- '.format(level_names[new_level])
link = self.get_link(label)
if (new_level <= self._split_level):
link = re.split('#', link)[0]
self.nav_list.append((new_entry.capitalize(), link))
toc += '
- {} '.format(link, new_entry.capitalize()) new_entry = False level = new_level continue if re.match(r'\d ', line): # found an new toc entry new_level = int(line[:2]) new_entry = line[2:].strip() if (new_level == 1): # found a new part self._part_number += 1 new_entry = self.get_part_prefix() + new_entry self.toc = self.toc_tmpl.format(toc) def get_page_name(self, s): """Return the name of a HTML page based on a string.""" return re.sub('[/_ \(\)]', '-', s.lower().strip()) + '.html' def get_link(self, s): """Build the link to an anchor, including the page name if required. Returns an empty string if the label wasn't found.""" s = s.lower().strip() # the labels for plotting styles don't use the 'with' part. if s.startswith('with'): s = s[5:] if (not s in self._labels): return '' if (self._labels[s][0] == self._current_page): return self._labels[s][1] return ''.join(self._labels[s]) def get_relative_link(self, s): return '#' + self.get_label(s) def get_label(self, s): # s contains only characters and white spaces. return s.strip().replace(' ', '-') def get_line(self): """Return the content of the current line.""" if (self._pos < self._lines): return self.docstr[self._pos] else: return '' def peek_line(self, offset=1): """Return the content of the line at a position relative to the current.""" return self.docstr[self._pos + offset] def get_part_prefix(self): """Add a prefix for the top level sections, the 'parts'.""" return 'Part {}: '.format(self._part_number) def escape_html(self, s): return s.replace('&', '&').replace('<', '<').replace('>', '>') """Following are all formatting methods, which may be accessed only from self.format_doc().""" def format_section(self): level = int(self.get_line()[:2]) title = self.get_line()[2:].strip().capitalize() self.format_top_nav(title, level) # iterate over all following labels to include them inside the heading. labels = '' while ((self._pos + 1 < self._lines) and self.peek_line().startswith('?')): self._pos += 1 labels += self.label_tmpl.format(self.get_label(self.get_line()[1:])) # Add a prefix if (level == 1): title = self.get_part_prefix() + labels + title else: title = labels + title print(self.header_tmpl.format(level, title, level), file=self.outfile) def format_top_nav(self, title, level): if (level > self._split_level): return None for i, v in enumerate(self.nav_list): if (v[0] == title): break top_nav = '
'
if (i > 0):
v = self.nav_list[i-1]
top_nav += '
'
print(top_nav, file=self.outfile)
def format_label(self):
print(self.label_tmpl.format(self.get_label(self.get_line()[1:])), file=self.outfile)
def format_table(self):
"""Extract the HTML table code."""
table_str = ''
while (not self.peek_line().startswith('@')):
self._pos += 1
if self.get_line().startswith('^'):
table_str += self.get_line()[1:]
print(table_str, file=self.outfile)
def format_codelisting(self):
code_str = ''
spaces = len(self.get_line()) - len(self.get_line().lstrip())
while (self.get_line().startswith(' ')):
code_str += self.get_line()[spaces:]
self._pos += 1
self._pos -= 1
print(self.code_tmpl.format(self.escape_html(code_str)), file=self.outfile)
def format_list(self):
s = ''
while (not self.peek_line().startswith('#end')):
self._pos += 1
s += '{}
'.format(self.link_tmpl.format(v[1], 'Prev: '+v[0]))
if (i < len(self.nav_list)-1):
v = self.nav_list[i+1]
top_nav += ''.format(self.link_tmpl.format(v[1], 'Next: '+v[0]))
top_nav += '{}
', file=self.outfile)
def close_and_finish_file(self):
if self.outfile:
print('
', file=self.outfile)
print(HTML_FOOTER, file=self.outfile)
self.outfile.close()
self.outfile = None
def format_doc(self):
self._part_number = 0
page_pattern = r'[0-{}] (.*?)$'.format(self._split_level)
if (self._split_level == 0):
self._current_page = self.get_page_name(self._single_filename)
self.open_and_init_file(self._current_page)
self._pos = 0
while (self._pos < self._lines):
line = self.get_line()
# skip the commented lines before the first section.
if (line.startswith('C')):
self._pos += 1
continue
if re.match(page_pattern, line):
self._current_page = self.get_page_name(line[2:].strip())
self.close_and_finish_file()
self.open_and_init_file(self._current_page)
# now do the actual parsing
if (re.match(r'\d+ ', line)):
if (int(line[:2]) == 1):
self._part_number += 1
self.format_section()
elif (line.startswith('?')):
self.format_label()
elif (line.startswith('@start table')):
if (not self.peek_line(-1).startswith('^')):
self.format_table()
elif line.startswith('#start'):
self.format_list()
elif line.startswith('^'):
self.format_raw_html()
elif line.startswith('F'):
self.format_image()
elif line.startswith(' '):
self.format_codelisting()
elif line.startswith(' '):
self.format_paragraph()
# currently we ignore all '#' (not the '#start' one), '%' and '='
self._pos += 1
HTML_HEADER = r"""