             del blocks[i]
             j = i
             i -= 1
             while j < len(blocks) and blocks[j]['indent'] > indent:
                 if prune:
                     del blocks[j]
                 else:
                     blocks[j]['indent'] -= adjustment
                     j += 1
         i += 1
     return blocks, pruned
 
 _sectionre = re.compile(r"""^([-=`:.'"~^_*+#])\1+$""")
 
 def findtables(blocks):
     '''Find simple tables
 
        Only simple one-line table elements are supported
     '''
 
     for block in blocks:
         # Searching for a block that looks like this:
         #
         # === ==== ===
         #  A    B   C
         # === ==== ===  <- optional
         #  1    2   3
         #  x    y   z
         # === ==== ===
         if (block['type'] == 'paragraph' and
             len(block['lines']) > 2 and
             _tablere.match(block['lines'][0]) and
             block['lines'][0] == block['lines'][-1]):
             block['type'] = 'table'
             block['header'] = False
             div = block['lines'][0]
 
             # column markers are ASCII so we can calculate column
             # position in bytes
             columns = [x for x in xrange(len(div))
                        if div[x] == '=' and (x == 0 or div[x - 1] == ' ')]
             rows = []
             for l in block['lines'][1:-1]:
                 if l == div:
                     block['header'] = True
                     continue
                 row = []
                 # we measure columns not in bytes or characters but in
                 # colwidth which makes things tricky
                 pos = columns[0] # leading whitespace is bytes
                 for n, start in enumerate(columns):
                     if n + 1 < len(columns):
                         width = columns[n + 1] - start
                         v = encoding.getcols(l, pos, width) # gather columns
                         pos += len(v) # calculate byte position of end
                         row.append(v.strip())
                     else:
                         row.append(l[pos:].strip())
                 rows.append(row)
 
             block['table'] = rows
 
     return blocks
 
 def findsections(blocks):
     """Finds sections.
 
     The blocks must have a 'type' field, i.e., they should have been
     run through findliteralblocks first.
     """
     for block in blocks:
         # Searching for a block that looks like this:
         #
         # +------------------------------+
         # | Section title                |
         # | -------------                |
         # +------------------------------+
         if (block['type'] == 'paragraph' and
             len(block['lines']) == 2 and
             encoding.colwidth(block['lines'][0]) == len(block['lines'][1]) and
             _sectionre.match(block['lines'][1])):
             block['underline'] = block['lines'][1][0]
             block['type'] = 'section'
             del block['lines'][1]
     return blocks
 
 def inlineliterals(blocks):
     substs = [('``', '"')]
     for b in blocks:
         if b['type'] in ('paragraph', 'section'):
             b['lines'] = [replace(l, substs) for l in b['lines']]
     return blocks
 
 def hgrole(blocks):
     substs = [(':hg:`', '"hg '), ('`', '"')]
     for b in blocks:
         if b['type'] in ('paragraph', 'section'):
             # Turn :hg:`command` into "hg command". This also works
             # when there is a line break in the command and relies on
             # the fact that we have no stray back-quotes in the input
             # (run the blocks through inlineliterals first).
             b['lines'] = [replace(l, substs) for l in b['lines']]
     return blocks
 
 def addmargins(blocks):
     """Adds empty blocks for vertical spacing.
 
     This groups bullets, options, and definitions together with no vertical
     space between them, and adds an empty block between all other blocks.
     """
     i = 1
     while i < len(blocks):
         if (blocks[i]['type'] == blocks[i - 1]['type'] and
             blocks[i]['type'] in ('bullet', 'option', 'field')):
             i += 1
         else:
             blocks.insert(i, dict(lines=[''], indent=0, type='margin'))
             i += 2
     return blocks
 
 def prunecomments(blocks):
     """Remove comments."""
     i = 0
     while i < len(blocks):
         b = blocks[i]
         if b['type'] == 'paragraph' and (b['lines'][0].startswith('.. ') or
                                          b['lines'] == ['..']):
             del blocks[i]
             if i < len(blocks) and blocks[i]['type'] == 'margin':
                 del blocks[i]
         else:
             i += 1
     return blocks
 
 _admonitionre = re.compile(r"\.\. (admonition|attention|caution|danger|"
                            r"error|hint|important|note|tip|warning)::",
                            flags=re.IGNORECASE)
 
 def findadmonitions(blocks):
     """
     Makes the type of the block an admonition block if
     the first line is an admonition directive
     """
     i = 0
     while i < len(blocks):
         m = _admonitionre.match(blocks[i]['lines'][0])
         if m:
             blocks[i]['type'] = 'admonition'
             admonitiontitle = blocks[i]['lines'][0][3:m.end() - 2].lower()
 
             firstline = blocks[i]['lines'][0][m.end() + 1:]
             if firstline:
                 blocks[i]['lines'].insert(1, '   ' + firstline)
 
             blocks[i]['admonitiontitle'] = admonitiontitle
             del blocks[i]['lines'][0]
         i = i + 1
     return blocks
 
 _admonitiontitles = {'attention': _('Attention:'),
                      'caution': _('Caution:'),
                      'danger': _('!Danger!')  ,
                      'error': _('Error:'),
                      'hint': _('Hint:'),
                      'important': _('Important:'),
                      'note': _('Note:'),
                      'tip': _('Tip:'),
                      'warning': _('Warning!')}
 
 def formatoption(block, width):
     desc = ' '.join(map(str.strip, block['lines']))
     colwidth = encoding.colwidth(block['optstr'])
     usablewidth = width - 1
     hanging = block['optstrwidth']
     initindent = '%s%s  ' % (block['optstr'], ' ' * ((hanging - colwidth)))
     hangindent = ' ' * (encoding.colwidth(initindent) + 1)
     return ' %s\n' % (util.wrap(desc, usablewidth,
                                            initindent=initindent,
                                            hangindent=hangindent))
 
 def formatblock(block, width):
     """Format a block according to width."""
     if width <= 0:
         width = 78
     indent = ' ' * block['indent']
     if block['type'] == 'admonition':
         admonition = _admonitiontitles[block['admonitiontitle']]
         hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip())
 
         defindent = indent + hang * ' '
         text = ' '.join(map(str.strip, block['lines']))
         return '%s\n%s\n' % (indent + admonition,
                              util.wrap(text, width=width,
                                        initindent=defindent,
                                        hangindent=defindent))
     if block['type'] == 'margin':
         return '\n'
     if block['type'] == 'literal':
         indent += '  '
         return indent + ('\n' + indent).join(block['lines']) + '\n'
     if block['type'] == 'section':
         underline = encoding.colwidth(block['lines'][0]) * block['underline']
         return "%s%s\n%s%s\n" % (indent, block['lines'][0],indent, underline)
     if block['type'] == 'table':
         table = block['table']
         # compute column widths
         widths = [max([encoding.colwidth(e) for e in c]) for c in zip(*table)]
         text = ''
         span = sum(widths) + len(widths) - 1
         indent = ' ' * block['indent']
         hang = ' ' * (len(indent) + span - widths[-1])
 
         for row in table:
             l = []
             for w, v in zip(widths, row):
                 pad = ' ' * (w - encoding.colwidth(v))
                 l.append(v + pad)
             l = ' '.join(l)
             l = util.wrap(l, width=width, initindent=indent, hangindent=hang)
             if not text and block['header']:
                 text = l + '\n' + indent + '-' * (min(width, span)) + '\n'
             else:
                 text += l + "\n"
         return text
     if block['type'] == 'definition':
         term = indent + block['lines'][0]
         hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip())
         defindent = indent + hang * ' '
         text = ' '.join(map(str.strip, block['lines'][1:]))
         return '%s\n%s\n' % (term, util.wrap(text, width=width,
                                              initindent=defindent,
                                              hangindent=defindent))
     subindent = indent
     if block['type'] == 'bullet':
         if block['lines'][0].startswith('| '):
             # Remove bullet for line blocks and add no extra
             # indention.
             block['lines'][0] = block['lines'][0][2:]
         else:
             m = _bulletre.match(block['lines'][0])
             subindent = indent + m.end() * ' '
     elif block['type'] == 'field':
         keywidth = block['keywidth']
         key = block['key']
 
         subindent = indent + _fieldwidth * ' '
         if len(key) + 2 > _fieldwidth:
             # key too large, use full line width
             key = key.ljust(width)
         elif keywidth + 2 < _fieldwidth:
             # all keys are small, add only two spaces
             key = key.ljust(keywidth + 2)
             subindent = indent + (keywidth + 2) * ' '
         else:
             # mixed sizes, use fieldwidth for this one
             key = key.ljust(_fieldwidth)
         block['lines'][0] = key + block['lines'][0]
     elif block['type'] == 'option':
         return formatoption(block, width)
 
     text = ' '.join(map(str.strip, block['lines']))
     return util.wrap(text, width=width,
                      initindent=indent,
                      hangindent=subindent) + '\n'
 
+def formathtml(blocks):
+    """Format RST blocks as HTML"""
+
+    out = []
+    headernest = ''
+    listnest = []
+
+    def openlist(start, level):
+        if not listnest or listnest[-1][0] != start:
+            listnest.append((start, level))
+            out.append('<%s>\n' % start)
+
+    blocks = [b for b in blocks if b['type'] != 'margin']
+
+    for pos, b in enumerate(blocks):
+        btype = b['type']
+        level = b['indent']
+        lines = b['lines']
+
+        if btype == 'admonition':
+            admonition = _admonitiontitles[b['admonitiontitle']]
+            text = ' '.join(map(str.strip, lines))
+            out.append('<p>\n<b>%s</b> %s\n</p>\n' % (admonition, text))
+        elif btype == 'paragraph':
+            out.append('<p>\n%s\n</p>\n' % '\n'.join(lines))
+        elif btype == 'margin':
+            pass
+        elif btype == 'literal':
+            out.append('<pre>\n%s\n</pre>\n' % '\n'.join(lines))
+        elif btype == 'section':
+            i = b['underline']
+            if i not in headernest:
+                headernest += i
+            level = headernest.index(i) + 1
+            out.append('<h%d>%s</h%d>\n' % (level, lines[0], level))
+        elif btype == 'table':
+            table = b['table']
+            t = []
+            for row in table:
+                l = []
+                for v in zip(row):
+                    if not t:
+                        l.append('<th>%s</th>' % v)
+                    else:
+                        l.append('<td>%s</td>' % v)
+                t.append(' <tr>%s</tr>\n' % ''.join(l))
+            out.append('<table>\n%s</table>\n' % ''.join(t))
+        elif btype == 'definition':
+            openlist('dl', level)
+            term = lines[0]
+            text = ' '.join(map(str.strip, lines[1:]))
+            out.append(' <dt>%s\n <dd>%s\n' % (term, text))
+        elif btype == 'bullet':
+            bullet, head = lines[0].split(' ', 1)
+            if bullet == '-':
+                openlist('ul', level)
+            else:
+                openlist('ol', level)
+            out.append(' <li> %s\n' % ' '.join([head] + lines[1:]))
+        elif btype == 'field':
+            openlist('dl', level)
+            key = b['key']
+            text = ' '.join(map(str.strip, lines))
+            out.append(' <dt>%s\n <dd>%s\n' % (key, text))
+        elif btype == 'option':
+            openlist('dl', level)
+            opt = b['optstr']
+            desc = ' '.join(map(str.strip, lines))
+            out.append(' <dt>%s\n <dd>%s\n' % (opt, desc))
+
+        # close lists if indent level of next block is lower
+        if listnest:
+            start, level = listnest[-1]
+            if pos == len(blocks) - 1:
+                out.append('</%s>\n' % start)
+                listnest.pop()
+            else:
+                nb = blocks[pos + 1]
+                ni = nb['indent']
+                if (ni < level or
+                    (ni == level and
+                     nb['type'] not in 'definition bullet field option')):
+                    out.append('</%s>\n' % start)
+                    listnest.pop()
+
+    return ''.join(out)
+
 def parse(text, indent=0, keep=None):
     """Parse text into a list of blocks"""
     pruned = []
     blocks = findblocks(text)
     for b in blocks:
         b['indent'] += indent
     blocks = findliteralblocks(blocks)
     blocks = findtables(blocks)
     blocks, pruned = prunecontainers(blocks, keep or [])
     blocks = findsections(blocks)
     blocks = inlineliterals(blocks)
     blocks = hgrole(blocks)
     blocks = splitparagraphs(blocks)
     blocks = updatefieldlists(blocks)
     blocks = updateoptionlists(blocks)
     blocks = addmargins(blocks)
     blocks = prunecomments(blocks)
     blocks = findadmonitions(blocks)
     return blocks, pruned
 
 def formatblocks(blocks, width):
     text = ''.join(formatblock(b, width) for b in blocks)
     return text
 
 def format(text, width, indent=0, keep=None):
     """Parse and format the text according to width."""
     blocks, pruned = parse(text, indent, keep or [])
     text = ''.join(formatblock(b, width) for b in blocks)
     if keep is None:
         return text
     else:
         return text, pruned
 
 def getsections(blocks):
     '''return a list of (section name, nesting level, blocks) tuples'''
     nest = ""
     level = 0
     secs = []
     for b in blocks:
         if b['type'] == 'section':
             i = b['underline']
             if i not in nest:
                 nest += i
             level = nest.index(i) + 1
             nest = nest[:level]
             secs.append((b['lines'][0], level, [b]))
         else:
             if not secs:
                 # add an initial empty section
                 secs = [('', 0, [])]
             secs[-1][2].append(b)
     return secs
 
 def decorateblocks(blocks, width):
     '''generate a list of (section name, line text) pairs for search'''
     lines = []
     for s in getsections(blocks):
         section = s[0]
         text = formatblocks(s[2], width)
         lines.append([(section, l) for l in text.splitlines(True)])
     return lines
 
 def maketable(data, indent=0, header=False):
     '''Generate an RST table for the given table data'''
 
     widths = [max(encoding.colwidth(e) for e in c) for c in zip(*data)]
     indent = ' ' * indent
     div = indent + ' '.join('=' * w for w in widths) + '\n'
 
     out = [div]
     for row in data:
         l = []
         for w, v in zip(widths, row):
             pad = ' ' * (w - encoding.colwidth(v))
             l.append(v + pad)
         out.append(indent + ' '.join(l) + "\n")
     if header and len(data) > 1:
         out.insert(2, div)
     out.append(div)
     return ''.join(out)

tests/test-minirst.py

@@ -1,244 +1,247 b''
 from pprint import pprint
 from mercurial import minirst
 
 def debugformat(title, text, width, **kwargs):
     print "%s formatted to fit within %d characters:" % (title, width)
+    formatted = minirst.format(text, width, **kwargs)
+    html = minirst.formathtml(minirst.parse(text, **kwargs)[0])
     print "-" * 70
-    formatted = minirst.format(text, width, **kwargs)
     if type(formatted) == tuple:
         print formatted[0]
         print "-" * 70
         pprint(formatted[1])
     else:
         print formatted
     print "-" * 70
+    print html
+    print "-" * 70
     print
 
 paragraphs = """
 This is some text in the first paragraph.
 
   A small indented paragraph.
   It is followed by some lines
   containing random whitespace.
  \n  \n   \nThe third and final paragraph.
 """
 
 debugformat('paragraphs', paragraphs, 60)
 debugformat('paragraphs', paragraphs, 30)
 
 
 definitions = """
 A Term
   Definition. The indented
   lines make up the definition.
 Another Term
   Another definition. The final line in the
    definition determines the indentation, so
     this will be indented with four spaces.
 
   A Nested/Indented Term
     Definition.
 """
 
 debugformat('definitions', definitions, 60)
 debugformat('definitions', definitions, 30)
 
 
 literals = r"""
 The fully minimized form is the most
 convenient form::
 
   Hello
     literal
       world
 
 In the partially minimized form a paragraph
 simply ends with space-double-colon. ::
 
   ////////////////////////////////////////
   long un-wrapped line in a literal block
   \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
 
 ::
 
   This literal block is started with '::',
     the so-called expanded form. The paragraph
       with '::' disappears in the final output.
 """
 
 debugformat('literals', literals, 60)
 debugformat('literals', literals, 30)
 
 
 lists = """
 - This is the first list item.
 
   Second paragraph in the first list item.
 
 - List items need not be separated
   by a blank line.
 - And will be rendered without
   one in any case.
 
 We can have indented lists:
 
   - This is an indented list item
 
   - Another indented list item::
 
       - A literal block in the middle
             of an indented list.
 
       (The above is not a list item since we are in the literal block.)
 
 ::
 
   Literal block with no indentation (apart from
   the two spaces added to all literal blocks).
 
 1. This is an enumerated list (first item).
 2. Continuing with the second item.
 
 (1) foo
 (2) bar
 
 1) Another
 2) List
 
 Line blocks are also a form of list:
 
 | This is the first line.
   The line continues here.
 | This is the second line.
 """
 
 debugformat('lists', lists, 60)
 debugformat('lists', lists, 30)
 
 
 options = """
 There is support for simple option lists,
 but only with long options:
 
 -X, --exclude  filter  an option with a short and long option with an argument
 -I, --include          an option with both a short option and a long option
 --all                  Output all.
 --both                 Output both (this description is
                        quite long).
 --long                 Output all day long.
 
 --par                 This option has two paragraphs in its description.
                       This is the first.
 
                       This is the second.  Blank lines may be omitted between
                       options (as above) or left in (as here).
 
 
 The next paragraph looks like an option list, but lacks the two-space
 marker after the option. It is treated as a normal paragraph:
 
 --foo bar baz
 """
 
 debugformat('options', options, 60)
 debugformat('options', options, 30)
 
 
 fields = """
 :a: First item.
 :ab: Second item. Indentation and wrapping
      is handled automatically.
 
 Next list:
 
 :small: The larger key below triggers full indentation here.
 :much too large: This key is big enough to get its own line.
 """
 
 debugformat('fields', fields, 60)
 debugformat('fields', fields, 30)
 
 containers = """
 Normal output.
 
 .. container:: debug
 
    Initial debug output.
 
 .. container:: verbose
 
    Verbose output.
 
    .. container:: debug
 
       Debug output.
 """
 
 debugformat('containers (normal)', containers, 60)
 debugformat('containers (verbose)', containers, 60, keep=['verbose'])
 debugformat('containers (debug)', containers, 60, keep=['debug'])
 debugformat('containers (verbose debug)', containers, 60,
             keep=['verbose', 'debug'])
 
 roles = """Please see :hg:`add`."""
 debugformat('roles', roles, 60)
 
 
 sections = """
 Title
 =====
 
 Section
 -------
 
 Subsection
 ''''''''''
 
 Markup: ``foo`` and :hg:`help`
 ------------------------------
 """
 debugformat('sections', sections, 20)
 
 
 admonitions = """
 .. note::
    This is a note
 
    - Bullet 1
    - Bullet 2
 
    .. warning:: This is a warning Second
       input line of warning
 
 .. danger::
    This is danger
 """
 
 debugformat('admonitions', admonitions, 30)
 
 comments = """
 Some text.
 
 .. A comment
 
    .. An indented comment
 
    Some indented text.
 
 ..
 
 Empty comment above
 """
 
 debugformat('comments', comments, 30)
 
 
 data = [['a', 'b', 'c'],
          ['1', '2', '3'],
          ['foo', 'bar', 'baz this list is very very very long man']]
 
 table = minirst.maketable(data, 2, True)
 
 print table
 
 debugformat('table', table, 30)

tests/test-minirst.py.out

@@ -1,428 +1,785 b''
 paragraphs formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 This is some text in the first paragraph.
 
   A small indented paragraph. It is followed by some lines
   containing random whitespace.
 
 The third and final paragraph.
 
 ----------------------------------------------------------------------
+<p>
+This is some text in the first paragraph.
+</p>
+<p>
+A small indented paragraph.
+It is followed by some lines
+containing random whitespace.
+</p>
+<p>
+The third and final paragraph.
+</p>
+
+----------------------------------------------------------------------
 
 paragraphs formatted to fit within 30 characters:
 ----------------------------------------------------------------------
 This is some text in the first
 paragraph.
 
   A small indented paragraph.
   It is followed by some lines
   containing random
   whitespace.
 
 The third and final paragraph.
 
 ----------------------------------------------------------------------
+<p>
+This is some text in the first paragraph.
+</p>
+<p>
+A small indented paragraph.
+It is followed by some lines
+containing random whitespace.
+</p>
+<p>
+The third and final paragraph.
+</p>
+
+----------------------------------------------------------------------
 
 definitions formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 A Term
   Definition. The indented lines make up the definition.
 
 Another Term
     Another definition. The final line in the definition
     determines the indentation, so this will be indented
     with four spaces.
 
   A Nested/Indented Term
     Definition.
 
 ----------------------------------------------------------------------
+<dl>
+ <dt>A Term
+ <dd>Definition. The indented lines make up the definition.
+ <dt>Another Term
+ <dd>Another definition. The final line in the definition determines the indentation, so this will be indented with four spaces.
+ <dt>A Nested/Indented Term
+ <dd>Definition.
+</dl>
+
+----------------------------------------------------------------------
 
 definitions formatted to fit within 30 characters:
 ----------------------------------------------------------------------
 A Term
   Definition. The indented
   lines make up the
   definition.
 
 Another Term
     Another definition. The
     final line in the
     definition determines the
     indentation, so this will
     be indented with four
     spaces.
 
   A Nested/Indented Term
     Definition.
 
 ----------------------------------------------------------------------
+<dl>
+ <dt>A Term
+ <dd>Definition. The indented lines make up the definition.
+ <dt>Another Term
+ <dd>Another definition. The final line in the definition determines the indentation, so this will be indented with four spaces.
+ <dt>A Nested/Indented Term
+ <dd>Definition.
+</dl>
+
+----------------------------------------------------------------------
 
 literals formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 The fully minimized form is the most convenient form:
 
   Hello
     literal
       world
 
 In the partially minimized form a paragraph simply ends with
 space-double-colon.
 
   ////////////////////////////////////////
   long un-wrapped line in a literal block
   \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
 
   This literal block is started with '::',
     the so-called expanded form. The paragraph
       with '::' disappears in the final output.
 
 ----------------------------------------------------------------------
+<p>
+The fully minimized form is the most
+convenient form:
+</p>
+<pre>
+Hello
+  literal
+    world
+</pre>
+<p>
+In the partially minimized form a paragraph
+simply ends with space-double-colon.
+</p>
+<pre>
+////////////////////////////////////////
+long un-wrapped line in a literal block
+\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
+</pre>
+<pre>
+This literal block is started with '::',
+  the so-called expanded form. The paragraph
+    with '::' disappears in the final output.
+</pre>
+
+----------------------------------------------------------------------
 
 literals formatted to fit within 30 characters:
 ----------------------------------------------------------------------
 The fully minimized form is
 the most convenient form:
 
   Hello
     literal
       world
 
 In the partially minimized
 form a paragraph simply ends
 with space-double-colon.
 
   ////////////////////////////////////////
   long un-wrapped line in a literal block
   \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
 
   This literal block is started with '::',
     the so-called expanded form. The paragraph
       with '::' disappears in the final output.
 
 ----------------------------------------------------------------------
+<p>
+The fully minimized form is the most
+convenient form:
+</p>
+<pre>
+Hello
+  literal
+    world
+</pre>
+<p>
+In the partially minimized form a paragraph
+simply ends with space-double-colon.
+</p>
+<pre>
+////////////////////////////////////////
+long un-wrapped line in a literal block
+\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
+</pre>
+<pre>
+This literal block is started with '::',
+  the so-called expanded form. The paragraph
+    with '::' disappears in the final output.
+</pre>
+
+----------------------------------------------------------------------
 
 lists formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 - This is the first list item.
 
   Second paragraph in the first list item.
 
 - List items need not be separated by a blank line.
 - And will be rendered without one in any case.
 
 We can have indented lists:
 
   - This is an indented list item
   - Another indented list item:
 
       - A literal block in the middle
             of an indented list.
 
       (The above is not a list item since we are in the literal block.)
 
   Literal block with no indentation (apart from
   the two spaces added to all literal blocks).
 
 1. This is an enumerated list (first item).
 2. Continuing with the second item.
 (1) foo
 (2) bar
 1) Another
 2) List
 
 Line blocks are also a form of list:
 
 This is the first line. The line continues here.
 This is the second line.
 
 ----------------------------------------------------------------------
+<ul>
+ <li> This is the first list item.
+<p>
+Second paragraph in the first list item.
+</p>
+ <li> List items need not be separated   by a blank line.
+ <li> And will be rendered without   one in any case.
+</ul>
+<p>
+We can have indented lists:
+</p>
+<ul>
+ <li> This is an indented list item
+ <li> Another indented list item:
+<pre>
+- A literal block in the middle
+      of an indented list.
+</pre>
+<pre>
+(The above is not a list item since we are in the literal block.)
+</pre>
+</ul>
+<pre>
+Literal block with no indentation (apart from
+the two spaces added to all literal blocks).
+</pre>
+<ol>
+ <li> This is an enumerated list (first item).
+ <li> Continuing with the second item.
+ <li> foo
+ <li> bar
+ <li> Another
+ <li> List
+</ol>
+<p>
+Line blocks are also a form of list:
+</p>
+<ol>
+ <li> This is the first line.   The line continues here.
+ <li> This is the second line.
+</ol>
+
+----------------------------------------------------------------------
 
 lists formatted to fit within 30 characters:
 ----------------------------------------------------------------------
 - This is the first list item.
 
   Second paragraph in the
   first list item.
 
 - List items need not be
   separated by a blank line.
 - And will be rendered without
   one in any case.
 
 We can have indented lists:
 
   - This is an indented list
     item
   - Another indented list
     item:
 
       - A literal block in the middle
             of an indented list.
 
       (The above is not a list item since we are in the literal block.)
 
   Literal block with no indentation (apart from
   the two spaces added to all literal blocks).
 
 1. This is an enumerated list
    (first item).
 2. Continuing with the second
    item.
 (1) foo
 (2) bar
 1) Another
 2) List
 
 Line blocks are also a form of
 list:
 
 This is the first line. The
 line continues here.
 This is the second line.
 
 ----------------------------------------------------------------------
+<ul>
+ <li> This is the first list item.
+<p>
+Second paragraph in the first list item.
+</p>
+ <li> List items need not be separated   by a blank line.
+ <li> And will be rendered without   one in any case.
+</ul>
+<p>
+We can have indented lists:
+</p>
+<ul>
+ <li> This is an indented list item
+ <li> Another indented list item:
+<pre>
+- A literal block in the middle
+      of an indented list.
+</pre>
+<pre>
+(The above is not a list item since we are in the literal block.)
+</pre>
+</ul>
+<pre>
+Literal block with no indentation (apart from
+the two spaces added to all literal blocks).
+</pre>
+<ol>
+ <li> This is an enumerated list (first item).
+ <li> Continuing with the second item.
+ <li> foo
+ <li> bar
+ <li> Another
+ <li> List
+</ol>
+<p>
+Line blocks are also a form of list:
+</p>
+<ol>
+ <li> This is the first line.   The line continues here.
+ <li> This is the second line.
+</ol>
+
+----------------------------------------------------------------------
 
 options formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 There is support for simple option lists, but only with long
 options:
 
  -X --exclude filter  an option with a short and long option
                       with an argument
  -I --include         an option with both a short option and
                       a long option
     --all             Output all.
     --both            Output both (this description is quite
                       long).
     --long            Output all day long.
     --par             This option has two paragraphs in its
                       description. This is the first.
 
                       This is the second.  Blank lines may
                       be omitted between options (as above)
                       or left in (as here).
 
 The next paragraph looks like an option list, but lacks the
 two-space marker after the option. It is treated as a normal
 paragraph:
 
 --foo bar baz
 
 ----------------------------------------------------------------------
+<p>
+There is support for simple option lists,
+but only with long options:
+</p>
+<dl>
+ <dt>-X --exclude filter
+ <dd>an option with a short and long option with an argument
+ <dt>-I --include
+ <dd>an option with both a short option and a long option
+ <dt>   --all
+ <dd>Output all.
+ <dt>   --both
+ <dd>Output both (this description is quite long).
+ <dt>   --long
+ <dd>Output all day long.
+ <dt>   --par
+ <dd>This option has two paragraphs in its description. This is the first.
+<p>
+This is the second.  Blank lines may be omitted between
+options (as above) or left in (as here).
+</p>
+</dl>
+<p>
+The next paragraph looks like an option list, but lacks the two-space
+marker after the option. It is treated as a normal paragraph:
+</p>
+<p>
+--foo bar baz
+</p>
+
+----------------------------------------------------------------------
 
 options formatted to fit within 30 characters:
 ----------------------------------------------------------------------
 There is support for simple
 option lists, but only with
 long options:
 
  -X --exclude filter  an
                       option
                       with a
                       short
                       and
                       long
                       option
                       with an
                       argumen
                       t
  -I --include         an
                       option
                       with
                       both a
                       short
                       option
                       and a
                       long
                       option
     --all             Output
                       all.
     --both            Output
                       both
                       (this d
                       escript
                       ion is
                       quite
                       long).
     --long            Output
                       all day
                       long.
     --par             This
                       option
                       has two
                       paragra
                       phs in
                       its des
                       criptio
                       n. This
                       is the
                       first.
 
                       This is
                       the
                       second.
                       Blank
                       lines
                       may be
                       omitted
                       between
                       options
                       (as
                       above)
                       or left
                       in (as
                       here).
 
 The next paragraph looks like
 an option list, but lacks the
 two-space marker after the
 option. It is treated as a
 normal paragraph:
 
 --foo bar baz
 
 ----------------------------------------------------------------------
+<p>
+There is support for simple option lists,
+but only with long options:
+</p>
+<dl>
+ <dt>-X --exclude filter
+ <dd>an option with a short and long option with an argument
+ <dt>-I --include
+ <dd>an option with both a short option and a long option
+ <dt>   --all
+ <dd>Output all.
+ <dt>   --both
+ <dd>Output both (this description is quite long).
+ <dt>   --long
+ <dd>Output all day long.
+ <dt>   --par
+ <dd>This option has two paragraphs in its description. This is the first.
+<p>
+This is the second.  Blank lines may be omitted between
+options (as above) or left in (as here).
+</p>
+</dl>
+<p>
+The next paragraph looks like an option list, but lacks the two-space
+marker after the option. It is treated as a normal paragraph:
+</p>
+<p>
+--foo bar baz
+</p>
+
+----------------------------------------------------------------------
 
 fields formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 a   First item.
 ab  Second item. Indentation and wrapping is handled
     automatically.
 
 Next list:
 
 small       The larger key below triggers full indentation
             here.
 much too large
             This key is big enough to get its own line.
 
 ----------------------------------------------------------------------
+<dl>
+ <dt>a
+ <dd>First item.
+ <dt>ab
+ <dd>Second item. Indentation and wrapping is handled automatically.
+</dl>
+<p>
+Next list:
+</p>
+<dl>
+ <dt>small
+ <dd>The larger key below triggers full indentation here.
+ <dt>much too large
+ <dd>This key is big enough to get its own line.
+</dl>
+
+----------------------------------------------------------------------
 
 fields formatted to fit within 30 characters:
 ----------------------------------------------------------------------
 a   First item.
 ab  Second item. Indentation
     and wrapping is handled
     automatically.
 
 Next list:
 
 small       The larger key
             below triggers
             full indentation
             here.
 much too large
             This key is big
             enough to get its
             own line.
 
 ----------------------------------------------------------------------
+<dl>
+ <dt>a
+ <dd>First item.
+ <dt>ab
+ <dd>Second item. Indentation and wrapping is handled automatically.
+</dl>
+<p>
+Next list:
+</p>
+<dl>
+ <dt>small
+ <dd>The larger key below triggers full indentation here.
+ <dt>much too large
+ <dd>This key is big enough to get its own line.
+</dl>
+
+----------------------------------------------------------------------
 
 containers (normal) formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 Normal output.
 
 ----------------------------------------------------------------------
+<p>
+Normal output.
+</p>
+
+----------------------------------------------------------------------
 
 containers (verbose) formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 Normal output.
 
 Verbose output.
 
 ----------------------------------------------------------------------
 ['debug', 'debug']
 ----------------------------------------------------------------------
+<p>
+Normal output.
+</p>
+<p>
+Verbose output.
+</p>
+
+----------------------------------------------------------------------
 
 containers (debug) formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 Normal output.
 
 Initial debug output.
 
 ----------------------------------------------------------------------
 ['verbose']
 ----------------------------------------------------------------------
+<p>
+Normal output.
+</p>
+<p>
+Initial debug output.
+</p>
+
+----------------------------------------------------------------------
 
 containers (verbose debug) formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 Normal output.
 
 Initial debug output.
 
 Verbose output.
 
 Debug output.
 
 ----------------------------------------------------------------------
 []
 ----------------------------------------------------------------------
+<p>
+Normal output.
+</p>
+<p>
+Initial debug output.
+</p>
+<p>
+Verbose output.
+</p>
+<p>
+Debug output.
+</p>
+
+----------------------------------------------------------------------
 
 roles formatted to fit within 60 characters:
 ----------------------------------------------------------------------
 Please see "hg add".
 
 ----------------------------------------------------------------------
+<p>
+Please see "hg add".
+</p>
+
+----------------------------------------------------------------------
 
 sections formatted to fit within 20 characters:
 ----------------------------------------------------------------------
 Title
 =====
 
 Section
 -------
 
 Subsection
 ''''''''''
 
 Markup: "foo" and "hg help"
 ---------------------------
 
 ----------------------------------------------------------------------
+<h1>Title</h1>
+<h2>Section</h2>
+<h3>Subsection</h3>
+<h2>Markup: "foo" and "hg help"</h2>
+
+----------------------------------------------------------------------
 
 admonitions formatted to fit within 30 characters:
 ----------------------------------------------------------------------
 Note:
    This is a note
 
    - Bullet 1
    - Bullet 2
 
    Warning!
       This is a warning Second
       input line of warning
 
 !Danger!
    This is danger
 
 ----------------------------------------------------------------------
+<p>
+<b>Note:</b> This is a note
+</p>
+<ul>
+ <li> Bullet 1
+ <li> Bullet 2
+</ul>
+<p>
+<b>Warning!</b> This is a warning Second input line of warning
+</p>
+<p>
+<b>!Danger!</b> This is danger
+</p>
+
+----------------------------------------------------------------------
 
 comments formatted to fit within 30 characters:
 ----------------------------------------------------------------------
 Some text.
 
    Some indented text.
 
 Empty comment above
 
 ----------------------------------------------------------------------
+<p>
+Some text.
+</p>
+<p>
+Some indented text.
+</p>
+<p>
+Empty comment above
+</p>
+
+----------------------------------------------------------------------
 
   === === ========================================
   a   b   c                                       
   === === ========================================
   1   2   3                                       
   foo bar baz this list is very very very long man
   === === ========================================
 
 table formatted to fit within 30 characters:
 ----------------------------------------------------------------------
   a   b   c
   ------------------------------
   1   2   3
   foo bar baz this list is
           very very very long
           man
 
 ----------------------------------------------------------------------
+<table>
+ <tr><th>a</th><th>b</th><th>c</th></tr>
+ <tr><td>1</td><td>2</td><td>3</td></tr>
+ <tr><td>foo</td><td>bar</td><td>baz this list is very very very long man</td></tr>
+</table>
 
+----------------------------------------------------------------------
+