def dump(self, tree):
assert isinstance(tree, ParseTree)
assert self.linker, 'rst dumper needs a linker object'
self.linker.set_usebase(True)
output = TextBuffer()
self.dump_children(tree.getroot(), output)
return output.get_lines(end_with_newline=not tree.ispartial)
def dump(self):
text = TextBuffer(
[u'<?xml version="1.0" encoding="utf-8"?>\n', u'<section>\n'])
for node in self._nodetree:
text += self._dump_node(node)
text.append(u'</section>\n')
return text.get_lines()
def dump(self, tree): assert isinstance(tree, ParseTree) assert self.linker, 'HTML dumper needs a linker object' self.linker.set_usebase(True) output = TextBuffer() self._dump_children(tree.getroot(), output, istoplevel=True) return output.get_lines()
def dump(self): text = TextBuffer([ u'<?xml version="1.0" encoding="utf-8"?>\n', u'<section>\n' ]) for node in self._nodetree: text += self._dump_node(node) text.append(u'</section>\n') return text.get_lines()
def process(self, dict):
'''Processes the template and returns a list of lines.
The dict is used to get / set template parameters.
'''
if not isinstance(dict, TemplateDict):
dict = TemplateDict(dict)
output = TextBuffer(self.tokens.process(dict))
return output.get_lines()
def process(self, dict): '''Processes the template and returns a list of lines. The dict is used to get / set template parameters. ''' if not isinstance(dict, TemplateDict): dict = TemplateDict(dict) output = TextBuffer(self.tokens.process(dict)) return output.get_lines()
def dump(self, tree):
assert isinstance(tree, ParseTree)
assert self.linker, 'LaTeX dumper needs a linker object'
self.linker.set_usebase(False)
self.end_tag_pending = []
output = TextBuffer()
self.dump_children(tree.getroot(), output)
self.end_tag_pending.reverse()
for et in self.end_tag_pending:
output.append(sec_end_tag[et] + '\n')
return output.get_lines()
def dump(self, tree):
assert isinstance(tree, ParseTree)
assert self.linker, 'LaTeX dumper needs a linker object'
self.linker.set_usebase(False)
self.document_type = self.template_options.get('document_type')
# Option set in template - potentially tainted value
if not self.document_type in ('report', 'article','book'):
logger.warn('No document type set in template, assuming "report"')
self.document_type = 'report' # arbitrary default
else:
logger.info('used document type: %s'%self.document_type)
output = TextBuffer()
self.dump_children(tree.getroot(), output)
return output.get_lines(end_with_newline=not tree.ispartial)
def dump(self, tree): assert isinstance(tree, ParseTree) output = TextBuffer() for element in tree.getiterator(): if not element.text is None: output.append(element.text) if not element.tail is None: output.append(element.tail) return output.get_lines()
def dump_children(self, list, output, list_level=-1, list_type=None, list_iter='0'):
if list.text:
output.append(list.text)
for element in list.getchildren():
if element.tag in ('p', 'div'):
#~ print element.tag
#~ print element.text
indent = 0
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
myoutput = TextBuffer()
self.dump_children(element, myoutput) # recurs
output.extend('\t'*indent)
output.extend(myoutput)
elif element.tag == 'h':
level = int(element.attrib['level'])
if level < 1: level = 1
elif level > 4: level = 4
char = level_tag[level-1]
heading = element.text
line = char * len(heading)
output.append(heading + '\n')
output.append(line)
elif element.tag in ('ul', 'ol'):
indent = int(element.attrib.get('indent', 0))
start = element.attrib.get('start')
myoutput = TextBuffer()
self.dump_children(element, myoutput, list_level=list_level+1, list_type=element.tag, list_iter=start) # recurs
if list_level == -1:
output.extend(myoutput)
else:
output.extend(myoutput)
elif element.tag == 'li':
if 'indent' in element.attrib:
# HACK for raw trees from pageview
list_level = int(element.attrib['indent'])
if list_type == 'ol':
bullet = str(list_iter) + '.'
list_iter = increase_list_iter(list_iter) or '1' # fallback if iter not valid
else:
bullet = bullet_types[element.attrib.get('bullet', BULLET)]
output.append('\t'*list_level+bullet+' ')
self.dump_children(element, output, list_level=list_level) # recurs
output.append('\n')
elif element.tag == 'pre':
myoutput = TextBuffer()
myoutput.append("::\n\n")
text = [ '\t' + t for t in element.text.split('\n')]
myoutput.append('\n'.join(text))
output.extend(myoutput)
elif element.tag == 'link':
assert 'href' in element.attrib, \
'BUG: link %s "%s"' % (element.attrib, element.text)
href = self.linker.link(element.attrib['href'])
text = element.text
output.append('`%s <%s>`_' % (text, href))
elif element.tag in ('sub', 'sup'):
if element.text:
tag = dumper_tags[element.tag]
output.append("%s`%s`\ " % (tag, element.text))
elif element.tag in ('mark', 'strike'):
if element.text: output.append(element.text)
elif element.tag in ('strong', 'emphasis'):
if element.text:
tag = dumper_tags[element.tag]
msg = tag + element.text + tag + ' '
if output: msg = ' ' + msg
output.append(msg)
elif element.tag == 'img':
src = self.linker.img(element.attrib['src'])
output.append('.. image:: %s' % src)
elif element.tag in dumper_tags:
if element.text:
tag = dumper_tags[element.tag]
output.append(' ' + tag + element.text + tag + ' ')
else:
assert False, 'Unknown node type: %s' % element
if element.tail:
output.append(element.tail)
def dump_children(self, list, output, list_level = -1):
if list.text:
output.append(tex_encode(list.text))
for element in list.getchildren():
text = tex_encode(element.text)
if element.tag in ('p', 'div'):
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
else:
indent = 0
myoutput = TextBuffer()
self.dump_children(element,myoutput)
if indent:
myoutput.prefix_lines('\t'*indent)
output.extend(myoutput)
elif element.tag == 'h':
level = int(element.attrib['level'])
if level < 1: level = 1
elif level > 5: level = 5
output.append(sectioning[self.document_type][level]%(text))
elif element.tag == 'ul':
output.append('\\begin{itemize}\n')
self.dump_children(element,output,list_level=list_level+1)
output.append('\\end{itemize}')
elif element.tag == 'ol':
start = element.attrib.get('start', 1)
if start in string.lowercase:
type = 'a'
start = string.lowercase.index(start) + 1
elif start in string.uppercase:
type = 'A'
start = string.uppercase.index(start) + 1
else:
type = '1'
start = int(start)
output.append('\\begin{enumerate}[%s]\n' % type)
if start > 1:
output.append('\setcounter{enumi}{%i}\n' % (start-1))
self.dump_children(element,output,list_level=list_level+1)
output.append('\\end{enumerate}')
elif element.tag == 'li':
if 'bullet' in element.attrib:
bullet = bullet_types[element.attrib['bullet']]
else:
bullet = bullet_types[BULLET]
output.append('\t'*list_level+bullet)
self.dump_children(element, output, list_level=list_level) # recurse
output.append('\n')
elif element.tag == 'pre':
indent = 0
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
myoutput = TextBuffer()
myoutput.append(element.text)
if indent:
myoutput.prefix_lines(' ' * indent)
output.append('\n\\begin{lstlisting}\n')
output.extend(myoutput)
output.append('\n\\end{lstlisting}\n')
elif element.tag == 'sub':
output.append('$_{%s}$' % element.text)
elif element.tag == 'sup':
output.append('$^{%s}$' % element.text)
elif element.tag == 'img':
#we try to get images about the same visual size, therefore need to specify dot density
#96 dpi seems to be common for computer monitors
dpi = 96
done = False
if 'type' in element.attrib and element.attrib['type'] == 'equation':
try:
# Try to find the source, otherwise fall back to image
src = element.attrib['src'][:-4] + '.tex'
file = self.linker.resolve_file(src)
if file is not None:
equation = file.read().strip()
else:
equation = None
except FileNotFoundError:
logger.warn('Could not find latex equation: %s', src)
else:
if equation:
output.append('\\begin{math}\n')
output.extend(equation)
output.append('\n\\end{math}')
done = True
if not done:
if 'width' in element.attrib and not 'height' in element.attrib:
options = 'width=%fin, keepaspectratio=true' \
% ( float(element.attrib['width']) / dpi )
elif 'height' in element.attrib and not 'width' in element.attrib:
options = 'height=%fin, keepaspectratio=true' \
% ( float(element.attrib['height']) / dpi )
else:
options = ''
#~ imagepath = File(self.linker.link(element.attrib['src'])).path
imagepath = self.linker.link(element.attrib['src'])
if imagepath.startswith('file://'):
imagepath = File(imagepath).path # avoid URIs here
image = '\\includegraphics[%s]{%s}' % (options, imagepath)
if 'href' in element.attrib:
href = self.linker.link(element.attrib['href'])
output.append('\\href{%s}{%s}' % (href, image))
else:
output.append(image)
elif element.tag == 'link':
href = self.linker.link(element.attrib['href'])
output.append('\\href{%s}{%s}' % (href, text))
elif element.tag == 'emphasis':
output.append('\\emph{'+text+'}')
elif element.tag == 'strong':
output.append('\\textbf{'+text+'}')
elif element.tag == 'mark':
output.append('\\uline{'+text+'}')
elif element.tag == 'strike':
output.append('\\sout{'+text+'}')
elif element.tag == 'code':
success = False
#Here we try several possible delimiters for the inline verb command of LaTeX
for delim in '+*|$&%!-_':
if not delim in text:
success = True
output.append('\\lstinline'+delim+text+delim)
break
if not success:
assert False, 'Found no suitable delimiter for verbatim text: %s' % element
pass
elif element.tag == 'tag':
# LaTeX doesn't have anything similar to tags afaik
output.append(text)
else:
assert False, 'Unknown node type: %s' % element
if element.tail:
output.append(tex_encode(element.tail))
def dump_children(self, list, output, list_level=-1, list_type=None, list_iter='0'):
if list.text:
output.append(list.text)
for element in list.getchildren():
if element.tag in ('p', 'div'):
indent = 0
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
myoutput = TextBuffer()
self.dump_children(element, myoutput) # recurs
if indent:
myoutput.prefix_lines('\t'*indent)
output.extend(myoutput)
elif element.tag == 'h':
## Copy from Markdown
level = int(element.attrib['level'])
if level < 1: level = 1
elif level > 5: level = 5
if level in (1, 2):
# setext-style headers for lvl 1 & 2
if level == 1: char = '='
else: char = '-'
heading = element.text
line = char * len(heading)
output.append(heading + '\n')
output.append(line)
else:
# atx-style headers for deeper levels
tag = '#' * level
output.append(tag + ' ' + element.text)
elif element.tag in ('ul', 'ol'):
indent = int(element.attrib.get('indent', 0))
start = element.attrib.get('start')
myoutput = TextBuffer()
self.dump_children(element, myoutput, list_level=list_level+1, list_type=element.tag, list_iter=start) # recurs
if indent:
myoutput.prefix_lines('\t'*indent)
output.extend(myoutput)
elif element.tag == 'li':
if 'indent' in element.attrib:
# HACK for raw trees from pageview
list_level = int(element.attrib['indent'])
if list_type == 'ol':
bullet = str(list_iter) + '.'
list_iter = increase_list_iter(list_iter) or '1' # fallback if iter not valid
else:
bullet = bullet_types[element.attrib.get('bullet', BULLET)]
output.append('\t'*list_level+bullet+' ')
self.dump_children(element, output, list_level=list_level) # recurs
output.append('\n')
elif element.tag == 'img':
src = element.attrib['src']
opts = []
for k, v in element.attrib.items():
if k == 'src' or k.startswith('_'):
continue
else:
opts.append('%s=%s' % (k, v))
if opts:
src += '?%s' % '&'.join(opts)
if element.text:
output.append(element.text)
else:
output.append(src)
elif element.tag == 'link':
assert 'href' in element.attrib, \
'BUG: link %s "%s"' % (element.attrib, element.text)
href = element.attrib['href']
if element.text:
output.append(element.text)
else:
output.append(href)
elif element.tag == 'pre':
indent = 0
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
myoutput = TextBuffer()
myoutput.append(element.text)
if indent:
myoutput.prefix_lines('\t'*indent)
output.extend(myoutput)
elif element.text:
output.append(element.text)
else:
pass
if element.tail:
output.append(element.tail)
def dump_children(self, list, output, list_level=-1, list_type=None, list_iter='0'):
if list.text:
output.append(list.text)
for element in list.getchildren():
if element.tag in ('p', 'div'):
indent = 0
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
myoutput = TextBuffer()
self.dump_children(element, myoutput) # recurs
# OPEN ISSUE: no indent for para
#if indent:
# myoutput.prefix_lines('\t'*indent)
output.extend(myoutput)
elif element.tag == 'h':
level = int(element.attrib['level'])
if level < 1: level = 1
elif level > 5: level = 5
if level in (1, 2):
# setext-style headers for lvl 1 & 2
if level == 1: char = '='
else: char = '-'
heading = element.text
line = char * len(heading)
output.append(heading + '\n')
output.append(line)
else:
# atx-style headers for deeper levels
tag = '#' * level
output.append(tag + ' ' + element.text)
elif element.tag in ('ul', 'ol'):
indent = int(element.attrib.get('indent', 0))
start = element.attrib.get('start')
myoutput = TextBuffer()
self.dump_children(element, myoutput, list_level=list_level+1, list_type=element.tag, list_iter=start) # recurs
# OPEN ISSUE: no indent for para
#if indent:
# myoutput.prefix_lines('\t'*indent)
if list_level == -1:
# Need empty lines around lists in markdown
output.append('\n')
output.extend(myoutput)
output.append('\n')
else:
output.extend(myoutput)
elif element.tag == 'li':
if 'indent' in element.attrib:
# HACK for raw trees from pageview
list_level = int(element.attrib['indent'])
if list_type == 'ol':
bullet = str(list_iter) + '.'
list_iter = increase_list_iter(list_iter) or '1' # fallback if iter not valid
else:
bullet = bullet_types[element.attrib.get('bullet', BULLET)]
output.append('\t'*list_level+bullet+' ')
self.dump_children(element, output, list_level=list_level) # recurs
output.append('\n')
elif element.tag == 'pre':
indent = 0
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
myoutput = TextBuffer()
myoutput.append(element.text)
# OPEN ISSUE: no indent for para
#if indent:
# myoutput.prefix_lines('\t'*indent)
myoutput.prefix_lines('\t') # verbatim is always indented
output.extend(myoutput)
elif element.tag == 'link':
assert 'href' in element.attrib, \
'BUG: link %s "%s"' % (element.attrib, element.text)
href = self.linker.link(element.attrib['href'])
text = element.text or href
if href == text and url_re.match(href):
output.append('<' + href + '>')
else:
output.append('[%s](%s)' % (text, href))
elif element.tag == 'img':
src = self.linker.img(element.attrib['src'])
# OPEN ISSUE: image properties used in zim not supported in pandoc
#opts = []
#items = element.attrib.items()
# we sort params only because unit tests don't like random output
#items.sort()
#for k, v in items:
# if k == 'src' or k.startswith('_'):
# continue
# elif v: # skip None, "" and 0
# opts.append('%s=%s' % (k, v))
#if opts:
# src += '?%s' % '&'.join(opts)
text = element.text or ''
output.append('' % (text, src))
elif element.tag in dumper_tags:
if element.text:
tag = dumper_tags[element.tag]
output.append(tag + element.text + tag)
else:
assert False, 'Unknown node type: %s' % element
if element.tail:
output.append(element.tail)
def dump_children(self, list, output, list_level=-1):
if list.text:
output.append(tex_encode(list.text))
for element in list.getchildren():
text = tex_encode(element.text)
if element.tag in ('p', 'div'):
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
else:
indent = 0
myoutput = TextBuffer()
self.dump_children(element, myoutput)
if indent:
myoutput.prefix_lines('\t' * indent)
output.extend(myoutput)
elif element.tag == 'h':
level = int(element.attrib['level'])
if level < 1: level = 1
elif level > 5: level = 5
if self.end_tag_pending:
if self.end_tag_pending[-1] == level and level == 5:
output.append(sec_end_tag[level] + '\n')
self.end_tag_pending.pop()
if level <= 4:
self.end_tag_pending.reverse()
for et in self.end_tag_pending:
output.append(sec_end_tag[et] + '\n')
self.end_tag_pending = []
if level >= 4:
if level == 4: output.append('%' + '_' * 78 + '\n')
self.end_tag_pending.append(level)
output.append(sectioning[level] % (text))
elif element.tag == 'ul':
output.append('\\begin{itemize}\n')
self.dump_children(element, output, list_level=list_level + 1)
output.append('\\end{itemize}')
elif element.tag == 'ol':
start = element.attrib.get('start', 1)
if start in string.lowercase:
type = 'a'
start = string.lowercase.index(start) + 1
elif start in string.uppercase:
type = 'A'
start = string.uppercase.index(start) + 1
else:
type = '1'
start = int(start)
output.append('\\begin{enumerate}[%s]\n' % type)
if start > 1:
output.append('\setcounter{enumi}{%i}\n' % (start - 1))
self.dump_children(element, output, list_level=list_level + 1)
output.append('\\end{enumerate}')
elif element.tag == 'li':
if 'bullet' in element.attrib:
bullet = bullet_types[element.attrib['bullet']]
else:
bullet = bullet_types[BULLET]
output.append('\t' * list_level + bullet)
self.dump_children(element, output,
list_level=list_level) # recurse
output.append('\n')
elif element.tag == 'pre':
indent = 0
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
myoutput = TextBuffer()
myoutput.append(element.text)
if indent:
myoutput.prefix_lines(' ' * indent)
output.append('\n\\begin{lstlisting}\n')
output.extend(myoutput)
output.append('\n\\end{lstlisting}\n')
elif element.tag == 'sub':
output.append('$_{%s}$' % element.text)
elif element.tag == 'sup':
output.append('$^{%s}$' % element.text)
elif element.tag == 'img':
print element.attrib
if list_level == -1: output.append('\\begin{center}\n')
#we try to get images about the same visual size, therefore need to specify dot density
#96 dpi seems to be common for computer monitors
dpi = 96
done = False
if 'type' in element.attrib and element.attrib[
'type'] == 'equation':
try:
# Try to find the source, otherwise fall back to image
src = element.attrib['src'][:-4] + '.tex'
file = self.linker.resolve_file(src)
if file is not None:
equation = file.read().strip()
else:
equation = None
except FileNotFoundError:
logger.warn('Could not find latex equation: %s', src)
else:
if equation:
output.append('\\begin{math}\n')
output.extend(equation)
output.append('\n\\end{math}')
done = True
if not done:
# if 'width' in element.attrib and not 'height' in element.attrib:
# options = 'width=%fin, keepaspectratio=true' \
# % ( float(element.attrib['width']) / dpi )
# elif 'height' in element.attrib and not 'width' in element.attrib:
# options = 'height=%fin, keepaspectratio=true' \
# % ( float(element.attrib['height']) / dpi )
# else:
# options = ''
imagepath = File(self.linker.link(
element.attrib['src'])).path
# imagepath = self.linker.link(element.attrib['src'])
# choose refering to width or height by w/h ratio.
# if height bigger than width, set to 0.8 by textheight.
# Otherwise if width bigger than height, set to 0.9 by
# textwidth
img_width, img_height = Image.open(imagepath).size
ratio = img_width / img_height
if ratio > 1:
options = 'width=0.90\\textwidth'
else:
options = 'height=0.80\\textheight'
image = '\\includegraphics[%s]{%s}' % (options, imagepath)
if 'href' in element.attrib:
href = self.linker.link(element.attrib['href'])
output.append('\\href{%s}{%s}' % (href, image))
else:
output.append(image)
if list_level == -1: output.append('\n\\end{center}\n')
elif element.tag == 'link':
href = self.linker.link(element.attrib['href'])
output.append('\\href{%s}{%s}' % (href, text))
elif element.tag == 'emphasis':
output.append('\\emph{' + text + '}')
elif element.tag == 'strong':
output.append('\\textbf{' + text + '}')
elif element.tag == 'mark':
output.append('\\uline{' + text + '}')
elif element.tag == 'strike':
output.append('\\sout{' + text + '}')
elif element.tag == 'code':
success = False
#Here we try several possible delimiters for the inline verb command of LaTeX
for delim in '+*|$&%!-_':
if not delim in text:
success = True
output.append('\\lstinline' + delim + text + delim)
break
if not success:
assert False, 'Found no suitable delimiter for verbatim text: %s' % element
pass
elif element.tag == 'tag':
# LaTeX doesn't have anything similar to tags afaik
output.append(text)
else:
assert False, 'Unknown node type: %s' % element
if element.tail:
output.append(tex_encode(element.tail))
def dump_children(self, list, output, list_level=-1):
if list.text:
output.append(list.text)
for element in list.getchildren():
if element.tag == 'p':
indent = 0
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
myoutput = TextBuffer()
self.dump_children(element, myoutput) # recurs
if indent:
myoutput.prefix_lines('\t'*indent)
output.extend(myoutput)
elif element.tag == 'ul':
self.dump_children(element, output, list_level=list_level+1) # recurs
elif element.tag == 'h':
level = int(element.attrib['level'])
if level < 1: level = 1
elif level > 5: level = 5
tag = '='*(7 - level)
output.append(tag+' '+element.text+' '+tag)
elif element.tag == 'li':
if 'indent' in element.attrib:
list_level = int(element.attrib['indent'])
if 'bullet' in element.attrib:
bullet = bullet_types[element.attrib['bullet']]
else:
bullet = '*'
output.append('\t'*list_level+bullet+' ')
self.dump_children(element, output, list_level=list_level) # recurs
output.append('\n')
elif element.tag == 'pre':
output.append("'''\n"+element.text+"'''\n")
elif element.tag == 'img':
src = element.attrib['src']
opts = []
for k, v in element.attrib.items():
if k == 'src' or k.startswith('_'):
continue
else:
opts.append('%s=%s' % (k, v))
if opts:
src += '?%s' % '&'.join(opts)
if element.text:
output.append('{{'+src+'|'+element.text+'}}')
else:
output.append('{{'+src+'}}')
elif element.tag == 'link':
assert 'href' in element.attrib, \
'BUG: link %s "%s"' % (element.attrib, element.text)
href = element.attrib['href']
if href == element.text:
if url_re.match(href):
output.append(href)
else:
output.append('[['+href+']]')
else:
output.append('[['+href+'|'+element.text+']]')
elif element.tag in dumper_tags:
tag = dumper_tags[element.tag]
output.append(tag+element.text+tag)
else:
assert False, 'Unknown node type: %s' % element
if element.tail:
output.append(element.tail)
def dump_children(self, list, output, list_level=-1, list_type=None, list_iter='0'):
if list.text:
output.append(list.text)
for element in list.getchildren():
if element.tag in ('p', 'div'):
indent = 0
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
myoutput = TextBuffer()
self.dump_children(element, myoutput) # recurs
if indent:
myoutput.prefix_lines('\t'*indent)
output.extend(myoutput)
elif element.tag == 'h':
level = int(element.attrib['level'])
if level < 1: level = 1
elif level > 5: level = 5
tag = '='*(7 - level)
output.append(tag+' '+element.text+' '+tag)
elif element.tag in ('ul', 'ol'):
indent = int(element.attrib.get('indent', 0))
start = element.attrib.get('start')
myoutput = TextBuffer()
self.dump_children(element, myoutput, list_level=list_level+1, list_type=element.tag, list_iter=start) # recurs
if indent:
myoutput.prefix_lines('\t'*indent)
output.extend(myoutput)
elif element.tag == 'li':
if 'indent' in element.attrib:
# HACK for raw trees from pageview
list_level = int(element.attrib['indent'])
if list_type == 'ol':
bullet = str(list_iter) + '.'
list_iter = increase_list_iter(list_iter) or '1' # fallback if iter not valid
elif 'bullet' in element.attrib: # ul OR raw tree from pageview...
if element.attrib['bullet'] in bullet_types:
bullet = bullet_types[element.attrib['bullet']]
else:
bullet = element.attrib['bullet'] # Assume it is numbered..
else: # ul
bullet = '*'
output.append('\t'*list_level+bullet+' ')
self.dump_children(element, output, list_level=list_level) # recurs
output.append('\n')
elif element.tag == 'pre':
indent = 0
if 'indent' in element.attrib:
indent = int(element.attrib['indent'])
myoutput = TextBuffer()
myoutput.append("'''\n"+element.text+"'''\n")
if indent:
myoutput.prefix_lines('\t'*indent)
output.extend(myoutput)
elif element.tag == 'img':
src = element.attrib['src']
opts = []
items = element.attrib.items()
# we sort params only because unit tests don't like random output
items.sort()
for k, v in items:
if k == 'src' or k.startswith('_'):
continue
elif v: # skip None, "" and 0
opts.append('%s=%s' % (k, v))
if opts:
src += '?%s' % '&'.join(opts)
if element.text:
output.append('{{'+src+'|'+element.text+'}}')
else:
output.append('{{'+src+'}}')
elif element.tag == 'sub':
output.append("_{%s}" % element.text)
elif element.tag == 'sup':
output.append("^{%s}" % element.text)
elif element.tag == 'link':
assert 'href' in element.attrib, \
'BUG: link %s "%s"' % (element.attrib, element.text)
href = element.attrib['href']
if href == element.text:
if url_re.match(href):
output.append(href)
else:
output.append('[['+href+']]')
else:
if element.text:
output.append('[['+href+'|'+element.text+']]')
else:
output.append('[['+href+']]')
elif element.tag in dumper_tags:
if element.text:
tag = dumper_tags[element.tag]
output.append(tag+element.text+tag)
else:
assert False, 'Unknown node type: %s' % element
if element.tail:
output.append(element.tail)
def dump(self, tree):
#~ print 'DUMP WIKI', tree.tostring()
assert isinstance(tree, ParseTree)
output = TextBuffer()
self.dump_children(tree.getroot(), output)
return output.get_lines(end_with_newline=not tree.ispartial)
