#!/usr/bin/env python3
from bs4 import BeautifulSoup
from bs4.element import NavigableString,PreformattedString,Comment, PageElement
import subprocess as s
import os,sys,re
def BeautifulSoupSucks(inp):
return BeautifulSoup(inp,"html.parser")
nl = re.compile('\n+')
NLPAT = '\s*\n\s*'
nlStart = re.compile('^'+NLPAT+'(.*)',re.DOTALL)
nlEnd = re.compile('(.*?)'+NLPAT+'$',re.DOTALL)
def inline(e):
return e.name in {'a','i','b','u','s','span','q','small'}
def needsMD(cur):
if not hasattr(cur,'attrs'): return False
if 'md' in cur.attrs:
del cur.attrs['md']
return True
# also mdify class="author" because I forget that one.
if not 'class' in cur.attrs: return False
return 'author' in cur.attrs['class']
def depth(e):
c = 0
while e.parent:
c += 1
e = e.parent
return c
things = "\\*_/"
outer = '\t\n .?!“”'
inner = '^' + "".join(set(things + outer).difference(set(".?!")))
md = re.compile("(["+outer+"])(?P["+things+"]+)(["+inner+"]+)(?P=head)(["+outer+"])")
def realtags(m):
try:
head,mod,contents,tail = m.groups()
except ValueError:
print("uhhh", m.groups())
raise
if mod == '**':
tag = 'b'
elif mod in {'*','/','_'}:
tag = 'i'
else:
raise RuntimeError("huh?",mod)
return head + '<' + tag + '>' + contents + '' + tail
def get(source,stdin=False):
if stdin:
source = md.sub(realtags, source)
source = '' + source + ''
doc = BeautifulSoupSucks(source)
else:
with open(source) as inp:
source = md.sub(realtags,inp.read())
source = '' + source + ''
doc = BeautifulSoupSucks(source)
return doc
def fuckyou(e):
space = re.compile("[\n\t]+")
return repr(space.sub(' ',str(e))[:20])
def process(doc):
class Thingy:
before = None
cur = None
ended = True
def __init__(self, tag, start):
self.start = start
self.tag = tag
assert self.ended
def end(self):
if not self.cur: return
self.cur.append('\n')
self.ended = True
def current(self):
if not self.ended:
return self.cur
self.ended = False
self.cur = doc.new_tag(self.tag)
assert self.cur
if self.before is None:
self.start.insert(0,self.cur)
else:
self.before.insert_after(self.cur)
self.before = self.cur
return self.cur
class ListThingy(Thingy):
def __init__(self, start, ordered=True):
top = doc.new_tag('ol' if ordered else 'ul')
start.append(top)
super().__init__('li', top)
def end(self):
if self.ended: return
self.cur.insert_before('\n')
self.ended = True
# 1) yield each line, plus elements in between
# 2) group those into newline delineated (before,e,after) lists
# 3) act on those lists
def lines(e):
group = []
seen = set()
def commit():
nonlocal group
ret = group
group = []
return ret
es = []
while e:
nexte = e.next_sibling
es.append(e)
e.extract()
e = nexte
for e in es:
# if e is text, yield lines
# otherwise, yield element
if isinstance(e,Comment):
group.append(e)
elif isinstance(e,NavigableString):
m = nlStart.match(e)
if m and m.group(1):
if group:
yield commit()
e = m.group(1)
lines = nl.split(e)
end_newline = lines and lines[-1] == ""
lines = [line for line in lines if lines]
if lines:
for line in lines[:-1]:
# if not nlStart, group won't be empty!
if not line: continue
group.append(line)
yield commit()
if end_newline:
if lines[-1]:
group.append(lines[-1])
yield commit()
else:
if lines[-1]:
group.append(lines[-1])
else:
if group and e.name not in {'i','b','span','u','s','em','a'}:
yield commit()
group.append(e)
if group:
yield commit()
def mdify(cur):
cur_list = None
try: cur.contents[0]
except IndexError: return
for group in lines(cur.contents[0]):
def derp_list():
head = group[0][2:]
tail = group[1:]
li = doc.new_tag('li')
li.append(head)
for e in tail:
li.append(e)
cur_list.append(li)
def derp_p():
if len(group) == 1 and not isinstance(group[0],str):
cur.append(group[0])
return
p = doc.new_tag('p')
gotsome = False
for e in group:
gotsome = gotsome or e
p.append(e)
if gotsome:
cur.append(p)
if cur_list is None:
pythonsucks = True
def maybe_listaderp():
nonlocal cur_list
if not group: return False
if not isinstance(group[0],str):
return False
if group[0].startswith('- '):
cur_list = doc.new_tag('ol')
return True
elif group[0].startswith('* '):
cur_list = doc.new_tag('ul')
return True
return False
if maybe_listaderp():
assert(cur_list)
cur.append(cur_list)
derp_list()
else:
derp_p()
else:
if isinstance(group[0],str):
if group[0].startswith('- ') or group[0].startswith('* '):
derp_list()
else:
cur_list = None
derp_p()
cur = doc.find('body')
while hasattr(cur,'contents') and len(cur.contents) == 1:
cur = cur.contents[0]
# stackless traversal, starting depth first and
# going up from there.
# descend into the deepest contents first
# then check if it's md and replace if so
# then go to the next sibling if exists
# otherwise go up to the parent if exist
# otherwise, done!
def mdifyAll():
nonlocal cur
while cur:
if hasattr(cur,'contents') and len(cur.contents):
cur = cur.contents[0]
continue
if needsMD(cur):
mdify(cur)
if cur.next_sibling:
cur = cur.next_sibling
elif cur.parent:
# like the outer loop, but w/out going down contents
# go up the parents until we find a sibling
cur = cur.parent
while cur:
nextcur = cur.next_sibling
if cur.name == 'body':
mdify(cur)
return
elif needsMD(cur):
mdify(cur)
if nextcur:
cur = nextcur
break
else:
cur = cur.parent
else:
break
mdifyAll()
return doc
def markdown(template,source,defaultTitle=None):
tmpl = BeautifulSoupSucks(template)
mk = get(source,False)
title = mk.find('title') or defaultTitle
if title:
title.extract()
title = str(title.string)
tmpl.find('title').append(title)
tmpl.find('intitle').replace_with(title)
mk = process(mk)
bod = mk.find('body')
if bod:
bod = bod.contents
else:
bod = [mk]
content = tmpl.find('content')
while bod:
derp = bod[-1]
derp.extract()
assert derp.parent is None
content.insert_after(derp)
content.decompose()
return tmpl,title
defaultTemplate = '''
'''
def getTemplate(default='template.xhtml'):
template = os.environ.get('template')
if not template:
template = default
if os.path.exists(template):
with open(template,'rt') as inp:
return inp.read()
with open(template,'wt') as out:
out.write(defaultTemplate)
return defaultTemplate
if __name__ == '__main__':
doc,title = markdown(getTemplate(),sys.argv[1])
if 'pretty' in os.environ:
print(doc.prettify())
else:
print(str(doc))