0.8.45

Changelog.yaml

@@ -19,11 +19,67 @@
 #   new recipes:
 #     - title: 
 
+- version: 0.8.45
+  date: 2012-03-30
+
+  new features:
+    - title: "E-book viewer: Allow the up and down keys to scroll past section boundaries"
+
+    - title: "calibredb: Allow specification of basic metadata on the command line when adding books."
+      tickets: [951063]
+
+    - title: "Driver for Samsung Galaxy Plus GT-I9001"
+
+    - title: "KF8 Input: Support KF8 format Amazon book samples."
+      tickets: [963418]
+
+    - title: "When a new plugin is added to calibre for the first time, have its icon (if any) show up even when a device is connected (this can be changed by the user at the time of plugin installation)"
+
+    - title: "Add keyboard shortcuts for Bold, Italic and Underline to the comments editor in the edit metadata dialog"
+      tickets: [963559] 
+ 
+  bug fixes:
+    - title: "E-book viewer: Fix last read position (and bookmarks in general) being inaccurate for some books."
+      description: "The technique for marking locations in books used by the viewer has changed. The new technique should be much more accurate than the last one, especially when the font size at which the book is being viewed is changed. Note that this change means that bookmarks created with this release of calibre will not be read by previous calibre versions. On a technical note, the viewer now uses the CFI specification from the EPUB 3 standard for bookmarks."
+      type: major
+
+    - title: "Workarounds for a few regressions in the user interface in 0.8.44 caused by the update to Qt 4.8.0"
+
+    - title: "Books list: Preserve the horizontal scroll position when sorting by a column"
+
+    - title: "Fix saving to disk and then adding the book back not restoring tags-like custom columns"
+
+    - title: "Linux installer: Fix completion for ebook-convert not working." 
+      tickets: [967834]
+
+    - title: "MOBI Output: Recognize type=text in addition to type=start guide elements"
+
+    - title: "Get Books: Updates to Nexto, Ebookpoint and Woblink stores"
+
+    - title: "Fix unable to clear username/password in Fetch news dialog"
+
+    - title: "PDF Output: Fix margin specifications not being applied"
+
+    - title: "Linux installer: Manually preserve the defaults.list mimetype association file to workaround buggy xdg-desktop-menu implementations in some distros."
+      tickets: [926559]
+
+    - title: "E-book viewer: Fix regression that caused the ebook viewer to stop functioning if it is launched from the main calibre program and then the main calibre program is closed."
+      tickets: [963960]
+
+
+  improved recipes:
+    - Our Daily Bread
+
+  new recipes:
+    - title: NRC Handelsblad (free)
+      author: veezh
+
 - version: 0.8.44
   date: 2012-03-23
 
   new features:
-    - title: "E-book viewer: A whole new full screen mode, with no toolbars to distract from the text and the ability to set the width of the column of text via Preferences in the ebook viewer."
+    - title: "E-book viewer: A whole new full screen mode."
+      description: "The new mode has no toolbars to distract from the text and the ability to set the width of the column of text via Preferences in the ebook viewer. Click the Fullscreen button on the toolbar in the viewer to enter fullscreen mode (or press the F11 or Ctrl+Shit+F keys)"
       type: major
       tickets: [959830]
 

recipes/le_monde.recipe

@@ -1,8 +1,9 @@
 __license__   = 'GPL v3'
-__copyright__ = '2011'
+__copyright__ = '2012'
 '''
 lemonde.fr
 '''
+import re
 from calibre.web.feeds.recipes import BasicNewsRecipe
 
 class LeMonde(BasicNewsRecipe):
@@ -24,7 +25,7 @@ class LeMonde(BasicNewsRecipe):
                     .ariane{font-size:xx-small;}
                     .source{font-size:xx-small;}
                     #.href{font-size:xx-small;}
-                    .LM_caption{color:#666666; font-size:x-small;}
+                    #.figcaption style{color:#666666; font-size:x-small;}
                     #.main-article-info{font-family:Arial,Helvetica,sans-serif;}
                     #full-contents{font-size:small; font-family:Arial,Helvetica,sans-serif;font-weight:normal;}
                     #match-stats-summary{font-size:small; font-family:Arial,Helvetica,sans-serif;font-weight:normal;}
@@ -40,8 +41,88 @@ class LeMonde(BasicNewsRecipe):
 
     remove_empty_feeds = True
 
-    auto_cleanup = True
+    filterDuplicates = True
 
+    def preprocess_html(self, soup):
+        for alink in soup.findAll('a'):
+            if alink.string is not None:
+               tstr = alink.string
+               alink.replaceWith(tstr)
+        return soup
+
+    preprocess_regexps = [
+        (re.compile(r'([0-9])%'), lambda m: m.group(1) + ' %'),
+        (re.compile(r'([0-9])([0-9])([0-9]) ([0-9])([0-9])([0-9])'), lambda m: m.group(1) + m.group(2) + m.group(3) + ' ' + m.group(4) + m.group(5) + m.group(6)),
+        (re.compile(r'([0-9]) ([0-9])([0-9])([0-9])'), lambda m: m.group(1) + ' ' + m.group(2) + m.group(3) + m.group(4)),
+        (re.compile(r'<span>'), lambda match: ' <span>'),
+        (re.compile(r'\("'), lambda match: '(&laquo;&nbsp;'),
+        (re.compile(r'"\)'), lambda match: '&nbsp;&raquo;)'),
+        (re.compile(r'&ldquo;'), lambda match: '(&laquo;&nbsp;'),
+        (re.compile(r'&rdquo;'), lambda match: '&nbsp;&raquo;)'),
+        (re.compile(r'>\''), lambda match: '>&lsquo;'),
+        (re.compile(r' \''), lambda match: ' &lsquo;'),
+        (re.compile(r' &quot;'), lambda match: ' &laquo;&nbsp;'),
+        (re.compile(r'>&quot;'), lambda match: '>&laquo;&nbsp;'),
+        (re.compile(r'&quot;<'), lambda match: '&nbsp;&raquo;<'),
+        (re.compile(r'&quot; '), lambda match: '&nbsp;&raquo; '),
+        (re.compile(r'&quot;,'), lambda match: '&nbsp;&raquo;,'),
+        (re.compile(r'\''), lambda match: '&rsquo;'),
+        (re.compile(r'"<em>'), lambda match: '<em>&laquo;&nbsp;'),
+        (re.compile(r'"<em>"</em><em>'), lambda match: '<em>&laquo;&nbsp;'),
+        (re.compile(r'"<a href='), lambda match: '&laquo;&nbsp;<a href='),
+        (re.compile(r'</em>"'), lambda match: '&nbsp;&raquo;</em>'),
+        (re.compile(r'</a>"'), lambda match: '&nbsp;&raquo;</a>'),
+        (re.compile(r'"</'), lambda match: '&nbsp;&raquo;</'),
+        (re.compile(r'>"'), lambda match: '>&laquo;&nbsp;'),
+        (re.compile(r'"<'), lambda match: '&nbsp;&raquo;<'),
+        (re.compile(r'&rsquo;"'), lambda match: '&rsquo;«&nbsp;'),
+        (re.compile(r' "'), lambda match: ' &laquo;&nbsp;'),
+        (re.compile(r'" '), lambda match: '&nbsp;&raquo; '),
+        (re.compile(r'"\.'), lambda match: '&nbsp;&raquo;.'),
+        (re.compile(r'",'), lambda match: '&nbsp;&raquo;,'),
+        (re.compile(r'"\?'), lambda match: '&nbsp;&raquo;?'),
+        (re.compile(r'":'), lambda match: '&nbsp;&raquo;:'),
+        (re.compile(r'";'), lambda match: '&nbsp;&raquo;;'),
+        (re.compile(r'"\!'), lambda match: '&nbsp;&raquo;!'),
+        (re.compile(r' :'), lambda match: '&nbsp;:'),
+        (re.compile(r' ;'), lambda match: '&nbsp;;'),
+        (re.compile(r' \?'), lambda match: '&nbsp;?'),
+        (re.compile(r' \!'), lambda match: '&nbsp;!'),
+        (re.compile(r'\s»'), lambda match: '&nbsp;»'),
+        (re.compile(r'«\s'), lambda match: '«&nbsp;'),
+        (re.compile(r' %'), lambda match: '&nbsp;%'),
+        (re.compile(r'\.jpg&nbsp;&raquo; width='), lambda match: '.jpg'),
+        (re.compile(r'\.png&nbsp;&raquo; width='), lambda match: '.png'),
+        (re.compile(r' &ndash; '), lambda match: '&nbsp;&ndash; '),
+        (re.compile(r'figcaption style="display:none"'), lambda match: 'figcaption'),
+        (re.compile(r' – '), lambda match: '&nbsp;&ndash; '),
+        (re.compile(r' - '), lambda match: '&nbsp;&ndash; '),
+        (re.compile(r' -,'), lambda match: '&nbsp;&ndash;,'),
+        (re.compile(r'&raquo;:'), lambda match: '&raquo;&nbsp;:'),
+        ]
+
+
+    keep_only_tags    = [
+                       dict(name='div', attrs={'class':['global']})
+                        ]
+
+    remove_tags = [
+                       dict(name='div', attrs={'class':['bloc_base meme_sujet']}),
+                       dict(name='p', attrs={'class':['lire']})
+                        ]
+
+    remove_tags_after = [dict(id='fb-like')]
+
+    def get_article_url(self, article):
+          url = article.get('guid', None)
+          if '/chat/' in url or '.blog' in url or '/video/' in url or '/sport/' in url or '/portfolio/' in url or '/visuel/' in url :
+              url = None
+          return url
+
+#    def get_article_url(self, article):
+#        link = article.get('link')
+#        if 'blog' not in link and ('chat' not in link):
+#             return link
 
     feeds          = [
                       ('A la une', 'http://www.lemonde.fr/rss/une.xml'),
@@ -66,11 +147,3 @@ class LeMonde(BasicNewsRecipe):
            cover_url = link_item.img['src']
 
         return cover_url
-
-    def get_article_url(self, article):
-        url = article.get('guid', None)
-        if '/chat/' in url or '.blog' in url or '/video/' in url or '/sport/' in url or '/portfolio/' in url or '/visuel/' in url :
-            url = None
-        return url
-
-

recipes/marketing_magazine.recipe

@@ -1,4 +1,6 @@
 __license__   = 'GPL v3'
+__author__    = 'faber1971'
+description   = 'Collection of Italian marketing websites - v1.04 (17, March 2012)'
 
 from calibre.web.feeds.news import BasicNewsRecipe
 
@@ -9,12 +11,9 @@ class AdvancedUserRecipe1327062445(BasicNewsRecipe):
     auto_cleanup = True
     remove_javascript = True
     no_stylesheets = True
+    conversion_options = {'linearize_tables': True}
     remove_tags = [
                      dict(name='ul', attrs={'id':'ads0'})
                   ]
     masthead_url            = 'http://www.simrendeogun.com/wp-content/uploads/2011/06/New-Marketing-Magazine-Logo.jpg'
-    __author__    = 'faber1971'
+    feeds          = [(u'My Marketing', u'http://feed43.com/0537744466058428.xml'), (u'My Marketing_', u'http://feed43.com/8126723074604845.xml'), (u'Venturini', u'http://robertoventurini.blogspot.com/feeds/posts/default?alt=rss'), (u'Ninja Marketing', u'http://feeds.feedburner.com/NinjaMarketing'), (u'Comunitàzione', u'http://www.comunitazione.it/feed/novita.asp'), (u'Brandforum news', u'http://www.brandforum.it/rss/news'), (u'Brandforum papers', u'http://www.brandforum.it/rss/papers'), (u'MarketingArena', u'http://feeds.feedburner.com/marketingarena'), (u'minimarketing', u'http://feeds.feedburner.com/minimarketingit'), (u'Marketing Journal', u'http://feeds.feedburner.com/marketingjournal/jPwA'), (u'Disambiguando', u'http://giovannacosenza.wordpress.com/feed/')]
-    description   = 'Collection of Italian marketing websites - v1.03 (20, February 2012)'
-    language = 'it'
-
-    feeds          = [(u'My Marketing', u'http://feed43.com/0537744466058428.xml'), (u'My Marketing_', u'http://feed43.com/8126723074604845.xml'), (u'Venturini', u'http://robertoventurini.blogspot.com/feeds/posts/default?alt=rss'), (u'Ninja Marketing', u'http://feeds.feedburner.com/NinjaMarketing'), (u'Comunitàzione', u'http://www.comunitazione.it/feed/novita.asp'), (u'Brandforum news', u'http://www.brandforum.it/rss/news'), (u'Brandforum papers', u'http://www.brandforum.it/rss/papers'), (u'MarketingArena', u'http://feeds.feedburner.com/marketingarena'), (u'minimarketing', u'http://feeds.feedburner.com/minimarketingit'), (u'Disambiguando', u'http://giovannacosenza.wordpress.com/feed/')]

recipes/nrc_handelsblad.recipe

@@ -0,0 +1,76 @@
+__license__   = 'GPL v3'
+__copyright__ = '2012'
+'''
+nrc.nl
+'''
+from calibre.web.feeds.recipes import BasicNewsRecipe
+
+class NRC(BasicNewsRecipe):
+    title                  = 'NRC Handelsblad'
+    __author__             = 'veezh'
+    description            = 'Nieuws (no subscription needed)'
+    oldest_article         = 1
+    max_articles_per_feed  = 100
+    no_stylesheets         = True
+    #delay                  = 1
+    use_embedded_content   = False
+    encoding               = 'utf-8'
+    publisher              = 'nrc.nl'
+    category               = 'news, Netherlands, world'
+    language               = 'nl'
+    timefmt = ''
+    #publication_type       = 'newsportal'
+    extra_css = '''
+                    h1{font-size:130%;}
+                    #h2{font-size:100%;font-weight:normal;}
+                    #.href{font-size:xx-small;}
+                    .bijschrift{color:#666666; font-size:x-small;}
+                    #.main-article-info{font-family:Arial,Helvetica,sans-serif;}
+                    #full-contents{font-size:small; font-family:Arial,Helvetica,sans-serif;font-weight:normal;}
+                    #match-stats-summary{font-size:small; font-family:Arial,Helvetica,sans-serif;font-weight:normal;}
+                '''
+    #preprocess_regexps     = [(re.compile(r'<!--.*?-->', re.DOTALL), lambda m: '')]
+    conversion_options = {
+                              'comments'        : description
+                             ,'tags'            : category
+                             ,'language'        : language
+                             ,'publisher'       : publisher
+                             ,'linearize_tables': True
+                          }
+
+    remove_empty_feeds = True
+
+    filterDuplicates = True
+
+    def preprocess_html(self, soup):
+        for alink in soup.findAll('a'):
+            if alink.string is not None:
+               tstr = alink.string
+               alink.replaceWith(tstr)
+        return soup
+
+    keep_only_tags = [dict(name='div', attrs={'class':'article'})]
+    remove_tags_after = [dict(id='broodtekst')]
+
+#    keep_only_tags    = [
+#                       dict(name='div', attrs={'class':['label']})
+#                        ]
+
+#    remove_tags_after = [dict(name='dl', attrs={'class':['tags']})]
+
+#    def get_article_url(self, article):
+#        link = article.get('link')
+#        if 'blog' not in link and ('chat' not in link):
+#             return link
+
+    feeds          = [
+#                      ('Nieuws', 'http://www.nrc.nl/rss.php'),
+                      ('Binnenland', 'http://www.nrc.nl/nieuws/categorie/binnenland/rss.php'),
+                      ('Buitenland', 'http://www.nrc.nl/nieuws/categorie/buitenland/rss.php'),
+                      ('Economie', 'http://www.nrc.nl/nieuws/categorie/economie/rss.php'),
+                      ('Wetenschap', 'http://www.nrc.nl/nieuws/categorie/wetenschap/rss.php'),
+                      ('Cultuur', 'http://www.nrc.nl/nieuws/categorie/cultuur/rss.php'),
+                      ('Boeken', 'http://www.nrc.nl/boeken/rss.php'),
+                      ('Tech', 'http://www.nrc.nl/tech/rss.php/'),
+                      ('Klimaat', 'http://www.nrc.nl/klimaat/rss.php/'),
+                      ]

recipes/ourdailybread.recipe

@@ -14,6 +14,7 @@ class OurDailyBread(BasicNewsRecipe):
     language              = 'en'
     max_articles_per_feed = 100
     no_stylesheets        = True
+    auto_cleanup = True
     use_embedded_content  = False
     category              = 'ODB, Daily Devotional, Bible, Christian Devotional, Devotional, RBC Ministries, Our Daily Bread, Devotionals, Daily Devotionals, Christian Devotionals, Faith, Bible Study, Bible Studies, Scripture, RBC, religion'
     encoding              = 'utf-8'
@@ -25,12 +26,12 @@ class OurDailyBread(BasicNewsRecipe):
                             ,'linearize_tables' : True
                          }
 
-    keep_only_tags    = [dict(attrs={'class':'module-content'})]
+    #keep_only_tags    = [dict(attrs={'class':'module-content'})]
-    remove_tags       = [
+    #remove_tags       = [
-                           dict(attrs={'id':'article-zoom'})
+                           #dict(attrs={'id':'article-zoom'})
-                          ,dict(attrs={'class':'listen-now-box'})
+                          #,dict(attrs={'class':'listen-now-box'})
-                        ]
+                        #]
-    remove_tags_after = dict(attrs={'class':'readable-area'})
+    #remove_tags_after = dict(attrs={'class':'readable-area'})
 
     extra_css = '''
                 .text{font-family:Arial,Helvetica,sans-serif;font-size:x-small;}

recipes/rue89.recipe

@@ -6,6 +6,7 @@ Rue89
 
 __author__ = '2010-2012, Louis Gesbert <meta at antislash dot info>'
 
+import re
 from calibre.web.feeds.news import BasicNewsRecipe
 
 class Rue89(BasicNewsRecipe):
@@ -15,23 +16,24 @@ class Rue89(BasicNewsRecipe):
     title = u'Rue89'
     language = 'fr'
     oldest_article = 7
-    max_articles_per_feed = 12
+    max_articles_per_feed = 50
 
     use_embedded_content = False
 
     # From http://www.rue89.com/les-flux-rss-de-rue89
     feeds = [
         (u'La Une',    u'http://www.rue89.com/feed'),
-        (u'Rue69',     u'http://www.rue89.com/rue69/feed'),
+        # Other feeds disabled, 'La Une' seems to include them all
-        (u'Eco',       u'http://www.rue89.com/rue89-eco/feed'),
+        # (u'Rue69',     u'http://www.rue89.com/rue69/feed'),
-        (u'Planète',   u'http://www.rue89.com/rue89-planete/feed'),
+        # (u'Eco',       u'http://www.rue89.com/rue89-eco/feed'),
-        (u'Sport',     u'http://www.rue89.com/rue89-sport/feed'),
+        # (u'Planète',   u'http://www.rue89.com/rue89-planete/feed'),
-        (u'Culture',   u'http://www.rue89.com/culture/feed'),
+        # (u'Sport',     u'http://www.rue89.com/rue89-sport/feed'),
-        (u'Hi-tech',   u'http://www.rue89.com/hi-tech/feed'),
+        # (u'Culture',   u'http://www.rue89.com/culture/feed'),
-        (u'Media',     u'http://www.rue89.com/medias/feed'),
+        # (u'Hi-tech',   u'http://www.rue89.com/hi-tech/feed'),
-        (u'Monde',     u'http://www.rue89.com/monde/feed'),
+        # (u'Media',     u'http://www.rue89.com/medias/feed'),
-        (u'Politique', u'http://www.rue89.com/politique/feed'),
+        # (u'Monde',     u'http://www.rue89.com/monde/feed'),
-        (u'Societe',   u'http://www.rue89.com/societe/feed'),
+        # (u'Politique', u'http://www.rue89.com/politique/feed'),
+        # (u'Societe',   u'http://www.rue89.com/societe/feed'),
     ]
 
     # Follow redirection from feedsportal.com
@@ -41,19 +43,36 @@ class Rue89(BasicNewsRecipe):
     def print_version(self, url):
         return url + '?imprimer=1'
 
-    no_stylesheets = True
-
     conversion_options = { 'smarten_punctuation' : True }
 
     keep_only_tags = [
-        dict(name='div', attrs={'id':'article'}),
+        dict(name='div', attrs={'id':'content'}),
     ]
 
     remove_tags_after = [
         dict(name='div', attrs={'id':'plus_loin'}),
+        dict(name='div', attrs={'class':'stats'}),
     ]
 
     remove_tags = [
         dict(name='div', attrs={'id':'article_tools'}),
         dict(name='div', attrs={'id':'plus_loin'}),
+        dict(name='div', attrs={'class':'stats'}),
+        dict(name='div', attrs={'class':'tools'}),
     ]
+
+    extra_css = "#content { padding: 0 0; }"
+
+    # Without this, parsing of video articles returns strange results
+    preprocess_regexps = [
+        (re.compile(r'<script.*?</script>', re.IGNORECASE|re.DOTALL), ''),
+    ]
+
+    def preprocess_html(self, soup):
+        # Remove whole article if it's a "zapnet" (video)
+        if soup.find('h1', {'class':'zapnet_title'}):
+            return None
+        # Reduce h2 titles to h3
+        for title in soup.findAll('h2'):
+            title.name = 'h3'
+        return soup

setup/installer/__init__.py

@@ -48,7 +48,7 @@ class Push(Command):
         threads = []
         for host in (
             r'Owner@winxp:/cygdrive/c/Documents\ and\ Settings/Owner/calibre',
-            'kovid@leopard_test:calibre',
+            'kovid@ox:calibre',
             r'kovid@win7:/cygdrive/c/Users/kovid/calibre',
             ):
             rcmd = BASE_RSYNC + EXCLUDES + ['.', host]

setup/iso_639/sr.po

@@ -8,14 +8,14 @@ msgstr ""
 "Report-Msgid-Bugs-To: Debian iso-codes team <pkg-isocodes-"
 "devel@lists.alioth.debian.org>\n"
 "POT-Creation-Date: 2011-11-25 14:01+0000\n"
-"PO-Revision-Date: 2012-03-21 15:46+0000\n"
+"PO-Revision-Date: 2012-03-25 12:19+0000\n"
-"Last-Translator: Иван Старчевић <ivanstar61@gmail.com>\n"
+"Last-Translator: Radan Putnik <srastral@gmail.com>\n"
 "Language-Team: Serbian <gnu@prevod.org>\n"
 "MIME-Version: 1.0\n"
 "Content-Type: text/plain; charset=UTF-8\n"
 "Content-Transfer-Encoding: 8bit\n"
-"X-Launchpad-Export-Date: 2012-03-22 04:56+0000\n"
+"X-Launchpad-Export-Date: 2012-03-26 04:37+0000\n"
-"X-Generator: Launchpad (build 14981)\n"
+"X-Generator: Launchpad (build 15008)\n"
 "Language: sr\n"
 
 #. name for aaa
@@ -888,7 +888,7 @@ msgstr "Ака-Кеде"
 
 #. name for aky
 msgid "Aka-Kol"
-msgstr ""
+msgstr "ака-кол"
 
 #. name for akz
 msgid "Alabama"
@@ -968,11 +968,11 @@ msgstr "Алтајски;Јужни"
 
 #. name for alu
 msgid "'Are'are"
-msgstr ""
+msgstr "ареаре"
 
 #. name for alw
 msgid "Alaba-K’abeena"
-msgstr ""
+msgstr "алаба-кабеена"
 
 #. name for alx
 msgid "Amol"
@@ -1004,7 +1004,7 @@ msgstr ""
 
 #. name for amf
 msgid "Hamer-Banna"
-msgstr ""
+msgstr "хаммер-банна"
 
 #. name for amg
 msgid "Amarag"
@@ -1104,7 +1104,7 @@ msgstr "Ансус"
 
 #. name for ane
 msgid "Xârâcùù"
-msgstr ""
+msgstr "ксаракуу"
 
 #. name for anf
 msgid "Animere"
@@ -1156,7 +1156,7 @@ msgstr "Јарава(Индија)"
 
 #. name for anr
 msgid "Andh"
-msgstr ""
+msgstr "андх"
 
 #. name for ans
 msgid "Anserma"
@@ -1256,7 +1256,7 @@ msgstr "Таикат"
 
 #. name for aot
 msgid "A'tong"
-msgstr ""
+msgstr "атонг"
 
 #. name for aox
 msgid "Atorada"
@@ -1284,7 +1284,7 @@ msgstr ""
 
 #. name for apf
 msgid "Agta; Pahanan"
-msgstr ""
+msgstr "агта (паханан)"
 
 #. name for apg
 msgid "Ampanang"
@@ -1392,7 +1392,7 @@ msgstr "Атакапа"
 
 #. name for aqr
 msgid "Arhâ"
-msgstr ""
+msgstr "арга"
 
 #. name for aqz
 msgid "Akuntsu"
@@ -1424,7 +1424,7 @@ msgstr "арагонски"
 
 #. name for arh
 msgid "Arhuaco"
-msgstr ""
+msgstr "архуако"
 
 #. name for ari
 msgid "Arikara"
@@ -1504,7 +1504,7 @@ msgstr ""
 
 #. name for asd
 msgid "Asas"
-msgstr ""
+msgstr "асас"
 
 #. name for ase
 msgid "American Sign Language"
@@ -1532,7 +1532,7 @@ msgstr "Нсари"
 
 #. name for ask
 msgid "Ashkun"
-msgstr ""
+msgstr "ашкун"
 
 #. name for asl
 msgid "Asilulu"
@@ -1604,11 +1604,11 @@ msgstr "Заива"
 
 #. name for atc
 msgid "Atsahuaca"
-msgstr ""
+msgstr "атсахуака"
 
 #. name for atd
 msgid "Manobo; Ata"
-msgstr ""
+msgstr "манобо (Ата)"
 
 #. name for ate
 msgid "Atemble"
@@ -1648,7 +1648,7 @@ msgstr "Атон"
 
 #. name for atp
 msgid "Atta; Pudtol"
-msgstr ""
+msgstr "атта (Пудтол)"
 
 #. name for atq
 msgid "Aralle-Tabulahan"
@@ -1660,7 +1660,7 @@ msgstr ""
 
 #. name for ats
 msgid "Gros Ventre"
-msgstr ""
+msgstr "грос-вентре"
 
 #. name for att
 msgid "Atta; Pamplona"
@@ -1692,7 +1692,7 @@ msgstr "Арта"
 
 #. name for aua
 msgid "Asumboa"
-msgstr ""
+msgstr "асумбоа"
 
 #. name for aub
 msgid "Alugu"
@@ -1712,7 +1712,7 @@ msgstr ""
 
 #. name for aug
 msgid "Aguna"
-msgstr ""
+msgstr "агуна"
 
 #. name for auh
 msgid "Aushi"
@@ -1752,7 +1752,7 @@ msgstr ""
 
 #. name for auq
 msgid "Anus"
-msgstr ""
+msgstr "анус"
 
 #. name for aur
 msgid "Aruek"
@@ -1852,7 +1852,7 @@ msgstr "Авети"
 
 #. name for awh
 msgid "Awbono"
-msgstr ""
+msgstr "авбоно"
 
 #. name for awi
 msgid "Aekyom"
@@ -1860,7 +1860,7 @@ msgstr ""
 
 #. name for awk
 msgid "Awabakal"
-msgstr ""
+msgstr "авабакал"
 
 #. name for awm
 msgid "Arawum"
@@ -1884,7 +1884,7 @@ msgstr ""
 
 #. name for awt
 msgid "Araweté"
-msgstr ""
+msgstr "аравете"
 
 #. name for awu
 msgid "Awyu; Central"
@@ -1912,7 +1912,7 @@ msgstr "Абипон"
 
 #. name for axg
 msgid "Arára; Mato Grosso"
-msgstr ""
+msgstr "арара (Мату-Гросу)"
 
 #. name for axk
 msgid "Yaka (Central African Republic)"
@@ -1924,7 +1924,7 @@ msgstr ""
 
 #. name for axx
 msgid "Xaragure"
-msgstr ""
+msgstr "ксарагуре"
 
 #. name for aya
 msgid "Awar"

src/calibre/constants.py

@@ -4,7 +4,7 @@ __license__   = 'GPL v3'
 __copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net'
 __docformat__ = 'restructuredtext en'
 __appname__   = u'calibre'
-numeric_version = (0, 8, 44)
+numeric_version = (0, 8, 45)
 __version__   = u'.'.join(map(unicode, numeric_version))
 __author__    = u"Kovid Goyal <kovid@kovidgoyal.net>"
 

src/calibre/customize/builtins.py

@@ -1460,7 +1460,7 @@ class StoreNextoStore(StoreBase):
     actual_plugin = 'calibre.gui2.store.stores.nexto_plugin:NextoStore'
 
     headquarters = 'PL'
-    formats = ['EPUB', 'PDF']
+    formats = ['EPUB', 'MOBI', 'PDF']
     affiliate = True
 
 class StoreOpenBooksStore(StoreBase):

src/calibre/devices/android/driver.py

@@ -187,7 +187,7 @@ class ANDROID(USBMS):
             'UMS', '.K080', 'P990', 'LTE', 'MB853', 'GT-S5660_CARD', 'A107',
             'GT-I9003_CARD', 'XT912', 'FILE-CD_GADGET', 'RK29_SDK', 'MB855',
             'XT910', 'BOOK_A10', 'USB_2.0_DRIVER', 'I9100T', 'P999DW',
-            'KTABLET_PC', 'INGENIC']
+            'KTABLET_PC', 'INGENIC', 'GT-I9001_CARD']
     WINDOWS_CARD_A_MEM = ['ANDROID_PHONE', 'GT-I9000_CARD', 'SGH-I897',
             'FILE-STOR_GADGET', 'SGH-T959', 'SAMSUNG_ANDROID', 'GT-P1000_CARD',
             'A70S', 'A101IT', '7', 'INCREDIBLE', 'A7EB', 'SGH-T849_CARD',
@@ -195,7 +195,7 @@ class ANDROID(USBMS):
             'ANDROID_MID', 'P990_SD_CARD', '.K080', 'LTE_CARD', 'MB853',
             'A1-07___C0541A4F', 'XT912', 'MB855', 'XT910', 'BOOK_A10_CARD',
             'USB_2.0_DRIVER', 'I9100T', 'P999DW_SD_CARD', 'KTABLET_PC',
-            'FILE-CD_GADGET']
+            'FILE-CD_GADGET', 'GT-I9001_CARD']
 
     OSX_MAIN_MEM = 'Android Device Main Memory'
 

src/calibre/ebooks/conversion/utils.py

@@ -527,11 +527,17 @@ class HeuristicProcessor(object):
         if re.findall('(<|>)', replacement_break):
             if re.match('^<hr', replacement_break):
                 if replacement_break.find('width') != -1:
-                    width = int(re.sub('.*?width(:|=)(?P<wnum>\d+).*', '\g<wnum>', replacement_break))
+                    try:
-                    replacement_break = re.sub('(?i)(width=\d+\%?|width:\s*\d+(\%|px|pt|em)?;?)', '', replacement_break)
+                        width = int(re.sub('.*?width(:|=)(?P<wnum>\d+).*', '\g<wnum>', replacement_break))
-                    divpercent = (100 - width) / 2
+                    except:
-                    hr_open = re.sub('45', str(divpercent), hr_open)
+                        scene_break = hr_open+'<hr style="height: 3px; background:#505050" /></div>'
-                    scene_break = hr_open+replacement_break+'</div>'
+                        self.log.warn('Invalid replacement scene break'
+                                ' expression, using default')
+                    else:
+                        replacement_break = re.sub('(?i)(width=\d+\%?|width:\s*\d+(\%|px|pt|em)?;?)', '', replacement_break)
+                        divpercent = (100 - width) / 2
+                        hr_open = re.sub('45', str(divpercent), hr_open)
+                        scene_break = hr_open+replacement_break+'</div>'
                 else:
                     scene_break = hr_open+'<hr style="height: 3px; background:#505050" /></div>'
             elif re.match('^<img', replacement_break):

src/calibre/ebooks/metadata/book/json_codec.py

@@ -108,6 +108,8 @@ def decode_is_multiple(fm):
             else:
                 im = {'cache_to_list': '|', 'ui_to_list': ',',
                       'list_to_ui': ', '}
+        elif im is None:
+            im = {}
         fm['is_multiple'] = im
 
 class JsonCodec(object):

src/calibre/ebooks/mobi/debug/headers.py

@@ -205,7 +205,10 @@ class EXTHHeader(object):
 
     @property
     def kf8_header_index(self):
-        return self.get(121, None)
+        ans = self.get(121, None)
+        if ans == NULL_INDEX:
+            ans = None
+        return ans
 
     def __str__(self):
         ans = ['*'*20 + ' EXTH Header '+ '*'*20]
@@ -328,7 +331,7 @@ class MOBIHeader(object): # {{{
             (self.sect_idx, self.skel_idx, self.datp_idx, self.oth_idx
                     ) = struct.unpack_from(b'>4L', self.raw, 248)
             self.unknown9 = self.raw[264:self.length]
-            if self.meta_orth_indx != self.sect_idx:
+            if self.meta_orth_indx not in {NULL_INDEX, self.sect_idx}:
                 raise ValueError('KF8 header has different Meta orth and '
                         'section indices')
 
@@ -467,9 +470,15 @@ class MOBIFile(object):
         if mh.file_version >= 8:
             self.kf8_type = 'standalone'
         elif mh.has_exth and mh.exth.kf8_header_index is not None:
-            self.kf8_type = 'joint'
             kf8i = mh.exth.kf8_header_index
-            mh8 = MOBIHeader(self.records[kf8i], kf8i)
+            try:
+                rec = self.records[kf8i-1]
+            except IndexError:
+                pass
+            else:
+                if rec.raw == b'BOUNDARY':
+                    self.kf8_type = 'joint'
+                    mh8 = MOBIHeader(self.records[kf8i], kf8i)
         self.mobi8_header = mh8
 
         if 'huff' in self.mobi_header.compression.lower():

src/calibre/ebooks/mobi/debug/mobi8.py

@@ -7,9 +7,10 @@ __license__   = 'GPL v3'
 __copyright__ = '2012, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-import sys, os
+import sys, os, imghdr
 
 from calibre.ebooks.mobi.debug.headers import TextRecord
+from calibre.ebooks.mobi.utils import read_font_record
 
 class MOBIFile(object):
 
@@ -30,6 +31,7 @@ class MOBIFile(object):
                 first_text_record+offset+h8.number_of_text_records])]
 
         self.raw_text = b''.join(r.raw for r in self.text_records)
+        self.extract_resources()
 
     def print_header(self, f=sys.stdout):
         print (str(self.mf.palmdb).encode('utf-8'), file=f)
@@ -41,6 +43,42 @@ class MOBIFile(object):
         print (file=f)
         print (str(self.mf.mobi8_header).encode('utf-8'), file=f)
 
+    def extract_resources(self):
+        self.resource_map = []
+        known_types = {b'FLIS', b'FCIS', b'SRCS',
+                    b'\xe9\x8e\r\n', b'RESC', b'BOUN', b'FDST', b'DATP',
+                    b'AUDI', b'VIDE'}
+
+        for i, rec in enumerate(self.resource_records):
+            sig = rec.raw[:4]
+            payload = rec.raw
+            ext = 'dat'
+            prefix = 'binary'
+            suffix = ''
+            if sig in {b'HUFF', b'CDIC', b'INDX'}: continue
+            # TODO: Ignore CNCX records as well
+            if sig == b'FONT':
+                font = read_font_record(rec.raw)
+                if font['err']:
+                    raise ValueError('Failed to read font record: %s Headers: %s'%(
+                        font['err'], font['headers']))
+                payload = (font['font_data'] if font['font_data'] else
+                        font['raw_data'])
+                prefix, ext = 'fonts', font['ext']
+            elif sig not in known_types:
+                q = imghdr.what(None, rec.raw)
+                if q:
+                    prefix, ext = 'images', q
+
+            if prefix == 'binary':
+                if sig == b'\xe9\x8e\r\n':
+                    suffix = '-EOF'
+                elif sig in known_types:
+                    suffix = '-' + sig.decode('ascii')
+
+            self.resource_map.append(('%s/%06d%s.%s'%(prefix, i, suffix, ext),
+                payload))
+
 
 def inspect_mobi(mobi_file, ddir):
     f = MOBIFile(mobi_file)
@@ -51,12 +89,14 @@ def inspect_mobi(mobi_file, ddir):
     with open(alltext, 'wb') as of:
         of.write(f.raw_text)
 
-    for tdir, attr in [('text_records', 'text_records'), ('images',
+    for x in ('text_records', 'images', 'fonts', 'binary'):
-        'image_records'), ('binary', 'binary_records'), ('font',
+        os.mkdir(os.path.join(ddir, x))
-            'font_records')]:
+
-        tdir = os.path.join(ddir, tdir)
+    for rec in f.text_records:
-        os.mkdir(tdir)
+        rec.dump(os.path.join(ddir, 'text_records'))
-        for rec in getattr(f, attr, []):
+
-            rec.dump(tdir)
+    for href, payload in f.resource_map:
+        with open(os.path.join(ddir, href), 'wb') as f:
+            f.write(payload)
 
 

src/calibre/ebooks/mobi/reader/headers.py

@@ -11,7 +11,7 @@ import struct, re, os
 from calibre import replace_entities
 from calibre.utils.date import parse_date
 from calibre.ebooks.mobi import MobiError
-from calibre.ebooks.metadata import MetaInformation
+from calibre.ebooks.metadata import MetaInformation, check_isbn
 from calibre.ebooks.mobi.langcodes import main_language, sub_language, mobi2iana
 
 NULL_INDEX = 0xffffffff
@@ -75,10 +75,14 @@ class EXTHHeader(object): # {{{
                 self.mi.author_sort = au.strip()
         elif idx == 101:
             self.mi.publisher = content.decode(codec, 'ignore').strip()
+            if self.mi.publisher in {'Unknown', _('Unknown')}:
+                self.mi.publisher = None
         elif idx == 103:
             self.mi.comments  = content.decode(codec, 'ignore')
         elif idx == 104:
-            self.mi.isbn      = content.decode(codec, 'ignore').strip().replace('-', '')
+            raw = check_isbn(content.decode(codec, 'ignore').strip().replace('-', ''))
+            if raw:
+                self.mi.isbn = raw
         elif idx == 105:
             if not self.mi.tags:
                 self.mi.tags = []
@@ -92,12 +96,24 @@ class EXTHHeader(object): # {{{
                 pass
         elif idx == 108:
             self.mi.book_producer = content.decode(codec, 'ignore').strip()
+        elif idx == 112: # dc:source set in some EBSP amazon samples
+            try:
+                content = content.decode(codec).strip()
+                isig = 'urn:isbn:'
+                if content.lower().startswith(isig):
+                    raw = check_isbn(content[len(isig):])
+                    if raw and not self.mi.isbn:
+                        self.mi.isbn = raw
+            except:
+                pass
         elif idx == 113:
             pass # ASIN or UUID
         elif idx == 116:
             self.start_offset, = struct.unpack(b'>L', content)
         elif idx == 121:
             self.kf8_header, = struct.unpack(b'>L', content)
+            if self.kf8_header == NULL_INDEX:
+                self.kf8_header = None
         #else:
         #    print 'unhandled metadata record', idx, repr(content)
 # }}}

src/calibre/ebooks/mobi/reader/index.py

@@ -39,10 +39,41 @@ def parse_indx_header(data):
     words = (
             'len', 'nul1', 'type', 'gen', 'start', 'count', 'code',
             'lng', 'total', 'ordt', 'ligt', 'nligt', 'ncncx'
-    )
+    ) + tuple('unknown%d'%i for i in xrange(27)) + ('ocnt', 'oentries',
+            'ordt1', 'ordt2', 'tagx')
     num = len(words)
     values = struct.unpack(bytes('>%dL' % num), data[4:4*(num+1)])
-    return dict(zip(words, values))
+    ans = dict(zip(words, values))
+    ordt1, ordt2 = ans['ordt1'], ans['ordt2']
+    ans['ordt1_raw'], ans['ordt2_raw'] = [], []
+    ans['ordt_map'] = ''
+
+    if ordt1 > 0 and data[ordt1:ordt1+4] == b'ORDT':
+        # I dont know what this is, but using it seems to be unnecessary, so
+        # just leave it as the raw bytestring
+        ans['ordt1_raw'] = data[ordt1+4:ordt1+4+ans['oentries']]
+    if ordt2 > 0 and data[ordt2:ordt2+4] == b'ORDT':
+        ans['ordt2_raw'] = raw = bytearray(data[ordt2+4:ordt2+4+2*ans['oentries']])
+        if ans['code'] == 65002:
+            # This appears to be EBCDIC-UTF (65002) encoded. I can't be
+            # bothered to write a decoder for this (see
+            # http://www.unicode.org/reports/tr16/) Just how stupid is Amazon?
+            # Instead, we use a weird hack that seems to do the trick for all
+            # the books with this type of ORDT record that I have come across.
+            # Some EBSP book samples in KF8 format from Amazon have this type
+            # of encoding.
+            # Basically we try to interpret every second byte as a printable
+            # ascii character. If we cannot, we map to the ? char.
+
+            parsed = bytearray(ans['oentries'])
+            for i in xrange(0, 2*ans['oentries'], 2):
+                parsed[i//2] = raw[i+1] if 0x20 < raw[i+1] < 0x7f else ord(b'?')
+            ans['ordt_map'] = bytes(parsed).decode('ascii')
+        else:
+            ans['ordt_map'] = '?'*ans['oentries']
+
+    return ans
+
 
 class CNCX(object): # {{{
 
@@ -163,7 +194,7 @@ def get_tag_map(control_byte_count, tagx, data, strict=False):
     return ans
 
 def parse_index_record(table, data, control_byte_count, tags, codec,
-        strict=False):
+        ordt_map, strict=False):
     header = parse_indx_header(data)
     idxt_pos = header['start']
     if data[idxt_pos:idxt_pos+4] != b'IDXT':
@@ -184,12 +215,11 @@ def parse_index_record(table, data, control_byte_count, tags, codec,
     for j in xrange(entry_count):
         start, end = idx_positions[j:j+2]
         rec = data[start:end]
-        ident, consumed = decode_string(rec, codec=codec)
+        ident, consumed = decode_string(rec, codec=codec, ordt_map=ordt_map)
         rec = rec[consumed:]
         tag_map = get_tag_map(control_byte_count, tags, rec, strict=strict)
         table[ident] = tag_map
 
-
 def read_index(sections, idx, codec):
     table, cncx = OrderedDict(), CNCX([], codec)
 
@@ -203,12 +233,13 @@ def read_index(sections, idx, codec):
         cncx_records = [x[0] for x in sections[off:off+indx_header['ncncx']]]
         cncx = CNCX(cncx_records, codec)
 
-    tag_section_start = indx_header['len']
+    tag_section_start = indx_header['tagx']
     control_byte_count, tags = parse_tagx_section(data[tag_section_start:])
 
     for i in xrange(idx + 1, idx + 1 + indx_count):
         # Index record
         data = sections[i][0]
-        parse_index_record(table, data, control_byte_count, tags, codec)
+        parse_index_record(table, data, control_byte_count, tags, codec,
+                indx_header['ordt_map'])
     return table, cncx
 

src/calibre/ebooks/mobi/reader/mobi8.py

@@ -285,7 +285,11 @@ class Mobi8Reader(object):
     def create_guide(self):
         guide = Guide()
         for ref_type, ref_title, fileno in self.guide:
-            elem = self.elems[fileno]
+            try:
+                elem = self.elems[fileno]
+            except IndexError:
+                # Happens for thumbnailstandard in Amazon book samples
+                continue
             fi = self.get_file_info(elem.insert_pos)
             idtext = self.get_id_tag(elem.insert_pos).decode(self.header.codec)
             linktgt = fi.filename

src/calibre/ebooks/mobi/utils.py

@@ -15,10 +15,12 @@ from calibre.ebooks import normalize
 
 IMAGE_MAX_SIZE = 10 * 1024 * 1024
 
-def decode_string(raw, codec='utf-8'):
+def decode_string(raw, codec='utf-8', ordt_map=''):
     length, = struct.unpack(b'>B', raw[0])
     raw = raw[1:1+length]
     consumed = length+1
+    if ordt_map:
+        return ''.join(ordt_map[ord(x)] for x in raw), consumed
     return raw.decode(codec), consumed
 
 def decode_hex_number(raw, codec='utf-8'):

src/calibre/ebooks/mobi/writer2/serializer.py

@@ -161,8 +161,8 @@ class Serializer(object):
                 self.serialize_text(ref.title, quot=True)
                 buf.write(b'" ')
                 if (ref.title.lower() == 'start' or
-                    (ref.type and ref.type.lower() in ('start',
+                    (ref.type and ref.type.lower() in {'start',
-                        'other.start'))):
+                        'other.start', 'text'})):
                     self._start_href = ref.href
             self.serialize_href(ref.href)
             # Space required or won't work, I kid you not

src/calibre/ebooks/oeb/display/__init__.py

@@ -0,0 +1,11 @@
+#!/usr/bin/env python
+# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
+from __future__ import (unicode_literals, division, absolute_import,
+                        print_function)
+
+__license__   = 'GPL v3'
+__copyright__ = '2012, Kovid Goyal <kovid@kovidgoyal.net>'
+__docformat__ = 'restructuredtext en'
+
+
+

src/calibre/ebooks/oeb/display/webview.py

@@ -0,0 +1,59 @@
+#!/usr/bin/env python
+# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
+from __future__ import (unicode_literals, division, absolute_import,
+                        print_function)
+
+__license__   = 'GPL v3'
+__copyright__ = '2012, Kovid Goyal <kovid@kovidgoyal.net>'
+__docformat__ = 'restructuredtext en'
+
+import re
+
+from calibre import guess_type
+
+class EntityDeclarationProcessor(object): # {{{
+
+    def __init__(self, html):
+        self.declared_entities = {}
+        for match in re.finditer(r'<!\s*ENTITY\s+([^>]+)>', html):
+            tokens = match.group(1).split()
+            if len(tokens) > 1:
+                self.declared_entities[tokens[0].strip()] = tokens[1].strip().replace('"', '')
+        self.processed_html = html
+        for key, val in self.declared_entities.iteritems():
+            self.processed_html = self.processed_html.replace('&%s;'%key, val)
+# }}}
+
+def self_closing_sub(match):
+    tag = match.group(1)
+    if tag.lower().strip() == 'br':
+        return match.group()
+    return '<%s %s></%s>'%(match.group(1), match.group(2), match.group(1))
+
+def load_html(path, view, codec='utf-8', mime_type=None,
+        pre_load_callback=lambda x:None):
+    from PyQt4.Qt import QUrl, QByteArray
+    if mime_type is None:
+        mime_type = guess_type(path)[0]
+    with open(path, 'rb') as f:
+        html = f.read().decode(codec, 'replace')
+
+    html = EntityDeclarationProcessor(html).processed_html
+    has_svg = re.search(r'<[:a-zA-Z]*svg', html) is not None
+    if 'xhtml' in mime_type:
+        self_closing_pat = re.compile(r'<([a-z1-6]+)\s+([^>]+)/>',
+                re.IGNORECASE)
+        html = self_closing_pat.sub(self_closing_sub, html)
+
+    html = re.sub(ur'<\s*title\s*/\s*>', u'', html, flags=re.IGNORECASE)
+    loading_url = QUrl.fromLocalFile(path)
+    pre_load_callback(loading_url)
+
+    if has_svg:
+        view.setContent(QByteArray(html.encode(codec)), mime_type,
+                loading_url)
+    else:
+        view.setHtml(html, loading_url)
+
+
+

src/calibre/ebooks/oeb/iterator.py

@@ -26,6 +26,8 @@ from calibre.constants import filesystem_encoding
 TITLEPAGE = CoverManager.SVG_TEMPLATE.decode('utf-8').replace(\
         '__ar__', 'none').replace('__viewbox__', '0 0 600 800'
         ).replace('__width__', '600').replace('__height__', '800')
+BM_FIELD_SEP = u'*|!|?|*'
+BM_LEGACY_ESC = u'esc-text-%&*#%(){}ads19-end-esc'
 
 def character_count(html):
     '''
@@ -273,27 +275,62 @@ class EbookIterator(object):
 
     def parse_bookmarks(self, raw):
         for line in raw.splitlines():
+            bm = None
             if line.count('^') > 0:
                 tokens = line.rpartition('^')
                 title, ref = tokens[0], tokens[2]
-                self.bookmarks.append((title, ref))
+                try:
+                    spine, _, pos = ref.partition('#')
+                    spine = int(spine.strip())
+                except:
+                    continue
+                bm = {'type':'legacy', 'title':title, 'spine':spine, 'pos':pos}
+            elif BM_FIELD_SEP in line:
+                try:
+                    title, spine, pos = line.strip().split(BM_FIELD_SEP)
+                    spine = int(spine)
+                except:
+                    continue
+                # Unescape from serialization
+                pos = pos.replace(BM_LEGACY_ESC, u'^')
+                # Check for pos being a scroll fraction
+                try:
+                    pos = float(pos)
+                except:
+                    pass
+                bm = {'type':'cfi', 'title':title, 'pos':pos, 'spine':spine}
+
+            if bm:
+                self.bookmarks.append(bm)
 
     def serialize_bookmarks(self, bookmarks):
         dat = []
-        for title, bm in bookmarks:
+        for bm in bookmarks:
-            dat.append(u'%s^%s'%(title, bm))
+            if bm['type'] == 'legacy':
-        return (u'\n'.join(dat) +'\n').encode('utf-8')
+                rec = u'%s^%d#%s'%(bm['title'], bm['spine'], bm['pos'])
+            else:
+                pos = bm['pos']
+                if isinstance(pos, (int, float)):
+                    pos = unicode(pos)
+                else:
+                    pos = pos.replace(u'^', BM_LEGACY_ESC)
+                rec = BM_FIELD_SEP.join([bm['title'], unicode(bm['spine']), pos])
+            dat.append(rec)
+        return (u'\n'.join(dat) +u'\n')
 
     def read_bookmarks(self):
         self.bookmarks = []
         bmfile = os.path.join(self.base, 'META-INF', 'calibre_bookmarks.txt')
         raw = ''
         if os.path.exists(bmfile):
-            raw = open(bmfile, 'rb').read().decode('utf-8')
+            with open(bmfile, 'rb') as f:
+                raw = f.read()
         else:
             saved = self.config['bookmarks_'+self.pathtoebook]
             if saved:
                 raw = saved
+        if not isinstance(raw, unicode):
+            raw = raw.decode('utf-8')
         self.parse_bookmarks(raw)
 
     def save_bookmarks(self, bookmarks=None):
@@ -306,18 +343,15 @@ class EbookIterator(object):
                 zf = open(self.pathtoebook, 'r+b')
             except IOError:
                 return
-            safe_replace(zf, 'META-INF/calibre_bookmarks.txt', StringIO(dat),
+            safe_replace(zf, 'META-INF/calibre_bookmarks.txt',
+                    StringIO(dat.encode('utf-8')),
                     add_missing=True)
         else:
             self.config['bookmarks_'+self.pathtoebook] = dat
 
     def add_bookmark(self, bm):
-        dups = []
+        self.bookmarks = [x for x in self.bookmarks if x['title'] !=
-        for x in self.bookmarks:
+                bm['title']]
-            if x[0] == bm[0]:
-                dups.append(x)
-        for x in dups:
-            self.bookmarks.remove(x)
         self.bookmarks.append(bm)
         self.save_bookmarks()
 

src/calibre/ebooks/oeb/transforms/guide.py

@@ -8,10 +8,9 @@ __docformat__ = 'restructuredtext en'
 
 
 class Clean(object):
-    '''Clean up guide, leaving only a pointer to the cover'''
+    '''Clean up guide, leaving only known values '''
 
     def __call__(self, oeb, opts):
-        from calibre.ebooks.oeb.base import urldefrag
         self.oeb, self.log, self.opts = oeb, oeb.log, opts
 
         if 'cover' not in self.oeb.guide:
@@ -32,10 +31,15 @@ class Clean(object):
                 ref.type = 'cover'
                 self.oeb.guide.refs['cover'] = ref
 
+        if ('start' in self.oeb.guide and 'text' not in self.oeb.guide):
+            # Prefer text to start as per the OPF 2.0 spec
+            x = self.oeb.guide['start']
+            self.oeb.guide.add('text', x.title, x.href)
+            self.oeb.guide.remove('start')
+
         for x in list(self.oeb.guide):
-            href = urldefrag(self.oeb.guide[x].href)[0]
+            if x.lower() not in {'cover', 'titlepage', 'masthead', 'toc',
-            if x.lower() not in ('cover', 'titlepage', 'masthead', 'toc',
+                    'title-page', 'copyright-page', 'text'}:
-                    'title-page', 'copyright-page', 'start'):
                 item = self.oeb.guide[x]
                 if item.title and item.title.lower() == 'start':
                     continue

src/calibre/ebooks/pdf/writer.py

@@ -18,10 +18,11 @@ from calibre.ebooks.pdf.pageoptions import unit, paper_size, \
 from calibre.ebooks.metadata import authors_to_string
 from calibre.ptempfile import PersistentTemporaryFile
 from calibre import __appname__, __version__, fit_image
+from calibre.ebooks.oeb.display.webview import load_html
 
 from PyQt4 import QtCore
-from PyQt4.Qt import QUrl, QEventLoop, QObject, \
+from PyQt4.Qt import (QEventLoop, QObject,
-    QPrinter, QMetaObject, QSizeF, Qt, QPainter, QPixmap
+    QPrinter, QMetaObject, QSizeF, Qt, QPainter, QPixmap)
 from PyQt4.QtWebKit import QWebView
 
 from pyPdf import PdfFileWriter, PdfFileReader
@@ -70,7 +71,7 @@ def get_pdf_printer(opts, for_comic=False):
                 opts.margin_right, opts.margin_bottom, QPrinter.Point)
     printer.setOrientation(orientation(opts.orientation))
     printer.setOutputFormat(QPrinter.PdfFormat)
-    printer.setFullPage(True)
+    printer.setFullPage(for_comic)
     return printer
 
 def get_printer_page_size(opts, for_comic=False):
@@ -156,8 +157,7 @@ class PDFWriter(QObject): # {{{
         self.combine_queue.append(os.path.join(self.tmp_path, '%i.pdf' % (len(self.combine_queue) + 1)))
 
         self.logger.debug('Processing %s...' % item)
-
+        load_html(item, self.view)
-        self.view.load(QUrl.fromLocalFile(item))
 
     def _render_html(self, ok):
         if ok:
@@ -168,9 +168,14 @@ class PDFWriter(QObject): # {{{
             # We have to set the engine to Native on OS X after the call to set
             # filename. Setting a filename with .pdf as the extension causes
             # Qt to set the format to use Qt's PDF engine even if native was
-            # previously set on the printer.
+            # previously set on the printer. Qt's PDF engine produces image
+            # based PDFs on OS X, so we cannot use it.
             if isosx:
                 printer.setOutputFormat(QPrinter.NativeFormat)
+            self.view.page().mainFrame().evaluateJavaScript('''
+                document.body.style.backgroundColor = "white";
+
+                ''')
             self.view.print_(printer)
             printer.abort()
         else:

src/calibre/gui2/actions/copy_to_library.py

@@ -81,8 +81,8 @@ class Worker(Thread): # {{{
             if prefs['add_formats_to_existing']:
                 identical_book_list = newdb.find_identical_books(mi)
                 if identical_book_list: # books with same author and nearly same title exist in newdb
-                    self.auto_merged_ids[x] = _('%s by %s')%(mi.title,
+                    self.auto_merged_ids[x] = _('%(title)s by %(author)s')%\
-                            mi.format_field('authors')[1])
+                    dict(title=mi.title, author=mi.format_field('authors')[1])
                     automerged = True
                     seen_fmts = set()
                     for identical_book in identical_book_list:

src/calibre/gui2/comments_editor.py

@@ -9,10 +9,10 @@ import re, os
 
 from lxml import html
 
-from PyQt4.Qt import QApplication, QFontInfo, QSize, QWidget, QPlainTextEdit, \
+from PyQt4.Qt import (QApplication, QFontInfo, QSize, QWidget, QPlainTextEdit,
-    QToolBar, QVBoxLayout, QAction, QIcon, Qt, QTabWidget, QUrl, \
+    QToolBar, QVBoxLayout, QAction, QIcon, Qt, QTabWidget, QUrl,
-    QSyntaxHighlighter, QColor, QChar, QColorDialog, QMenu, QInputDialog, \
+    QSyntaxHighlighter, QColor, QChar, QColorDialog, QMenu, QInputDialog,
-    QHBoxLayout
+    QHBoxLayout, QKeySequence)
 from PyQt4.QtWebKit import QWebView, QWebPage
 
 from calibre.ebooks.chardet import xml_to_unicode
@@ -32,6 +32,7 @@ class PageAction(QAction): # {{{
                 type=Qt.QueuedConnection)
         self.page_action.changed.connect(self.update_state,
                 type=Qt.QueuedConnection)
+        self.update_state()
 
     @property
     def page_action(self):
@@ -66,6 +67,12 @@ class EditorWidget(QWebView): # {{{
 
         self.comments_pat = re.compile(r'<!--.*?-->', re.DOTALL)
 
+        extra_shortcuts = {
+                'ToggleBold': 'Bold',
+                'ToggleItalic': 'Italic',
+                'ToggleUnderline': 'Underline',
+        }
+
         for wac, name, icon, text, checkable in [
                 ('ToggleBold', 'bold', 'format-text-bold', _('Bold'), True),
                 ('ToggleItalic', 'italic', 'format-text-italic', _('Italic'),
@@ -106,6 +113,9 @@ class EditorWidget(QWebView): # {{{
             ]:
             ac = PageAction(wac, icon, text, checkable, self)
             setattr(self, 'action_'+name, ac)
+            ss = extra_shortcuts.get(wac, None)
+            if ss:
+                ac.setShortcut(QKeySequence(getattr(QKeySequence, ss)))
 
         self.action_color = QAction(QIcon(I('format-text-color')), _('Foreground color'),
                 self)

src/calibre/gui2/complete.py

@@ -6,8 +6,8 @@ __copyright__ = '2011, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
 
-from PyQt4.Qt import QLineEdit, QAbstractListModel, Qt, \
+from PyQt4.Qt import (QLineEdit, QAbstractListModel, Qt,
-        QApplication, QCompleter
+        QApplication, QCompleter, QMetaObject)
 
 from calibre.utils.icu import sort_key, lower
 from calibre.gui2 import NONE
@@ -182,14 +182,27 @@ class MultiCompleteComboBox(EnComboBox):
     def set_add_separator(self, what):
         self.lineEdit().set_add_separator(what)
 
-
+    def show_initial_value(self, what):
+        '''
+        Show an initial value. Handle the case of the initial value being blank
+        correctly (on Qt 4.8.0 having a blank value causes the first value from
+        the completer to be shown, when the event loop runs).
+        '''
+        what = unicode(what)
+        le = self.lineEdit()
+        if not what.strip():
+            QMetaObject.invokeMethod(self, 'clearEditText',
+                    Qt.QueuedConnection)
+        else:
+            self.setEditText(what)
+            le.selectAll()
 
 if __name__ == '__main__':
     from PyQt4.Qt import QDialog, QVBoxLayout
     app = QApplication([])
     d = QDialog()
     d.setLayout(QVBoxLayout())
-    le = MultiCompleteLineEdit(d)
+    le = MultiCompleteComboBox(d)
     d.layout().addWidget(le)
     le.all_items = ['one', 'otwo', 'othree', 'ooone', 'ootwo', 'oothree']
     d.exec_()

src/calibre/gui2/dialogs/choose_plugin_toolbars.py

@@ -9,8 +9,8 @@ __docformat__ = 'restructuredtext en'
 __license__   = 'GPL v3'
 
 
-from PyQt4.Qt import QDialog, QVBoxLayout, QLabel, QDialogButtonBox,  \
+from PyQt4.Qt import (QDialog, QVBoxLayout, QLabel, QDialogButtonBox,
-            QListWidget, QAbstractItemView
+            QListWidget, QAbstractItemView)
 from PyQt4 import QtGui
 
 class ChoosePluginToolbarsDialog(QDialog):
@@ -39,6 +39,9 @@ class ChoosePluginToolbarsDialog(QDialog):
         self._locations_list.setSizePolicy(sizePolicy)
         for key, text in locations:
             self._locations_list.addItem(text)
+            if key in {'toolbar', 'toolbar-device'}:
+                self._locations_list.item(self._locations_list.count()-1
+                        ).setSelected(True)
         self._layout.addWidget(self._locations_list)
 
         self._footer_label = QLabel(

src/calibre/gui2/dialogs/scheduler.py

@@ -11,9 +11,9 @@ from datetime import timedelta
 import calendar, textwrap
 from collections import OrderedDict
 
-from PyQt4.Qt import QDialog, Qt, QTime, QObject, QMenu, QHBoxLayout, \
+from PyQt4.Qt import (QDialog, Qt, QTime, QObject, QMenu, QHBoxLayout,
-        QAction, QIcon, QMutex, QTimer, pyqtSignal, QWidget, QGridLayout, \
+        QAction, QIcon, QMutex, QTimer, pyqtSignal, QWidget, QGridLayout,
-        QCheckBox, QTimeEdit, QLabel, QLineEdit, QDoubleSpinBox
+        QCheckBox, QTimeEdit, QLabel, QLineEdit, QDoubleSpinBox)
 
 from calibre.gui2.dialogs.scheduler_ui import Ui_Dialog
 from calibre.gui2 import config as gconf, error_dialog
@@ -317,6 +317,8 @@ class SchedulerDialog(QDialog, Ui_Dialog):
                     return False
             if un or pw:
                 self.recipe_model.set_account_info(urn, un, pw)
+            else:
+                self.recipe_model.clear_account_info(urn)
 
         if self.schedule.isChecked():
             schedule_type, schedule = \

src/calibre/gui2/library/delegates.py

@@ -128,8 +128,7 @@ class TextDelegate(QStyledItemDelegate): # {{{
             for item in sorted(complete_items, key=sort_key):
                 editor.addItem(item)
             ct = index.data(Qt.DisplayRole).toString()
-            editor.setEditText(ct)
+            editor.show_initial_value(ct)
-            editor.lineEdit().selectAll()
         else:
             editor = EnLineEdit(parent)
         return editor
@@ -170,8 +169,7 @@ class CompleteDelegate(QStyledItemDelegate): # {{{
             for item in sorted(all_items, key=sort_key):
                 editor.addItem(item)
             ct = index.data(Qt.DisplayRole).toString()
-            editor.setEditText(ct)
+            editor.show_initial_value(ct)
-            editor.lineEdit().selectAll()
         else:
             editor = EnLineEdit(parent)
         return editor
@@ -190,8 +188,7 @@ class LanguagesDelegate(QStyledItemDelegate): # {{{
         editor = LanguagesEdit(parent=parent)
         editor.init_langs(index.model().db)
         ct = index.data(Qt.DisplayRole).toString()
-        editor.setEditText(ct)
+        editor.show_initial_value(ct)
-        editor.lineEdit().selectAll()
         return editor
 
     def setModelData(self, editor, model, index):

src/calibre/gui2/library/views.py

@@ -262,9 +262,11 @@ class BooksView(QTableView): # {{{
         self.selected_ids = [idc(r) for r in selected_rows]
 
     def sorting_done(self, indexc):
+        pos = self.horizontalScrollBar().value()
         self.select_rows(self.selected_ids, using_ids=True, change_current=True,
             scroll=True)
         self.selected_ids = []
+        self.horizontalScrollBar().setValue(pos)
 
     def sort_by_named_field(self, field, order, reset=True):
         if field in self.column_map:

src/calibre/gui2/metadata/single_download.py

@@ -882,6 +882,11 @@ class FullFetch(QDialog): # {{{
         self.covers_widget.chosen.connect(self.ok_clicked)
         self.stack.addWidget(self.covers_widget)
 
+        # Workaround for Qt 4.8.0 bug that causes the frame of the window to go
+        # off the top of the screen if a max height is not set for the
+        # QWebView. Seems to only happen on windows, but keep it for all
+        # platforms just in case.
+        self.identify_widget.comments_view.setMaximumHeight(500)
         self.resize(850, 550)
 
         self.finished.connect(self.cleanup)

src/calibre/gui2/store/search_result.py

@@ -29,4 +29,4 @@ class SearchResult(object):
         self.plugin_author = ''
 
     def __eq__(self, other):
-        return self.title == other.title and self.author == other.author and self.store_name == other.store_name
+        return self.title == other.title and self.author == other.author and self.store_name == other.store_name and self.formats == other.formats

src/calibre/gui2/store/stores/ebookpoint_plugin.py

@@ -3,7 +3,7 @@
 from __future__ import (unicode_literals, division, absolute_import, print_function)
 
 __license__ = 'GPL 3'
-__copyright__ = '2011, Tomasz Długosz <tomek3d@gmail.com>'
+__copyright__ = '2011-2012, Tomasz Długosz <tomek3d@gmail.com>'
 __docformat__ = 'restructuredtext en'
 
 import re
@@ -64,9 +64,7 @@ class EbookpointStore(BasicStoreConfig, StorePlugin):
                 author = ''.join(data.xpath('.//p[@class="author"]/text()'))
                 price = ''.join(data.xpath('.//p[@class="price"]/ins/text()'))
 
-                with closing(br.open(id.strip(), timeout=timeout)) as nf:
+                formats = ', '.join(data.xpath('.//div[@class="ikony"]/span/text()'))
-                    idata = html.fromstring(nf.read())
-                    formats = ', '.join(idata.xpath('//dd[@class="radio-line"]/label/text()'))
 
                 counter -= 1
 
@@ -77,6 +75,6 @@ class EbookpointStore(BasicStoreConfig, StorePlugin):
                 s.price = re.sub(r'\.',',',price)
                 s.detail_item = id.strip()
                 s.drm = SearchResult.DRM_UNLOCKED
-                s.formats = formats.upper().strip()
+                s.formats = formats.upper()
 
                 yield s

src/calibre/gui2/store/stores/nexto_plugin.py

@@ -68,8 +68,8 @@ class NextoStore(BasicStoreConfig, StorePlugin):
                     title = ''.join(data.xpath('.//a[@class="title"]/text()'))
                     title = re.sub(r' - ebook$', '', title)
                     formats = ', '.join(data.xpath('.//ul[@class="formats_available"]/li//b/text()'))
-                    DrmFree = re.search(r'bez.DRM', formats)
+                    DrmFree = re.search(r'znak', formats)
-                    formats = re.sub(r'\(.+\)', '', formats)
+                    formats = re.sub(r'\ ?\(.+?\)', '', formats)
 
                     author = ''
                     with closing(br.open('http://www.nexto.pl/' + id.strip(), timeout=timeout/4)) as nf:

src/calibre/gui2/store/stores/woblink_plugin.py

@@ -6,6 +6,7 @@ __license__ = 'GPL 3'
 __copyright__ = '2011-2012, Tomasz Długosz <tomek3d@gmail.com>'
 __docformat__ = 'restructuredtext en'
 
+import copy
 import re
 import urllib
 from contextlib import closing
@@ -43,9 +44,9 @@ class WoblinkStore(BasicStoreConfig, StorePlugin):
         url = 'http://woblink.com/publication?query=' + urllib.quote_plus(query.encode('utf-8'))
         if max_results > 10:
             if max_results > 20:
-                url += '&limit=' + str(30)
+                url += '&limit=30'
             else:
-                url += '&limit=' + str(20)
+                url += '&limit=20'
 
         br = browser()
 
@@ -66,15 +67,6 @@ class WoblinkStore(BasicStoreConfig, StorePlugin):
                 price = ''.join(data.xpath('.//div[@class="prices"]/span[1]/span/text()'))
                 price = re.sub('\.', ',', price)
                 formats = [ form[8:-4].split('_')[0] for form in data.xpath('.//p[3]/img/@src')]
-                if 'epub' in formats:
-                    formats.remove('epub')
-                    formats.append('WOBLINK')
-                    if 'E Ink' in data.xpath('.//div[@class="prices"]/img/@title'):
-                        formats.insert(0, 'EPUB')
-                if 'pdf' in formats:
-                    formats[formats.index('pdf')] = 'PDF' 
-
-                counter -= 1
 
                 s = SearchResult()
                 s.cover_url = 'http://woblink.com' + cover_url
@@ -82,7 +74,28 @@ class WoblinkStore(BasicStoreConfig, StorePlugin):
                 s.author = author.strip()
                 s.price = price + ' zł'
                 s.detail_item = id.strip()
-                s.drm = SearchResult.DRM_UNKNOWN if 'MOBI' in formats else SearchResult.DRM_LOCKED
-                s.formats = ', '.join(formats)
                 
-                yield s
+                # MOBI should be send first,
+                if 'MOBI' in formats:
+                    t = copy.copy(s)
+                    t.title += ' MOBI'
+                    t.drm = SearchResult.DRM_UNLOCKED
+                    t.formats = 'MOBI'
+                    formats.remove('MOBI')
+                    
+                    counter -= 1
+                    yield t
+                    
+                # and the remaining formats (if any) next
+                if formats:
+                    if 'epub' in formats:
+                        formats.remove('epub')
+                        formats.append('WOBLINK')
+                        if 'E Ink' in data.xpath('.//div[@class="prices"]/img/@title'):
+                            formats.insert(0, 'EPUB')
+                    
+                    s.drm = SearchResult.DRM_LOCKED
+                    s.formats = ', '.join(formats).upper()
+                    
+                    counter -= 1
+                    yield s

src/calibre/gui2/update.py

@@ -151,7 +151,7 @@ class UpdateMixin(object):
             plt = u''
             if has_plugin_updates:
                 plt = _(' (%d plugin updates)')%plugin_updates
-            msg = (u'<span style="color:red; font-weight: bold">%s: '
+            msg = (u'<span style="color:green; font-weight: bold">%s: '
                     u'<a href="update:%s">%s%s</a></span>') % (
                         _('Update found'), version, calibre_version, plt)
         else:

src/calibre/gui2/viewer/bookmarkmanager.py

@@ -31,6 +31,7 @@ class BookmarkManager(QDialog, Ui_BookmarkManager):
             bookmarks = self.bookmarks[:]
         self._model = BookmarkTableModel(self, bookmarks)
         self.bookmarks_table.setModel(self._model)
+        self.bookmarks_table.resizeColumnsToContents()
 
     def delete_bookmark(self):
         indexes = self.bookmarks_table.selectionModel().selectedIndexes()
@@ -80,7 +81,7 @@ class BookmarkManager(QDialog, Ui_BookmarkManager):
             if not bad:
                 bookmarks = self._model.bookmarks[:]
                 for bm in imported:
-                    if bm not in bookmarks and bm[0] != 'calibre_current_page_bookmark':
+                    if bm not in bookmarks and bm['title'] != 'calibre_current_page_bookmark':
                         bookmarks.append(bm)
                 self.set_bookmarks(bookmarks)
 
@@ -105,13 +106,14 @@ class BookmarkTableModel(QAbstractTableModel):
 
     def data(self, index, role):
         if role in (Qt.DisplayRole, Qt.EditRole):
-            ans = self.bookmarks[index.row()][0]
+            ans = self.bookmarks[index.row()]['title']
             return NONE if ans is None else QVariant(ans)
         return NONE
 
     def setData(self, index, value, role):
         if role == Qt.EditRole:
-            self.bookmarks[index.row()] = (unicode(value.toString()).strip(), self.bookmarks[index.row()][1])
+            bm = self.bookmarks[index.row()]
+            bm['title'] = unicode(value.toString()).strip()
             self.emit(SIGNAL("dataChanged(QModelIndex, QModelIndex)"), index, index)
             return True
         return False

src/calibre/gui2/viewer/documentview.py

@@ -4,14 +4,14 @@ __copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net'
 __docformat__ = 'restructuredtext en'
 
 # Imports {{{
-import os, math, re, glob, sys, zipfile
+import os, math, glob, zipfile
 from base64 import b64encode
 from functools import partial
 
 from PyQt4.Qt import (QSize, QSizePolicy, QUrl, SIGNAL, Qt,
                      QPainter, QPalette, QBrush, QFontDatabase, QDialog,
                      QColor, QPoint, QImage, QRegion, QVariant, QIcon,
-                     QFont, pyqtSignature, QAction, QByteArray, QMenu,
+                     QFont, pyqtSignature, QAction, QMenu,
                      pyqtSignal, QSwipeGesture, QApplication)
 from PyQt4.QtWebKit import QWebPage, QWebView, QWebSettings
 
@@ -21,10 +21,11 @@ from calibre.gui2.viewer.config_ui import Ui_Dialog
 from calibre.gui2.viewer.flip import SlideFlip
 from calibre.gui2.shortcuts import Shortcuts, ShortcutConfig
 from calibre.constants import iswindows
-from calibre import prints, guess_type
+from calibre import prints
 from calibre.gui2.viewer.keys import SHORTCUTS
 from calibre.gui2.viewer.javascript import JavaScriptLoader
 from calibre.gui2.viewer.position import PagePosition
+from calibre.ebooks.oeb.display.webview import load_html
 
 # }}}
 
@@ -312,10 +313,14 @@ class Document(QWebPage): # {{{
         self.javascript('goto_reference("%s")'%ref)
 
     def goto_bookmark(self, bm):
-        bm = bm.strip()
+        if bm['type'] == 'legacy':
-        if bm.startswith('>'):
+            bm = bm['pos']
-            bm = bm[1:].strip()
+            bm = bm.strip()
-        self.javascript('scroll_to_bookmark("%s")'%bm)
+            if bm.startswith('>'):
+                bm = bm[1:].strip()
+            self.javascript('scroll_to_bookmark("%s")'%bm)
+        elif bm['type'] == 'cfi':
+            self.page_position.to_pos(bm['pos'])
 
     def javascript(self, string, typ=None):
         ans = self.mainFrame().evaluateJavaScript(string)
@@ -366,40 +371,9 @@ class Document(QWebPage): # {{{
     def elem_outer_xml(self, elem):
         return unicode(elem.toOuterXml())
 
-    def find_bookmark_element(self):
-        mf = self.mainFrame()
-        doc_pos = self.ypos
-        min_delta, min_elem = sys.maxint, None
-        for y in range(10, -500, -10):
-            for x in range(-50, 500, 10):
-                pos = QPoint(x, y)
-                result = mf.hitTestContent(pos)
-                if result.isNull(): continue
-                elem = result.enclosingBlockElement()
-                if elem.isNull(): continue
-                try:
-                    ypos = self.element_ypos(elem)
-                except:
-                    continue
-                delta = abs(ypos - doc_pos)
-                if delta < 25:
-                    return elem
-                if delta < min_delta:
-                    min_elem, min_delta = elem, delta
-        return min_elem
-
-
     def bookmark(self):
-        elem = self.find_bookmark_element()
+        pos = self.page_position.current_pos
-
+        return {'type':'cfi', 'pos':pos}
-        if elem is None or self.element_ypos(elem) < 100:
-            bm = 'body|%f'%(float(self.ypos)/(self.height*0.7))
-        else:
-            bm = unicode(elem.evaluateJavaScript(
-                'calculate_bookmark(%d, this)'%self.ypos).toString())
-            if not bm:
-                bm = 'body|%f'%(float(self.ypos)/(self.height*0.7))
-        return bm
 
     @property
     def at_bottom(self):
@@ -474,19 +448,6 @@ class Document(QWebPage): # {{{
 
 # }}}
 
-class EntityDeclarationProcessor(object): # {{{
-
-    def __init__(self, html):
-        self.declared_entities = {}
-        for match in re.finditer(r'<!\s*ENTITY\s+([^>]+)>', html):
-            tokens = match.group(1).split()
-            if len(tokens) > 1:
-                self.declared_entities[tokens[0].strip()] = tokens[1].strip().replace('"', '')
-        self.processed_html = html
-        for key, val in self.declared_entities.iteritems():
-            self.processed_html = self.processed_html.replace('&%s;'%key, val)
-# }}}
-
 class DocumentView(QWebView): # {{{
 
     magnification_changed = pyqtSignal(object)
@@ -497,8 +458,6 @@ class DocumentView(QWebView): # {{{
         self.is_auto_repeat_event = False
         self.debug_javascript = debug_javascript
         self.shortcuts =  Shortcuts(SHORTCUTS, 'shortcuts/viewer')
-        self.self_closing_pat = re.compile(r'<([a-z1-6]+)\s+([^>]+)/>',
-                re.IGNORECASE)
         self.setSizePolicy(QSizePolicy(QSizePolicy.Expanding, QSizePolicy.Expanding))
         self._size_hint = QSize(510, 680)
         self.initial_pos = 0.0
@@ -689,30 +648,16 @@ class DocumentView(QWebView): # {{{
     def path(self):
         return os.path.abspath(unicode(self.url().toLocalFile()))
 
-    def self_closing_sub(self, match):
-        tag = match.group(1)
-        if tag.lower().strip() == 'br':
-            return match.group()
-        return '<%s %s></%s>'%(match.group(1), match.group(2), match.group(1))
-
     def load_path(self, path, pos=0.0):
         self.initial_pos = pos
-        mt = getattr(path, 'mime_type', None)
-        if mt is None:
-            mt = guess_type(path)[0]
-        html = open(path, 'rb').read().decode(path.encoding, 'replace')
-        html = EntityDeclarationProcessor(html).processed_html
-        has_svg = re.search(r'<[:a-zA-Z]*svg', html) is not None
 
-        if 'xhtml' in mt:
+        def callback(lu):
-            html = self.self_closing_pat.sub(self.self_closing_sub, html)
+            self.loading_url = lu
-        if self.manager is not None:
+            if self.manager is not None:
-            self.manager.load_started()
+                self.manager.load_started()
-        self.loading_url = QUrl.fromLocalFile(path)
+
-        if has_svg:
+        load_html(path, self, codec=path.encoding, mime_type=getattr(path,
-            self.setContent(QByteArray(html.encode(path.encoding)), mt, QUrl.fromLocalFile(path))
+            'mime_type', None), pre_load_callback=callback)
-        else:
-            self.setHtml(html, self.loading_url)
         self.turn_off_internal_scrollbars()
 
     def initialize_scrollbar(self):
@@ -1010,8 +955,12 @@ class DocumentView(QWebView): # {{{
             finally:
                 self.is_auto_repeat_event = False
         elif key == 'Down':
+            if self.document.at_bottom:
+                self.manager.next_document()
             self.scroll_by(y=15)
         elif key == 'Up':
+            if self.document.at_top:
+                self.manager.previous_document()
             self.scroll_by(y=-15)
         elif key == 'Left':
             self.scroll_by(x=-15)

src/calibre/gui2/viewer/main.py

@@ -27,6 +27,7 @@ from calibre.ebooks.metadata import MetaInformation
 from calibre.customize.ui import available_input_formats
 from calibre.gui2.viewer.dictionary import Lookup
 from calibre import as_unicode, force_unicode, isbytestring
+from calibre.ptempfile import reset_base_dir
 
 vprefs = JSONConfig('viewer')
 
@@ -512,17 +513,18 @@ class EbookViewer(MainWindow, Ui_EbookViewer):
                     self.load_path(self.iterator.spine[spine_index])
 
     def goto_bookmark(self, bm):
-        m = bm[1].split('#')
+        spine_index = bm['spine']
-        if len(m) > 1:
+        if spine_index > -1 and self.current_index == spine_index:
-            spine_index, m = int(m[0]), m[1]
+            if self.resize_in_progress:
-            if spine_index > -1 and self.current_index == spine_index:
+                self.view.document.page_position.set_pos(bm['pos'])
-                self.view.goto_bookmark(m)
             else:
-                self.pending_bookmark = bm
+                self.view.goto_bookmark(bm)
-                if spine_index < 0 or spine_index >= len(self.iterator.spine):
+        else:
-                    spine_index = 0
+            self.pending_bookmark = bm
-                    self.pending_bookmark = None
+            if spine_index < 0 or spine_index >= len(self.iterator.spine):
-                self.load_path(self.iterator.spine[spine_index])
+                spine_index = 0
+                self.pending_bookmark = None
+            self.load_path(self.iterator.spine[spine_index])
 
     def toc_clicked(self, index):
         item = self.toc_model.itemFromIndex(index)
@@ -699,6 +701,14 @@ class EbookViewer(MainWindow, Ui_EbookViewer):
         self.view.load_path(path, pos=pos)
 
     def viewport_resize_started(self, event):
+        old, curr = event.size(), event.oldSize()
+        if not self.window_mode_changed and old.width() == curr.width():
+            # No relayout changes, so page position does not need to be saved
+            # This is needed as Qt generates a viewport resized event that
+            # changes only the height after a file has been loaded. This can
+            # cause the last read position bookmark to become slightly
+            # inaccurate
+            return
         if not self.resize_in_progress:
             # First resize, so save the current page position
             self.resize_in_progress = True
@@ -746,9 +756,10 @@ class EbookViewer(MainWindow, Ui_EbookViewer):
                 _('Enter title for bookmark:'), text=bm)
         title = unicode(title).strip()
         if ok and title:
-            pos = self.view.bookmark()
+            bm = self.view.bookmark()
-            bookmark = '%d#%s'%(self.current_index, pos)
+            bm['spine'] = self.current_index
-            self.iterator.add_bookmark((title, bookmark))
+            bm['title'] = title
+            self.iterator.add_bookmark(bm)
             self.set_bookmarks(self.iterator.bookmarks)
 
     def set_bookmarks(self, bookmarks):
@@ -758,12 +769,12 @@ class EbookViewer(MainWindow, Ui_EbookViewer):
         current_page = None
         self.existing_bookmarks = []
         for bm in bookmarks:
-            if bm[0] == 'calibre_current_page_bookmark' and \
+            if bm['title'] == 'calibre_current_page_bookmark':
-                                self.get_remember_current_page_opt():
+                if self.get_remember_current_page_opt():
-                current_page = bm
+                    current_page = bm
             else:
-                self.existing_bookmarks.append(bm[0])
+                self.existing_bookmarks.append(bm['title'])
-                self.bookmarks_menu.addAction(bm[0], partial(self.goto_bookmark, bm))
+                self.bookmarks_menu.addAction(bm['title'], partial(self.goto_bookmark, bm))
         return current_page
 
     def manage_bookmarks(self):
@@ -783,9 +794,10 @@ class EbookViewer(MainWindow, Ui_EbookViewer):
             return
         if hasattr(self, 'current_index'):
             try:
-                pos = self.view.bookmark()
+                bm = self.view.bookmark()
-                bookmark = '%d#%s'%(self.current_index, pos)
+                bm['spine'] = self.current_index
-                self.iterator.add_bookmark(('calibre_current_page_bookmark', bookmark))
+                bm['title'] = 'calibre_current_page_bookmark'
+                self.iterator.add_bookmark(bm)
             except:
                 traceback.print_exc()
 
@@ -947,6 +959,7 @@ View an ebook.
 def main(args=sys.argv):
     # Ensure viewer can continue to function if GUI is closed
     os.environ.pop('CALIBRE_WORKER_TEMP_DIR', None)
+    reset_base_dir()
 
     parser = option_parser()
     opts, args = parser.parse_args(args)

src/calibre/gui2/viewer/position.py

@@ -67,10 +67,16 @@ class PagePosition(object):
 
     def restore(self):
         if self._cpos is None: return
-        if isinstance(self._cpos, (int, float)):
+        self.to_pos(self._cpos)
-            self.document.scroll_fraction = self._cpos
-        else:
-            self.scroll_to_cfi(self._cpos)
         self._cpos = None
 
+    def to_pos(self, pos):
+        if isinstance(pos, (int, float)):
+            self.document.scroll_fraction = pos
+        else:
+            self.scroll_to_cfi(pos)
+
+    def set_pos(self, pos):
+        self._cpos = pos
+
 

src/calibre/library/check_library.py

@@ -47,8 +47,8 @@ class CheckLibrary(object):
         self.is_case_sensitive = db.is_case_sensitive
 
         self.all_authors = frozenset([x[1] for x in db.all_authors()])
-        self.all_ids = frozenset([id for id in db.all_ids()])
+        self.all_ids = frozenset([id_ for id_ in db.all_ids()])
-        self.all_dbpaths = frozenset(self.dbpath(id) for id in self.all_ids)
+        self.all_dbpaths = frozenset(self.dbpath(id_) for id_ in self.all_ids)
         self.all_lc_dbpaths = frozenset([f.lower() for f in self.all_dbpaths])
 
         self.db_id_regexp = re.compile(r'^.* \((\d+)\)$')
@@ -73,8 +73,8 @@ class CheckLibrary(object):
 
         self.failed_folders = []
 
-    def dbpath(self, id):
+    def dbpath(self, id_):
-        return self.db.path(id, index_is_id=True)
+        return self.db.path(id_, index_is_id=True)
 
     @property
     def errors_occurred(self):
@@ -116,21 +116,21 @@ class CheckLibrary(object):
                     self.invalid_titles.append((auth_dir, db_path, 0))
                     continue
 
-                id = m.group(1)
+                id_ = m.group(1)
-                # Third check: the id must be in the DB and the paths must match
+                # Third check: the id_ must be in the DB and the paths must match
                 if self.is_case_sensitive:
-                    if int(id) not in self.all_ids or \
+                    if int(id_) not in self.all_ids or \
                             db_path not in self.all_dbpaths:
                         self.extra_titles.append((title_dir, db_path, 0))
                         continue
                 else:
-                    if int(id) not in self.all_ids or \
+                    if int(id_) not in self.all_ids or \
                             db_path.lower() not in self.all_lc_dbpaths:
                         self.extra_titles.append((title_dir, db_path, 0))
                         continue
 
                 # Record the book to check its formats
-                self.book_dirs.append((db_path, title_dir, id))
+                self.book_dirs.append((db_path, title_dir, id_))
                 found_titles = True
 
             # Fourth check: author directories that contain no titles
@@ -145,6 +145,21 @@ class CheckLibrary(object):
                 # Sort-of check: exception processing directory
                 self.failed_folders.append((title_path, traceback.format_exc(), []))
 
+        # Check for formats and covers in db for book dirs that are gone
+        for id_ in self.all_ids:
+            path = self.dbpath(id_)
+            if not os.path.exists(os.path.join(lib, path)):
+                title_dir = os.path.basename(path)
+                book_formats = frozenset([x for x in
+                            self.db.format_files(id_, index_is_id=True)])
+                for fmt in book_formats:
+                    self.missing_formats.append((title_dir,
+                            os.path.join(path, fmt[0]+'.'+fmt[1].lower()), id_))
+                if self.db.has_cover(id_):
+                    self.missing_covers.append((title_dir,
+                            os.path.join(path, 'cover.jpg'), id_))
+
+
     def is_ebook_file(self, filename):
         ext = os.path.splitext(filename)[1]
         if not ext:
@@ -226,8 +241,8 @@ class CheckLibrary(object):
         if self.db.has_cover(book_id):
             if 'cover.jpg' not in filenames:
                 self.missing_covers.append((title_dir,
-                        os.path.join(db_path, title_dir, 'cover.jpg'), book_id))
+                        os.path.join(db_path, 'cover.jpg'), book_id))
         else:
             if 'cover.jpg' in filenames:
                 self.extra_covers.append((title_dir,
-                        os.path.join(db_path, title_dir, 'cover.jpg'), book_id))
+                        os.path.join(db_path, 'cover.jpg'), book_id))

src/calibre/library/cli.py

@@ -204,7 +204,8 @@ class DevNull(object):
         pass
 NULL = DevNull()
 
-def do_add(db, paths, one_book_per_directory, recurse, add_duplicates):
+def do_add(db, paths, one_book_per_directory, recurse, add_duplicates, otitle,
+        oauthors, oisbn, otags, oseries, oseries_index):
     orig = sys.stdout
     #sys.stdout = NULL
     try:
@@ -231,6 +232,11 @@ def do_add(db, paths, one_book_per_directory, recurse, add_duplicates):
                 mi.title = os.path.splitext(os.path.basename(book))[0]
             if not mi.authors:
                 mi.authors = [_('Unknown')]
+            for x in ('title', 'authors', 'isbn', 'tags', 'series'):
+                val = locals()['o'+x]
+                if val: setattr(mi, x[1:], val)
+            if oseries:
+                mi.series_index = oseries_index
 
             formats.append(format)
             metadata.append(mi)
@@ -302,39 +308,56 @@ the directory related options below.
     parser.add_option('-e', '--empty', action='store_true', default=False,
                     help=_('Add an empty book (a book with no formats)'))
     parser.add_option('-t', '--title', default=None,
-            help=_('Set the title of the added empty book'))
+            help=_('Set the title of the added book(s)'))
     parser.add_option('-a', '--authors', default=None,
-            help=_('Set the authors of the added empty book'))
+            help=_('Set the authors of the added book(s)'))
     parser.add_option('-i', '--isbn', default=None,
-            help=_('Set the ISBN of the added empty book'))
+            help=_('Set the ISBN of the added book(s)'))
+    parser.add_option('-T', '--tags', default=None,
+            help=_('Set the tags of the added book(s)'))
+    parser.add_option('-s', '--series', default=None,
+            help=_('Set the series of the added book(s)'))
+    parser.add_option('-S', '--series-index', default=1.0, type=float,
+            help=_('Set the series number of the added book(s)'))
+
 
     return parser
 
-def do_add_empty(db, title, authors, isbn):
+def do_add_empty(db, title, authors, isbn, tags, series, series_index):
-    from calibre.ebooks.metadata import MetaInformation, string_to_authors
+    from calibre.ebooks.metadata import MetaInformation
     mi = MetaInformation(None)
     if title is not None:
         mi.title = title
     if authors:
-        mi.authors = string_to_authors(authors)
+        mi.authors = authors
     if isbn:
         mi.isbn = isbn
+    if tags:
+        mi.tags = tags
+    if series:
+        mi.series, mi.series_index = series, series_index
     db.import_book(mi, [])
     write_dirtied(db)
     send_message()
 
 def command_add(args, dbpath):
+    from calibre.ebooks.metadata import string_to_authors
     parser = add_option_parser()
     opts, args = parser.parse_args(sys.argv[:1] + args)
+    aut = string_to_authors(opts.authors) if opts.authors else []
+    tags = [x.strip() for x in opts.tags.split(',')] if opts.tags else []
     if opts.empty:
-        do_add_empty(get_db(dbpath, opts), opts.title, opts.authors, opts.isbn)
+        do_add_empty(get_db(dbpath, opts), opts.title, aut, opts.isbn, tags,
+                opts.series, opts.series_index)
         return 0
     if len(args) < 2:
         parser.print_help()
         print
         print >>sys.stderr, _('You must specify at least one file to add')
         return 1
-    do_add(get_db(dbpath, opts), args[1:], opts.one_book_per_directory, opts.recurse, opts.duplicates)
+    do_add(get_db(dbpath, opts), args[1:], opts.one_book_per_directory,
+            opts.recurse, opts.duplicates, opts.title, opts.authors, opts.isbn,
+            tags, opts.series, opts.series_index)
     return 0
 
 def do_remove(db, ids):

src/calibre/linux.py

@@ -40,6 +40,46 @@ entry_points = {
                             ],
       }
 
+class PreserveMIMEDefaults(object):
+
+    def __init__(self):
+        self.initial_values = {}
+
+    def __enter__(self):
+        def_data_dirs = '/usr/local/share:/usr/share'
+        paths = os.environ.get('XDG_DATA_DIRS', def_data_dirs)
+        paths = paths.split(':')
+        paths.append(os.environ.get('XDG_DATA_HOME', os.path.expanduser(
+            '~/.local/share')))
+        paths = list(filter(os.path.isdir, paths))
+        if not paths:
+            # Env var had garbage in it, ignore it
+            paths = def_data_dirs.split(':')
+        paths = list(filter(os.path.isdir, paths))
+        self.paths = {os.path.join(x, 'applications/defaults.list') for x in
+                paths}
+        self.initial_values = {}
+        for x in self.paths:
+            try:
+                with open(x, 'rb') as f:
+                    self.initial_values[x] = f.read()
+            except:
+                self.initial_values[x] = None
+
+    def __exit__(self, *args):
+        for path, val in self.initial_values.iteritems():
+            if val is None:
+                try:
+                    os.remove(path)
+                except:
+                    pass
+            elif os.path.exists(path):
+                with open(path, 'r+b') as f:
+                    if f.read() != val:
+                        f.seek(0)
+                        f.truncate()
+                        f.write(val)
+
 # Uninstall script {{{
 UNINSTALL = '''\
 #!{python}
@@ -202,6 +242,10 @@ class PostInstall:
             if not os.path.exists(os.path.dirname(f)):
                 os.makedirs(os.path.dirname(f))
             self.manifest.append(f)
+            complete = 'calibre-complete'
+            if getattr(sys, 'frozen_path', None):
+                complete = os.path.join(getattr(sys, 'frozen_path'), complete)
+
             self.info('Installing bash completion to', f)
             with open(f, 'wb') as f:
                 f.write('# calibre Bash Shell Completion\n')
@@ -286,8 +330,8 @@ class PostInstall:
                 }
                 complete -o nospace  -F _ebook_device ebook-device
 
-                complete -o nospace -C calibre-complete ebook-convert
+                complete -o nospace -C %s ebook-convert
-                '''))
+                ''')%complete)
         except TypeError as err:
             if 'resolve_entities' in str(err):
                 print 'You need python-lxml >= 2.0.5 for calibre'
@@ -333,57 +377,55 @@ class PostInstall:
 
     def setup_desktop_integration(self): # {{{
         try:
-
             self.info('Setting up desktop integration...')
 
+            with TemporaryDirectory() as tdir, CurrentDir(tdir), \
+                                PreserveMIMEDefaults():
+                render_img('mimetypes/lrf.png', 'calibre-lrf.png')
+                check_call('xdg-icon-resource install --noupdate --context mimetypes --size 128 calibre-lrf.png application-lrf', shell=True)
+                self.icon_resources.append(('mimetypes', 'application-lrf', '128'))
+                check_call('xdg-icon-resource install --noupdate --context mimetypes --size 128 calibre-lrf.png text-lrs', shell=True)
+                self.icon_resources.append(('mimetypes', 'application-lrs',
+                '128'))
+                render_img('lt.png', 'calibre-gui.png')
+                check_call('xdg-icon-resource install --noupdate --size 128 calibre-gui.png calibre-gui', shell=True)
+                self.icon_resources.append(('apps', 'calibre-gui', '128'))
+                render_img('viewer.png', 'calibre-viewer.png')
+                check_call('xdg-icon-resource install --size 128 calibre-viewer.png calibre-viewer', shell=True)
+                self.icon_resources.append(('apps', 'calibre-viewer', '128'))
 
-            with TemporaryDirectory() as tdir:
+                mimetypes = set([])
-                with CurrentDir(tdir):
+                for x in all_input_formats():
-                    render_img('mimetypes/lrf.png', 'calibre-lrf.png')
+                    mt = guess_type('dummy.'+x)[0]
-                    check_call('xdg-icon-resource install --noupdate --context mimetypes --size 128 calibre-lrf.png application-lrf', shell=True)
+                    if mt and 'chemical' not in mt and 'ctc-posml' not in mt:
-                    self.icon_resources.append(('mimetypes', 'application-lrf', '128'))
+                        mimetypes.add(mt)
-                    check_call('xdg-icon-resource install --noupdate --context mimetypes --size 128 calibre-lrf.png text-lrs', shell=True)
-                    self.icon_resources.append(('mimetypes', 'application-lrs',
-                    '128'))
-                    render_img('lt.png', 'calibre-gui.png')
-                    check_call('xdg-icon-resource install --noupdate --size 128 calibre-gui.png calibre-gui', shell=True)
-                    self.icon_resources.append(('apps', 'calibre-gui', '128'))
-                    render_img('viewer.png', 'calibre-viewer.png')
-                    check_call('xdg-icon-resource install --size 128 calibre-viewer.png calibre-viewer', shell=True)
-                    self.icon_resources.append(('apps', 'calibre-viewer', '128'))
 
-                    mimetypes = set([])
+                def write_mimetypes(f):
-                    for x in all_input_formats():
+                    f.write('MimeType=%s;\n'%';'.join(mimetypes))
-                        mt = guess_type('dummy.'+x)[0]
-                        if mt and 'chemical' not in mt and 'ctc-posml' not in mt:
-                            mimetypes.add(mt)
 
-                    def write_mimetypes(f):
+                f = open('calibre-lrfviewer.desktop', 'wb')
-                        f.write('MimeType=%s;\n'%';'.join(mimetypes))
+                f.write(VIEWER)
-
+                f.close()
-                    f = open('calibre-lrfviewer.desktop', 'wb')
+                f = open('calibre-ebook-viewer.desktop', 'wb')
-                    f.write(VIEWER)
+                f.write(EVIEWER)
-                    f.close()
+                write_mimetypes(f)
-                    f = open('calibre-ebook-viewer.desktop', 'wb')
+                f.close()
-                    f.write(EVIEWER)
+                f = open('calibre-gui.desktop', 'wb')
-                    write_mimetypes(f)
+                f.write(GUI)
-                    f.close()
+                write_mimetypes(f)
-                    f = open('calibre-gui.desktop', 'wb')
+                f.close()
-                    f.write(GUI)
+                des = ('calibre-gui.desktop', 'calibre-lrfviewer.desktop',
-                    write_mimetypes(f)
+                        'calibre-ebook-viewer.desktop')
-                    f.close()
+                for x in des:
-                    des = ('calibre-gui.desktop', 'calibre-lrfviewer.desktop',
+                    cmd = ['xdg-desktop-menu', 'install', '--noupdate', './'+x]
-                            'calibre-ebook-viewer.desktop')
+                    check_call(' '.join(cmd), shell=True)
-                    for x in des:
+                    self.menu_resources.append(x)
-                        cmd = ['xdg-desktop-menu', 'install', '--noupdate', './'+x]
+                check_call(['xdg-desktop-menu', 'forceupdate'])
-                        check_call(' '.join(cmd), shell=True)
+                f = open('calibre-mimetypes', 'wb')
-                        self.menu_resources.append(x)
+                f.write(MIME)
-                    check_call(['xdg-desktop-menu', 'forceupdate'])
+                f.close()
-                    f = open('calibre-mimetypes', 'wb')
+                self.mime_resources.append('calibre-mimetypes')
-                    f.write(MIME)
+                check_call('xdg-mime install ./calibre-mimetypes', shell=True)
-                    f.close()
-                    self.mime_resources.append('calibre-mimetypes')
-                    check_call('xdg-mime install ./calibre-mimetypes', shell=True)
         except Exception:
             if self.opts.fatal_errors:
                 raise

src/calibre/ptempfile.py

@@ -74,6 +74,11 @@ def base_dir():
 
     return _base_dir
 
+def reset_base_dir():
+    global _base_dir
+    _base_dir = None
+    base_dir()
+
 def force_unicode(x):
     # Cannot use the implementation in calibre.__init__ as it causes a circular
     # dependency