def _init_block_parser(self):
# Top level parser, to break up block level items
p = RuleParser(
Rule(VERBATIM_BLOCK,
r'''
^(?P<pre_indent>\t*) \'\'\' \s*? # 3 "'"
( (?:^.*\n)*? ) # multi-line text
^(?P=pre_indent) \'\'\' \s*? \n # another 3 "'" with matching indent
''',
process=self.parse_pre),
Rule(OBJECT,
r'''
^(?P<obj_indent>\t*) \{\{\{ \s*? (\S+:.*\n) # "{{{ object_type: attrib=..."
( (?:^.*\n)*? ) # multi-line body
^(?P=obj_indent) \}\}\} \s*? \n # "}}}" with matching indent
''',
process=self.parse_object),
Rule(
HEADING,
r'^( ==+ [\ \t]+ \S.*? ) [\ \t]* =* \n', # "==== heading ===="
process=self.parse_heading),
# standard table format
Rule(TABLE,
r'''
^(\|.*\|) \s*? \n # starting and ending with |
^( (?:\| [ \|\-:]+ \| \s*? \n)? ) # column align
( (?:^\|.*\| \s*? \n)+ ) # multi-lines: starting and ending with |
''',
process=self.parse_table),
# line format
Rule(LINE, r'(?<=\n)-{5,}(?=\n)',
process=self.parse_line) # \n----\n
)
p.process_unmatched = self.parse_para
return p
def _init_intermediate_parser(self):
# Intermediate level, breaks up lists and indented blocks
# TODO: deprecate this by taking lists out of the para
# and make a new para for each indented block
p = RuleParser(
Rule('X-Bullet-List',
r'''(
^ %s .* \n # Line starting with bullet
(?:
^ \t* %s .* \n # Line with same or more indent and bullet
)* # .. repeat
)''' % (bullet_pattern, bullet_pattern),
process=self.parse_list),
Rule('X-Indented-Bullet-List',
r'''(
^(?P<list_indent>\t+) %s .* \n # Line with indent and bullet
(?:
^(?P=list_indent) \t* %s .* \n # Line with same or more indent and bullet
)* # .. repeat
)''' % (bullet_pattern, bullet_pattern),
process=self.parse_list),
Rule('X-Indented-Block',
r'''(
^(?P<block_indent>\t+) .* \n # Line with indent
(?:
^(?P=block_indent) (?!\t|%s) .* \n # Line with _same_ indent, no bullet
)* # .. repeat
)''' % bullet_pattern,
process=self.parse_indent),
)
p.process_unmatched = self.inline_parser
return p