Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Add support for the <base> tag to web2disk

Browse Source
This commit is contained in:
Kovid Goyal 2008-06-02 15:33:55 -07:00
parent ed5cc70502
commit 732961c07b
1 changed files with 8 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
src/calibre/web/fetch/simple.py
View File

@ -334,19 +334,23 @@ class RecursiveFetcher(object, LoggingInterface):
dsrc = xml_to_unicode(dsrc, self.verbose)[0] dsrc = xml_to_unicode(dsrc, self.verbose)[0]
soup = self.get_soup(dsrc) soup = self.get_soup(dsrc)
newbaseurl = f.geturl()
base = soup.find('base', href=True)
if base is not None:
newbaseurl = base['href']
self.log_debug('Processing images...') self.log_debug('Processing images...')
self.process_images(soup, f.geturl()) self.process_images(soup, newbaseurl)
if self.download_stylesheets: if self.download_stylesheets:
self.process_stylesheets(soup, f.geturl()) self.process_stylesheets(soup, newbaseurl)
res = os.path.join(linkdiskpath, basename(iurl)) res = os.path.join(linkdiskpath, basename(iurl))
self.downloaded_paths.append(res) self.downloaded_paths.append(res)
self.filemap[nurl] = res self.filemap[nurl] = res
if recursion_level < self.max_recursions: if recursion_level < self.max_recursions:
self.log_debug('Processing links...') self.log_debug('Processing links...')
self.process_links(soup, iurl, recursion_level+1) self.process_links(soup, newbaseurl, recursion_level+1)
else: else:
self.process_return_links(soup, iurl) self.process_return_links(soup, newbaseurl)
self.log_debug('Recursion limit reached. Skipping links in %s', iurl) self.log_debug('Recursion limit reached. Skipping links in %s', iurl)
if callable(self.postprocess_html_ext): if callable(self.postprocess_html_ext):