bazarr/libs/markdown/blockparser.py

101 lines
3.4 KiB
Python

from __future__ import unicode_literals
from __future__ import absolute_import
from . import util
from . import odict
class State(list):
""" Track the current and nested state of the parser.
This utility class is used to track the state of the BlockParser and
support multiple levels if nesting. It's just a simple API wrapped around
a list. Each time a state is set, that state is appended to the end of the
list. Each time a state is reset, that state is removed from the end of
the list.
Therefore, each time a state is set for a nested block, that state must be
reset when we back out of that level of nesting or the state could be
corrupted.
While all the methods of a list object are available, only the three
defined below need be used.
"""
def set(self, state):
""" Set a new state. """
self.append(state)
def reset(self):
""" Step back one step in nested state. """
self.pop()
def isstate(self, state):
""" Test that top (current) level is of given state. """
if len(self):
return self[-1] == state
else:
return False
class BlockParser:
""" Parse Markdown blocks into an ElementTree object.
A wrapper class that stitches the various BlockProcessors together,
looping through them and creating an ElementTree object.
"""
def __init__(self, markdown):
self.blockprocessors = odict.OrderedDict()
self.state = State()
self.markdown = markdown
def parseDocument(self, lines):
""" Parse a markdown document into an ElementTree.
Given a list of lines, an ElementTree object (not just a parent
Element) is created and the root element is passed to the parser
as the parent. The ElementTree object is returned.
This should only be called on an entire document, not pieces.
"""
# Create a ElementTree from the lines
self.root = util.etree.Element(self.markdown.doc_tag)
self.parseChunk(self.root, '\n'.join(lines))
return util.etree.ElementTree(self.root)
def parseChunk(self, parent, text):
""" Parse a chunk of markdown text and attach to given etree node.
While the ``text`` argument is generally assumed to contain multiple
blocks which will be split on blank lines, it could contain only one
block. Generally, this method would be called by extensions when
block parsing is required.
The ``parent`` etree Element passed in is altered in place.
Nothing is returned.
"""
self.parseBlocks(parent, text.split('\n\n'))
def parseBlocks(self, parent, blocks):
""" Process blocks of markdown text and attach to given etree node.
Given a list of ``blocks``, each blockprocessor is stepped through
until there are no blocks left. While an extension could potentially
call this method directly, it's generally expected to be used
internally.
This is a public method as an extension may need to add/alter
additional BlockProcessors which call this method to recursively
parse a nested block.
"""
while blocks:
for processor in self.blockprocessors.values():
if processor.test(parent, blocks[0]):
if processor.run(parent, blocks) is not False:
# run returns True or None
break