def _html_to_wiki(self, html):
converter = HTMLWiki([])
try:
wiki = converter.from_html(html)
except converter.ParseError:
raise forms.ValidationError("Could not parse the HTML file.")
return wiki
def clean_wikitext(self):
markup = self.data['markup']
if markup != 'html':
# not HTML input: it's okay as-is
return self.data['wikitext']
html = self.data['wikitext']
converter = HTMLWiki([])
try:
wiki = converter.from_html(html)
except converter.ParseError:
raise forms.ValidationError("Could not parse the HTML file.")
return wiki
def convert_content(request, course_slug, page_label=None):
"""
Convert between wikicreole and HTML (AJAX called in editor when switching editing modes)
"""
if request.method != 'POST':
return ForbiddenResponse(request, 'POST only')
if 'to' not in request.POST:
return ForbiddenResponse(request, 'must send "to" language')
if 'data' not in request.POST:
return ForbiddenResponse(request, 'must sent source "data"')
offering = get_object_or_404(CourseOffering, slug=course_slug)
to = request.POST['to']
data = request.POST['data']
if to == 'html':
# convert wikitext to HTML
# temporarily change the current version to get the result (but don't save)
if page_label:
page = get_object_or_404(Page, offering=offering, label=page_label)
pv = page.current_version()
else:
# create temporary Page for conversion during creation
p = Page(offering=offering)
pv = PageVersion(page=p)
pv.wikitext = data
pv.diff_from = None
result = {'data': pv.html_contents()}
return HttpResponse(json.dumps(result),
content_type="application/json")
else:
# convert HTML to wikitext
converter = HTMLWiki([])
try:
wiki = converter.from_html(data)
except converter.ParseError:
wiki = ''
result = {'data': wiki}
return HttpResponse(json.dumps(result),
content_type="application/json")
def convert_content(request, course_slug, page_label=None):
"""
Convert between wikicreole and HTML (AJAX called in editor when switching editing modes)
"""
if request.method != 'POST':
return ForbiddenResponse(request, 'POST only')
if 'to' not in request.POST:
return ForbiddenResponse(request, 'must send "to" language')
if 'data' not in request.POST:
return ForbiddenResponse(request, 'must sent source "data"')
offering = get_object_or_404(CourseOffering, slug=course_slug)
to = request.POST['to']
data = request.POST['data']
if to == 'html':
# convert wikitext to HTML
# temporarily change the current version to get the result (but don't save)
if page_label:
page = get_object_or_404(Page, offering=offering, label=page_label)
pv = page.current_version()
else:
# create temporary Page for conversion during creation
p = Page(offering=offering)
pv = PageVersion(page=p)
pv.wikitext = data
pv.diff_from = None
result = {'data': pv.html_contents()}
return HttpResponse(json.dumps(result), content_type="application/json")
else:
# convert HTML to wikitext
converter = HTMLWiki([])
try:
wiki = converter.from_html(data)
except converter.ParseError:
wiki = ''
result = {'data': wiki}
return HttpResponse(json.dumps(result), content_type="application/json")
def __init__(self, offering, editor, *args, **kwargs):
super(SiteImportForm, self).__init__(*args, **kwargs)
self.converter = HTMLWiki([])
self.offering = offering
self.editor = editor
class SiteImportForm(forms.Form):
url = forms.URLField(required=True,
label='URL',
widget=forms.TextInput(attrs={'size': 70}))
can_read = forms.ChoiceField(choices=READ_ACL_CHOICES,
required=True,
initial="ALL")
can_write = forms.ChoiceField(choices=WRITE_ACL_CHOICES,
required=True,
initial="STAF")
def __init__(self, offering, editor, *args, **kwargs):
super(SiteImportForm, self).__init__(*args, **kwargs)
self.converter = HTMLWiki([])
self.offering = offering
self.editor = editor
def _labelize(self, url, title):
path = urlparse.urlsplit(url).path
if path:
parts = path.split('/')
if len(parts) >= 1 and parts[-1]:
return parts[-1]
elif len(parts) >= 2 and parts[-2]:
return parts[-2]
def _import_page(self, url):
try:
fh = urllib2.urlopen(url, timeout=20)
if 'content-type' in fh.headers:
ctype = fh.headers['content-type'].split(';')[0]
is_html = ctype in ['text/html', 'application/xhtml+xml']
else:
is_html = False
html = fh.read()
fh.close()
except:
raise forms.ValidationError('Could not fetch "%s".' % (url))
if not is_html:
raise forms.ValidationError('Not HTML at "%s".' % (url))
try:
wiki, title, urls = self.converter.from_html_full(html)
except self.converter.ParseError:
raise forms.ValidationError("Could not parse the HTML file %s." %
(url))
label = self._labelize(url, title)
if not title:
title = label
page = Page(offering=self.offering, label=label)
pv = PageVersion(page=page,
editor=self.editor,
title=title,
wikitext=wiki,
comment="imported content")
#print (page, pv, pv.title)
#print [urlparse.urljoin(url, u) for u in urls]
return page, pv, urls
def clean_url(self):
url = self.cleaned_data['url']
if not url:
return None
baseurl = urlparse.urljoin(url, "./")
needed = set([url])
done = set()
found = {}
errors = []
while needed:
if len(found) >= 20:
break
url = needed.pop()
try:
page, pv, newurls = self._import_page(url)
except forms.ValidationError as e:
errors.append(e.messages[0])
done.add(url)
newurls = set((urlparse.urljoin(url, u) for u in newurls))
newurls = set((u for u in newurls if u.startswith(baseurl)))
needed = needed | newurls - done
found[page.label] = (page, pv)
#print ">>>", found, errors
return found, errors