Check all links in the book

src/calibre/ebooks/oeb/polish/check/links.py

@@ -0,0 +1,61 @@
+#!/usr/bin/env python
+# vim:fileencoding=utf-8
+from __future__ import (unicode_literals, division, absolute_import,
+                        print_function)
+
+__license__ = 'GPL v3'
+__copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
+
+from collections import defaultdict
+from urlparse import urlparse
+
+from calibre.ebooks.oeb.base import OEB_DOCS, OEB_STYLES
+from calibre.ebooks.oeb.polish.container import guess_type
+from calibre.ebooks.oeb.polish.check.base import BaseError, WARN
+
+class BadLink(BaseError):
+
+    HELP = _('The resource pointed to by this link does not exist. You should'
+             ' either fix, or remove the link.')
+    level = WARN
+
+class FileLink(BadLink):
+
+    HELP = _('This link uses the file:// URL scheme. This does not work with many ebook readers.'
+             ' Remove the file:// prefix and make sure the link points to a file inside the book.')
+
+class LocalLink(BadLink):
+
+    HELP = _('This link points to a file outside the book. It will not work if the'
+             ' book is read on any computer other than the one it was created on.'
+             ' Either fix or remove the link.')
+
+def check_links(container):
+    links_map = defaultdict(set)
+    xml_types = {guess_type('a.opf'), guess_type('a.ncx')}
+    errors = []
+    a = errors.append
+
+    def fl(x):
+        x = repr(x)
+        if x.startswith('u'):
+            x = x[1:]
+        return x
+
+    for name, mt in container.mime_map.iteritems():
+        if mt in OEB_DOCS or mt in OEB_STYLES or mt in xml_types:
+            for href, lnum, col in container.iterlinks(name):
+                tname = container.href_to_name(href, name)
+                if tname is not None:
+                    if container.exists(tname):
+                        links_map[tname].add(name)
+                    else:
+                        a(BadLink(_('The linked resource %s does not exist') % fl(href), name, lnum, col))
+                else:
+                    purl = urlparse(href)
+                    if purl.scheme == 'file':
+                        a(FileLink(_('The link %s is a file:// URL') % fl(href), name, lnum, col))
+                    elif purl.path and purl.path.startswith('/') and purl.scheme in {'', 'file'}:
+                        a(LocalLink(_('The link %s points to a file outside the book') % fl(href), name, lnum, col))
+
+    return errors

src/calibre/ebooks/oeb/polish/check/main.py

@@ -14,6 +14,7 @@ from calibre.ebooks.oeb.polish.cover import is_raster_image
 from calibre.ebooks.oeb.polish.check.base import run_checkers
 from calibre.ebooks.oeb.polish.check.parsing import check_xml_parsing
 from calibre.ebooks.oeb.polish.check.images import check_raster_images
+from calibre.ebooks.oeb.polish.check.links import check_links
 
 XML_TYPES = frozenset(map(guess_type, ('a.xml', 'a.svg', 'a.opf', 'a.ncx')))
 
@@ -37,6 +38,8 @@ def run_checks(container):
     errors.extend(run_checkers(check_xml_parsing, html_items))
     errors.extend(run_checkers(check_raster_images, raster_images))
 
+    errors += check_links(container)
+
     return errors
 
 def fix_errors(container, errors):

src/calibre/ebooks/oeb/polish/container.py

@@ -9,6 +9,7 @@ __docformat__ = 'restructuredtext en'
 
 import os, logging, sys, hashlib, uuid, re, shutil
 from collections import defaultdict
+from bisect import bisect
 from io import BytesIO
 from urlparse import urlparse
 from future_builtins import zip
@@ -293,8 +294,17 @@ class Container(object):  # {{{
             if get_line_numbers:
                 with self.open(name) as f:
                     raw = self.decode(f.read())
+                    new_lines = tuple(m.start() + 1 for m in re.finditer(r'\n', raw))
+                    def position(pos):
+                        lnum = bisect(new_lines, pos)
+                        try:
+                            offset = abs(pos - new_lines[lnum - 1])
+                        except IndexError:
+                            offset = pos
+                        return (lnum + 1, offset)
                     for link, offset in itercsslinks(raw):
-                        yield link, 0, offset
+                        lnum, col = position(offset)
+                        yield link, lnum, col
             else:
                 for link in getUrls(self.parsed(name)):
                     yield link