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

pelican-import, fix urls of attached files

Browse source 
With this patch wp-attach modifies ```<img``` and  ```<a href``` urls to point at the attachment path - so  ```http://example.com/wp-content/uploads/blah.png```   becomes    ```wp-content/uploads/blah.png```

This needs wp-content in pelicanconf STATIC_FILES

I thought I could use {filename} but since the importer is generates source html this doesn't work.
It would be nice if there was a way to customise the output path, but not sure how that would work.
This commit is contained in:
Stuart Axon 2017-10-12 00:45:59 +01:00 • committed by GitHub
commit 97c7089d28
1 changed files with 23 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

30
pelican/tools/pelican_import.py
View file

@ -12,6 +12,7 @@ import time
from codecs import open
import collections
from six.moves.urllib.error import URLError
from six.moves.urllib.parse import urlparse
from six.moves.urllib.request import urlretrieve
@ -29,11 +30,22 @@ except ImportError:
logger = logging.getLogger(__name__)
def decode_wp_content(content, br=True):
def decode_wp_content(content, attached_files, br=True):
pre_tags = {}
if content.strip() == "":
return ""
if attached_files:
for path, urls in attached_files.items():
for url in urls:
content = re.sub(r'(<a\s+[^>]*href=")%s(")' % url,
r'\1%s\2' % path,
content)
content = re.sub(r'(<img\s+[^>]*src=")%s(")' % url,
r'\1%s\2' % path,
content)
content += "\n"
if "<pre" in content:
pre_parts = content.split("</pre>")
@ -651,10 +663,13 @@ def get_attachments(xml):
def download_attachments(output_path, urls):
"""Downloads WordPress attachments and returns a list of paths to
attachments that can be associated with a post (relative path to output
directory). Files that fail to download, will not be added to posts"""
locations = []
"""Downloads WordPress attachments and returns a returns a dict {url:path} of
attachments that can be associated with a post (relative path to output
directory). Files that fail to download, will not be added to posts
{relpath: {set of urls}}
"""
locations = collections.defaultdict(set)
for url in urls:
path = urlparse(url).path
# teardown path and rebuild to negate any errors with
@ -671,7 +686,8 @@ def download_attachments(output_path, urls):
print('downloading {}'.format(filename))
try:
urlretrieve(url, os.path.join(full_path, filename))
locations.append(os.path.join(localpath, filename))
relpath = os.path.join(localpath, filename)
locations[relpath].add(url)
except (URLError, IOError) as e:
# Python 2.7 throws an IOError rather Than URLError
logger.warning("No file could be downloaded from %s\n%s", url, e)
@ -720,7 +736,7 @@ def fields2pelican(
# Replace newlines with paragraphs wrapped with <p> so
# HTML is valid before conversion
if in_markup == 'wp-html':
new_content = decode_wp_content(content)
new_content = decode_wp_content(content, attached_files)
else:
paragraphs = content.splitlines()
paragraphs = ['<p>{0}</p>'.format(p) for p in paragraphs]