def createText(parent, text, sources=()):
global nodeDict
t = TextNode()
t.parent = lookupNode(parent, Slot)
t.text = text
t.save()
for s in sources:
t.sources.add(s)
t.save()
nodeDict[t.getTextPath()] = t
return t
def submit_slot_with_text(request):
short_title = request.POST['short_title']
if not short_title:
return "ERROR: Short Title darf nicht leer sein."
parent_id = int(request.POST['parent_id'])
# lookup if slot exists
slots = Slot.objects.filter(parent__id=parent_id, short_title=short_title)
if slots:
s = slots[0]
else :
s = Slot()
s.short_title = request.POST['short_title']
s.parent_id = parent_id
s.save()
t = TextNode()
t.text = request.POST['text']
t.parent = s
t.save()
add_auto_upvote(request.user, t)
return HttpResponseRedirect(t.getTextPath())
def parse(s, parent_slot):
#make sure we start with a heading 1
m = h1_start.match(s) # TODO: match short titles and warn about
if m :
title = m.groups("title")[0]
title = title.partition("ยง")[0] # silently remove attempt to set short_title in H1
s = h1_start.sub("", s)
else :
# TODO: warn about missing title
title = parent_slot.short_title
# do we need a StructureNode or will a TextNode do?
if not general_h.search(s) :
# TextNode
t = TextNode()
t.text = "= %s =\n"%title.strip() + s.strip()
t.parent = parent_slot
t.save()
return t
# else : StructureNode
node = StructureNode()
node.parent = parent_slot
node.save()
# determine used header depth:
level = 0
for i in range(2, 7):
m = getHeadingMatcher(i)
if m.search(s):
level = i
break
assert 1 < level < 7
split_doc = m.split(s)
# now the text before, between and after headings is split_doc[0::3]
# the text of the headings are split_doc[1::3]
# and the short titles (or None if omitted) are split_doc[2::3]
# what do we do now?
# leading text is used to construct an "Einleitung" Slot and TextNode
introduction = Slot()
introduction.parent = node
introduction.short_title = "Einleitung"
introduction.save()
introduction_text = TextNode()
introduction_text.parent = introduction
intro_text = split_doc[0]
# assert that no headings are in intro-text
if general_h.search(intro_text):
# TODO: Warn!
intro_text = general_h.sub(r"~\1", intro_text)
#general_h.
introduction_text.text = "= %s =\n"%title + intro_text
introduction_text.save()
# iterate the headings, short_titles, and corresponding texts:
short_title_set = set()
for title, short_title, text in zip(split_doc[1::3], split_doc[2::3], split_doc[3::3]):
# check if short_title is valid/unique/exists
if not short_title or len(short_title.strip()) == 0:
short_title=title[:min(15, len(title))]
# make short titles valid
short_title = short_title[:min(20, len(short_title))]
short_title = strip_accents(short_title)
short_title = invalid_symbols.sub('',short_title)
if short_title in short_title_set:
i = 1
while short_title + str(i) in short_title_set:
i += 1
short_title += str(i)
short_title_set.add(short_title)
slot = Slot()
slot.parent = node
slot.short_title = short_title.strip().replace(" ", "_")
slot.save()
parse("= %s =\n"%title.strip() + text.strip(), slot)
return node
