Merge branch 'kovidgoyal/master'

Changelog.yaml

@@ -20,6 +20,50 @@
 #   new recipes:
 #     - title: 
 
+- version: 0.9.37
+  date: 2013-06-28
+
+  new features:
+    - title: "Conversion: Add option to embed all referenced fonts"
+      type: major
+      description: "Add an option to embed all fonts that are referenced in the input document but are not already embedded. This will search your system for the referenced font, and if found, the font will be embedded.  Only works if the output format supports font embedding (for example: EPUB or AZW3). The option is under the Look & Feel section of the conversion dialog."
+
+    - title: "ToC Editor: When generating a ToC from files, if the file has no text, do not skip it. Instead create an entry using the filename of the file."
+
+    - title: "AZW3 Input: Add support for the page-progression-direction that is used to indicate page turns should happen from right to left. The attribute is passed into EPUB when converting."
+      tickets: [1194766]
+
+    - title: "ebook-convert: Add a --from-opf option to read metadata from OPF files directly, instead of having to run ebook-meta --from-opf after conversion"
+ 
+  bug fixes:
+    - title: "PDF Output: Fix Table of Contents being added to the end of the PDF even without the Add Table of Contents option being enabled."
+      tickets: [1194836]
+
+    - title: "When auto-merging books on add, also merge identifiers."
+
+    - title: "Fix an error when using the Template Editor to create a template that uses custom columns."
+      tickets: [1193763]
+
+    - title: "LRF Output: Fix " entities in attribute values causing problems"
+
+    - title: "News download: Apply the default page margin conversion settings. Also, when converting to PDF, apply the pdf conversion defaults."
+      tickets: [1193912]
+
+    - title: "Fix a regression that broke scanning for books on all devices that used the Aluratek Color driver."
+      tickets: [1192940]
+
+    - title: "fetch-ebbok-metadata: Fix --opf argument erroneously requiring a value"
+
+    - title: "When waiting before sending email, log the wait."
+      tickets: [1195173]
+
+  improved recipes:
+    - taz.de (RSS)
+    - Miradas al sur
+    - Frontline
+    - La Nacion (Costa Rica)
+
+
 - version: 0.9.36
   date: 2013-06-21
 

recipes/democracy_now.recipe

@@ -0,0 +1,45 @@
+# vim:fileencoding=utf-8
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class DemocracyNowRecipe(BasicNewsRecipe):
+    title = u'Democracy now!'
+    __author__ = u'Antoine Beaupré'
+    description = 'A daily TV/radio news program, hosted by Amy Goodman and Juan Gonzalez, airing on over 1,100 stations, pioneering the largest community media collaboration in the United States.'  # noqa
+    language = 'en'
+    cover_url = 'http://www.democracynow.org/images/dn-logo-for-podcast.png'
+
+    oldest_article = 1
+    max_articles_per_feed = 10
+    publication_type = 'magazine'
+
+    auto_cleanup = False
+    use_embedded_content = False
+    no_stylesheets = True
+    remove_javascript = True
+
+    feeds = [
+        (u'Daily news', u'http://www.democracynow.org/democracynow.rss')]
+
+    keep_only_tags = [dict(name='div', attrs={'id': 'page'}), ]
+    remove_tags = [dict(name='div', attrs={'id': 'topics_list'}),
+                   dict(name='div', attrs={'id': 'header'}),
+                   dict(name='div', attrs={'id': 'footer'}),
+                   dict(name='div', attrs={'id': 'right'}),
+                   dict(name='div', attrs={'id': 'left-panel'}),
+                   dict(name='div', attrs={'id': 'top-video-content'}),
+                   dict(name='div', attrs={'id': 'google-news-date'}),
+                   dict(name='div', attrs={'id': 'story-donate'}),
+                   dict(
+                   name='div', attrs={'id': 'transcript-expand-collapse'}),
+                   dict(name='span', attrs={'class': 'show-links'}),
+                   dict(name='span', attrs={'class': 'storyNav'}),
+                   dict(name='div', attrs={'class': 'headline_share'}),
+                   dict(name='div', attrs={'class': 'mediaBar'}),
+                   dict(name='div', attrs={'class': 'shareAndPrinterBar'}),
+                   dict(name='div', attrs={'class': 'utility-navigation'}),
+                   dict(name='div', attrs={'class': 'bottomContentNav'}),
+                   dict(name='div', attrs={'class': 'recentShows'}),
+                   dict(
+                   name='div', attrs={'class': 'printer-and-transcript-links'}),
+                   ]
+

recipes/gosc_niedzielny.recipe

@@ -47,13 +47,7 @@ class GN(BasicNewsRecipe):
                 return feeds
 
         def find_articles(self, main_block):
-                for a in main_block.findAll('div', attrs={'class':'prev_doc2'}):
+                for a in main_block.findAll('div', attrs={'class':['prev_doc2', 'sr-document']}):
-						art = a.find('a')
-						yield {
-                                'title' : self.tag_to_string(art),
-                                'url'   : 'http://www.gosc.pl' + art['href']
-                                }
-                for a in main_block.findAll('div', attrs={'class':'sr-document'}):
 						art = a.find('a')
 						yield {
                                 'title' : self.tag_to_string(art),

recipes/iprofesional.recipe

@@ -1,5 +1,4 @@
-__license__   = 'GPL v3'
+__copyright__ = '2011-2013, Darko Miletic <darko.miletic at gmail.com>'
-__copyright__ = '2011, Darko Miletic <darko.miletic at gmail.com>'
 '''
 www.iprofesional.com
 '''
@@ -19,13 +18,15 @@ class iProfesional(BasicNewsRecipe):
     use_embedded_content  = False
     language              = 'es_AR'
     remove_empty_feeds    = True
-    publication_type      = 'nesportal'
+    publication_type      = 'newsportal'
-    masthead_url          = 'http://www.iprofesional.com/img/logo-iprofesional.png'
+    masthead_url          = 'http://www.iprofesional.com/img/header/logoiprofesional.png'
     extra_css             = """
-                               body{font-family: Arial,Helvetica,sans-serif }
+                               body{font-family: 'Droid Sans',Arial,sans-serif }
                                img{margin-bottom: 0.4em; display:block}
-                               .titulo-interior{font-family: Georgia,"Times New Roman",Times,serif}
+                               .titulo{font-family: WhitneyBoldWhitneyBold,Arial,Helvetica,sans-serif; color: blue}
-                               .autor-nota{font-size: small; font-weight: bold; font-style: italic; color: gray}
+                               .fecha-archivo{font-weight: bold; color: rgb(205, 150, 24)}
+                               .description{font-weight: bold; color: gray }
+                               .firma{font-size: small}
                             """
 
     conversion_options = {
@@ -35,27 +36,21 @@ class iProfesional(BasicNewsRecipe):
                         , 'language'  : language
                         }
 
-    keep_only_tags = [dict(attrs={'class':['fecha','interior-nota']})]
+    keep_only_tags = [dict(attrs={'class':'desarrollo'})]
-
+    remove_tags = [dict(name=['meta','link','base','embed','object','iframe'])]
-    remove_tags = [
-                     dict(name=['meta','link','base','embed','object','iframe'])
-                    ,dict(attrs={'class':['menu-imprimir','guardarNota','IN-widget','fin','permalink']})
-                  ]
-    remove_attributes=['lang','xmlns:og','xmlns:fb']
-
 
     feeds = [
               (u'Ultimas noticias'  , u'http://feeds.feedburner.com/iprofesional-principales-noticias')
-             ,(u'Finanzas'          , u'http://feeds.feedburner.com/iprofesional-finanzas'            )
+             ,(u'Finanzas'          , u'http://feeds.feedburner.com/iprofesional-finanzas')
-             ,(u'Impuestos'         , u'http://feeds.feedburner.com/iprofesional-impuestos'           )
+             ,(u'Impuestos'         , u'http://feeds.feedburner.com/iprofesional-impuestos')
-             ,(u'Negocios'          , u'http://feeds.feedburner.com/iprofesional-economia'            )
+             ,(u'Negocios'          , u'http://feeds.feedburner.com/iprofesional-economia')
-             ,(u'Comercio Exterior' , u'http://feeds.feedburner.com/iprofesional-comercio-exterior'   )
+             ,(u'Comercio Exterior' , u'http://feeds.feedburner.com/iprofesional-comercio-exterior')
-             ,(u'Tecnologia'        , u'http://feeds.feedburner.com/iprofesional-tecnologia'          )
+             ,(u'Tecnologia'        , u'http://feeds.feedburner.com/iprofesional-tecnologia')
-             ,(u'Management'        , u'http://feeds.feedburner.com/iprofesional-managment'           )
+             ,(u'Management'        , u'http://feeds.feedburner.com/iprofesional-managment')
-             ,(u'Marketing'         , u'http://feeds.feedburner.com/iprofesional-marketing'           )
+             ,(u'Marketing'         , u'http://feeds.feedburner.com/iprofesional-marketing')
-             ,(u'Legales'           , u'http://feeds.feedburner.com/iprofesional-legales'             )
+             ,(u'Legales'           , u'http://feeds.feedburner.com/iprofesional-legales')
-             ,(u'Autos'             , u'http://feeds.feedburner.com/iprofesional-autos'               )
+             ,(u'Autos'             , u'http://feeds.feedburner.com/iprofesional-autos')
-             ,(u'Vinos'             , u'http://feeds.feedburner.com/iprofesional-vinos-bodegas'       )
+             ,(u'Vinos'             , u'http://feeds.feedburner.com/iprofesional-vinos-bodegas')
             ]
 
     def preprocess_html(self, soup):
@@ -64,16 +59,17 @@ class iProfesional(BasicNewsRecipe):
         for item in soup.findAll('a'):
             limg = item.find('img')
             if item.string is not None:
-               str = item.string
+                str = item.string
-               item.replaceWith(str)
+                item.replaceWith(str)
             else:
-               if limg:
+                if limg:
-                  item.name = 'div'
+                    item.name = 'div'
-                  item.attrs = []
+                    item.attrs = []
-               else:
+                else:
-                   str = self.tag_to_string(item)
+                    str = self.tag_to_string(item)
-                   item.replaceWith(str)
+                    item.replaceWith(str)
         for item in soup.findAll('img'):
-            if not item.has_key('alt'):
+            if 'alt' not in item:
-               item['alt'] = 'image'
+                item['alt'] = 'image'
         return soup
+

recipes/miradasalsur.recipe

@@ -4,9 +4,7 @@ sur.infonews.com
 '''
 
 import datetime
-from calibre import strftime
 from calibre.web.feeds.news import BasicNewsRecipe
-from calibre.ebooks.BeautifulSoup import Tag
 
 class MiradasAlSur(BasicNewsRecipe):
     title                 = 'Miradas al Sur'
@@ -25,7 +23,7 @@ class MiradasAlSur(BasicNewsRecipe):
     extra_css             = """
                              body{font-family: Arial,Helvetica,sans-serif}
                              h1{font-family: Georgia,Times,serif}
-                             .field-field-story-author{color: gray; font-size: small}                             
+                             .field-field-story-author{color: gray; font-size: small}
                             """
     conversion_options = {
                           'comment'   : description
@@ -34,22 +32,22 @@ class MiradasAlSur(BasicNewsRecipe):
                         , 'language'  : language
                         , 'series'    : title
                         }
-                            
+
     keep_only_tags = [dict(name='div', attrs={'id':['content-header', 'content-area']})]
-    remove_tags    = [ 
+    remove_tags    = [
-                      dict(name=['link','meta','iframe','embed','object']), 
+                      dict(name=['link','meta','iframe','embed','object']),
                       dict(name='form', attrs={'class':'fivestar-widget'}),
                       dict(attrs={'class':lambda x: x and 'terms-inline' in x.split()})
                      ]
 
     feeds = [
-             (u'Politica'           , u'http://sur.infonews.com/taxonomy/term/1/0/feed' ),
+             (u'Politica'           , u'http://sur.infonews.com/taxonomy/term/1/0/feed'),
-             (u'Internacional'      , u'http://sur.infonews.com/taxonomy/term/2/0/feed' ),
+             (u'Internacional'      , u'http://sur.infonews.com/taxonomy/term/2/0/feed'),
              (u'Informe Especial'   , u'http://sur.infonews.com/taxonomy/term/14/0/feed'),
-             (u'Delitos y pesquisas', u'http://sur.infonews.com/taxonomy/term/6/0/feed' ),
+             (u'Delitos y pesquisas', u'http://sur.infonews.com/taxonomy/term/6/0/feed'),
-             (u'Lesa Humanidad'     , u'http://sur.infonews.com/taxonomy/term/7/0/feed' ),
+             (u'Lesa Humanidad'     , u'http://sur.infonews.com/taxonomy/term/7/0/feed'),
-             (u'Cultura'            , u'http://sur.infonews.com/taxonomy/term/8/0/feed' ),
+             (u'Cultura'            , u'http://sur.infonews.com/taxonomy/term/8/0/feed'),
-             (u'Deportes'           , u'http://sur.infonews.com/taxonomy/term/9/0/feed' ),
+             (u'Deportes'           , u'http://sur.infonews.com/taxonomy/term/9/0/feed'),
              (u'Contratapa'         , u'http://sur.infonews.com/taxonomy/term/10/0/feed'),
             ]
 
@@ -60,10 +58,10 @@ class MiradasAlSur(BasicNewsRecipe):
         cdate = datetime.date.today()
         todayweekday = cdate.isoweekday()
         if (todayweekday != 7):
-            cdate -= datetime.timedelta(days=todayweekday)        
+            cdate -= datetime.timedelta(days=todayweekday)
-        cover_page_url = cdate.strftime('http://sur.infonews.com/ediciones/%Y-%m-%d/tapa');
+        cover_page_url = cdate.strftime('http://sur.infonews.com/ediciones/%Y-%m-%d/tapa')
         soup = self.index_to_soup(cover_page_url)
         cover_item = soup.find('img', attrs={'class':lambda x: x and 'imagecache-tapa_edicion_full' in x.split()})
         if cover_item:
-           cover_url = cover_item['src']
+            cover_url = cover_item['src']
         return cover_url

recipes/taz_rss.recipe

@@ -1,23 +1,43 @@
-__license__   = 'GPL v3'
+__license__ = 'GPL v3'
-__copyright__ = '2010, Alexander Schremmer <alex@alexanderweb.de>'
+__copyright__ = '2013, Alexander Schremmer <alex@alexanderweb.de>, Robert Riemann <robert@riemann.cc>'
 
+import re
 from calibre.web.feeds.news import BasicNewsRecipe
 
 class TazRSSRecipe(BasicNewsRecipe):
-    title = u'Taz.de (die tageszeitung) RSS Feed - German'
+    title = u'Taz - die Tageszeitung'
-    __author__ = 'Alexander Schremmer'
+    description = u'Taz.de - die tageszeitung'
+    __author__ = 'Alexander Schremmer, Robert Riemann'
     language = 'de'
     lang = 'de-DE'
     oldest_article = 7
     max_articles_per_feed = 100
     publisher = 'taz Entwicklungs GmbH & Co. Medien KG'
+    # masthead_url = u'http://galeria-autonomica.de/wp-content/uploads/a_taz-logo.gif'
+    masthead_url = u'http://upload.wikimedia.org/wikipedia/de/thumb/1/15/Die-Tageszeitung-Logo.svg/500px-Die-Tageszeitung-Logo.svg.png'
 
     conversion_options = {'publisher': publisher,
                           'language': lang,
                         }
-
+    feeds = [
-    feeds          = [(u'TAZ main feed', u'http://www.taz.de/rss.xml')]
+             (u'Schlagzeilen', u'http://www.taz.de/!p3270;rss/'),
+             (u'Politik', u'http://www.taz.de/Politik/!p2;rss/'),
+             (u'Zukunft', u'http://www.taz.de/Zukunft/!p4;rss/'),
+             (u'Netz', u'http://www.taz.de/Netz/!p5;rss/'),
+             (u'Debatte', u'http://www.taz.de/Debatte/!p9;rss/'),
+             (u'Leben', u'http://www.taz.de/Leben/!p10;rss/'),
+             (u'Sport', u'http://www.taz.de/Sport/!p12;rss/'),
+             (u'Wahrheit', u'http://www.taz.de/Wahrheit/!p13;rss/'),
+             (u'Berlin', u'http://www.taz.de/Berlin/!p14;rss/'),
+             (u'Nord', u'http://www.taz.de/Nord/!p11;rss/')
+            ]
     keep_only_tags = [dict(name='div', attrs={'class': 'sect sect_article'})]
     remove_tags = [
-         dict(name=['div'], attrs={'class': 'artikelwerbung'}),
+                   dict(name=['div'], attrs={'class': 'artikelwerbung'}),
-         dict(name=['ul'], attrs={'class': 'toolbar'}),]
+                   dict(name=['ul'], attrs={'class': 'toolbar'}),
+                   # remove: taz paywall
+                   dict(name=['div'], attrs={'id': 'tzi_paywall'}),
+                   # remove: Artikel zum Thema (not working on Kindle)
+                   dict(name=['div'], attrs={'class': re.compile(r".*\bsect_seealso\b.*")}),
+                   dict(name=['div'], attrs={'class': 'sectfoot'})
+                  ]

setup/file_hosting_servers.rst

@@ -0,0 +1,51 @@
+Provisioning a file hosting server
+====================================
+
+Create the ssh authorized keys file.
+
+Edit /etc/ssh/sshd_config and change PermitRootLogin to without-password.
+Restart sshd.
+
+hostname whatever
+Edit /etc/hosts and put in FQDN in the appropriate places, for example::
+    27.0.1.1       download.calibre-ebook.com download
+    46.28.49.116 download.calibre-ebook.com download
+
+dpkg-reconfigure tzdata
+set timezone to Asia/Kolkata
+service cron restart
+
+apt-get install vim nginx zsh python-lxml python-mechanize iotop htop smartmontools mosh
+chsh -s /bin/zsh
+
+mkdir -p /root/staging /root/work/vim /srv/download /srv/manual
+
+scp .zshrc .vimrc  server:
+scp -r ~/work/vim/zsh-syntax-highlighting server:work/vim
+
+If the server has a backup hard-disk, mount it at /mnt/backup and edit /etc/fstab so that it is auto-mounted.
+Then, add the following to crontab::
+    @daily     /usr/bin/rsync -ha /srv /mnt/backup
+    @daily     /usr/bin/rsync -ha /etc /mnt/backup
+
+Nginx
+------
+
+Copy over /etc/nginx/sites-available/default from another file server. When
+copying, remember to use cat instead of cp to preserve hardlinks (the file is a
+hardlink to /etc/nginx/sites-enabled/default)
+
+rsync /srv from another file server
+
+service nginx start
+
+Services
+---------
+
+SSH into sourceforge and downloadbestsoftware so that their host keys are
+stored.
+
+   ssh -oStrictHostKeyChecking=no kovid@www.downloadbestsoft-mirror1.com
+   ssh -oStrictHostKeyChecking=no kovidgoyal,calibre@frs.sourceforge.net
+   ssh -oStrictHostKeyChecking=no files.calibre-ebook.com (and whatever other mirrors are present)
+

setup/hosting.py

@@ -7,16 +7,14 @@ __license__   = 'GPL v3'
 __copyright__ = '2011, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-import os, time, sys, traceback, subprocess, urllib2, re, base64, httplib
+import os, time, sys, traceback, subprocess, urllib2, re, base64, httplib, shutil
 from argparse import ArgumentParser, FileType
 from subprocess import check_call
 from tempfile import NamedTemporaryFile
 from collections import OrderedDict
 
-import mechanize
-from lxml import html
-
 def login_to_google(username, password):  # {{{
+    import mechanize
     br = mechanize.Browser()
     br.addheaders = [('User-agent',
         'Mozilla/5.0 (X11; Linux x86_64; rv:9.0) Gecko/20100101 Firefox/9.0')]
@@ -246,6 +244,7 @@ class GoogleCode(Base):  # {{{
         return login_to_google(self.username, self.gmail_password)
 
     def get_files_hosted_by_google_code(self):
+        from lxml import html
         self.info('Getting existing files in google code:', self.gc_project)
         raw = urllib2.urlopen(self.files_list).read()
         root = html.fromstring(raw)
@@ -380,11 +379,115 @@ class SourceForge(Base):  # {{{
 
 # }}}
 
+def generate_index():  # {{{
+    os.chdir('/srv/download')
+    releases = set()
+    for x in os.listdir('.'):
+        if os.path.isdir(x) and '.' in x:
+            releases.add(tuple((int(y) for y in x.split('.'))))
+    rmap = OrderedDict()
+    for rnum in sorted(releases, reverse=True):
+        series = rnum[:2] if rnum[0] == 0 else rnum[:1]
+        if series not in rmap:
+            rmap[series] = []
+        rmap[series].append(rnum)
+
+    template = '''<!DOCTYPE html>\n<html lang="en"> <head> <meta charset="utf-8"> <title>{title}</title> <style type="text/css"> {style} </style> </head> <body> <h1>{title}</h1> <p>{msg}</p> {body} </body> </html> '''  # noqa
+    style = '''
+    body { font-family: sans-serif; background-color: #eee; }
+    a { text-decoration: none; }
+    a:visited { color: blue }
+    a:hover { color: red }
+    ul { list-style-type: none }
+    li { padding-bottom: 1ex }
+    dd li { text-indent: 0; margin: 0 }
+    dd ul { padding: 0; margin: 0 }
+    dt { font-weight: bold }
+    dd { margin-bottom: 2ex }
+    '''
+    body = []
+    for series in rmap:
+        body.append('<li><a href="{0}.html" title="Releases in the {0}.x series">{0}.x</a>\xa0\xa0\xa0<span style="font-size:smaller">[{1} releases]</span></li>'.format(  # noqa
+                '.'.join(map(type(''), series)), len(rmap[series])))
+    body = '<ul>{0}</ul>'.format(' '.join(body))
+    index = template.format(title='Previous calibre releases', style=style, msg='Choose a series of calibre releases', body=body)
+    with open('index.html', 'wb') as f:
+        f.write(index.encode('utf-8'))
+
+    for series, releases in rmap.iteritems():
+        sname = '.'.join(map(type(''), series))
+        body = [
+            '<li><a href="{0}/" title="Release {0}">{0}</a></li>'.format('.'.join(map(type(''), r)))
+            for r in releases]
+        body = '<ul class="release-list">{0}</ul>'.format(' '.join(body))
+        index = template.format(title='Previous calibre releases (%s.x)' % sname, style=style,
+                                msg='Choose a calibre release', body=body)
+        with open('%s.html' % sname, 'wb') as f:
+            f.write(index.encode('utf-8'))
+
+        for r in releases:
+            rname = '.'.join(map(type(''), r))
+            os.chdir(rname)
+            try:
+                body = []
+                files = os.listdir('.')
+                windows = [x for x in files if x.endswith('.msi')]
+                if windows:
+                    windows = ['<li><a href="{0}" title="{1}">{1}</a></li>'.format(
+                        x, 'Windows 64-bit Installer' if '64bit' in x else 'Windows 32-bit Installer')
+                        for x in windows]
+                    body.append('<dt>Windows</dt><dd><ul>{0}</ul></dd>'.format(' '.join(windows)))
+                portable = [x for x in files if '-portable-' in x]
+                if portable:
+                    body.append('<dt>Calibre Portable</dt><dd><a href="{0}" title="{1}">{1}</a></dd>'.format(
+                        portable[0], 'Calibre Portable Installer'))
+                osx = [x for x in files if x.endswith('.dmg')]
+                if osx:
+                    body.append('<dt>Apple Mac</dt><dd><a href="{0}" title="{1}">{1}</a></dd>'.format(
+                        osx[0], 'OS X Disk Image (.dmg)'))
+                linux = [x for x in files if x.endswith('.bz2')]
+                if linux:
+                    linux = ['<li><a href="{0}" title="{1}">{1}</a></li>'.format(
+                        x, 'Linux 64-bit binary' if 'x86_64' in x else 'Linux 32-bit binary')
+                        for x in linux]
+                    body.append('<dt>Linux</dt><dd><ul>{0}</ul></dd>'.format(' '.join(linux)))
+                source = [x for x in files if x.endswith('.xz') or x.endswith('.gz')]
+                if source:
+                    body.append('<dt>Source Code</dt><dd><a href="{0}" title="{1}">{1}</a></dd>'.format(
+                        source[0], 'Source code (all platforms)'))
+
+                body = '<dl>{0}</dl>'.format(''.join(body))
+                index = template.format(title='calibre release (%s)' % rname, style=style,
+                                msg='', body=body)
+                with open('index.html', 'wb') as f:
+                    f.write(index.encode('utf-8'))
+            finally:
+                os.chdir('..')
+
+# }}}
+
 def upload_to_servers(files, version):  # {{{
-    for server, rdir in {'files':'/usr/share/nginx/html'}.iteritems():
+    base = '/srv/download/'
+    dest = os.path.join(base, version)
+    if not os.path.exists(dest):
+        os.mkdir(dest)
+    for src in files:
+        shutil.copyfile(src, os.path.join(dest, os.path.basename(src)))
+    cwd = os.getcwd()
+    try:
+        generate_index()
+    finally:
+        os.chdir(cwd)
+
+    for server, rdir in {'files':'/srv/download/'}.iteritems():
         print('Uploading to server:', server)
         server = '%s.calibre-ebook.com' % server
-        rdir = '%s/%s/' % (rdir, version)
+        # Copy the generated index files
+        print ('Copying generated index')
+        check_call(['rsync', '-hza', '-e', 'ssh -x', '--include', '*.html',
+                    '--filter', '-! */', base, 'root@%s:%s' % (server, rdir)])
+        # Copy the release files
+        rdir = '%s%s/' % (rdir, version)
         for x in files:
             start = time.time()
             print ('Uploading', x)
@@ -400,6 +503,7 @@ def upload_to_servers(files, version):  # {{{
                 else:
                     break
             print ('Uploaded in', int(time.time() - start), 'seconds\n\n')
+
 # }}}
 
 def upload_to_dbs(files, version):  # {{{
@@ -530,3 +634,4 @@ if __name__ == '__main__':
 # }}}
 
 
+

setup/publish.py

@@ -113,6 +113,6 @@ class TagRelease(Command):
 
     def run(self, opts):
         self.info('Tagging release')
-        subprocess.check_call('git tag -a {0} -m "version-{0}"'.format(__version__).split())
+        subprocess.check_call('git tag -a v{0} -m "version-{0}"'.format(__version__).split())
-        subprocess.check_call('git push origin {0}'.format(__version__).split())
+        subprocess.check_call('git push origin v{0}'.format(__version__).split())
 

setup/translations.py

@@ -21,7 +21,12 @@ def qt_sources():
 class POT(Command):  # {{{
 
     description = 'Update the .pot translation template and upload it'
-    LP_BASE = os.path.join(os.path.dirname(os.path.dirname(Command.SRC)), 'calibre-translations')
+    LP_BASE = os.path.join(os.path.dirname(Command.SRC))
+    if not os.path.exists(os.path.join(LP_BASE, 'setup', 'iso_639')):
+        # We are in a git checkout, translations are assumed to be in a
+        # directory called calibre-translations at the same level as the
+        # calibre directory.
+        LP_BASE = os.path.join(os.path.dirname(os.path.dirname(Command.SRC)), 'calibre-translations')
     LP_SRC = os.path.join(LP_BASE, 'src')
     LP_PATH = os.path.join(LP_SRC, os.path.join(__appname__, 'translations'))
     LP_ISO_PATH = os.path.join(LP_BASE, 'setup', 'iso_639')
@@ -317,21 +322,24 @@ class GetTranslations(Translations):  # {{{
 
 class ISO639(Command):  # {{{
 
-    description = 'Compile translations for ISO 639 codes'
+    description = 'Compile language code maps for performance'
     DEST = os.path.join(os.path.dirname(POT.SRC), 'resources', 'localization',
             'iso639.pickle')
 
     def run(self, opts):
-        src = POT.LP_ISO_PATH
+        src = self.j(self.d(self.SRC), 'setup', 'iso_639_3.xml')
         if not os.path.exists(src):
             raise Exception(src + ' does not exist')
         dest = self.DEST
+        base = self.d(dest)
+        if not os.path.exists(base):
+            os.makedirs(base)
         if not self.newer(dest, [src, __file__]):
             self.info('Pickled code is up to date')
             return
         self.info('Pickling ISO-639 codes to', dest)
         from lxml import etree
-        root = etree.fromstring(open(self.j(src, 'iso_639_3.xml'), 'rb').read())
+        root = etree.fromstring(open(src, 'rb').read())
         by_2 = {}
         by_3b = {}
         by_3t = {}
@@ -345,7 +353,7 @@ class ISO639(Command):  # {{{
             threet = x.get('id')
             threeb = x.get('part2_code', None)
             if threeb is None:
-                # Only recognize langauges in ISO-639-2
+                # Only recognize languages in ISO-639-2
                 continue
             name = x.get('name')
 

setup/upload.py

@@ -19,10 +19,9 @@ from setup import Command, __version__, installer_name, __appname__
 PREFIX = "/var/www/calibre-ebook.com"
 DOWNLOADS = PREFIX+"/htdocs/downloads"
 BETAS = DOWNLOADS +'/betas'
-USER_MANUAL = '/var/www/localhost/htdocs/'
 HTML2LRF = "calibre/ebooks/lrf/html/demo"
 TXT2LRF  = "src/calibre/ebooks/lrf/txt/demo"
-STAGING_HOST = '67.207.135.179'
+STAGING_HOST = 'download.calibre-ebook.com'
 STAGING_USER = 'root'
 STAGING_DIR = '/root/staging'
 
@@ -135,19 +134,26 @@ class UploadInstallers(Command):  # {{{
         available = set(glob.glob('dist/*'))
         files = {x:installer_description(x) for x in
                 all_possible.intersection(available)}
+        sizes = {os.path.basename(x):os.path.getsize(x) for x in files}
+        self.record_sizes(sizes)
         tdir = mkdtemp()
         backup = os.path.join('/mnt/external/calibre/%s' % __version__)
         if not os.path.exists(backup):
             os.mkdir(backup)
         try:
             self.upload_to_staging(tdir, backup, files)
-            self.upload_to_sourceforge()
             self.upload_to_calibre()
+            self.upload_to_sourceforge()
             self.upload_to_dbs()
             # self.upload_to_google(opts.replace)
         finally:
             shutil.rmtree(tdir, ignore_errors=True)
 
+    def record_sizes(self, sizes):
+        print ('\nRecording dist sizes')
+        args = ['%s:%s:%s' % (__version__, fname, size) for fname, size in sizes.iteritems()]
+        check_call(['ssh', 'divok', 'dist_sizes'] + args)
+
     def upload_to_staging(self, tdir, backup, files):
         os.mkdir(tdir+'/dist')
         hosting = os.path.join(os.path.dirname(os.path.abspath(__file__)),
@@ -155,9 +161,9 @@ class UploadInstallers(Command):  # {{{
         shutil.copyfile(hosting, os.path.join(tdir, 'hosting.py'))
 
         for f in files:
-            for x in (tdir, backup):
+            for x in (tdir+'/dist', backup):
-                dest = os.path.join(x, f)
+                dest = os.path.join(x, os.path.basename(f))
-                shutil.copyfile(f, dest)
+                shutil.copy2(f, x)
                 os.chmod(dest, stat.S_IREAD|stat.S_IWRITE|stat.S_IRGRP|stat.S_IROTH)
 
         with open(os.path.join(tdir, 'fmap'), 'wb') as fo:
@@ -219,9 +225,9 @@ class UploadUserManual(Command):  # {{{
         for x in glob.glob(self.j(path, '*')):
             self.build_plugin_example(x)
 
-        check_call(' '.join(['rsync', '-z', '-r', '--progress',
+        for host in ('download', 'files'):
-            'manual/.build/html/',
+            check_call(' '.join(['rsync', '-z', '-r', '--progress',
-                    'bugs:%s'%USER_MANUAL]), shell=True)
+                'manual/.build/html/', '%s:/srv/manual/' % host]), shell=True)
 # }}}
 
 class UploadDemo(Command):  # {{{
@@ -249,8 +255,6 @@ class UploadToServer(Command):  # {{{
     description = 'Upload miscellaneous data to calibre server'
 
     def run(self, opts):
-        check_call('ssh divok rm -f %s/calibre-\*.tar.xz'%DOWNLOADS, shell=True)
-        # check_call('scp dist/calibre-*.tar.xz divok:%s/'%DOWNLOADS, shell=True)
         check_call('gpg --armor --detach-sign dist/calibre-*.tar.xz',
                 shell=True)
         check_call('scp dist/calibre-*.tar.xz.asc divok:%s/signatures/'%DOWNLOADS,

src/calibre/__init__.py

@@ -310,9 +310,9 @@ def get_parsed_proxy(typ='http', debug=True):
     proxy = proxies.get(typ, None)
     if proxy:
         pattern = re.compile((
-            '(?:ptype://)?' \
+            '(?:ptype://)?'
-            '(?:(?P<user>\w+):(?P<pass>.*)@)?' \
+            '(?:(?P<user>\w+):(?P<pass>.*)@)?'
-            '(?P<host>[\w\-\.]+)' \
+            '(?P<host>[\w\-\.]+)'
             '(?::(?P<port>\d+))?').replace('ptype', typ)
         )
 
@@ -535,7 +535,7 @@ def entity_to_unicode(match, exceptions=[], encoding='cp1252',
     ent = match.group(1)
     if ent in exceptions:
         return '&'+ent+';'
-    if ent in {'apos', 'squot'}: # squot is generated by some broken CMS software
+    if ent in {'apos', 'squot'}:  # squot is generated by some broken CMS software
         return check("'")
     if ent == 'hellips':
         ent = 'hellip'
@@ -565,7 +565,7 @@ def entity_to_unicode(match, exceptions=[], encoding='cp1252',
         return '&'+ent+';'
 
 _ent_pat = re.compile(r'&(\S+?);')
-xml_entity_to_unicode = partial(entity_to_unicode, result_exceptions = {
+xml_entity_to_unicode = partial(entity_to_unicode, result_exceptions={
     '"' : '&quot;',
     "'" : '&apos;',
     '<' : '&lt;',
@@ -670,8 +670,8 @@ def human_readable(size, sep=' '):
     """ Convert a size in bytes into a human readable form """
     divisor, suffix = 1, "B"
     for i, candidate in enumerate(('B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB')):
-        if size < 1024**(i+1):
+        if size < (1 << ((i + 1) * 10)):
-            divisor, suffix = 1024**(i), candidate
+            divisor, suffix = (1 << (i * 10)), candidate
             break
     size = str(float(size)/divisor)
     if size.find(".") > -1:

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, 9, 36)
+numeric_version = (0, 9, 37)
 __version__   = u'.'.join(map(unicode, numeric_version))
 __author__    = u"Kovid Goyal <kovid@kovidgoyal.net>"
 

src/calibre/customize/conversion.py

@@ -77,7 +77,7 @@ class OptionRecommendation(object):
                                                     self.option.choices:
             raise ValueError('OpRec: %s: Recommended value not in choices'%
                              self.option.name)
-        if not (isinstance(self.recommended_value, (int, float, str, unicode))\
+        if not (isinstance(self.recommended_value, (int, float, str, unicode))
             or self.recommended_value is None):
             raise ValueError('OpRec: %s:'%self.option.name +
                              repr(self.recommended_value) +
@@ -139,8 +139,10 @@ class InputFormatPlugin(Plugin):
     file_types     = set([])
 
     #: If True, this input plugin generates a collection of images,
-    #: one per HTML file. You can obtain access to the images via
+    #: one per HTML file. This can be set dynamically, in the convert method
-    #: convenience method, :meth:`get_image_collection`.
+    #: if the input files can be both image collections and non-image collections.
+    #: If you set this to True, you must implement the get_images() method that returns
+    #: a list of images.
     is_image_collection = False
 
     #: Number of CPU cores used by this plugin
@@ -238,7 +240,6 @@ class InputFormatPlugin(Plugin):
             ret = self.convert(stream, options, file_ext,
                                log, accelerators)
 
-
         return ret
 
     def postprocess_book(self, oeb, opts, log):
@@ -313,7 +314,6 @@ class OutputFormatPlugin(Plugin):
         Plugin.__init__(self, *args)
         self.report_progress = DummyReporter()
 
-
     def convert(self, oeb_book, output, input_plugin, opts, log):
         '''
         Render the contents of `oeb_book` (which is an instance of
@@ -363,3 +363,4 @@ class OutputFormatPlugin(Plugin):
 
 
 
+

src/calibre/devices/android/driver.py

@@ -96,7 +96,7 @@ class ANDROID(USBMS):
 
             # Google
             0x18d1 : {
-                0x0001 : [0x0223, 0x230, 0x9999],
+                0x0001 : [0x0222, 0x0223, 0x230, 0x9999],
                 0x0002 : [0x9999],
                 0x0003 : [0x0230, 0x9999],
                 0x4e11 : [0x0100, 0x226, 0x227],
@@ -219,7 +219,7 @@ class ANDROID(USBMS):
             'POCKET', 'ONDA_MID', 'ZENITHIN', 'INGENIC', 'PMID701C', 'PD',
             'PMP5097C', 'MASS', 'NOVO7', 'ZEKI', 'COBY', 'SXZ', 'USB_2.0',
             'COBY_MID', 'VS', 'AINOL', 'TOPWISE', 'PAD703', 'NEXT8D12',
-            'MEDIATEK', 'KEENHI', 'TECLAST', 'SURFTAB', 'XENTA',]
+            'MEDIATEK', 'KEENHI', 'TECLAST', 'SURFTAB', 'XENTA', 'OBREEY_S']
     WINDOWS_MAIN_MEM = ['ANDROID_PHONE', 'A855', 'A853', 'A953', 'INC.NEXUS_ONE',
             '__UMS_COMPOSITE', '_MB200', 'MASS_STORAGE', '_-_CARD', 'SGH-I897',
             'GT-I9000', 'FILE-STOR_GADGET', 'SGH-T959_CARD', 'SGH-T959', 'SAMSUNG_ANDROID',
@@ -241,7 +241,7 @@ class ANDROID(USBMS):
             'S5830I_CARD', 'MID7042', 'LINK-CREATE', '7035', 'VIEWPAD_7E',
             'NOVO7', 'MB526', '_USB#WYK7MSF8KE', 'TABLET_PC', 'F', 'MT65XX_MS',
             'ICS', 'E400', '__FILE-STOR_GADG', 'ST80208-1', 'GT-S5660M_CARD', 'XT894', '_USB',
-            'PROD_TAB13-201',
+            'PROD_TAB13-201', 'URFPAD2',
     ]
     WINDOWS_CARD_A_MEM = ['ANDROID_PHONE', 'GT-I9000_CARD', 'SGH-I897',
             'FILE-STOR_GADGET', 'SGH-T959_CARD', 'SGH-T959', 'SAMSUNG_ANDROID', 'GT-P1000_CARD',
@@ -254,7 +254,7 @@ class ANDROID(USBMS):
             'UMS_COMPOSITE', 'PRO', '.KOBO_VOX', 'SGH-T989_CARD', 'SGH-I727',
             'USB_FLASH_DRIVER', 'ANDROID', 'MID7042', '7035', 'VIEWPAD_7E',
             'NOVO7', 'ADVANCED', 'TABLET_PC', 'F', 'E400_SD_CARD', 'ST80208-1', 'XT894',
-            '_USB', 'PROD_TAB13-201',
+            '_USB', 'PROD_TAB13-201', 'URFPAD2'
     ]
 
     OSX_MAIN_MEM = 'Android Device Main Memory'

src/calibre/devices/idevice/libimobiledevice.py

@@ -418,6 +418,14 @@ class libiMobileDevice():
         if False:
             self._idevice_set_debug_level(DEBUG)
 
+    def mkdir(self, path):
+        '''
+        Mimic mkdir(), creating a directory at path. Does not create
+        intermediate folders
+        '''
+        self._log_location("'%s'" % path)
+        return self._afc_make_directory(path)
+
     def mount_ios_app(self, app_name=None, app_id=None):
         '''
         Convenience method to get iDevice ready to talk to app_name or app_id
@@ -1007,6 +1015,27 @@ class libiMobileDevice():
                     self.log(" %s: %s" % (key, file_stats[key]))
         return file_stats
 
+    def _afc_make_directory(self, path):
+        '''
+        Creates a directory on the device. Does not create intermediate dirs.
+
+        Args:
+         client: (AFC_CLIENT_T) The client to use to make a directory
+         dir:    (const char *) The directory's fully-qualified path
+
+        Result:
+         error:  AFC_E_SUCCESS on success or an AFC_E_* error value
+        '''
+        self._log_location("%s" % repr(path))
+
+        error = self.lib.afc_make_directory(byref(self.afc),
+                                            str(path)) & 0xFFFF
+        if error:
+            if self.verbose:
+                self.log(" ERROR: %s" % self._afc_error(error))
+
+        return error
+
     def _afc_read_directory(self, directory=''):
         '''
         Gets a directory listing of the directory requested

src/calibre/devices/misc.py

@@ -227,16 +227,17 @@ class TREKSTOR(USBMS):
     VENDOR_ID   = [0x1e68]
     PRODUCT_ID  = [0x0041, 0x0042, 0x0052, 0x004e, 0x0056,
             0x0067,  # This is for the Pyrus Mini
+            0x006f,  # This is for the Pyrus Maxi
             0x003e,  # This is for the EBOOK_PLAYER_5M https://bugs.launchpad.net/bugs/792091
             0x5cL,  # This is for the 4ink http://www.mobileread.com/forums/showthread.php?t=191318
             ]
-    BCD         = [0x0002, 0x100]
+    BCD         = [0x0002, 0x100, 0x0222]
 
     EBOOK_DIR_MAIN = 'Ebooks'
 
     VENDOR_NAME = 'TREKSTOR'
     WINDOWS_MAIN_MEM = WINDOWS_CARD_A_MEM = ['EBOOK_PLAYER_7',
-            'EBOOK_PLAYER_5M', 'EBOOK-READER_3.0', 'EREADER_PYRUS', 'PYRUS_MINI']
+            'EBOOK_PLAYER_5M', 'EBOOK-READER_3.0', 'EREADER_PYRUS', 'PYRUS_MINI', 'PYRUS_MAXI']
     SUPPORTS_SUB_DIRS = True
     SUPPORTS_SUB_DIRS_DEFAULT = False
 

src/calibre/ebooks/conversion/cli.py

@@ -27,7 +27,7 @@ specified as the first two arguments to the command.
 The output ebook format is guessed from the file extension of \
 output_file. output_file can also be of the special format .EXT where \
 EXT is the output file extension. In this case, the name of the output \
-file is derived the name of the input file. Note that the filenames must \
+file is derived from the name of the input file. Note that the filenames must \
 not start with a hyphen. Finally, if output_file has no extension, then \
 it is treated as a directory and an "open ebook" (OEB) consisting of HTML \
 files is written to that directory. These files are the files that would \
@@ -94,6 +94,8 @@ def option_recommendation_to_cli_option(add_option, rec):
     if opt.long_switch == 'verbose':
         attrs['action'] = 'count'
         attrs.pop('type', '')
+    if opt.name == 'read_metadata_from_opf':
+        switches.append('--from-opf')
     if opt.name in DEFAULT_TRUE_OPTIONS and rec.recommended_value is True:
         switches = ['--disable-'+opt.long_switch]
     add_option(Option(*switches, **attrs))
@@ -136,7 +138,7 @@ def add_pipeline_options(parser, plumber):
                   [
                       'base_font_size', 'disable_font_rescaling',
                       'font_size_mapping', 'embed_font_family',
-                      'subset_embedded_fonts',
+                      'subset_embedded_fonts', 'embed_all_fonts',
                       'line_height', 'minimum_line_height',
                       'linearize_tables',
                       'extra_css', 'filter_css',
@@ -190,7 +192,7 @@ def add_pipeline_options(parser, plumber):
                   ),
 
               'METADATA' : (_('Options to set metadata in the output'),
-                            plumber.metadata_option_names,
+                            plumber.metadata_option_names + ['read_metadata_from_opf'],
                             ),
               'DEBUG': (_('Options to help with debugging the conversion'),
                         [
@@ -320,7 +322,7 @@ def main(args=sys.argv):
         opts.search_replace = read_sr_patterns(opts.search_replace, log)
 
     recommendations = [(n.dest, getattr(opts, n.dest),
-                        OptionRecommendation.HIGH) \
+                        OptionRecommendation.HIGH)
                                         for n in parser.options_iter()
                                         if n.dest]
     plumber.merge_ui_recommendations(recommendations)
@@ -342,3 +344,4 @@ def main(args=sys.argv):
 
 if __name__ == '__main__':
     sys.exit(main())
+

src/calibre/ebooks/conversion/plumber.py

@@ -134,8 +134,7 @@ OptionRecommendation(name='output_profile',
             help=_('Specify the output profile. The output profile '
                    'tells the conversion system how to optimize the '
                    'created document for the specified device. In some cases, '
-                   'an output profile is required to produce documents that '
+                   'an output profile can be used to optimize the output for a particular device, but this is rarely necessary. '
-                   'will work on a device. For example EPUB on the SONY reader. '
                    'Choices are:') +
                            ', '.join([x.short_name for x in output_profiles()])
         ),
@@ -205,6 +204,17 @@ OptionRecommendation(name='embed_font_family',
             'with some output formats, principally EPUB and AZW3.')
         ),
 
+OptionRecommendation(name='embed_all_fonts',
+        recommended_value=False, level=OptionRecommendation.LOW,
+        help=_(
+            'Embed every font that is referenced in the input document '
+            'but not already embedded. This will search your system for the '
+            'fonts, and if found, they will be embedded. Embedding will only work '
+            'if the format you are converting to supports embedded fonts, such as '
+            'EPUB, AZW3 or PDF. Please ensure that you have the proper license for embedding '
+            'the fonts used in this document.'
+        )),
+
 OptionRecommendation(name='subset_embedded_fonts',
         recommended_value=False, level=OptionRecommendation.LOW,
         help=_(
@@ -965,6 +975,9 @@ OptionRecommendation(name='search_replace',
         if self.for_regex_wizard and hasattr(self.opts, 'no_process'):
             self.opts.no_process = True
         self.flush()
+        if self.opts.embed_all_fonts or self.opts.embed_font_family:
+            # Start the threaded font scanner now, for performance
+            from calibre.utils.fonts.scanner import font_scanner  # noqa
         import cssutils, logging
         cssutils.log.setLevel(logging.WARN)
         get_types_map()  # Ensure the mimetypes module is intialized
@@ -1129,6 +1142,10 @@ OptionRecommendation(name='search_replace',
         RemoveFakeMargins()(self.oeb, self.log, self.opts)
         RemoveAdobeMargins()(self.oeb, self.log, self.opts)
 
+        if self.opts.embed_all_fonts:
+            from calibre.ebooks.oeb.transforms.embed_fonts import EmbedFonts
+            EmbedFonts()(self.oeb, self.log, self.opts)
+
         if self.opts.subset_embedded_fonts and self.output_plugin.file_type != 'pdf':
             from calibre.ebooks.oeb.transforms.subset import SubsetFonts
             SubsetFonts()(self.oeb, self.log, self.opts)

src/calibre/ebooks/docx/cleanup.py

@@ -8,6 +8,8 @@ __copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
 
 import os
 
+from calibre.ebooks.docx.names import ancestor
+
 def mergeable(previous, current):
     if previous.tail or current.tail:
         return False
@@ -97,6 +99,16 @@ def before_count(root, tag, limit=10):
             return limit
 
 def cleanup_markup(log, root, styles, dest_dir, detect_cover):
+    # Move <hr>s outside paragraphs, if possible.
+    for hr in root.xpath('//span/hr'):
+        p = ancestor(hr, 'p')
+        descendants = tuple(p.iterdescendants())
+        if descendants[-1] is hr:
+            parent = p.getparent()
+            idx = parent.index(p)
+            parent.insert(idx+1, hr)
+            hr.tail = '\n\t'
+
     # Merge consecutive spans that have the same styling
     current_run = []
     for span in root.xpath('//span'):
@@ -165,3 +177,4 @@ def cleanup_markup(log, root, styles, dest_dir, detect_cover):
                     return path
 
 
+

src/calibre/ebooks/docx/container.py

@@ -183,7 +183,7 @@ class DOCX(object):
             root = fromstring(raw)
             for item in root.xpath('//*[local-name()="Relationships"]/*[local-name()="Relationship" and @Type and @Target]'):
                 target = item.get('Target')
-                if item.get('TargetMode', None) != 'External':
+                if item.get('TargetMode', None) != 'External' and not target.startswith('#'):
                     target = '/'.join((base, target.lstrip('/')))
                 typ = item.get('Type')
                 Id = item.get('Id')

src/calibre/ebooks/docx/fields.py

@@ -0,0 +1,108 @@
+#!/usr/bin/env python
+# vim:fileencoding=utf-8
+from __future__ import (unicode_literals, division, absolute_import,
+                        print_function)
+
+__license__ = 'GPL v3'
+__copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
+
+import re
+
+from calibre.ebooks.docx.names import XPath, get
+
+class Field(object):
+
+    def __init__(self, start):
+        self.start = start
+        self.end = None
+        self.contents = []
+        self.instructions = []
+
+    def add_instr(self, elem):
+        raw = elem.text
+        if not raw:
+            return
+        name, rest = raw.strip().partition(' ')[0::2]
+        self.instructions.append((name, rest.strip()))
+
+WORD, FLAG = 0, 1
+scanner = re.Scanner([
+    (r'\\\S{1}', lambda s, t: (t, FLAG)),  # A flag of the form \x
+    (r'"[^"]*"', lambda s, t: (t[1:-1], WORD)),  # Quoted word
+    (r'[^\s\\"]\S*', lambda s, t: (t, WORD)),  # A non-quoted word, must not start with a backslash or a space or a quote
+    (r'\s+', None),
+], flags=re.DOTALL)
+
+
+def parse_hyperlink(raw, log):
+    ans = {}
+    last_option = None
+    raw = raw.replace('\\\\', '\x01').replace('\\"', '\x02')
+    for token, token_type in scanner.scan(raw)[0]:
+        token = token.replace('\x01', '\\').replace('\x02', '"')
+        if not ans:
+            if token_type is not WORD:
+                log('Invalid hyperlink, first token is not a URL (%s)' % raw)
+                return ans
+            ans['url'] = token
+        if token_type is FLAG:
+            last_option = {'l':'anchor', 'm':'image-map', 'n':'target', 'o':'title', 't':'target'}.get(token[1], None)
+            if last_option is not None:
+                ans[last_option] = None
+        elif token_type is WORD:
+            if last_option is not None:
+                ans[last_option] = token
+    return ans
+
+
+class Fields(object):
+
+    def __init__(self):
+        self.fields = []
+
+    def __call__(self, doc, log):
+        stack = []
+        for elem in XPath(
+            '//*[name()="w:p" or name()="w:r" or name()="w:instrText" or (name()="w:fldChar" and (@w:fldCharType="begin" or @w:fldCharType="end"))]')(doc):
+            if elem.tag.endswith('}fldChar'):
+                typ = get(elem, 'w:fldCharType')
+                if typ == 'begin':
+                    stack.append(Field(elem))
+                    self.fields.append(stack[-1])
+                else:
+                    try:
+                        stack.pop().end = elem
+                    except IndexError:
+                        pass
+            elif elem.tag.endswith('}instrText'):
+                if stack:
+                    stack[-1].add_instr(elem)
+            else:
+                if stack:
+                    stack[-1].contents.append(elem)
+
+        # Parse hyperlink fields
+        self.hyperlink_fields = []
+        for field in self.fields:
+            if len(field.instructions) == 1 and field.instructions[0][0] == 'HYPERLINK':
+                hl = parse_hyperlink(field.instructions[0][1], log)
+                if hl:
+                    if 'target' in hl and hl['target'] is None:
+                        hl['target'] = '_blank'
+                    all_runs = []
+                    current_runs = []
+                    # We only handle spans in a single paragraph
+                    # being wrapped in <a>
+                    for x in field.contents:
+                        if x.tag.endswith('}p'):
+                            if current_runs:
+                                all_runs.append(current_runs)
+                            current_runs = []
+                        elif x.tag.endswith('}r'):
+                            current_runs.append(x)
+                    if current_runs:
+                        all_runs.append(current_runs)
+                    for runs in all_runs:
+                        self.hyperlink_fields.append((hl, runs))
+
+

src/calibre/ebooks/docx/images.py

@@ -8,7 +8,7 @@ __copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
 
 import os
 
-from lxml.html.builder import IMG
+from lxml.html.builder import IMG, HR
 
 from calibre.ebooks.docx.names import XPath, get, barename
 from calibre.utils.filenames import ascii_filename
@@ -96,6 +96,7 @@ class Images(object):
         self.used = {}
         self.names = set()
         self.all_images = set()
+        self.links = []
 
     def __call__(self, relationships_by_id):
         self.rid_map = relationships_by_id
@@ -125,8 +126,18 @@ class Images(object):
         self.all_images.add('images/' + name)
         return name
 
-    def pic_to_img(self, pic, alt=None):
+    def pic_to_img(self, pic, alt, parent):
         name = None
+        link = None
+        for hl in XPath('descendant::a:hlinkClick[@r:id]')(parent):
+            link = {'id':get(hl, 'r:id')}
+            tgt = hl.get('tgtFrame', None)
+            if tgt:
+                link['target'] = tgt
+            title = hl.get('tooltip', None)
+            if title:
+                link['title'] = title
+
         for pr in XPath('descendant::pic:cNvPr')(pic):
             name = pr.get('name', None)
             if name:
@@ -138,6 +149,8 @@ class Images(object):
                     src = self.generate_filename(rid, name)
                     img = IMG(src='images/%s' % src)
                     img.set('alt', alt or 'Image')
+                    if link is not None:
+                        self.links.append((img, link))
                     return img
 
     def drawing_to_html(self, drawing, page):
@@ -145,7 +158,7 @@ class Images(object):
         for inline in XPath('./wp:inline')(drawing):
             style, alt = get_image_properties(inline)
             for pic in XPath('descendant::pic:pic')(inline):
-                ans = self.pic_to_img(pic, alt)
+                ans = self.pic_to_img(pic, alt, inline)
                 if ans is not None:
                     if style:
                         ans.set('style', '; '.join('%s: %s' % (k, v) for k, v in style.iteritems()))
@@ -156,13 +169,33 @@ class Images(object):
             style, alt = get_image_properties(anchor)
             self.get_float_properties(anchor, style, page)
             for pic in XPath('descendant::pic:pic')(anchor):
-                ans = self.pic_to_img(pic, alt)
+                ans = self.pic_to_img(pic, alt, anchor)
                 if ans is not None:
                     if style:
                         ans.set('style', '; '.join('%s: %s' % (k, v) for k, v in style.iteritems()))
                     yield ans
 
     def pict_to_html(self, pict, page):
+        # First see if we have an <hr>
+        is_hr = len(pict) == 1 and get(pict[0], 'o:hr') in {'t', 'true'}
+        if is_hr:
+            style = {}
+            hr = HR()
+            try:
+                pct = float(get(pict[0], 'o:hrpct'))
+            except (ValueError, TypeError, AttributeError):
+                pass
+            else:
+                if pct > 0:
+                    style['width'] = '%.3g%%' % pct
+            align = get(pict[0], 'o:hralign', 'center')
+            if align in {'left', 'right'}:
+                style['margin-left'] = '0' if align == 'left' else 'auto'
+                style['margin-right'] = 'auto' if align == 'left' else '0'
+            if style:
+                hr.set('style', '; '.join(('%s:%s' % (k, v) for k, v in style.iteritems())))
+            yield hr
+
         for imagedata in XPath('descendant::v:imagedata[@r:id]')(pict):
             rid = get(imagedata, 'r:id')
             if rid in self.rid_map:

src/calibre/ebooks/docx/styles.py

@@ -403,6 +403,11 @@ class Styles(object):
                         ps.margin_top = 0
             last_para = p
 
+    def apply_section_page_breaks(self, paras):
+        for p in paras:
+            ps = self.resolve_paragraph(p)
+            ps.pageBreakBefore = True
+
     def register(self, css, prefix):
         h = hash(frozenset(css.iteritems()))
         ans, _ = self.classes.get(h, (None, None))

src/calibre/ebooks/docx/to_html.py

@@ -26,6 +26,7 @@ from calibre.ebooks.docx.footnotes import Footnotes
 from calibre.ebooks.docx.cleanup import cleanup_markup
 from calibre.ebooks.docx.theme import Theme
 from calibre.ebooks.docx.toc import create_toc
+from calibre.ebooks.docx.fields import Fields
 from calibre.ebooks.metadata.opf2 import OPFCreator
 from calibre.utils.localization import canonicalize_lang, lang_as_iso639_1
 
@@ -52,6 +53,7 @@ class Convert(object):
         self.body = BODY()
         self.theme = Theme()
         self.tables = Tables()
+        self.fields = Fields()
         self.styles = Styles(self.tables)
         self.images = Images()
         self.object_map = OrderedDict()
@@ -79,6 +81,7 @@ class Convert(object):
     def __call__(self):
         doc = self.docx.document
         relationships_by_id, relationships_by_type = self.docx.document_relationships
+        self.fields(doc, self.log)
         self.read_styles(relationships_by_type)
         self.images(relationships_by_id)
         self.layers = OrderedDict()
@@ -96,7 +99,11 @@ class Convert(object):
                 p = self.convert_p(wp)
                 self.body.append(p)
                 paras.append(wp)
+        self.read_block_anchors(doc)
         self.styles.apply_contextual_spacing(paras)
+        # Apply page breaks at the start of every section, except the first
+        # section (since that will be the start of the file)
+        self.styles.apply_section_page_breaks(self.section_starts[1:])
 
         notes_header = None
         if self.footnotes.has_notes:
@@ -177,6 +184,7 @@ class Convert(object):
     def read_page_properties(self, doc):
         current = []
         self.page_map = OrderedDict()
+        self.section_starts = []
 
         for p in descendants(doc, 'w:p', 'w:tbl'):
             if p.tag.endswith('}tbl'):
@@ -186,8 +194,10 @@ class Convert(object):
             sect = tuple(descendants(p, 'w:sectPr'))
             if sect:
                 pr = PageProperties(sect)
-                for x in current + [p]:
+                paras = current + [p]
+                for x in paras:
                     self.page_map[x] = pr
+                self.section_starts.append(paras[0])
                 current = []
             else:
                 current.append(p)
@@ -287,6 +297,22 @@ class Convert(object):
             opf.render(of, ncx, 'toc.ncx')
         return os.path.join(self.dest_dir, 'metadata.opf')
 
+    def read_block_anchors(self, doc):
+        doc_anchors = frozenset(XPath('./w:body/w:bookmarkStart[@w:name]')(doc))
+        if doc_anchors:
+            current_bm = None
+            rmap = {v:k for k, v in self.object_map.iteritems()}
+            for p in descendants(doc, 'w:p', 'w:bookmarkStart[@w:name]'):
+                if p.tag.endswith('}p'):
+                    if current_bm and p in rmap:
+                        para = rmap[p]
+                        if 'id' not in para.attrib:
+                            para.set('id', generate_anchor(current_bm, frozenset(self.anchor_map.itervalues())))
+                        self.anchor_map[current_bm] = para.get('id')
+                        current_bm = None
+                elif p in doc_anchors:
+                    current_bm = get(p, 'w:name')
+
     def convert_p(self, p):
         dest = P()
         self.object_map[dest] = p
@@ -316,7 +342,13 @@ class Convert(object):
             elif x.tag.endswith('}bookmarkStart'):
                 anchor = get(x, 'w:name')
                 if anchor and anchor not in self.anchor_map:
+                    old_anchor = current_anchor
                     self.anchor_map[anchor] = current_anchor = generate_anchor(anchor, frozenset(self.anchor_map.itervalues()))
+                    if old_anchor is not None:
+                        # The previous anchor was not applied to any element
+                        for a, t in tuple(self.anchor_map.iteritems()):
+                            if t == old_anchor:
+                                self.anchor_map[a] = current_anchor
             elif x.tag.endswith('}hyperlink'):
                 current_hyperlink = x
 
@@ -396,6 +428,46 @@ class Convert(object):
             # hrefs that point nowhere give epubcheck a hernia. The element
             # should be styled explicitly by Word anyway.
             # span.set('href', '#')
+        rmap = {v:k for k, v in self.object_map.iteritems()}
+        for hyperlink, runs in self.fields.hyperlink_fields:
+            spans = [rmap[r] for r in runs if r in rmap]
+            if not spans:
+                continue
+            if len(spans) > 1:
+                span = self.wrap_elems(spans, SPAN())
+            span.tag = 'a'
+            tgt = hyperlink.get('target', None)
+            if tgt:
+                span.set('target', tgt)
+            tt = hyperlink.get('title', None)
+            if tt:
+                span.set('title', tt)
+            url = hyperlink['url']
+            if url in self.anchor_map:
+                span.set('href', '#' + self.anchor_map[url])
+                continue
+            span.set('href', url)
+
+        for img, link in self.images.links:
+            parent = img.getparent()
+            idx = parent.index(img)
+            a = A(img)
+            a.tail, img.tail = img.tail, None
+            parent.insert(idx, a)
+            tgt = link.get('target', None)
+            if tgt:
+                a.set('target', tgt)
+            tt = link.get('title', None)
+            if tt:
+                a.set('title', tt)
+            rid = link['id']
+            if rid in relationships_by_id:
+                dest = relationships_by_id[rid]
+                if dest.startswith('#'):
+                    if dest[1:] in self.anchor_map:
+                        a.set('href', '#' + self.anchor_map[dest[1:]])
+                else:
+                    a.set('href', dest)
 
     def convert_run(self, run):
         ans = SPAN()

src/calibre/ebooks/metadata/opf2.py

@@ -1047,6 +1047,14 @@ class OPF(object):  # {{{
                     if raw:
                         return raw.rpartition(':')[-1]
 
+    @property
+    def page_progression_direction(self):
+        spine = self.XPath('descendant::*[re:match(name(), "spine", "i")][1]')(self.root)
+        if spine:
+            for k, v in spine[0].attrib.iteritems():
+                if k == 'page-progression-direction' or k.endswith('}page-progression-direction'):
+                    return v
+
     def guess_cover(self):
         '''
         Try to guess a cover. Needed for some old/badly formed OPF files.
@@ -1185,6 +1193,7 @@ class OPFCreator(Metadata):
         '''
         Metadata.__init__(self, title='', other=other)
         self.base_path = os.path.abspath(base_path)
+        self.page_progression_direction = None
         if self.application_id is None:
             self.application_id = str(uuid.uuid4())
         if not isinstance(self.toc, TOC):
@@ -1356,6 +1365,8 @@ class OPFCreator(Metadata):
         spine = E.spine()
         if self.toc is not None:
             spine.set('toc', 'ncx')
+        if self.page_progression_direction is not None:
+            spine.set('page-progression-direction', self.page_progression_direction)
         if self.spine is not None:
             for ref in self.spine:
                 if ref.id is not None:

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

@@ -20,7 +20,7 @@ from calibre.ebooks.mobi.reader.ncx import read_ncx, build_toc
 from calibre.ebooks.mobi.reader.markup import expand_mobi8_markup
 from calibre.ebooks.metadata.opf2 import Guide, OPFCreator
 from calibre.ebooks.metadata.toc import TOC
-from calibre.ebooks.mobi.utils import read_font_record
+from calibre.ebooks.mobi.utils import read_font_record, read_resc_record
 from calibre.ebooks.oeb.parse_utils import parse_html
 from calibre.ebooks.oeb.base import XPath, XHTML, xml2text
 from calibre.utils.imghdr import what
@@ -65,6 +65,7 @@ class Mobi8Reader(object):
         self.mobi6_reader, self.log = mobi6_reader, log
         self.header = mobi6_reader.book_header
         self.encrypted_fonts = []
+        self.resc_data = {}
 
     def __call__(self):
         self.mobi6_reader.check_for_drm()
@@ -389,9 +390,11 @@ class Mobi8Reader(object):
             data = sec[0]
             typ = data[:4]
             href = None
-            if typ in {b'FLIS', b'FCIS', b'SRCS', b'\xe9\x8e\r\n',
+            if typ in {b'FLIS', b'FCIS', b'SRCS', b'\xe9\x8e\r\n', b'BOUN',
-                    b'RESC', b'BOUN', b'FDST', b'DATP', b'AUDI', b'VIDE'}:
+                       b'FDST', b'DATP', b'AUDI', b'VIDE'}:
                 pass  # Ignore these records
+            elif typ == b'RESC':
+                self.resc_data = read_resc_record(data)
             elif typ == b'FONT':
                 font = read_font_record(data)
                 href = "fonts/%05d.%s" % (fname_idx, font['ext'])
@@ -452,6 +455,9 @@ class Mobi8Reader(object):
         opf.create_manifest_from_files_in([os.getcwdu()], exclude=exclude)
         opf.create_spine(spine)
         opf.set_toc(toc)
+        ppd = self.resc_data.get('page-progression-direction', None)
+        if ppd:
+            opf.page_progression_direction = ppd
 
         with open('metadata.opf', 'wb') as of, open('toc.ncx', 'wb') as ncx:
             opf.render(of, ncx, 'toc.ncx')

src/calibre/ebooks/mobi/utils.py

@@ -7,7 +7,7 @@ __license__   = 'GPL v3'
 __copyright__ = '2011, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-import struct, string, zlib, os
+import struct, string, zlib, os, re
 from collections import OrderedDict
 from io import BytesIO
 
@@ -393,6 +393,15 @@ def mobify_image(data):
         data = im.export('gif')
     return data
 
+def read_resc_record(data):
+    ans = {}
+    match = re.search(br'''<spine [^>]*page-progression-direction=['"](.+?)['"]''', data)
+    if match is not None:
+        ppd = match.group(1).lower()
+        if ppd in {b'ltr', b'rtl'}:
+            ans['page-progression-direction'] = ppd.decode('ascii')
+    return ans
+
 # Font records {{{
 def read_font_record(data, extent=1040):
     '''

src/calibre/ebooks/oeb/base.py

@@ -1210,6 +1210,7 @@ class Spine(object):
     def __init__(self, oeb):
         self.oeb = oeb
         self.items = []
+        self.page_progression_direction = None
 
     def _linear(self, linear):
         if isinstance(linear, basestring):
@@ -1896,4 +1897,6 @@ class OEBBook(object):
                              attrib={'media-type': PAGE_MAP_MIME})
             spine.attrib['page-map'] = id
             results[PAGE_MAP_MIME] = (href, self.pages.to_page_map())
+        if self.spine.page_progression_direction in {'ltr', 'rtl'}:
+            spine.attrib['page-progression-direction'] = self.spine.page_progression_direction
         return results

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

@@ -0,0 +1,158 @@
+#!/usr/bin/env python
+# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:fdm=marker:ai
+from __future__ import (unicode_literals, division, absolute_import,
+                        print_function)
+
+__license__   = 'GPL v3'
+__copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
+__docformat__ = 'restructuredtext en'
+
+import sys
+
+from lxml import etree
+
+from calibre import prints
+from calibre.ebooks.oeb.base import XHTML
+from calibre.ebooks.oeb.polish.stats import normalize_font_properties
+from calibre.utils.filenames import ascii_filename
+
+props = {'font-family':None, 'font-weight':'normal', 'font-style':'normal', 'font-stretch':'normal'}
+
+def matching_rule(font, rules):
+    ff = font['font-family']
+    if not isinstance(ff, basestring):
+        ff = tuple(ff)[0]
+    family = icu_lower(ff)
+    wt = font['font-weight']
+    style = font['font-style']
+    stretch = font['font-stretch']
+
+    for rule in rules:
+        if rule['font-style'] == style and rule['font-stretch'] == stretch and rule['font-weight'] == wt:
+            ff = rule['font-family']
+            if not isinstance(ff, basestring):
+                ff = tuple(ff)[0]
+            if icu_lower(ff) == family:
+                return rule
+
+def embed_font(container, font, all_font_rules, report, warned):
+    rule = matching_rule(font, all_font_rules)
+    ff = font['font-family']
+    if not isinstance(ff, basestring):
+        ff = ff[0]
+    if rule is None:
+        from calibre.utils.fonts.scanner import font_scanner, NoFonts
+        if ff in warned:
+            return
+        try:
+            fonts = font_scanner.fonts_for_family(ff)
+        except NoFonts:
+            report(_('Failed to find fonts for family: %s, not embedding') % ff)
+            warned.add(ff)
+            return
+        wt = int(font.get('font-weight', '400'))
+        for f in fonts:
+            if f['weight'] == wt and f['font-style'] == font.get('font-style', 'normal') and f['font-stretch'] == font.get('font-stretch', 'normal'):
+                report('Embedding font %s from %s' % (f['full_name'], f['path']))
+                data = font_scanner.get_font_data(f)
+                fname = f['full_name']
+                ext = 'otf' if f['is_otf'] else 'ttf'
+                fname = ascii_filename(fname).replace(' ', '-').replace('(', '').replace(')', '')
+                item = container.generate_item('fonts/%s.%s'%(fname, ext), id_prefix='font')
+                name = container.href_to_name(item.get('href'), container.opf_name)
+                with container.open(name, 'wb') as out:
+                    out.write(data)
+                href = container.name_to_href(name)
+                rule = {k:f.get(k, v) for k, v in props.iteritems()}
+                rule['src'] = 'url(%s)' % href
+                rule['name'] = name
+                return rule
+        msg = _('Failed to find font matching: family: %s; weight: %s; style: %s; stretch: %s') % (
+            ff, font['font-weight'], font['font-style'], font['font-stretch'])
+        if msg not in warned:
+            warned.add(msg)
+            report(msg)
+    else:
+        name = rule['src']
+        href = container.name_to_href(name)
+        rule = {k:ff if k == 'font-family' else rule.get(k, v) for k, v in props.iteritems()}
+        rule['src'] = 'url(%s)' % href
+        rule['name'] = name
+        return rule
+
+def embed_all_fonts(container, stats, report):
+    all_font_rules = tuple(stats.all_font_rules.itervalues())
+    warned = set()
+    rules, nrules = [], []
+    modified = set()
+
+    for path in container.spine_items:
+        name = container.abspath_to_name(path)
+        fu = stats.font_usage_map.get(name, None)
+        fs = stats.font_spec_map.get(name, None)
+        fr = stats.font_rule_map.get(name, None)
+        if None in (fs, fu, fr):
+            continue
+        fs = {icu_lower(x) for x in fs}
+        for font in fu.itervalues():
+            if icu_lower(font['font-family']) not in fs:
+                continue
+            rule = matching_rule(font, fr)
+            if rule is None:
+                # This font was not already embedded in this HTML file, before
+                # processing started
+                rule = matching_rule(font, nrules)
+                if rule is None:
+                    rule = embed_font(container, font, all_font_rules, report, warned)
+                    if rule is not None:
+                        rules.append(rule)
+                        nrules.append(normalize_font_properties(rule.copy()))
+                        modified.add(name)
+                        stats.font_stats[rule['name']] = font['text']
+                else:
+                    # This font was previously embedded by this code, update its stats
+                    stats.font_stats[rule['name']] |= font['text']
+                    modified.add(name)
+
+    if not rules:
+        report(_('No embeddable fonts found'))
+        return
+
+    # Write out CSS
+    rules = [';\n\t'.join('%s: %s' % (
+        k, '"%s"' % v if k == 'font-family' else v) for k, v in rule.iteritems() if (k in props and props[k] != v and v != '400') or k == 'src')
+        for rule in rules]
+    css = '\n\n'.join(['@font-face {\n\t%s\n}' % r for r in rules])
+    item = container.generate_item('fonts.css', id_prefix='font_embed')
+    name = container.href_to_name(item.get('href'), container.opf_name)
+    with container.open(name, 'wb') as out:
+        out.write(css.encode('utf-8'))
+
+    # Add link to CSS in all files that need it
+    for spine_name in modified:
+        root = container.parsed(spine_name)
+        head = root.xpath('//*[local-name()="head"][1]')[0]
+        href = container.name_to_href(name, spine_name)
+        etree.SubElement(head, XHTML('link'), rel='stylesheet', type='text/css', href=href).tail = '\n'
+        container.dirty(spine_name)
+
+
+if __name__ == '__main__':
+    from calibre.ebooks.oeb.polish.container import get_container
+    from calibre.ebooks.oeb.polish.stats import StatsCollector
+    from calibre.utils.logging import default_log
+    default_log.filter_level = default_log.DEBUG
+    inbook = sys.argv[-1]
+    ebook = get_container(inbook, default_log)
+    report = []
+    stats = StatsCollector(ebook, do_embed=True)
+    embed_all_fonts(ebook, stats, report.append)
+    outbook, ext = inbook.rpartition('.')[0::2]
+    outbook += '_subset.'+ext
+    ebook.commit(outbook)
+    prints('\nReport:')
+    for msg in report:
+        prints(msg)
+    print()
+    prints('Output written to:', outbook)
+

src/calibre/ebooks/oeb/polish/font_stats.coffee

@@ -67,6 +67,18 @@ class FontStats
                 ans.push(usage)
         py_bridge.value = ans
 
+    get_font_families: () ->
+        ans = {}
+        for node in document.getElementsByTagName('*')
+            rules = document.defaultView.getMatchedCSSRules(node, '')
+            if rules
+                for rule in rules
+                    style = rule.style
+                    family = style.getPropertyValue('font-family')
+                    if family
+                        ans[family] = true
+        py_bridge.value = ans
+
 if window?
     window.font_stats = new FontStats()
 

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

@@ -14,6 +14,7 @@ from functools import partial
 from calibre.ebooks.oeb.polish.container import get_container
 from calibre.ebooks.oeb.polish.stats import StatsCollector
 from calibre.ebooks.oeb.polish.subset import subset_all_fonts
+from calibre.ebooks.oeb.polish.embed import embed_all_fonts
 from calibre.ebooks.oeb.polish.cover import set_cover
 from calibre.ebooks.oeb.polish.replace import smarten_punctuation
 from calibre.ebooks.oeb.polish.jacket import (
@@ -21,6 +22,7 @@ from calibre.ebooks.oeb.polish.jacket import (
 from calibre.utils.logging import Log
 
 ALL_OPTS = {
+    'embed': False,
     'subset': False,
     'opf': None,
     'cover': None,
@@ -47,6 +49,13 @@ changes needed for the desired effect.</p>
 <p>Note that polishing only works on files in the %s formats.</p>\
 ''')%_(' or ').join('<b>%s</b>'%x for x in SUPPORTED),
 
+'embed': _('''\
+<p>Embed all fonts that are referenced in the document and are not already embedded.
+This will scan your computer for the fonts, and if they are found, they will be
+embedded into the document.</p>
+<p>Please ensure that you have the proper license for embedding the fonts used in this document.</p>
+'''),
+
 'subset': _('''\
 <p>Subsetting fonts means reducing an embedded font to contain
 only the characters used from that font in the book. This
@@ -118,8 +127,8 @@ def polish(file_map, opts, log, report):
         ebook = get_container(inbook, log)
         jacket = None
 
-        if opts.subset:
+        if opts.subset or opts.embed:
-            stats = StatsCollector(ebook)
+            stats = StatsCollector(ebook, do_embed=opts.embed)
 
         if opts.opf:
             rt(_('Updating metadata'))
@@ -159,6 +168,11 @@ def polish(file_map, opts, log, report):
             smarten_punctuation(ebook, report)
             report('')
 
+        if opts.embed:
+            rt(_('Embedding referenced fonts'))
+            embed_all_fonts(ebook, stats, report)
+            report('')
+
         if opts.subset:
             rt(_('Subsetting embedded fonts'))
             subset_all_fonts(ebook, stats.font_stats, report)
@@ -197,6 +211,7 @@ def option_parser():
     parser = OptionParser(usage=USAGE)
     a = parser.add_option
     o = partial(a, default=False, action='store_true')
+    o('--embed-fonts', '-e', dest='embed', help=CLI_HELP['embed'])
     o('--subset-fonts', '-f', dest='subset', help=CLI_HELP['subset'])
     a('--cover', '-c', help=_(
         'Path to a cover image. Changes the cover specified in the ebook. '

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

@@ -7,10 +7,11 @@ __license__   = 'GPL v3'
 __copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-import json, sys, os
+import json, sys, os, logging
 from urllib import unquote
+from collections import defaultdict
 
-from cssutils import parseStyle
+from cssutils import CSSParser
 from PyQt4.Qt import (pyqtProperty, QString, QEventLoop, Qt, QSize, QTimer,
                       pyqtSlot)
 from PyQt4.QtWebKit import QWebPage, QWebView
@@ -41,14 +42,14 @@ def normalize_font_properties(font):
                    'extra-expanded', 'ultra-expanded'}:
         val = 'normal'
     font['font-stretch'] = val
+    return font
 
-widths = {x:i for i, x in enumerate(( 'ultra-condensed',
+widths = {x:i for i, x in enumerate(('ultra-condensed',
         'extra-condensed', 'condensed', 'semi-condensed', 'normal',
         'semi-expanded', 'expanded', 'extra-expanded', 'ultra-expanded'
         ))}
 
 def get_matching_rules(rules, font):
-    normalize_font_properties(font)
     matches = []
 
     # Filter on family
@@ -100,7 +101,7 @@ def get_matching_rules(rules, font):
             return m
     return []
 
-class Page(QWebPage): # {{{
+class Page(QWebPage):  # {{{
 
     def __init__(self, log):
         self.log = log
@@ -157,10 +158,12 @@ class Page(QWebPage): # {{{
 
 class StatsCollector(object):
 
-    def __init__(self, container):
+    def __init__(self, container, do_embed=False):
         self.container = container
         self.log = self.logger = container.log
+        self.do_embed = do_embed
         must_use_qt()
+        self.parser = CSSParser(loglevel=logging.CRITICAL, log=logging.getLogger('calibre.css'))
 
         self.loop = QEventLoop()
         self.view = QWebView()
@@ -173,6 +176,10 @@ class StatsCollector(object):
 
         self.render_queue = list(container.spine_items)
         self.font_stats = {}
+        self.font_usage_map = {}
+        self.font_spec_map = {}
+        self.font_rule_map = {}
+        self.all_font_rules = {}
 
         QTimer.singleShot(0, self.render_book)
 
@@ -235,27 +242,35 @@ class StatsCollector(object):
         rules = []
         for rule in font_face_rules:
             ff = rule.get('font-family', None)
-            if not ff: continue
+            if not ff:
-            style = parseStyle('font-family:%s'%ff, validate=False)
+                continue
+            style = self.parser.parseStyle('font-family:%s'%ff, validate=False)
             ff = [x.value for x in
                   style.getProperty('font-family').propertyValue]
             if not ff or ff[0] == 'inherit':
                 continue
             rule['font-family'] = frozenset(icu_lower(f) for f in ff)
             src = rule.get('src', None)
-            if not src: continue
+            if not src:
-            style = parseStyle('background-image:%s'%src, validate=False)
+                continue
+            style = self.parser.parseStyle('background-image:%s'%src, validate=False)
             src = style.getProperty('background-image').propertyValue[0].uri
             name = self.href_to_name(src, '@font-face rule')
+            if name is None:
+                continue
             rule['src'] = name
             normalize_font_properties(rule)
             rule['width'] = widths[rule['font-stretch']]
             rule['weight'] = int(rule['font-weight'])
             rules.append(rule)
 
-        if not rules:
+        if not rules and not self.do_embed:
             return
 
+        self.font_rule_map[self.container.abspath_to_name(self.current_item)] = rules
+        for rule in rules:
+            self.all_font_rules[rule['src']] = rule
+
         for rule in rules:
             if rule['src'] not in self.font_stats:
                 self.font_stats[rule['src']] = set()
@@ -265,19 +280,48 @@ class StatsCollector(object):
         if not isinstance(font_usage, list):
             raise Exception('Unknown error occurred while reading font usage')
         exclude = {'\n', '\r', '\t'}
+        self.font_usage_map[self.container.abspath_to_name(self.current_item)] = fu = defaultdict(dict)
+        bad_fonts = {'serif', 'sans-serif', 'monospace', 'cursive', 'fantasy', 'sansserif', 'inherit'}
         for font in font_usage:
             text = set()
             for t in font['text']:
                 text |= frozenset(t)
             text.difference_update(exclude)
-            if not text: continue
+            if not text:
+                continue
+            normalize_font_properties(font)
             for rule in get_matching_rules(rules, font):
                 self.font_stats[rule['src']] |= text
+            if self.do_embed:
+                ff = [icu_lower(x) for x in font.get('font-family', [])]
+                if ff and ff[0] not in bad_fonts:
+                    keys = {'font-weight', 'font-style', 'font-stretch', 'font-family'}
+                    key = frozenset(((k, ff[0] if k == 'font-family' else v) for k, v in font.iteritems() if k in keys))
+                    val = fu[key]
+                    if not val:
+                        val.update({k:(font[k][0] if k == 'font-family' else font[k]) for k in keys})
+                        val['text'] = set()
+                    val['text'] |= text
+        self.font_usage_map[self.container.abspath_to_name(self.current_item)] = dict(fu)
+
+        if self.do_embed:
+            self.page.evaljs('window.font_stats.get_font_families()')
+            font_families = self.page.bridge_value
+            if not isinstance(font_families, dict):
+                raise Exception('Unknown error occurred while reading font families')
+            self.font_spec_map[self.container.abspath_to_name(self.current_item)] = fs = set()
+            for raw in font_families.iterkeys():
+                style = self.parser.parseStyle('font-family:' + raw, validate=False).getProperty('font-family')
+                for x in style.propertyValue:
+                    x = x.value
+                    if x and x.lower() not in bad_fonts:
+                        fs.add(x)
 
 if __name__ == '__main__':
     from calibre.ebooks.oeb.polish.container import get_container
     from calibre.utils.logging import default_log
     default_log.filter_level = default_log.DEBUG
     ebook = get_container(sys.argv[-1], default_log)
-    print (StatsCollector(ebook).font_stats)
+    print (StatsCollector(ebook, do_embed=True).font_stats)
+
 

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

@@ -281,15 +281,18 @@ def find_text(node):
 
 def from_files(container):
     toc = TOC()
-    for spinepath in container.spine_items:
+    for i, spinepath in enumerate(container.spine_items):
         name = container.abspath_to_name(spinepath)
         root = container.parsed(name)
         body = XPath('//h:body')(root)
         if not body:
             continue
         text = find_text(body[0])
-        if text:
+        if not text:
-            toc.add(text, name)
+            text = name.rpartition('/')[-1]
+            if i == 0 and text.rpartition('.')[0].lower() in {'titlepage', 'cover'}:
+                text = _('Cover')
+        toc.add(text, name)
     return toc
 
 def add_id(container, name, loc):

src/calibre/ebooks/oeb/reader.py

@@ -330,6 +330,9 @@ class OEBReader(object):
         if len(spine) == 0:
             raise OEBError("Spine is empty")
         self._spine_add_extra()
+        for val in xpath(opf, '/o2:package/o2:spine/@page-progression-direction'):
+            if val in {'ltr', 'rtl'}:
+                spine.page_progression_direction = val
 
     def _guide_from_opf(self, opf):
         guide = self.oeb.guide

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

@@ -0,0 +1,233 @@
+#!/usr/bin/env python
+# vim:fileencoding=utf-8
+from __future__ import (unicode_literals, division, absolute_import,
+                        print_function)
+
+__license__ = 'GPL v3'
+__copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
+
+import logging
+from collections import defaultdict
+
+import cssutils
+from lxml import etree
+
+from calibre import guess_type
+from calibre.ebooks.oeb.base import XPath, CSS_MIME, XHTML
+from calibre.ebooks.oeb.transforms.subset import get_font_properties, find_font_face_rules, elem_style
+from calibre.utils.filenames import ascii_filename
+from calibre.utils.fonts.scanner import font_scanner, NoFonts
+
+def used_font(style, embedded_fonts):
+    ff = [unicode(f) for f in style.get('font-family', []) if unicode(f).lower() not in {
+        'serif', 'sansserif', 'sans-serif', 'fantasy', 'cursive', 'monospace'}]
+    if not ff:
+        return False, None
+    lnames = {unicode(x).lower() for x in ff}
+
+    matching_set = []
+
+    # Filter on font-family
+    for ef in embedded_fonts:
+        flnames = {x.lower() for x in ef.get('font-family', [])}
+        if not lnames.intersection(flnames):
+            continue
+        matching_set.append(ef)
+    if not matching_set:
+        return True, None
+
+    # Filter on font-stretch
+    widths = {x:i for i, x in enumerate(('ultra-condensed',
+            'extra-condensed', 'condensed', 'semi-condensed', 'normal',
+            'semi-expanded', 'expanded', 'extra-expanded', 'ultra-expanded'
+            ))}
+
+    width = widths[style.get('font-stretch', 'normal')]
+    for f in matching_set:
+        f['width'] = widths[style.get('font-stretch', 'normal')]
+
+    min_dist = min(abs(width-f['width']) for f in matching_set)
+    if min_dist > 0:
+        return True, None
+    nearest = [f for f in matching_set if abs(width-f['width']) ==
+        min_dist]
+    if width <= 4:
+        lmatches = [f for f in nearest if f['width'] <= width]
+    else:
+        lmatches = [f for f in nearest if f['width'] >= width]
+    matching_set = (lmatches or nearest)
+
+    # Filter on font-style
+    fs = style.get('font-style', 'normal')
+    matching_set = [f for f in matching_set if f.get('font-style', 'normal') == fs]
+
+    # Filter on font weight
+    fw = int(style.get('font-weight', '400'))
+    matching_set = [f for f in matching_set if f.get('weight', 400) == fw]
+
+    if not matching_set:
+        return True, None
+    return True, matching_set[0]
+
+
+class EmbedFonts(object):
+
+    '''
+    Embed all referenced fonts, if found on system. Must be called after CSS flattening.
+    '''
+
+    def __call__(self, oeb, log, opts):
+        self.oeb, self.log, self.opts = oeb, log, opts
+        self.sheet_cache = {}
+        self.find_style_rules()
+        self.find_embedded_fonts()
+        self.parser = cssutils.CSSParser(loglevel=logging.CRITICAL, log=logging.getLogger('calibre.css'))
+        self.warned = set()
+        self.warned2 = set()
+
+        for item in oeb.spine:
+            if not hasattr(item.data, 'xpath'):
+                continue
+            sheets = []
+            for href in XPath('//h:link[@href and @type="text/css"]/@href')(item.data):
+                sheet = self.oeb.manifest.hrefs.get(item.abshref(href), None)
+                if sheet is not None:
+                    sheets.append(sheet)
+            if sheets:
+                self.process_item(item, sheets)
+
+    def find_embedded_fonts(self):
+        '''
+        Find all @font-face rules and extract the relevant info from them.
+        '''
+        self.embedded_fonts = []
+        for item in self.oeb.manifest:
+            if not hasattr(item.data, 'cssRules'):
+                continue
+            self.embedded_fonts.extend(find_font_face_rules(item, self.oeb))
+
+    def find_style_rules(self):
+        '''
+        Extract all font related style information from all stylesheets into a
+        dict mapping classes to font properties specified by that class. All
+        the heavy lifting has already been done by the CSS flattening code.
+        '''
+        rules = defaultdict(dict)
+        for item in self.oeb.manifest:
+            if not hasattr(item.data, 'cssRules'):
+                continue
+            for i, rule in enumerate(item.data.cssRules):
+                if rule.type != rule.STYLE_RULE:
+                    continue
+                props = {k:v for k,v in
+                        get_font_properties(rule).iteritems() if v}
+                if not props:
+                    continue
+                for sel in rule.selectorList:
+                    sel = sel.selectorText
+                    if sel and sel.startswith('.'):
+                        # We dont care about pseudo-selectors as the worst that
+                        # can happen is some extra characters will remain in
+                        # the font
+                        sel = sel.partition(':')[0]
+                        rules[sel[1:]].update(props)
+
+        self.style_rules = dict(rules)
+
+    def get_page_sheet(self):
+        if self.page_sheet is None:
+            manifest = self.oeb.manifest
+            id_, href = manifest.generate('page_css', 'page_styles.css')
+            self.page_sheet = manifest.add(id_, href, CSS_MIME, data=self.parser.parseString('', validate=False))
+            head = self.current_item.xpath('//*[local-name()="head"][1]')
+            if head:
+                href = self.current_item.relhref(href)
+                l = etree.SubElement(head[0], XHTML('link'),
+                    rel='stylesheet', type=CSS_MIME, href=href)
+                l.tail = '\n'
+            else:
+                self.log.warn('No <head> cannot embed font rules')
+        return self.page_sheet
+
+    def process_item(self, item, sheets):
+        ff_rules = []
+        self.current_item = item
+        self.page_sheet = None
+        for sheet in sheets:
+            if 'page_css' in sheet.id:
+                ff_rules.extend(find_font_face_rules(sheet, self.oeb))
+                self.page_sheet = sheet
+
+        base = {'font-family':['serif'], 'font-weight': '400',
+                'font-style':'normal', 'font-stretch':'normal'}
+
+        for body in item.data.xpath('//*[local-name()="body"]'):
+            self.find_usage_in(body, base, ff_rules)
+
+    def find_usage_in(self, elem, inherited_style, ff_rules):
+        style = elem_style(self.style_rules, elem.get('class', '') or '', inherited_style)
+        for child in elem:
+            self.find_usage_in(child, style, ff_rules)
+        has_font, existing = used_font(style, ff_rules)
+        if not has_font:
+            return
+        if existing is None:
+            in_book = used_font(style, self.embedded_fonts)[1]
+            if in_book is None:
+                # Try to find the font in the system
+                added = self.embed_font(style)
+                if added is not None:
+                    ff_rules.append(added)
+                    self.embedded_fonts.append(added)
+            else:
+                # TODO: Create a page rule from the book rule (cannot use it
+                # directly as paths might be different)
+                item = in_book['item']
+                sheet = self.parser.parseString(in_book['rule'].cssText, validate=False)
+                rule = sheet.cssRules[0]
+                page_sheet = self.get_page_sheet()
+                href = page_sheet.abshref(item.href)
+                rule.style.setProperty('src', 'url(%s)' % href)
+                ff_rules.append(find_font_face_rules(sheet, self.oeb)[0])
+                page_sheet.data.insertRule(rule, len(page_sheet.data.cssRules))
+
+    def embed_font(self, style):
+        ff = [unicode(f) for f in style.get('font-family', []) if unicode(f).lower() not in {
+            'serif', 'sansserif', 'sans-serif', 'fantasy', 'cursive', 'monospace'}]
+        if not ff:
+            return
+        ff = ff[0]
+        if ff in self.warned or ff == 'inherit':
+            return
+        try:
+            fonts = font_scanner.fonts_for_family(ff)
+        except NoFonts:
+            self.log.warn('Failed to find fonts for family:', ff, 'not embedding')
+            self.warned.add(ff)
+            return
+        try:
+            weight = int(style.get('font-weight', '400'))
+        except (ValueError, TypeError, AttributeError):
+            w = style['font-weight']
+            if w not in self.warned2:
+                self.log.warn('Invalid weight in font style: %r' % w)
+                self.warned2.add(w)
+            return
+        for f in fonts:
+            if f['weight'] == weight and f['font-style'] == style.get('font-style', 'normal') and f['font-stretch'] == style.get('font-stretch', 'normal'):
+                self.log('Embedding font %s from %s' % (f['full_name'], f['path']))
+                data = font_scanner.get_font_data(f)
+                name = f['full_name']
+                ext = 'otf' if f['is_otf'] else 'ttf'
+                name = ascii_filename(name).replace(' ', '-').replace('(', '').replace(')', '')
+                fid, href = self.oeb.manifest.generate(id=u'font', href=u'fonts/%s.%s'%(name, ext))
+                item = self.oeb.manifest.add(fid, href, guess_type('dummy.'+ext)[0], data=data)
+                item.unload_data_from_memory()
+                page_sheet = self.get_page_sheet()
+                href = page_sheet.relhref(item.href)
+                css = '''@font-face { font-family: "%s"; font-weight: %s; font-style: %s; font-stretch: %s; src: url(%s) }''' % (
+                    f['font-family'], f['font-weight'], f['font-style'], f['font-stretch'], href)
+                sheet = self.parser.parseString(css, validate=False)
+                page_sheet.data.insertRule(sheet.cssRules[0], len(page_sheet.data.cssRules))
+                return find_font_face_rules(sheet, self.oeb)[0]
+

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

@@ -194,7 +194,7 @@ class CSSFlattener(object):
         for i, font in enumerate(faces):
             ext = 'otf' if font['is_otf'] else 'ttf'
             fid, href = self.oeb.manifest.generate(id=u'font',
-                href=u'%s.%s'%(ascii_filename(font['full_name']).replace(u' ', u'-'), ext))
+                href=u'fonts/%s.%s'%(ascii_filename(font['full_name']).replace(u' ', u'-'), ext))
             item = self.oeb.manifest.add(fid, href,
                     guess_type('dummy.'+ext)[0],
                     data=font_scanner.get_font_data(font))

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

@@ -339,6 +339,8 @@ class FlowSplitter(object):
                     # We want to keep the descendants of the split point in
                     # Tree 1
                     keep_descendants = True
+                    # We want the split point element, but not its tail
+                    elem.tail = '\n'
 
                 continue
             if hit_split_point:
@@ -357,6 +359,18 @@ class FlowSplitter(object):
         for elem in tuple(body2.iterdescendants()):
             if elem is split_point2:
                 if not before:
+                    # Keep the split point element's tail, if it contains non-whitespace
+                    # text
+                    tail = elem.tail
+                    if tail and not tail.isspace():
+                        parent = elem.getparent()
+                        idx = parent.index(elem)
+                        if idx == 0:
+                            parent.text = (parent.text or '') + tail
+                        else:
+                            sib = parent[idx-1]
+                            sib.tail = (sib.tail or '') + tail
+                    # Remove the element itself
                     nix_element(elem)
                 break
             if elem in ancestors:

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

@@ -12,6 +12,111 @@ from collections import defaultdict
 from calibre.ebooks.oeb.base import urlnormalize
 from calibre.utils.fonts.sfnt.subset import subset, NoGlyphs, UnsupportedFont
 
+def get_font_properties(rule, default=None):
+    '''
+    Given a CSS rule, extract normalized font properties from
+    it. Note that shorthand font property should already have been expanded
+    by the CSS flattening code.
+    '''
+    props = {}
+    s = rule.style
+    for q in ('font-family', 'src', 'font-weight', 'font-stretch',
+            'font-style'):
+        g = 'uri' if q == 'src' else 'value'
+        try:
+            val = s.getProperty(q).propertyValue[0]
+            val = getattr(val, g)
+            if q == 'font-family':
+                val = [x.value for x in s.getProperty(q).propertyValue]
+                if val and val[0] == 'inherit':
+                    val = None
+        except (IndexError, KeyError, AttributeError, TypeError, ValueError):
+            val = None if q in {'src', 'font-family'} else default
+        if q in {'font-weight', 'font-stretch', 'font-style'}:
+            val = unicode(val).lower() if (val or val == 0) else val
+            if val == 'inherit':
+                val = default
+        if q == 'font-weight':
+            val = {'normal':'400', 'bold':'700'}.get(val, val)
+            if val not in {'100', '200', '300', '400', '500', '600', '700',
+                    '800', '900', 'bolder', 'lighter'}:
+                val = default
+            if val == 'normal':
+                val = '400'
+        elif q == 'font-style':
+            if val not in {'normal', 'italic', 'oblique'}:
+                val = default
+        elif q == 'font-stretch':
+            if val not in {'normal', 'ultra-condensed', 'extra-condensed',
+                    'condensed', 'semi-condensed', 'semi-expanded',
+                    'expanded', 'extra-expanded', 'ultra-expanded'}:
+                val = default
+        props[q] = val
+    return props
+
+
+def find_font_face_rules(sheet, oeb):
+    '''
+    Find all @font-face rules in the given sheet and extract the relevant info from them.
+    sheet can be either a ManifestItem or a CSSStyleSheet.
+    '''
+    ans = []
+    try:
+        rules = sheet.data.cssRules
+    except AttributeError:
+        rules = sheet.cssRules
+
+    for i, rule in enumerate(rules):
+        if rule.type != rule.FONT_FACE_RULE:
+            continue
+        props = get_font_properties(rule, default='normal')
+        if not props['font-family'] or not props['src']:
+            continue
+
+        try:
+            path = sheet.abshref(props['src'])
+        except AttributeError:
+            path = props['src']
+        ff = oeb.manifest.hrefs.get(urlnormalize(path), None)
+        if not ff:
+            continue
+        props['item'] = ff
+        if props['font-weight'] in {'bolder', 'lighter'}:
+            props['font-weight'] = '400'
+        props['weight'] = int(props['font-weight'])
+        props['rule'] = rule
+        props['chars'] = set()
+        ans.append(props)
+
+    return ans
+
+
+def elem_style(style_rules, cls, inherited_style):
+    '''
+    Find the effective style for the given element.
+    '''
+    classes = cls.split()
+    style = inherited_style.copy()
+    for cls in classes:
+        style.update(style_rules.get(cls, {}))
+    wt = style.get('font-weight', None)
+    pwt = inherited_style.get('font-weight', '400')
+    if wt == 'bolder':
+        style['font-weight'] = {
+                '100':'400',
+                '200':'400',
+                '300':'400',
+                '400':'700',
+                '500':'700',
+                }.get(pwt, '900')
+    elif wt == 'lighter':
+        style['font-weight'] = {
+                '600':'400', '700':'400',
+                '800':'700', '900':'700'}.get(pwt, '100')
+
+    return style
+
+
 class SubsetFonts(object):
 
     '''
@@ -76,72 +181,15 @@ class SubsetFonts(object):
             self.log('Reduced total font size to %.1f%% of original'%
                     (totals[0]/totals[1] * 100))
 
-    def get_font_properties(self, rule, default=None):
-        '''
-        Given a CSS rule, extract normalized font properties from
-        it. Note that shorthand font property should already have been expanded
-        by the CSS flattening code.
-        '''
-        props = {}
-        s = rule.style
-        for q in ('font-family', 'src', 'font-weight', 'font-stretch',
-                'font-style'):
-            g = 'uri' if q == 'src' else 'value'
-            try:
-                val = s.getProperty(q).propertyValue[0]
-                val = getattr(val, g)
-                if q == 'font-family':
-                    val = [x.value for x in s.getProperty(q).propertyValue]
-                    if val and val[0] == 'inherit':
-                        val = None
-            except (IndexError, KeyError, AttributeError, TypeError, ValueError):
-                val = None if q in {'src', 'font-family'} else default
-            if q in {'font-weight', 'font-stretch', 'font-style'}:
-                val = unicode(val).lower() if (val or val == 0) else val
-                if val == 'inherit':
-                    val = default
-            if q == 'font-weight':
-                val = {'normal':'400', 'bold':'700'}.get(val, val)
-                if val not in {'100', '200', '300', '400', '500', '600', '700',
-                        '800', '900', 'bolder', 'lighter'}:
-                    val = default
-                if val == 'normal': val = '400'
-            elif q == 'font-style':
-                if val not in {'normal', 'italic', 'oblique'}:
-                    val = default
-            elif q == 'font-stretch':
-                if val not in { 'normal', 'ultra-condensed', 'extra-condensed',
-                        'condensed', 'semi-condensed', 'semi-expanded',
-                        'expanded', 'extra-expanded', 'ultra-expanded'}:
-                    val = default
-            props[q] = val
-        return props
-
     def find_embedded_fonts(self):
         '''
         Find all @font-face rules and extract the relevant info from them.
         '''
         self.embedded_fonts = []
         for item in self.oeb.manifest:
-            if not hasattr(item.data, 'cssRules'): continue
+            if not hasattr(item.data, 'cssRules'):
-            for i, rule in enumerate(item.data.cssRules):
+                continue
-                if rule.type != rule.FONT_FACE_RULE:
+            self.embedded_fonts.extend(find_font_face_rules(item, self.oeb))
-                    continue
-                props = self.get_font_properties(rule, default='normal')
-                if not props['font-family'] or not props['src']:
-                    continue
-
-                path = item.abshref(props['src'])
-                ff = self.oeb.manifest.hrefs.get(urlnormalize(path), None)
-                if not ff:
-                    continue
-                props['item'] = ff
-                if props['font-weight'] in {'bolder', 'lighter'}:
-                    props['font-weight'] = '400'
-                props['weight'] = int(props['font-weight'])
-                props['chars'] = set()
-                props['rule'] = rule
-                self.embedded_fonts.append(props)
 
     def find_style_rules(self):
         '''
@@ -151,12 +199,13 @@ class SubsetFonts(object):
         '''
         rules = defaultdict(dict)
         for item in self.oeb.manifest:
-            if not hasattr(item.data, 'cssRules'): continue
+            if not hasattr(item.data, 'cssRules'):
+                continue
             for i, rule in enumerate(item.data.cssRules):
                 if rule.type != rule.STYLE_RULE:
                     continue
                 props = {k:v for k,v in
-                        self.get_font_properties(rule).iteritems() if v}
+                        get_font_properties(rule).iteritems() if v}
                 if not props:
                     continue
                 for sel in rule.selectorList:
@@ -172,41 +221,17 @@ class SubsetFonts(object):
 
     def find_font_usage(self):
         for item in self.oeb.manifest:
-            if not hasattr(item.data, 'xpath'): continue
+            if not hasattr(item.data, 'xpath'):
+                continue
             for body in item.data.xpath('//*[local-name()="body"]'):
                 base = {'font-family':['serif'], 'font-weight': '400',
                         'font-style':'normal', 'font-stretch':'normal'}
                 self.find_usage_in(body, base)
 
-    def elem_style(self, cls, inherited_style):
-        '''
-        Find the effective style for the given element.
-        '''
-        classes = cls.split()
-        style = inherited_style.copy()
-        for cls in classes:
-            style.update(self.style_rules.get(cls, {}))
-        wt = style.get('font-weight', None)
-        pwt = inherited_style.get('font-weight', '400')
-        if wt == 'bolder':
-            style['font-weight'] = {
-                    '100':'400',
-                    '200':'400',
-                    '300':'400',
-                    '400':'700',
-                    '500':'700',
-                    }.get(pwt, '900')
-        elif wt == 'lighter':
-            style['font-weight'] = {
-                    '600':'400', '700':'400',
-                    '800':'700', '900':'700'}.get(pwt, '100')
-
-        return style
-
     def used_font(self, style):
         '''
         Given a style find the embedded font that matches it. Returns None if
-        no match is found ( can happen if not family matches).
+        no match is found (can happen if no family matches).
         '''
         ff = style.get('font-family', [])
         lnames = {unicode(x).lower() for x in ff}
@@ -222,7 +247,7 @@ class SubsetFonts(object):
             return None
 
         # Filter on font-stretch
-        widths = {x:i for i, x in enumerate(( 'ultra-condensed',
+        widths = {x:i for i, x in enumerate(('ultra-condensed',
                 'extra-condensed', 'condensed', 'semi-condensed', 'normal',
                 'semi-expanded', 'expanded', 'extra-expanded', 'ultra-expanded'
                 ))}
@@ -280,7 +305,7 @@ class SubsetFonts(object):
         return ans
 
     def find_usage_in(self, elem, inherited_style):
-        style = self.elem_style(elem.get('class', '') or '', inherited_style)
+        style = elem_style(self.style_rules, elem.get('class', '') or '', inherited_style)
         for child in elem:
             self.find_usage_in(child, style)
         font = self.used_font(style)
@@ -290,3 +315,4 @@ class SubsetFonts(object):
                 font['chars'] |= chars
 
 
+

src/calibre/ebooks/pdf/render/from_html.py

@@ -253,7 +253,7 @@ class PDFWriter(QObject):
             return self.loop.exit(1)
         try:
             if not self.render_queue:
-                if self.toc is not None and len(self.toc) > 0 and not hasattr(self, 'rendered_inline_toc'):
+                if self.opts.pdf_add_toc and self.toc is not None and len(self.toc) > 0 and not hasattr(self, 'rendered_inline_toc'):
                     return self.render_inline_toc()
                 self.loop.exit()
             else:

src/calibre/ebooks/pdf/render/links.py

@@ -8,9 +8,8 @@ __copyright__ = '2012, Kovid Goyal <kovid at kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
 import os
-from future_builtins import map
+from urlparse import urlparse
-from urlparse import urlparse, urlunparse
+from urllib2 import unquote
-from urllib2 import quote, unquote
 
 from calibre.ebooks.pdf.render.common import Array, Name, Dictionary, String
 
@@ -84,10 +83,8 @@ class Links(object):
                     action = Dictionary({
                         'Type':Name('Action'), 'S':Name('URI'),
                     })
-                    parts = (x.encode('utf-8') if isinstance(x, type(u'')) else
+                    # Do not try to normalize/quote/unquote this URL as if it
-                             x for x in purl)
+                    # has a query part, it will get corrupted
-                    url = urlunparse(map(quote, map(unquote,
-                                                    parts))).decode('ascii')
                     action['URI'] = String(url)
                     annot['A'] = action
             if 'A' in annot or 'Dest' in annot:

src/calibre/gui2/actions/edit_metadata.py

@@ -399,8 +399,7 @@ class EditMetadataAction(InterfaceAction):
         if safe_merge:
             if not confirm('<p>'+_(
                 'Book formats and metadata from the selected books '
-                'will be added to the <b>first selected book</b> (%s). '
+                'will be added to the <b>first selected book</b> (%s).<br> '
-                'ISBN will <i>not</i> be merged.<br><br> '
                 'The second and subsequently selected books will not '
                 'be deleted or changed.<br><br>'
                 'Please confirm you want to proceed.')%title
@@ -413,7 +412,7 @@ class EditMetadataAction(InterfaceAction):
                 'Book formats from the selected books will be merged '
                 'into the <b>first selected book</b> (%s). '
                 'Metadata in the first selected book will not be changed. '
-                'Author, Title, ISBN and all other metadata will <i>not</i> be merged.<br><br>'
+                'Author, Title and all other metadata will <i>not</i> be merged.<br><br>'
                 'After merger the second and subsequently '
                 'selected books, with any metadata they have will be <b>deleted</b>. <br><br>'
                 'All book formats of the first selected book will be kept '
@@ -427,8 +426,7 @@ class EditMetadataAction(InterfaceAction):
         else:
             if not confirm('<p>'+_(
                 'Book formats and metadata from the selected books will be merged '
-                'into the <b>first selected book</b> (%s). '
+                'into the <b>first selected book</b> (%s).<br><br>'
-                'ISBN will <i>not</i> be merged.<br><br>'
                 'After merger the second and '
                 'subsequently selected books will be <b>deleted</b>. <br><br>'
                 'All book formats of the first selected book will be kept '
@@ -490,11 +488,13 @@ class EditMetadataAction(InterfaceAction):
     def merge_metadata(self, dest_id, src_ids):
         db = self.gui.library_view.model().db
         dest_mi = db.get_metadata(dest_id, index_is_id=True)
+        merged_identifiers = db.get_identifiers(dest_id, index_is_id=True)
         orig_dest_comments = dest_mi.comments
         dest_cover = db.cover(dest_id, index_is_id=True)
         had_orig_cover = bool(dest_cover)
         for src_id in src_ids:
             src_mi = db.get_metadata(src_id, index_is_id=True)
+
             if src_mi.comments and orig_dest_comments != src_mi.comments:
                 if not dest_mi.comments:
                     dest_mi.comments = src_mi.comments
@@ -523,7 +523,15 @@ class EditMetadataAction(InterfaceAction):
             if not dest_mi.series:
                 dest_mi.series = src_mi.series
                 dest_mi.series_index = src_mi.series_index
+
+            src_identifiers = db.get_identifiers(src_id, index_is_id=True)
+            src_identifiers.update(merged_identifiers)
+            merged_identifiers = src_identifiers.copy()
+
+        if merged_identifiers:
+            dest_mi.set_identifiers(merged_identifiers)
         db.set_metadata(dest_id, dest_mi, ignore_errors=False)
+
         if not had_orig_cover and dest_cover:
             db.set_cover(dest_id, dest_cover)
 

src/calibre/gui2/actions/polish.py

@@ -45,6 +45,7 @@ class Polish(QDialog):  # {{{
                   ORIGINAL_* format before running it.</p>''')
             ),
 
+            'embed':_('<h3>Embed referenced fonts</h3>%s')%HELP['embed'],
             'subset':_('<h3>Subsetting fonts</h3>%s')%HELP['subset'],
 
             'smarten_punctuation':
@@ -75,6 +76,7 @@ class Polish(QDialog):  # {{{
 
         count = 0
         self.all_actions = OrderedDict([
+            ('embed', _('&Embed all referenced fonts')),
             ('subset', _('&Subset all embedded fonts')),
             ('smarten_punctuation', _('Smarten &punctuation')),
             ('metadata', _('Update &metadata in the book files')),

src/calibre/gui2/convert/look_and_feel.py

@@ -32,7 +32,7 @@ class LookAndFeelWidget(Widget, Ui_Form):
         Widget.__init__(self, parent,
                 ['change_justification', 'extra_css', 'base_font_size',
                     'font_size_mapping', 'line_height', 'minimum_line_height',
-                    'embed_font_family', 'subset_embedded_fonts',
+                    'embed_font_family', 'embed_all_fonts', 'subset_embedded_fonts',
                     'smarten_punctuation', 'unsmarten_punctuation',
                     'disable_font_rescaling', 'insert_blank_line',
                     'remove_paragraph_spacing',

src/calibre/gui2/convert/look_and_feel.ui

@@ -14,6 +14,70 @@
    <string>Form</string>
   </property>
   <layout class="QGridLayout" name="gridLayout">
+   <item row="12" column="1" colspan="2">
+    <widget class="QCheckBox" name="opt_keep_ligatures">
+     <property name="text">
+      <string>Keep &amp;ligatures</string>
+     </property>
+    </widget>
+   </item>
+   <item row="12" column="3">
+    <widget class="QCheckBox" name="opt_linearize_tables">
+     <property name="text">
+      <string>&amp;Linearize tables</string>
+     </property>
+    </widget>
+   </item>
+   <item row="1" column="0">
+    <widget class="QLabel" name="label_18">
+     <property name="text">
+      <string>Base &amp;font size:</string>
+     </property>
+     <property name="buddy">
+      <cstring>opt_base_font_size</cstring>
+     </property>
+    </widget>
+   </item>
+   <item row="9" column="3">
+    <widget class="QLabel" name="label_7">
+     <property name="text">
+      <string>&amp;Line size:</string>
+     </property>
+     <property name="alignment">
+      <set>Qt::AlignRight|Qt::AlignTrailing|Qt::AlignVCenter</set>
+     </property>
+     <property name="buddy">
+      <cstring>opt_insert_blank_line_size</cstring>
+     </property>
+    </widget>
+   </item>
+   <item row="5" column="1" colspan="2">
+    <widget class="EncodingComboBox" name="opt_input_encoding">
+     <property name="editable">
+      <bool>true</bool>
+     </property>
+    </widget>
+   </item>
+   <item row="8" column="0" colspan="2">
+    <widget class="QCheckBox" name="opt_remove_paragraph_spacing">
+     <property name="text">
+      <string>Remove &amp;spacing between paragraphs</string>
+     </property>
+    </widget>
+   </item>
+   <item row="8" column="3">
+    <widget class="QLabel" name="label_4">
+     <property name="text">
+      <string>&amp;Indent size:</string>
+     </property>
+     <property name="alignment">
+      <set>Qt::AlignRight|Qt::AlignTrailing|Qt::AlignVCenter</set>
+     </property>
+     <property name="buddy">
+      <cstring>opt_remove_paragraph_spacing_indent_size</cstring>
+     </property>
+    </widget>
+   </item>
    <item row="3" column="4">
     <widget class="QDoubleSpinBox" name="opt_line_height">
      <property name="suffix">
@@ -24,6 +88,57 @@
      </property>
     </widget>
    </item>
+   <item row="9" column="0" colspan="2">
+    <widget class="QCheckBox" name="opt_insert_blank_line">
+     <property name="text">
+      <string>Insert &amp;blank line between paragraphs</string>
+     </property>
+    </widget>
+   </item>
+   <item row="9" column="4">
+    <widget class="QDoubleSpinBox" name="opt_insert_blank_line_size">
+     <property name="suffix">
+      <string> em</string>
+     </property>
+     <property name="decimals">
+      <number>1</number>
+     </property>
+    </widget>
+   </item>
+   <item row="10" column="0">
+    <widget class="QLabel" name="label_5">
+     <property name="text">
+      <string>Text &amp;justification:</string>
+     </property>
+     <property name="buddy">
+      <cstring>opt_change_justification</cstring>
+     </property>
+    </widget>
+   </item>
+   <item row="10" column="2" colspan="3">
+    <widget class="QComboBox" name="opt_change_justification"/>
+   </item>
+   <item row="11" column="0">
+    <widget class="QCheckBox" name="opt_smarten_punctuation">
+     <property name="text">
+      <string>Smarten &amp;punctuation</string>
+     </property>
+    </widget>
+   </item>
+   <item row="11" column="1" colspan="4">
+    <widget class="QCheckBox" name="opt_asciiize">
+     <property name="text">
+      <string>&amp;Transliterate unicode characters to ASCII</string>
+     </property>
+    </widget>
+   </item>
+   <item row="12" column="0">
+    <widget class="QCheckBox" name="opt_unsmarten_punctuation">
+     <property name="text">
+      <string>&amp;UnSmarten punctuation</string>
+     </property>
+    </widget>
+   </item>
    <item row="3" column="3">
     <widget class="QLabel" name="label">
      <property name="text">
@@ -44,51 +159,6 @@
      </property>
     </widget>
    </item>
-   <item row="3" column="1">
-    <widget class="QDoubleSpinBox" name="opt_minimum_line_height">
-     <property name="suffix">
-      <string> %</string>
-     </property>
-     <property name="decimals">
-      <number>1</number>
-     </property>
-     <property name="maximum">
-      <double>900.000000000000000</double>
-     </property>
-    </widget>
-   </item>
-   <item row="1" column="1">
-    <widget class="QDoubleSpinBox" name="opt_base_font_size">
-     <property name="suffix">
-      <string> pt</string>
-     </property>
-     <property name="decimals">
-      <number>1</number>
-     </property>
-     <property name="minimum">
-      <double>0.000000000000000</double>
-     </property>
-     <property name="maximum">
-      <double>50.000000000000000</double>
-     </property>
-     <property name="singleStep">
-      <double>1.000000000000000</double>
-     </property>
-     <property name="value">
-      <double>15.000000000000000</double>
-     </property>
-    </widget>
-   </item>
-   <item row="2" column="0">
-    <widget class="QLabel" name="label_2">
-     <property name="text">
-      <string>Font size &amp;key:</string>
-     </property>
-     <property name="buddy">
-      <cstring>opt_font_size_mapping</cstring>
-     </property>
-    </widget>
-   </item>
    <item row="2" column="1" colspan="3">
     <layout class="QHBoxLayout" name="horizontalLayout">
      <item>
@@ -133,56 +203,72 @@
      </property>
     </widget>
    </item>
-   <item row="5" column="1" colspan="2">
+   <item row="3" column="1">
-    <widget class="EncodingComboBox" name="opt_input_encoding">
+    <widget class="QDoubleSpinBox" name="opt_minimum_line_height">
-     <property name="editable">
-      <bool>true</bool>
-     </property>
-    </widget>
-   </item>
-   <item row="7" column="0" colspan="2">
-    <widget class="QCheckBox" name="opt_remove_paragraph_spacing">
-     <property name="text">
-      <string>Remove &amp;spacing between paragraphs</string>
-     </property>
-    </widget>
-   </item>
-   <item row="7" column="3">
-    <widget class="QLabel" name="label_4">
-     <property name="text">
-      <string>&amp;Indent size:</string>
-     </property>
-     <property name="alignment">
-      <set>Qt::AlignRight|Qt::AlignTrailing|Qt::AlignVCenter</set>
-     </property>
-     <property name="buddy">
-      <cstring>opt_remove_paragraph_spacing_indent_size</cstring>
-     </property>
-    </widget>
-   </item>
-   <item row="7" column="4">
-    <widget class="QDoubleSpinBox" name="opt_remove_paragraph_spacing_indent_size">
-     <property name="toolTip">
-      <string>&lt;p&gt;When calibre removes inter paragraph spacing, it automatically sets a paragraph indent, to ensure that paragraphs can be easily distinguished. This option controls the width of that indent.</string>
-     </property>
-     <property name="specialValueText">
-      <string>No change</string>
-     </property>
      <property name="suffix">
-      <string> em</string>
+      <string> %</string>
+     </property>
+     <property name="decimals">
+      <number>1</number>
+     </property>
+     <property name="maximum">
+      <double>900.000000000000000</double>
+     </property>
+    </widget>
+   </item>
+   <item row="1" column="1">
+    <widget class="QDoubleSpinBox" name="opt_base_font_size">
+     <property name="suffix">
+      <string> pt</string>
      </property>
      <property name="decimals">
       <number>1</number>
      </property>
      <property name="minimum">
-      <double>-0.100000000000000</double>
+      <double>0.000000000000000</double>
+     </property>
+     <property name="maximum">
+      <double>50.000000000000000</double>
      </property>
      <property name="singleStep">
-      <double>0.100000000000000</double>
+      <double>1.000000000000000</double>
+     </property>
+     <property name="value">
+      <double>15.000000000000000</double>
      </property>
     </widget>
    </item>
-   <item row="12" column="0" colspan="5">
+   <item row="0" column="0" colspan="5">
+    <widget class="QCheckBox" name="opt_disable_font_rescaling">
+     <property name="text">
+      <string>&amp;Disable font size rescaling</string>
+     </property>
+    </widget>
+   </item>
+   <item row="6" column="1" colspan="2">
+    <widget class="FontFamilyChooser" name="opt_embed_font_family" native="true"/>
+   </item>
+   <item row="2" column="0">
+    <widget class="QLabel" name="label_2">
+     <property name="text">
+      <string>Font size &amp;key:</string>
+     </property>
+     <property name="buddy">
+      <cstring>opt_font_size_mapping</cstring>
+     </property>
+    </widget>
+   </item>
+   <item row="6" column="0">
+    <widget class="QLabel" name="label_10">
+     <property name="text">
+      <string>&amp;Embed font family:</string>
+     </property>
+     <property name="buddy">
+      <cstring>opt_embed_font_family</cstring>
+     </property>
+    </widget>
+   </item>
+   <item row="13" column="0" colspan="5">
     <widget class="QTabWidget" name="tabWidget">
      <property name="currentIndex">
       <number>0</number>
@@ -300,121 +386,42 @@
      </widget>
     </widget>
    </item>
-   <item row="8" column="0" colspan="2">
-    <widget class="QCheckBox" name="opt_insert_blank_line">
-     <property name="text">
-      <string>Insert &amp;blank line between paragraphs</string>
-     </property>
-    </widget>
-   </item>
    <item row="8" column="4">
-    <widget class="QDoubleSpinBox" name="opt_insert_blank_line_size">
+    <widget class="QDoubleSpinBox" name="opt_remove_paragraph_spacing_indent_size">
+     <property name="toolTip">
+      <string>&lt;p&gt;When calibre removes inter paragraph spacing, it automatically sets a paragraph indent, to ensure that paragraphs can be easily distinguished. This option controls the width of that indent.</string>
+     </property>
+     <property name="specialValueText">
+      <string>No change</string>
+     </property>
      <property name="suffix">
       <string> em</string>
      </property>
      <property name="decimals">
       <number>1</number>
      </property>
-    </widget>
+     <property name="minimum">
-   </item>
+      <double>-0.100000000000000</double>
-   <item row="9" column="0">
-    <widget class="QLabel" name="label_5">
-     <property name="text">
-      <string>Text &amp;justification:</string>
      </property>
-     <property name="buddy">
+     <property name="singleStep">
-      <cstring>opt_change_justification</cstring>
+      <double>0.100000000000000</double>
      </property>
     </widget>
    </item>
-   <item row="9" column="2" colspan="3">
+   <item row="7" column="3">
-    <widget class="QComboBox" name="opt_change_justification"/>
-   </item>
-   <item row="10" column="0">
-    <widget class="QCheckBox" name="opt_smarten_punctuation">
-     <property name="text">
-      <string>Smarten &amp;punctuation</string>
-     </property>
-    </widget>
-   </item>
-   <item row="10" column="1" colspan="4">
-    <widget class="QCheckBox" name="opt_asciiize">
-     <property name="text">
-      <string>&amp;Transliterate unicode characters to ASCII</string>
-     </property>
-    </widget>
-   </item>
-   <item row="11" column="0">
-    <widget class="QCheckBox" name="opt_unsmarten_punctuation">
-     <property name="text">
-      <string>&amp;UnSmarten punctuation</string>
-     </property>
-    </widget>
-   </item>
-   <item row="11" column="1" colspan="2">
-    <widget class="QCheckBox" name="opt_keep_ligatures">
-     <property name="text">
-      <string>Keep &amp;ligatures</string>
-     </property>
-    </widget>
-   </item>
-   <item row="11" column="3">
-    <widget class="QCheckBox" name="opt_linearize_tables">
-     <property name="text">
-      <string>&amp;Linearize tables</string>
-     </property>
-    </widget>
-   </item>
-   <item row="1" column="0">
-    <widget class="QLabel" name="label_18">
-     <property name="text">
-      <string>Base &amp;font size:</string>
-     </property>
-     <property name="buddy">
-      <cstring>opt_base_font_size</cstring>
-     </property>
-    </widget>
-   </item>
-   <item row="8" column="3">
-    <widget class="QLabel" name="label_7">
-     <property name="text">
-      <string>&amp;Line size:</string>
-     </property>
-     <property name="alignment">
-      <set>Qt::AlignRight|Qt::AlignTrailing|Qt::AlignVCenter</set>
-     </property>
-     <property name="buddy">
-      <cstring>opt_insert_blank_line_size</cstring>
-     </property>
-    </widget>
-   </item>
-   <item row="6" column="0">
-    <widget class="QLabel" name="label_10">
-     <property name="text">
-      <string>&amp;Embed font family:</string>
-     </property>
-     <property name="buddy">
-      <cstring>opt_embed_font_family</cstring>
-     </property>
-    </widget>
-   </item>
-   <item row="0" column="0" colspan="5">
-    <widget class="QCheckBox" name="opt_disable_font_rescaling">
-     <property name="text">
-      <string>&amp;Disable font size rescaling</string>
-     </property>
-    </widget>
-   </item>
-   <item row="6" column="1" colspan="2">
-    <widget class="FontFamilyChooser" name="opt_embed_font_family" native="true"/>
-   </item>
-   <item row="6" column="3" colspan="2">
     <widget class="QCheckBox" name="opt_subset_embedded_fonts">
      <property name="text">
       <string>&amp;Subset all embedded fonts</string>
      </property>
     </widget>
    </item>
+   <item row="7" column="0" colspan="3">
+    <widget class="QCheckBox" name="opt_embed_all_fonts">
+     <property name="text">
+      <string>&amp;Embed referenced fonts</string>
+     </property>
+    </widget>
+   </item>
   </layout>
  </widget>
  <customwidgets>

src/calibre/gui2/email.py

@@ -32,7 +32,7 @@ class Worker(Thread):
         self.func, self.args = func, args
 
     def run(self):
-        #time.sleep(1000)
+        # time.sleep(1000)
         try:
             self.func(*self.args)
         except Exception as e:
@@ -46,7 +46,7 @@ class Worker(Thread):
 class Sendmail(object):
 
     MAX_RETRIES = 1
-    TIMEOUT = 15 * 60 # seconds
+    TIMEOUT = 15 * 60  # seconds
 
     def __init__(self):
         self.calculate_rate_limit()
@@ -92,7 +92,11 @@ class Sendmail(object):
                 raise worker.exception
 
     def sendmail(self, attachment, aname, to, subject, text, log):
+        logged = False
         while time.time() - self.last_send_time <= self.rate_limit:
+            if not logged and self.rate_limit > 0:
+                log('Waiting %s seconds before sending, to avoid being marked as spam.\nYou can control this delay via Preferences->Tweaks' % self.rate_limit)
+                logged = True
             time.sleep(1)
         try:
             opts = email_config().parse()
@@ -162,7 +166,7 @@ def email_news(mi, remove, get_fmts, done, job_manager):
 plugboard_email_value = 'email'
 plugboard_email_formats = ['epub', 'mobi', 'azw3']
 
-class EmailMixin(object): # {{{
+class EmailMixin(object):  # {{{
 
     def send_by_mail(self, to, fmts, delete_from_library, subject='', send_ids=None,
             do_auto_convert=True, specific_format=None):
@@ -204,10 +208,10 @@ class EmailMixin(object): # {{{
                     if not components:
                         components = [mi.title]
                     subjects.append(os.path.join(*components))
-                a = authors_to_string(mi.authors if mi.authors else \
+                a = authors_to_string(mi.authors if mi.authors else
                         [_('Unknown')])
-                texts.append(_('Attached, you will find the e-book') + \
+                texts.append(_('Attached, you will find the e-book') +
-                        '\n\n' + t + '\n\t' + _('by') + ' ' + a + '\n\n' + \
+                        '\n\n' + t + '\n\t' + _('by') + ' ' + a + '\n\n' +
                         _('in the %s format.') %
                         os.path.splitext(f)[1][1:].upper())
                 prefix = ascii_filename(t+' - '+a)
@@ -227,7 +231,7 @@ class EmailMixin(object): # {{{
         auto = []
         if _auto_ids != []:
             for id in _auto_ids:
-                if specific_format == None:
+                if specific_format is None:
                     dbfmts = self.library_view.model().db.formats(id, index_is_id=True)
                     formats = [f.lower() for f in (dbfmts.split(',') if dbfmts else
                         [])]
@@ -298,8 +302,9 @@ class EmailMixin(object): # {{{
         sent_mails = email_news(mi, remove,
                 get_fmts, self.email_sent, self.job_manager)
         if sent_mails:
-            self.status_bar.show_message(_('Sent news to')+' '+\
+            self.status_bar.show_message(_('Sent news to')+' '+
                     ', '.join(sent_mails),  3000)
 
 # }}}
 
+

src/calibre/gui2/metadata/single_download.py

@@ -1006,7 +1006,7 @@ class FullFetch(QDialog):  # {{{
         l.addWidget(self.bb)
         self.bb.rejected.connect(self.reject)
         self.bb.accepted.connect(self.accept)
-        self.next_button = self.bb.addButton(_('Next'), self.bb.AcceptRole)
+        self.next_button = self.bb.addButton(_('Next'), self.bb.ActionRole)
         self.next_button.setDefault(True)
         self.next_button.setEnabled(False)
         self.next_button.setIcon(QIcon(I('ok.png')))
@@ -1019,7 +1019,7 @@ class FullFetch(QDialog):  # {{{
         self.log_button = self.bb.addButton(_('View log'), self.bb.ActionRole)
         self.log_button.clicked.connect(self.view_log)
         self.log_button.setIcon(QIcon(I('debug.png')))
-        self.ok_button.setVisible(False)
+        self.ok_button.setEnabled(False)
         self.prev_button.setVisible(False)
 
         self.identify_widget = IdentifyWidget(self.log, self)
@@ -1044,7 +1044,7 @@ class FullFetch(QDialog):  # {{{
 
     def book_selected(self, book, caches):
         self.next_button.setVisible(False)
-        self.ok_button.setVisible(True)
+        self.ok_button.setEnabled(True)
         self.prev_button.setVisible(True)
         self.book = book
         self.stack.setCurrentIndex(1)
@@ -1055,8 +1055,9 @@ class FullFetch(QDialog):  # {{{
 
     def back_clicked(self):
         self.next_button.setVisible(True)
-        self.ok_button.setVisible(False)
+        self.ok_button.setEnabled(False)
         self.prev_button.setVisible(False)
+        self.next_button.setFocus()
         self.stack.setCurrentIndex(0)
         self.covers_widget.cancel()
         self.covers_widget.reset_covers()
@@ -1081,6 +1082,7 @@ class FullFetch(QDialog):  # {{{
         self.next_button.setEnabled(True)
 
     def next_clicked(self, *args):
+        gprefs['metadata_single_gui_geom'] = bytearray(self.saveGeometry())
         self.identify_widget.get_result()
 
     def ok_clicked(self, *args):

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

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 
 from __future__ import (unicode_literals, division, absolute_import, print_function)
-store_version = 4 # Needed for dynamic plugin loading
+store_version = 5  # Needed for dynamic plugin loading
 
 __license__ = 'GPL 3'
 __copyright__ = '2011-2013, Tomasz Długosz <tomek3d@gmail.com>'
@@ -56,20 +56,20 @@ class WoblinkStore(BasicStoreConfig, StorePlugin):
         counter = max_results
         with closing(br.open(url, timeout=timeout)) as f:
             doc = html.fromstring(f.read())
-            for data in doc.xpath('//div[@class="book-item backgroundmix"]'):
+            for data in doc.xpath('//div[@class="nw_katalog_lista_ksiazka"]'):
                 if counter <= 0:
                     break
 
-                id = ''.join(data.xpath('.//td[@class="w10 va-t mYHaveItYes"]/a[1]/@href'))
+                id = ''.join(data.xpath('.//div[@class="nw_katalog_lista_ksiazka_okladka nw_okladka"]/a[1]/@href'))
                 if not id:
                     continue
 
-                cover_url = ''.join(data.xpath('.//td[@class="w10 va-t mYHaveItYes"]/a[1]/img/@src'))
+                cover_url = ''.join(data.xpath('.//div[@class="nw_katalog_lista_ksiazka_okladka nw_okladka"]/a[1]/img/@src'))
-                title = ''.join(data.xpath('.//h2[@class="title"]/a[1]/text()'))
+                title = ''.join(data.xpath('.//h2[@class="nw_katalog_lista_ksiazka_detale_tytul"]/a[1]/text()'))
-                author = ', '.join(data.xpath('.//td[@class="va-t"]/h3/a/text()'))
+                author = ', '.join(data.xpath('.//h3[@class="nw_katalog_lista_ksiazka_detale_autor"]/a/text()'))
-                price = ''.join(data.xpath('.//div[@class="prices"]/span[1]/strong/span/text()'))
+                price = ''.join(data.xpath('.//div[@class="nw_katalog_lista_ksiazka_opcjezakupu_cena"]/span/text()'))
                 price = re.sub('\.', ',', price)
-                formats = [ form[8:-4].split('.')[0] for form in data.xpath('.//p[3]/img/@src')]
+                formats = ', '.join(data.xpath('.//p[@class="nw_katalog_lista_ksiazka_detale_formaty"]/span/text()'))
 
                 s = SearchResult()
                 s.cover_url = 'http://woblink.com' + cover_url
@@ -77,25 +77,16 @@ class WoblinkStore(BasicStoreConfig, StorePlugin):
                 s.author = author.strip()
                 s.price = price + ' zł'
                 s.detail_item = id.strip()
+                s.formats = formats
 
-                if 'epub_drm' in formats:
+                if 'EPUB DRM' in formats:
                     s.drm = SearchResult.DRM_LOCKED
-                    s.formats = 'EPUB'
-
-                    counter -= 1
-                    yield s
-                elif 'pdf' in formats:
-                    s.drm = SearchResult.DRM_LOCKED
-                    s.formats = 'PDF'
 
                     counter -= 1
                     yield s
                 else:
                     s.drm = SearchResult.DRM_UNLOCKED
-                    if 'MOBI_nieb' in formats:
-                        formats.remove('MOBI_nieb')
-                        formats.append('MOBI')
-                    s.formats = ', '.join(formats).upper()
 
                     counter -= 1
                     yield s
+

src/calibre/gui2/tag_browser/ui.py

@@ -334,7 +334,7 @@ class TagBrowserWidget(QWidget):  # {{{
         search_layout = QHBoxLayout()
         self._layout.addLayout(search_layout)
         self.item_search = HistoryLineEdit(parent)
-        self.item_search.setMinimumContentsLength(10)
+        self.item_search.setMinimumContentsLength(5)
         self.item_search.setSizeAdjustPolicy(self.item_search.AdjustToMinimumContentsLengthWithIcon)
         try:
             self.item_search.lineEdit().setPlaceholderText(

src/calibre/library/catalogs/epub_mobi.py

@@ -149,7 +149,7 @@ class EPUB_MOBI(CatalogPlugin):
                           default=None,
                           dest='output_profile',
                           action=None,
-                          help=_("Specifies the output profile.  In some cases, an output profile is required to optimize the catalog for the device.  For example, 'kindle' or 'kindle_dx' creates a structured Table of Contents with Sections and Articles.\n"
+                          help=_("Specifies the output profile. In some cases, an output profile is required to optimize the catalog for the device.  For example, 'kindle' or 'kindle_dx' creates a structured Table of Contents with Sections and Articles.\n"
                           "Default: '%default'\n"
                           "Applies to: AZW3, ePub, MOBI output formats")),
                    Option('--prefix-rules',

src/calibre/library/database2.py

@@ -1500,8 +1500,9 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
         format = os.path.splitext(npath)[-1].lower().replace('.', '').upper()
         stream = lopen(npath, 'rb')
         format = check_ebook_format(stream, format)
-        retval = self.add_format(index, format, stream, replace=replace,
+        id = index if index_is_id else self.id(index)
-                               index_is_id=index_is_id, path=path, notify=notify)
+        retval = self.add_format(id, format, stream, replace=replace,
+                               index_is_id=True, path=path, notify=notify)
         run_plugins_on_postimport(self, id, format)
         return retval
 

src/calibre/utils/filenames.py

@@ -70,9 +70,11 @@ def shorten_components_to(length, components, more_to_take=0):
         else:
             if x is components[-1]:
                 b, e = os.path.splitext(x)
-                if e == '.': e = ''
+                if e == '.':
+                    e = ''
                 r = shorten_component(b, delta)+e
-                if r.startswith('.'): r = x[0]+r
+                if r.startswith('.'):
+                    r = x[0]+r
             else:
                 r = shorten_component(x, delta)
             r = r.strip()
@@ -115,7 +117,7 @@ def is_case_sensitive(path):
         os.remove(f1)
     return is_case_sensitive
 
-def case_preserving_open_file(path, mode='wb', mkdir_mode=0777):
+def case_preserving_open_file(path, mode='wb', mkdir_mode=0o777):
     '''
     Open the file pointed to by path with the specified mode. If any
     directories in path do not exist, they are created. Returns the
@@ -211,7 +213,8 @@ def samefile_windows(src, dst):
     handles = []
 
     def get_fileid(x):
-        if isbytestring(x): x = x.decode(filesystem_encoding)
+        if isbytestring(x):
+            x = x.decode(filesystem_encoding)
         try:
             h = win32file.CreateFile(x, 0, 0, None, win32file.OPEN_EXISTING,
                     win32file.FILE_FLAG_BACKUP_SEMANTICS, 0)
@@ -254,6 +257,24 @@ def samefile(src, dst):
             os.path.normcase(os.path.abspath(dst)))
     return samestring
 
+def windows_hardlink(src, dest):
+    import win32file, pywintypes
+    msg = u'Creating hardlink from %s to %s failed: %%s' % (src, dest)
+    try:
+        win32file.CreateHardLink(dest, src)
+    except pywintypes.error as e:
+        raise Exception(msg % e)
+    # We open and close dest, to ensure its directory entry is updated
+    # see http://blogs.msdn.com/b/oldnewthing/archive/2011/12/26/10251026.aspx
+    h = win32file.CreateFile(
+        dest, 0, win32file.FILE_SHARE_READ | win32file.FILE_SHARE_WRITE | win32file.FILE_SHARE_DELETE,
+        None, win32file.OPEN_EXISTING, 0, None)
+    sz = win32file.GetFileSize(h)
+    win32file.CloseHandle(h)
+
+    if sz != os.path.getsize(src):
+        raise Exception(msg % ('hardlink size: %d not the same as source size' % sz))
+
 class WindowsAtomicFolderMove(object):
 
     '''
@@ -270,14 +291,16 @@ class WindowsAtomicFolderMove(object):
         import win32file, winerror
         from pywintypes import error
 
-        if isbytestring(path): path = path.decode(filesystem_encoding)
+        if isbytestring(path):
+            path = path.decode(filesystem_encoding)
 
         if not os.path.exists(path):
             return
 
         for x in os.listdir(path):
             f = os.path.normcase(os.path.abspath(os.path.join(path, x)))
-            if not os.path.isfile(f): continue
+            if not os.path.isfile(f):
+                continue
             try:
                 # Ensure the file is not read-only
                 win32file.SetFileAttributes(f, win32file.FILE_ATTRIBUTE_NORMAL)
@@ -315,9 +338,7 @@ class WindowsAtomicFolderMove(object):
             else:
                 raise ValueError(u'The file %r does not exist'%path)
         try:
-            win32file.CreateHardLink(dest, path)
+            windows_hardlink(path, dest)
-            if os.path.getsize(dest) != os.path.getsize(path):
-                raise Exception('This apparently can happen on network shares. Sigh.')
             return
         except:
             pass
@@ -355,10 +376,8 @@ class WindowsAtomicFolderMove(object):
 
 def hardlink_file(src, dest):
     if iswindows:
-        import win32file
+        windows_hardlink(src, dest)
-        win32file.CreateHardLink(dest, src)
-        if os.path.getsize(dest) != os.path.getsize(src):
-            raise Exception('This apparently can happen on network shares. Sigh.')
         return
     os.link(src, dest)
 
+

src/calibre/utils/soupparser.py

@@ -62,7 +62,7 @@ def _parse(source, beautifulsoup, makeelement, **bsargs):
     if makeelement is None:
         makeelement = html.html_parser.makeelement
     if 'convertEntities' not in bsargs:
-        bsargs['convertEntities'] = 'html'
+        bsargs['convertEntities'] = 'xhtml'  # Changed by Kovid, otherwise ' is mangled, see https://bugs.launchpad.net/calibre/+bug/1197585
     tree = beautifulsoup(source, **bsargs)
     root = _convert_tree(tree, makeelement)
     # from ET: wrap the document in a html root element, if necessary

src/calibre/web/fetch/javascript.py

@@ -128,6 +128,8 @@ def download_resources(browser, resource_cache, output_dir):
             else:
                 img_counter += 1
                 ext = what(None, raw) or 'jpg'
+                if ext == 'jpeg':
+                    ext = 'jpg'  # Apparently Moon+ cannot handle .jpeg
                 href = 'img_%d.%s' % (img_counter, ext)
             dest = os.path.join(output_dir, href)
             resource_cache[h] = dest