legalmd/legalmd/legal_token.py

295 lines
8.3 KiB
Python

# legalmd: Markdown-based legal markup
# Copyright © 2019, 2021 Lee Yingtong Li (RunasSudo)
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
from itertools import zip_longest
import re
import mistletoe
mistletoe.block_token.remove_token(mistletoe.block_token.BlockCode)
class NumberedHeading(mistletoe.block_token.BlockToken):
pattern = re.compile(r'(#{1,6})\s*([0-9A-Z–]+|xx)\s+(.+)')
def __init__(self, match):
self.level, self.label, content = match
self.children = mistletoe.span_token.tokenize_inner(content)
@classmethod
def start(cls, line):
return cls.pattern.match(line)
@classmethod
def read(cls, lines):
line = next(lines)
match = cls.pattern.match(line)
level = len(match.group(1))
label = match.group(2)
content = match.group(3)
if content == '***':
content = '★★★'
return level, label, content
def full_label(self):
if not self.label:
return None
if not isinstance(self.parent, mistletoe.block_token.Document):
return None
return self.label
mistletoe.block_token.add_token(NumberedHeading)
class Subrules(mistletoe.block_token.BlockToken):
pattern = re.compile(r'\t*(\([0-9A-Za-z–]+\)|[0-9A-Za-z–]+\.)\s+.')
def __init__(self, children):
self.children = children
@classmethod
def start(cls, line):
return cls.pattern.match(line)
@classmethod
def read(cls, lines):
children = []
while True:
lines.anchor()
try:
line = next(lines)
except StopIteration:
break
if len(line.strip()) == 0:
continue
lines.reset()
if Note.pattern.match(line):
break
if not SubrulesItem.start(line):
break
children.append(SubrulesItem(*SubrulesItem.read(lines)))
return children
mistletoe.block_token.remove_token(mistletoe.block_token.List)
mistletoe.block_token.add_token(Subrules)
class SubrulesItem(mistletoe.block_token.BlockToken):
pattern = re.compile(r'(\t*)((?:\([0-9A-Za-z–]+\)|[0-9A-Za-z–]+\.)(?: \([0-9A-Za-z–]+\)| [0-9A-Za-z–]+\.)*)?\s*(.+)')
def __init__(self, level, label, children):
self.level = level
self.label = label
self.children = children
@classmethod
def start(cls, line):
match = cls.pattern.match(line)
if not match.group(1) and not match.group(2):
# Neither an indent nor a label
return False
if Definition.pattern.match(line):
return False
return True
@classmethod
def read(cls, lines):
line = next(lines)
lead_in, label, content = cls.pattern.match(line).group(1, 2, 3)
if content == '***':
content = '★★★'
level = len(lead_in)
children = mistletoe.span_token.tokenize_inner(content)
if label is None:
level -= 1
return level, label, children
def full_label(self):
if not self.label:
return None
if not isinstance(self.parent.parent, mistletoe.block_token.Document):
return None
labels = [self.label]
# Subrules items
cur_level = self.level
for child in reversed(self.parent.children[0:self.parent.children.index(self)]):
if child.level < cur_level and child.label:
labels.append(child.label)
cur_level = child.level
# Section
section = next((x for x in reversed(self.parent.parent.children[0:self.parent.parent.children.index(self.parent)]) if isinstance(x, NumberedHeading)), None)
if not section:
return None
labels.append(section.label)
return ''.join(reversed(labels))
class Note(mistletoe.block_token.BlockToken):
pattern = re.compile(r'(\t*)([0-9A-Z ]+):\s+(.+)')
def __init__(self, match):
self.level, self.label, self.children = match
@classmethod
def start(cls, line):
return cls.pattern.match(line)
@classmethod
def read(cls, lines):
line = next(lines)
match = cls.pattern.match(line)
level_str, label, content = match.group(1, 2, 3)
label = label[0].upper() + label[1:].lower()
return len(level_str), label, mistletoe.span_token.tokenize_inner(content)
mistletoe.block_token.add_token(Note)
class Definition(mistletoe.block_token.BlockToken):
pattern = re.compile(r'(\t*)(\*\*\*.+\*\*\*(?::|.*means|.*includes).*)')
def __init__(self, match):
self.level, self.children = match
@classmethod
def start(cls, line):
return cls.pattern.match(line)
@classmethod
def read(cls, lines):
line = next(lines)
match = cls.pattern.match(line)
level_str, content = match.group(1, 2)
return len(level_str), mistletoe.span_token.tokenize_inner(content)
mistletoe.block_token.add_token(Definition)
def convert_leading_tabs(string):
return string
mistletoe.block_token.Quote.convert_leading_tabs = convert_leading_tabs
class Table(mistletoe.block_token.BlockToken):
def __init__(self, lines):
if '---' in lines[1]:
self.column_align = [mistletoe.block_token.Table.parse_align(column)
for column in mistletoe.block_token.Table.split_delimiter(lines[1])]
self.header = TableRow(lines[0], 0, self.column_align)
self.children = [TableRow(line, rownum + 1, self.column_align) for rownum, line in enumerate(lines[2:])]
else:
self.column_align = [None]
self.children = [TableRow(line, rownum) for rownum, line in enumerate(lines)]
@staticmethod
def start(line):
return '|' in line
@staticmethod
def read(lines):
lines.anchor()
line_buffer = [next(lines)]
while lines.peek() is not None and '|' in lines.peek():
line_buffer.append(next(lines))
if len(line_buffer) < 2 or '---' not in line_buffer[1]:
lines.reset()
return None
return line_buffer
class TableRow(mistletoe.block_token.BlockToken):
def __init__(self, line, rownum, row_align=None):
self.rownum = rownum
self.row_align = row_align or [None]
cells = filter(None, line.strip().split('|'))
self.children = [TableCell(x[0].strip() if x[0] else '', self, colnum, x[1])
for colnum, x in enumerate(zip_longest(cells, self.row_align))]
self.label = self.children[0].label
class TableCell(mistletoe.block_token.BlockToken):
pattern1 = re.compile(r'\s*(?:([0-9A-Za-z]+)(?:\[([0-9]+)\])?\s+)?(.*)')
def __init__(self, inner, row, colnum, align=None):
self.row = row
self.colnum = colnum
self.align = align
if self.colnum == 0 or self.row.rownum == 0:
label, weight, content = self.pattern1.match(inner).group(1, 2, 3)
self.label = label.strip() if label else None
self.weight = int(weight) if weight else 1
self.children = mistletoe.span_token.tokenize_inner(content)
else:
self.label = None
self.weight = None
self.children = mistletoe.span_token.tokenize_inner(inner)
mistletoe.block_token.remove_token(mistletoe.block_token.Table)
mistletoe.block_token.add_token(Table)
class CrossReference(mistletoe.span_token.SpanToken):
pattern = re.compile(r'`(?:([A-Za-z.]+?)\s+)?([0-9A-Za-z\(\)]+?)`_')
pattern_parts = re.compile(r'^[0-9A-Za-z]+|\([0-9A-Za-z]+\)')
def __init__(self, match):
self.reference_type = match.group(1)
self.reference_num = match.group(2)
self.reference_parts = re.findall(self.pattern_parts, self.reference_num)
def get_reference(self):
parent = None
elem = self.parent
while elem is not None:
if isinstance(elem, SubrulesItem):
parent = elem
break
elem = elem.parent
if parent is None:
# Might be a freestanding rule
section = next((x for x in reversed(self.parent.parent.children[0:self.parent.parent.children.index(self.parent)]) if isinstance(x, NumberedHeading)), None)
if section:
parent = section
if parent and parent.full_label():
basenum = re.findall(self.pattern_parts, parent.full_label())
else:
basenum = []
# Try possibilities
for index in reversed(range(0, len(basenum) + 1)):
try_ref = ''.join(basenum[:index] + self.reference_parts)
if try_ref in self.doc.full_label_map:
return self.doc.full_label_map[try_ref]
return None
mistletoe.span_token.add_token(CrossReference)