github/pelican
1
0
Fork 1
mirror of https://github.com/getpelican/pelican.git synced 2025-10-15 20:28:56 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Get HtmlReader to work again

Browse source 
wrote unit tests and documentation, improved regular expression.
The HtmlReader is enabled by default now and parses metadata in html
files of the form:
<!-- key:value -->
This commit is contained in:
Florian Jacob 2012-09-02 10:09:08 +02:00
commit 39db9ddcfd
7 changed files with 72 additions and 15 deletions
Download patch file Download diff file Expand all files Collapse all files

31
pelican/readers.py
View file

@ -142,19 +142,30 @@ class MarkdownReader(Reader):
class HtmlReader(Reader):
file_extensions = ['html', 'htm']
_re = re.compile('\<\!\-\-\#\s?[A-z0-9_-]*\s?\:s?[A-z0-9\s_-]*\s?\-\-\>')
# re.DOTALL and .*? (minimal match of an arbitrary number of characters)
# allow multi-line metadata to be matched correctly
_re = re.compile('<\!--([^\:]*):(.*?)-->', re.DOTALL)
def read(self, filename):
"""Parse content and metadata of (x)HTML files"""
with open(filename) as content:
metadata = {'title': 'unnamed'}
for i in self._re.findall(content):
key = i.split(':')[0][5:].strip()
value = i.split(':')[-1][:-3].strip()
name = key.lower()
metadata[name] = self.process_metadata(name, value)
"""Parse content and metadata of (x)HTML files.
return content, metadata
Matches for metadata tags in the form <!-- name:value -->
Activated when you add 'html' to your MARKUP settings variable
"""
content = open(filename)
metadata = {'title': 'unnamed'}
for comment in self._re.findall(content):
key = comment[0].strip().lower()
value = comment[1].strip()
# remove identation from multi-line metadata
value = re.sub('[ \t]+', ' ', value)
value = re.sub(' ?\n ?', '\n', value)
metadata[key] = self.process_metadata(key, value)
return content, metadata
_EXTENSIONS = {}