0.9.13

Changelog.yaml

@@ -19,6 +19,57 @@
 #   new recipes:
 #     - title: 
 
+- version: 0.9.13
+  date: 2013-01-04
+
+  new features:
+    - title: "Complete rewrite of the PDF Output engine, to support links and fix various bugs"
+      type: major
+      description: "calibre now has a new PDF output engine that supports links in the text. It also fixes various bugs, detailed below. In order to implement support for links and fix these bugs, the engine had to be completely rewritten, so there may be some regressions."
+
+    - title: "Show disabled device plugins in Preferences->Ignored Devices"
+
+    - title: "Get Books: Fix Smashwords, Google books and B&N stores. Add Nook UK store"
+
+    - title: "Allow series numbers lower than -100 for custom series columns."
+      tickets: [1094475]
+
+    - title: "Add mass storage driver for rockhip based android smart phones"
+      tickets: [1087809]
+
+    - title: "Add a clear ratings button to the edit metadata dialog"
+ 
+  bug fixes:
+    - title: "PDF Output: Fix custom page sizes not working on OS X"
+
+    - title: "PDF Output: Fix embedding of many fonts not supported (note that embedding of OpenType fonts with Postscript outlines is still not supported on windows, though it is supported on other operating systems)"
+
+    - title: "PDF Output: Fix crashes converting some books to PDF on OS X"
+      tickets: [1087688]
+
+    - title: "HTML Input: Handle entities inside href attributes when following the links in an HTML file."
+      tickets: [1094203]
+
+    - title: "Content server: Fix custom icons not used for sub categories"
+      tickets: [1095016]
+
+    - title: "Force use of non-unicode constants in compiled templates. Fixes a problem with regular expression character classes and probably other things."
+
+    - title: "Kobo driver: Do not error out if there are invalid dates in the device database"
+      tickets: [1094597]
+
+    - title: "Content server: Fix for non-unicode hostnames when using mDNS"
+      tickets: [1094063]
+
+  improved recipes:
+    - Today's Zaman
+    - The Economist
+    - Foreign Affairs
+    - New York Times
+    - Alternet
+    - Harper's Magazine
+    - La Stampa
+
 - version: 0.9.12
   date: 2012-12-28
 

README

@@ -1,7 +1,7 @@
-calibre is an e-book library manager. It can view, convert and catalog e-books \
+calibre is an e-book library manager. It can view, convert and catalog e-books 
-in most of the major e-book formats. It can also talk to e-book reader \
+in most of the major e-book formats. It can also talk to e-book reader 
-devices. It can go out to the internet and fetch metadata for your books. \
+devices. It can go out to the internet and fetch metadata for your books. 
-It can download newspapers and convert them into e-books for convenient \
+It can download newspapers and convert them into e-books for convenient 
 reading. It is cross platform, running on Linux, Windows and OS X.
 
 For screenshots: https://calibre-ebook.com/demo
@@ -15,5 +15,5 @@ bzr branch lp:calibre
 To update your copy of the source code:
 bzr merge
 
-Tarballs of the source code for each release are now available \
+Tarballs of the source code for each release are now available 
 at http://code.google.com/p/calibre-ebook

manual/faq.rst

@@ -162,7 +162,8 @@ Follow these steps to find the problem:
     * If you are connecting an Apple iDevice (iPad, iPod Touch, iPhone), use the 'Connect to iTunes' method in the 'Getting started' instructions in `Calibre + Apple iDevices: Start here <http://www.mobileread.com/forums/showthread.php?t=118559>`_.
     * Make sure you are running the latest version of |app|. The latest version can always be downloaded from `the calibre website <http://calibre-ebook.com/download>`_.
     * Ensure your operating system is seeing the device. That is, the device should show up in Windows Explorer (in Windows) or Finder (in OS X).
-    * In |app|, go to Preferences->Plugins->Device Interface plugin and make sure the plugin for your device is enabled, the plugin icon next to it should be green when it is enabled.
+    * In |app|, go to Preferences->Ignored Devices and check that your device
+      is not being ignored
     * If all the above steps fail, go to Preferences->Miscellaneous and click debug device detection with your device attached and post the output as a ticket on `the calibre bug tracker <http://bugs.calibre-ebook.com>`_.
 
 My device is non-standard or unusual. What can I do to connect to it?
@@ -668,6 +669,9 @@ There are three possible things I know of, that can cause this:
       the blacklist of programs inside RoboForm to fix this. Or uninstall
       RoboForm.
 
+    * The Logitech SetPoint Settings application causes random crashes in
+      |app| when it is open. Close it before starting |app|.
+
 |app| is not starting on OS X?
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 

recipes/alternet.recipe

@@ -10,14 +10,12 @@ class Alternet(BasicNewsRecipe):
     category = 'News, Magazine'
     description = 'News magazine and online community'
     feeds          = [
-        (u'Front Page', u'http://feeds.feedblitz.com/alternet'),
+        (u'Front Page', u'http://feeds.feedblitz.com/alternet')
-        (u'Breaking News', u'http://feeds.feedblitz.com/alternet_breaking_news'),
-        (u'Top Ten Campaigns', u'http://feeds.feedblitz.com/alternet_top_10_campaigns'),
-        (u'Special Coverage Areas', u'http://feeds.feedblitz.com/alternet_coverage')
         ]
+
     remove_attributes = ['width', 'align','cellspacing']
     remove_javascript = True
-    use_embedded_content   = False
+    use_embedded_content   = True
     no_stylesheets = True
     language = 'en'
     encoding               = 'UTF-8'

recipes/economist.recipe

@@ -70,18 +70,6 @@ class Economist(BasicNewsRecipe):
         return br
     '''
 
-    def get_cover_url(self):
-        soup = self.index_to_soup('http://www.economist.com/printedition/covers')
-        div = soup.find('div', attrs={'class':lambda x: x and
-            'print-cover-links' in x})
-        a = div.find('a', href=True)
-        url = a.get('href')
-        if url.startswith('/'):
-            url = 'http://www.economist.com' + url
-        soup = self.index_to_soup(url)
-        div = soup.find('div', attrs={'class':'cover-content'})
-        img = div.find('img', src=True)
-        return img.get('src')
 
     def parse_index(self):
         return self.economist_parse_index()
@@ -92,7 +80,7 @@ class Economist(BasicNewsRecipe):
         if div is not None:
             img = div.find('img', src=True)
             if img is not None:
-                self.cover_url = img['src']
+                self.cover_url = re.sub('thumbnail','full',img['src'])
         feeds = OrderedDict()
         for section in soup.findAll(attrs={'class':lambda x: x and 'section' in
             x}):

recipes/economist_free.recipe

@@ -9,7 +9,7 @@ from calibre.web.feeds.news import BasicNewsRecipe
 from calibre.ebooks.BeautifulSoup import Tag, NavigableString
 from collections import OrderedDict
 
-import time, re
+import re
 
 class Economist(BasicNewsRecipe):
 
@@ -37,7 +37,6 @@ class Economist(BasicNewsRecipe):
             padding: 7px 0px 9px;
         }
         '''
-
     oldest_article = 7.0
     remove_tags = [
             dict(name=['script', 'noscript', 'title', 'iframe', 'cf_floatingcontent']),
@@ -46,7 +45,6 @@ class Economist(BasicNewsRecipe):
             {'class': lambda x: x and 'share-links-header' in x},
     ]
     keep_only_tags = [dict(id='ec-article-body')]
-    needs_subscription = False
     no_stylesheets = True
     preprocess_regexps = [(re.compile('</html>.*', re.DOTALL),
         lambda x:'</html>')]
@@ -55,27 +53,25 @@ class Economist(BasicNewsRecipe):
     # downloaded with connection reset by peer (104) errors.
     delay = 1
 
-    def get_cover_url(self):
+    needs_subscription = False
-        soup = self.index_to_soup('http://www.economist.com/printedition/covers')
+    '''
-        div = soup.find('div', attrs={'class':lambda x: x and
+    def get_browser(self):
-            'print-cover-links' in x})
+        br = BasicNewsRecipe.get_browser()
-        a = div.find('a', href=True)
+        if self.username and self.password:
-        url = a.get('href')
+            br.open('http://www.economist.com/user/login')
-        if url.startswith('/'):
+            br.select_form(nr=1)
-            url = 'http://www.economist.com' + url
+            br['name'] = self.username
-        soup = self.index_to_soup(url)
+            br['pass'] = self.password
-        div = soup.find('div', attrs={'class':'cover-content'})
+            res = br.submit()
-        img = div.find('img', src=True)
+            raw = res.read()
-        return img.get('src')
+            if '>Log out<' not in raw:
+                raise ValueError('Failed to login to economist.com. '
+                        'Check your username and password.')
+        return br
+    '''
+
 
     def parse_index(self):
-        try:
-            return self.economist_parse_index()
-        except:
-            raise
-            self.log.warn(
-                'Initial attempt to parse index failed, retrying in 30 seconds')
-            time.sleep(30)
         return self.economist_parse_index()
 
     def economist_parse_index(self):
@@ -84,7 +80,7 @@ class Economist(BasicNewsRecipe):
         if div is not None:
             img = div.find('img', src=True)
             if img is not None:
-                self.cover_url = img['src']
+                self.cover_url = re.sub('thumbnail','full',img['src'])
         feeds = OrderedDict()
         for section in soup.findAll(attrs={'class':lambda x: x and 'section' in
             x}):
@@ -151,154 +147,3 @@ class Economist(BasicNewsRecipe):
             div.insert(2, img)
             table.replaceWith(div)
         return soup
-
-'''
-from calibre.web.feeds.news import BasicNewsRecipe
-from calibre.utils.threadpool import ThreadPool, makeRequests
-from calibre.ebooks.BeautifulSoup import Tag, NavigableString
-import time, string, re
-from datetime import datetime
-from lxml import html
-
-class Economist(BasicNewsRecipe):
-
-    title = 'The Economist (RSS)'
-    language = 'en'
-
-    __author__ = "Kovid Goyal"
-    description = ('Global news and current affairs from a European'
-            ' perspective. Best downloaded on Friday mornings (GMT).'
-            ' Much slower than the print edition based version.')
-    extra_css      = '.headline {font-size: x-large;} \n h2 { font-size: small;  } \n h1 { font-size: medium;  }'
-    oldest_article = 7.0
-    cover_url = 'http://media.economist.com/sites/default/files/imagecache/print-cover-thumbnail/print-covers/currentcoverus_large.jpg'
-    #cover_url = 'http://www.economist.com/images/covers/currentcoverus_large.jpg'
-    remove_tags = [
-            dict(name=['script', 'noscript', 'title', 'iframe', 'cf_floatingcontent']),
-            dict(attrs={'class':['dblClkTrk', 'ec-article-info',
-                'share_inline_header', 'related-items']}),
-            {'class': lambda x: x and 'share-links-header' in x},
-    ]
-    keep_only_tags = [dict(id='ec-article-body')]
-    no_stylesheets = True
-    preprocess_regexps = [(re.compile('</html>.*', re.DOTALL),
-        lambda x:'</html>')]
-
-    def parse_index(self):
-        from calibre.web.feeds.feedparser import parse
-        if self.test:
-            self.oldest_article = 14.0
-        raw = self.index_to_soup(
-                'http://feeds.feedburner.com/economist/full_print_edition',
-                raw=True)
-        entries = parse(raw).entries
-        pool = ThreadPool(10)
-        self.feed_dict = {}
-        requests = []
-        for i, item in enumerate(entries):
-            title       = item.get('title', _('Untitled article'))
-            published = item.date_parsed
-            if not published:
-                published = time.gmtime()
-            utctime = datetime(*published[:6])
-            delta = datetime.utcnow() - utctime
-            if delta.days*24*3600 + delta.seconds > 24*3600*self.oldest_article:
-                self.log.debug('Skipping article %s as it is too old.'%title)
-                continue
-            link        = item.get('link', None)
-            description = item.get('description', '')
-            author      = item.get('author', '')
-
-            requests.append([i, link, title, description, author, published])
-        if self.test:
-            requests = requests[:4]
-        requests = makeRequests(self.process_eco_feed_article, requests, self.eco_article_found,
-                self.eco_article_failed)
-        for r in requests: pool.putRequest(r)
-        pool.wait()
-
-        return self.eco_sort_sections([(t, a) for t, a in
-            self.feed_dict.items()])
-
-    def eco_sort_sections(self, feeds):
-        if not feeds:
-            raise ValueError('No new articles found')
-        order = {
-            'The World This Week': 1,
-            'Leaders': 2,
-            'Letters': 3,
-            'Briefing': 4,
-            'Business': 5,
-            'Finance And Economics': 6,
-            'Science & Technology': 7,
-            'Books & Arts': 8,
-            'International': 9,
-            'United States': 10,
-            'Asia': 11,
-            'Europe': 12,
-            'The Americas': 13,
-            'Middle East & Africa': 14,
-            'Britain': 15,
-            'Obituary': 16,
-        }
-        return sorted(feeds, cmp=lambda x,y:cmp(order.get(x[0], 100),
-            order.get(y[0], 100)))
-
-    def process_eco_feed_article(self, args):
-        from calibre import browser
-        i, url, title, description, author, published = args
-        br = browser()
-        ret = br.open(url)
-        raw = ret.read()
-        url = br.geturl().split('?')[0]+'/print'
-        root = html.fromstring(raw)
-        matches = root.xpath('//*[@class = "ec-article-info"]')
-        feedtitle = 'Miscellaneous'
-        if matches:
-            feedtitle = string.capwords(html.tostring(matches[-1], method='text',
-                    encoding=unicode).split('|')[-1].strip())
-        return (i, feedtitle, url, title, description, author, published)
-
-    def eco_article_found(self, req, result):
-        from calibre.web.feeds import Article
-        i, feedtitle, link, title, description, author, published = result
-        self.log('Found print version for article:', title, 'in', feedtitle,
-                'at', link)
-
-        a = Article(i, title, link, author, description, published, '')
-
-        article = dict(title=a.title, description=a.text_summary,
-            date=time.strftime(self.timefmt, a.date), author=a.author, url=a.url)
-        if feedtitle not in self.feed_dict:
-            self.feed_dict[feedtitle] = []
-        self.feed_dict[feedtitle].append(article)
-
-    def eco_article_failed(self, req, tb):
-        self.log.error('Failed to download %s with error:'%req.args[0][2])
-        self.log.debug(tb)
-
-    def eco_find_image_tables(self, soup):
-        for x in soup.findAll('table', align=['right', 'center']):
-            if len(x.findAll('font')) in (1,2) and len(x.findAll('img')) == 1:
-                yield x
-
-    def postprocess_html(self, soup, first):
-        body = soup.find('body')
-        for name, val in body.attrs:
-            del body[name]
-        for table in list(self.eco_find_image_tables(soup)):
-            caption = table.find('font')
-            img = table.find('img')
-            div = Tag(soup, 'div')
-            div['style'] = 'text-align:left;font-size:70%'
-            ns = NavigableString(self.tag_to_string(caption))
-            div.insert(0, ns)
-            div.insert(1, Tag(soup, 'br'))
-            img.extract()
-            del img['width']
-            del img['height']
-            div.insert(2, img)
-            table.replaceWith(div)
-        return soup
-'''
-

recipes/foreignaffairs.recipe

@@ -11,21 +11,21 @@ class ForeignAffairsRecipe(BasicNewsRecipe):
     by Chen Wei weichen302@gmx.com, 2012-02-05'''
 
     __license__  = 'GPL v3'
-    __author__ = 'kwetal'
+    __author__ = 'Rick Shang, kwetal'
     language = 'en'
     version = 1.01
 
-    title = u'Foreign Affairs (Subcription or (free) Registration)'
+    title = u'Foreign Affairs (Subcription)'
     publisher = u'Council on Foreign Relations'
     category = u'USA, Foreign Affairs'
     description = u'The leading forum for serious discussion of American foreign policy and international affairs.'
 
     no_stylesheets = True
     remove_javascript = True
+    needs_subscription = True
 
     INDEX = 'http://www.foreignaffairs.com'
     FRONTPAGE = 'http://www.foreignaffairs.com/magazine'
-    INCLUDE_PREMIUM = False
 
 
     remove_tags = []
@@ -68,42 +68,56 @@ class ForeignAffairsRecipe(BasicNewsRecipe):
 
 
     def parse_index(self):
+
         answer = []
         soup = self.index_to_soup(self.FRONTPAGE)
-        sec_start = soup.findAll('div', attrs={'class':'panel-separator'})
+        #get dates
+        date = re.split('\s\|\s',self.tag_to_string(soup.head.title.string))[0]
+        self.timefmt =  u' [%s]'%date
+
+        sec_start = soup.findAll('div', attrs= {'class':'panel-pane'})
         for sec in sec_start:
-            content = sec.nextSibling
-            if content:
-                section = self.tag_to_string(content.find('h2'))
             articles = []
-
+            section = self.tag_to_string(sec.find('h2'))
-                tags = []
+            if 'Books' in section:
-                for div in content.findAll('div', attrs = {'class': re.compile(r'view-row\s+views-row-[0-9]+\s+views-row-[odd|even].*')}):
+                reviewsection=sec.find('div', attrs = {'class': 'item-list'})
-                    tags.append(div)
+                for subsection in reviewsection.findAll('div'):
-                for li in content.findAll('li'):
+                    subsectiontitle=self.tag_to_string(subsection.span.a)
-                    tags.append(li)
+                    subsectionurl=self.INDEX + subsection.span.a['href']
-
+                    soup1 = self.index_to_soup(subsectionurl)
-                for div in tags:
+                    for div in soup1.findAll('div', attrs = {'class': 'views-field-title'}):
-                    title = url = description = author = None
+                        if div.find('a') is not None:
-
+                            originalauthor=self.tag_to_string(div.findNext('div', attrs = {'class':'views-field-field-article-book-nid'}).div.a)
-                    if self.INCLUDE_PREMIUM:
+                            title=subsectiontitle+': '+self.tag_to_string(div.span.a)+' by '+originalauthor
-                        found_premium = False
+                            url=self.INDEX+div.span.a['href']
+                            atr=div.findNext('div', attrs = {'class': 'views-field-field-article-display-authors-value'})
+                            if atr is not None:
+                                author=self.tag_to_string(atr.span.a)
                             else:
-                        found_premium = div.findAll('span', attrs={'class':
+                                author=''
-                                                               'premium-icon'})
+                            desc=div.findNext('span', attrs = {'class': 'views-field-field-article-summary-value'})
-                    if not found_premium:
+                            if desc is not None:
-                        tag = div.find('div', attrs={'class': 'views-field-title'})
+                                description=self.tag_to_string(desc.div.p)
-
+                            else:
-                        if tag:
+                                description=''
-                            a = tag.find('a')
+                            articles.append({'title':title, 'date':None, 'url':url, 'description':description, 'author':author})
-                            if a:
+                    subsectiontitle=''
-                                title = self.tag_to_string(a)
+            else:
-                                url = self.INDEX + a['href']
+                for div in sec.findAll('div', attrs = {'class': 'views-field-title'}):
-                            author = self.tag_to_string(div.find('div', attrs = {'class': 'views-field-field-article-display-authors-value'}))
+                    if div.find('a') is not None:
-                            tag_summary = div.find('span', attrs = {'class': 'views-field-field-article-summary-value'})
+                        title=self.tag_to_string(div.span.a)
-                            description = self.tag_to_string(tag_summary)
+                        url=self.INDEX+div.span.a['href']
-                            articles.append({'title':title, 'date':None, 'url':url,
+                        atr=div.findNext('div', attrs = {'class': 'views-field-field-article-display-authors-value'})
-                                     'description':description, 'author':author})
+                        if atr is not None:
+                            author=self.tag_to_string(atr.span.a)
+                        else:
+                            author=''
+                        desc=div.findNext('span', attrs = {'class': 'views-field-field-article-summary-value'})
+                        if desc is not None:
+                            description=self.tag_to_string(desc.div.p)
+                        else:
+                            description=''
+                        articles.append({'title':title, 'date':None, 'url':url, 'description':description, 'author':author})
             if articles:
                     answer.append((section, articles))
         return answer
@@ -115,15 +129,17 @@ class ForeignAffairsRecipe(BasicNewsRecipe):
 
         return soup
 
-    needs_subscription = True
+
 
     def get_browser(self):
         br = BasicNewsRecipe.get_browser()
         if self.username is not None and self.password is not None:
-            br.open('https://www.foreignaffairs.com/user?destination=home')
+            br.open('https://www.foreignaffairs.com/user?destination=user%3Fop%3Dlo')
             br.select_form(nr = 1)
             br['name']   = self.username
             br['pass'] = self.password
             br.submit()
         return br
 
+    def cleanup(self):
+        self.browser.open('http://www.foreignaffairs.com/logout?destination=user%3Fop=lo')

recipes/harpers_full.recipe

@@ -8,7 +8,7 @@ If you have institutional subscription based on access IP you do not need to ent
 anything in username/password fields
 '''
 
-import time
+import time, re
 import urllib
 from calibre import strftime
 from calibre.web.feeds.news import BasicNewsRecipe
@@ -29,7 +29,6 @@ class Harpers_full(BasicNewsRecipe):
     needs_subscription    = 'optional'
     masthead_url          = 'http://harpers.org/wp-content/themes/harpers/images/pheader.gif'
     publication_type      = 'magazine'
-    INDEX                 = strftime('http://harpers.org/archive/%Y/%m')
     LOGIN                 = 'http://harpers.org/wp-content/themes/harpers/ajax_login.php'
     extra_css             = """
                                 body{font-family: adobe-caslon-pro,serif}
@@ -65,17 +64,28 @@ class Harpers_full(BasicNewsRecipe):
         return br
 
     def parse_index(self):
+        #find current issue
+        soup = self.index_to_soup('http://harpers.org/')
+        currentIssue=soup.find('div',attrs={'class':'mainNavi'}).find('li',attrs={'class':'curentIssue'})
+        currentIssue_url=self.tag_to_string(currentIssue.a['href'])
+        self.log(currentIssue_url)
+
+        #go to the current issue
+        soup1 = self.index_to_soup(currentIssue_url)
+        date = re.split('\s\|\s',self.tag_to_string(soup1.head.title.string))[0]
+        self.timefmt =  u' [%s]'%date
+
+        #get cover
+        coverurl='http://harpers.org/wp-content/themes/harpers/ajax_microfiche.php?img=harpers-'+re.split('harpers.org/',currentIssue_url)[1]+'gif/0001.gif'
+        soup2 = self.index_to_soup(coverurl)
+        self.cover_url = self.tag_to_string(soup2.find('img')['src'])
+        self.log(self.cover_url)
         articles = []
-        print 'Processing ' + self.INDEX
-        soup = self.index_to_soup(self.INDEX)
         count = 0
-        for item in soup.findAll('div', attrs={'class':'articleData'}):
+        for item in soup1.findAll('div', attrs={'class':'articleData'}):
             text_links = item.findAll('h2')
             for text_link in text_links:
                 if count == 0:
-                   lcover_url = item.find(attrs={'class':'dwpdf'})
-                   if lcover_url:
-                      self.cover_url = lcover_url.a['href']
                    count = 1
                 else:
                    url   = text_link.a['href']
@@ -87,7 +97,14 @@ class Harpers_full(BasicNewsRecipe):
                                      ,'url'        :url
                                      ,'description':''
                                     })
-        return [(soup.head.title.string, articles)]
+        return [(soup1.head.title.string, articles)]
 
     def print_version(self, url):
         return url + '?single=1'
+
+    def cleanup(self):
+        soup = self.index_to_soup('http://harpers.org/')
+        signouturl=self.tag_to_string(soup.find('li', attrs={'class':'subLogOut'}).findNext('li').a['href'])
+        self.log(signouturl)
+        self.browser.open(signouturl)
+

recipes/nytimes.recipe

@@ -15,6 +15,7 @@ from calibre.ebooks.BeautifulSoup import BeautifulSoup, Tag, BeautifulStoneSoup
 class NYTimes(BasicNewsRecipe):
 
     recursions=1 # set this to zero to omit Related articles lists
+    match_regexps=[r'/[12][0-9][0-9][0-9]/[0-9]+/'] # speeds up processing by preventing index page links from being followed
 
     # set getTechBlogs to True to include the technology blogs
     # set tech_oldest_article to control article age
@@ -24,6 +25,14 @@ class NYTimes(BasicNewsRecipe):
     tech_oldest_article = 14
     tech_max_articles_per_feed = 25
 
+    # set getPopularArticles to False if you don't want the Most E-mailed and Most Viewed articles
+    # otherwise you will get up to 20 of the most popular e-mailed and viewed articles (in each category)
+    getPopularArticles = True
+    popularPeriod = '1' # set this to the number of days to include in the measurement
+                        # e.g. 7 will get the most popular measured over the last 7 days
+                        # and 30 will get the most popular measured over 30 days.
+                        # you still only get up to 20 articles in each category
+
 
     # set headlinesOnly to True for the headlines-only version. If True, webEdition is ignored.
     headlinesOnly = True
@@ -153,7 +162,7 @@ class NYTimes(BasicNewsRecipe):
 
     timefmt = ''
 
-    simultaneous_downloads = 1
+    #simultaneous_downloads = 1 # no longer required to deal with ads
 
     cover_margins = (18,18,'grey99')
 
@@ -204,7 +213,8 @@ class NYTimes(BasicNewsRecipe):
                                         re.compile('^subNavigation'),
                                         re.compile('^leaderboard'),
                                         re.compile('^module'),
-                                        re.compile('commentCount')
+                                        re.compile('commentCount'),
+                                        'credit'
                                         ]}),
                     dict(name='div', attrs={'class':re.compile('toolsList')}),  # bits
                     dict(name='div', attrs={'class':re.compile('postNavigation')}),  # bits
@@ -291,11 +301,11 @@ class NYTimes(BasicNewsRecipe):
                 del ans[idx]
                 idx_max = idx_max-1
                 continue
-            if self.verbose:
+            if True: #self.verbose
                 self.log("Section %s: %d articles" % (ans[idx][0], len(ans[idx][1])) )
             for article in ans[idx][1]:
                 total_article_count += 1
-                if self.verbose:
+                if True: #self.verbose
                     self.log("\t%-40.40s... \t%-60.60s..." % (article['title'].encode('cp1252','replace'),
                               article['url'].encode('cp1252','replace')))
             idx = idx+1
@@ -351,23 +361,8 @@ class NYTimes(BasicNewsRecipe):
         br = BasicNewsRecipe.get_browser()
         return br
 
-##    This doesn't work (and probably never did). It either gets another serve of the advertisement,
-##    or if it gets the article then get_soup (from which it is invoked) traps trying to do xml decoding.
-##
-##    def skip_ad_pages(self, soup):
-##        # Skip ad pages served before actual article
-##        skip_tag = soup.find(True, {'name':'skip'})
-##        if skip_tag is not None:
-##            self.log.warn("Found forwarding link: %s" % skip_tag.parent['href'])
-##            url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
-##            url += '?pagewanted=all'
-##            self.log.warn("Skipping ad to article at '%s'" % url)
-##            return self.index_to_soup(url, raw=True)
-
-
     cover_tag = 'NY_NYT'
     def get_cover_url(self):
-        from datetime import timedelta, date
         cover = 'http://webmedia.newseum.org/newseum-multimedia/dfp/jpg'+str(date.today().day)+'/lg/'+self.cover_tag+'.jpg'
         br = BasicNewsRecipe.get_browser()
         daysback=1
@@ -390,6 +385,7 @@ class NYTimes(BasicNewsRecipe):
 
     masthead_url = 'http://graphics8.nytimes.com/images/misc/nytlogo379x64.gif'
 
+
     def short_title(self):
         return self.title
 
@@ -398,6 +394,7 @@ class NYTimes(BasicNewsRecipe):
         from contextlib import closing
         import copy
         from calibre.ebooks.chardet import xml_to_unicode
+        print("ARTICLE_TO_SOUP "+url_or_raw)
         if re.match(r'\w+://', url_or_raw):
             br = self.clone_browser(self.browser)
             open_func = getattr(br, 'open_novisit', br.open)
@@ -489,6 +486,67 @@ class NYTimes(BasicNewsRecipe):
                             description=description, author=author,
                             content=''))
 
+    def get_popular_articles(self,ans):
+        if self.getPopularArticles:
+            popular_articles = {}
+            key_list = []
+
+            def handleh3(h3tag):
+                try:
+                    url = h3tag.a['href']
+                except:
+                    return ('','','','')
+                url = re.sub(r'\?.*', '', url)
+                if self.exclude_url(url):
+                    return ('','','','')
+                url += '?pagewanted=all'
+                title = self.tag_to_string(h3tag.a,False)
+                h6tag = h3tag.findNextSibling('h6')
+                if h6tag is not None:
+                    author = self.tag_to_string(h6tag,False)
+                else:
+                    author = ''
+                ptag = h3tag.findNextSibling('p')
+                if ptag is not None:
+                    desc = self.tag_to_string(ptag,False)
+                else:
+                    desc = ''
+                return(title,url,author,desc)
+
+
+            have_emailed = False
+            emailed_soup = self.index_to_soup('http://www.nytimes.com/most-popular-emailed?period='+self.popularPeriod)
+            for h3tag in emailed_soup.findAll('h3'):
+                (title,url,author,desc) = handleh3(h3tag)
+                if url=='':
+                    continue
+                if not have_emailed:
+                    key_list.append('Most E-Mailed')
+                    popular_articles['Most E-Mailed'] = []
+                    have_emailed = True
+                popular_articles['Most E-Mailed'].append(
+                    dict(title=title, url=url, date=strftime('%a, %d %b'),
+                        description=desc, author=author,
+                        content=''))
+            have_viewed = False
+            viewed_soup = self.index_to_soup('http://www.nytimes.com/most-popular-viewed?period='+self.popularPeriod)
+            for h3tag in viewed_soup.findAll('h3'):
+                (title,url,author,desc) = handleh3(h3tag)
+                if url=='':
+                    continue
+                if not have_viewed:
+                    key_list.append('Most Viewed')
+                    popular_articles['Most Viewed'] = []
+                    have_viewed = True
+                popular_articles['Most Viewed'].append(
+                    dict(title=title, url=url, date=strftime('%a, %d %b'),
+                        description=desc, author=author,
+                        content=''))
+            viewed_ans = [(k, popular_articles[k]) for k in key_list if popular_articles.has_key(k)]
+            for x in viewed_ans:
+                ans.append(x)
+        return ans
+
     def get_tech_feeds(self,ans):
         if self.getTechBlogs:
             tech_articles = {}
@@ -550,7 +608,7 @@ class NYTimes(BasicNewsRecipe):
                         self.handle_article(lidiv)
 
         self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
-        return self.filter_ans(self.get_tech_feeds(self.ans))
+        return self.filter_ans(self.get_tech_feeds(self.get_popular_articles(self.ans)))
 
 
     def parse_todays_index(self):
@@ -583,7 +641,7 @@ class NYTimes(BasicNewsRecipe):
                         self.handle_article(lidiv)
 
         self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
-        return self.filter_ans(self.get_tech_feeds(self.ans))
+        return self.filter_ans(self.get_tech_feeds(self.get_popular_articles(self.ans)))
 
     def parse_headline_index(self):
 
@@ -657,7 +715,7 @@ class NYTimes(BasicNewsRecipe):
                         self.articles[section_name].append(dict(title=title, url=url, date=pubdate, description=description, author=author, content=''))
 
         self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
-        return self.filter_ans(self.get_tech_feeds(self.ans))
+        return self.filter_ans(self.get_tech_feeds(self.get_popular_articles(self.ans)))
 
     def parse_index(self):
         if self.headlinesOnly:
@@ -745,11 +803,12 @@ class NYTimes(BasicNewsRecipe):
 
 
     def preprocess_html(self, soup):
-        print("PREPROCESS TITLE="+self.tag_to_string(soup.title))
+        #print(strftime("%H:%M:%S")+" --  PREPROCESS TITLE="+self.tag_to_string(soup.title))
         skip_tag = soup.find(True, {'name':'skip'})
         if skip_tag is not None:
-            url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
+            #url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
-            url += '?pagewanted=all'
+            url = 'http://www.nytimes.com' + skip_tag.parent['href']
+            #url += '?pagewanted=all'
             self.log.warn("Skipping ad to article at '%s'" % url)
             sleep(5)
             soup = self.handle_tags(self.article_to_soup(url))
@@ -920,6 +979,7 @@ class NYTimes(BasicNewsRecipe):
             for aside in soup.findAll('div','aside'):
                 aside.extract()
             soup = self.strip_anchors(soup,True)
+            #print("RECURSIVE: "+self.tag_to_string(soup.title))
 
         if soup.find('div',attrs={'id':'blogcontent'}) is None:
             if first_fetch:
@@ -1005,7 +1065,7 @@ class NYTimes(BasicNewsRecipe):
                         if headline:
                                 tag = Tag(soup, "h2")
                                 tag['class'] = "headline"
-                                    tag.insert(0, self.fixChars(headline.renderContents()))
+                                tag.insert(0, self.fixChars(self.tag_to_string(headline,False)))
                                 soup.insert(0, tag)
                                 hrs = soup.findAll('hr')
                                 for hr in hrs:
@@ -1019,7 +1079,7 @@ class NYTimes(BasicNewsRecipe):
                 if bylineauthor:
                     tag = Tag(soup, "h6")
                     tag['class'] = "byline"
-                        tag.insert(0, self.fixChars(bylineauthor.renderContents()))
+                    tag.insert(0, self.fixChars(self.tag_to_string(bylineauthor,False)))
                     bylineauthor.replaceWith(tag)
             except:
                 self.log("ERROR:  fixing byline author format")
@@ -1030,7 +1090,7 @@ class NYTimes(BasicNewsRecipe):
                 if blogcredit:
                     tag = Tag(soup, "h6")
                     tag['class'] = "credit"
-                        tag.insert(0, self.fixChars(blogcredit.renderContents()))
+                    tag.insert(0, self.fixChars(self.tag_to_string(blogcredit,False)))
                     blogcredit.replaceWith(tag)
             except:
                 self.log("ERROR:  fixing credit format")
@@ -1084,7 +1144,7 @@ class NYTimes(BasicNewsRecipe):
                         divTag.replaceWith(tag)
             except:
                 self.log("ERROR:  Problem in Add class=authorId to <div> so we can format with CSS")
-
+        #print(strftime("%H:%M:%S")+" -- POSTPROCESS TITLE="+self.tag_to_string(soup.title))
         return soup
 
     def populate_article_metadata(self, article, soup, first):

recipes/nytimes_sub.recipe

@@ -15,6 +15,7 @@ from calibre.ebooks.BeautifulSoup import BeautifulSoup, Tag, BeautifulStoneSoup
 class NYTimes(BasicNewsRecipe):
 
     recursions=1 # set this to zero to omit Related articles lists
+    match_regexps=[r'/[12][0-9][0-9][0-9]/[0-9]+/'] # speeds up processing by preventing index page links from being followed
 
     # set getTechBlogs to True to include the technology blogs
     # set tech_oldest_article to control article age
@@ -24,6 +25,14 @@ class NYTimes(BasicNewsRecipe):
     tech_oldest_article = 14
     tech_max_articles_per_feed = 25
 
+    # set getPopularArticles to False if you don't want the Most E-mailed and Most Viewed articles
+    # otherwise you will get up to 20 of the most popular e-mailed and viewed articles (in each category)
+    getPopularArticles = True
+    popularPeriod = '1' # set this to the number of days to include in the measurement
+                        # e.g. 7 will get the most popular measured over the last 7 days
+                        # and 30 will get the most popular measured over 30 days.
+                        # you still only get up to 20 articles in each category
+
 
     # set headlinesOnly to True for the headlines-only version. If True, webEdition is ignored.
     headlinesOnly = False
@@ -32,7 +41,7 @@ class NYTimes(BasicNewsRecipe):
     # number of days old an article can be for inclusion. If oldest_web_article = None all articles
     # will be included. Note: oldest_web_article is ignored if webEdition = False
     webEdition = False
-    oldest_web_article = 7
+    oldest_web_article = None
 
     # download higher resolution images than the small thumbnails typically included in the article
     # the down side of having large beautiful images is the file size is much larger, on the order of 7MB per paper
@@ -153,7 +162,7 @@ class NYTimes(BasicNewsRecipe):
 
     timefmt = ''
 
-    simultaneous_downloads = 1
+    #simultaneous_downloads = 1 # no longer required to deal with ads
 
     cover_margins = (18,18,'grey99')
 
@@ -204,7 +213,8 @@ class NYTimes(BasicNewsRecipe):
                                         re.compile('^subNavigation'),
                                         re.compile('^leaderboard'),
                                         re.compile('^module'),
-                                        re.compile('commentCount')
+                                        re.compile('commentCount'),
+                                        'credit'
                                         ]}),
                     dict(name='div', attrs={'class':re.compile('toolsList')}),  # bits
                     dict(name='div', attrs={'class':re.compile('postNavigation')}),  # bits
@@ -291,11 +301,11 @@ class NYTimes(BasicNewsRecipe):
                 del ans[idx]
                 idx_max = idx_max-1
                 continue
-            if self.verbose:
+            if True: #self.verbose
                 self.log("Section %s: %d articles" % (ans[idx][0], len(ans[idx][1])) )
             for article in ans[idx][1]:
                 total_article_count += 1
-                if self.verbose:
+                if True: #self.verbose
                     self.log("\t%-40.40s... \t%-60.60s..." % (article['title'].encode('cp1252','replace'),
                               article['url'].encode('cp1252','replace')))
             idx = idx+1
@@ -351,23 +361,8 @@ class NYTimes(BasicNewsRecipe):
         br = BasicNewsRecipe.get_browser()
         return br
 
-##    This doesn't work (and probably never did). It either gets another serve of the advertisement,
-##    or if it gets the article then get_soup (from which it is invoked) traps trying to do xml decoding.
-##
-##    def skip_ad_pages(self, soup):
-##        # Skip ad pages served before actual article
-##        skip_tag = soup.find(True, {'name':'skip'})
-##        if skip_tag is not None:
-##            self.log.warn("Found forwarding link: %s" % skip_tag.parent['href'])
-##            url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
-##            url += '?pagewanted=all'
-##            self.log.warn("Skipping ad to article at '%s'" % url)
-##            return self.index_to_soup(url, raw=True)
-
-
     cover_tag = 'NY_NYT'
     def get_cover_url(self):
-        from datetime import timedelta, date
         cover = 'http://webmedia.newseum.org/newseum-multimedia/dfp/jpg'+str(date.today().day)+'/lg/'+self.cover_tag+'.jpg'
         br = BasicNewsRecipe.get_browser()
         daysback=1
@@ -390,6 +385,7 @@ class NYTimes(BasicNewsRecipe):
 
     masthead_url = 'http://graphics8.nytimes.com/images/misc/nytlogo379x64.gif'
 
+
     def short_title(self):
         return self.title
 
@@ -398,6 +394,7 @@ class NYTimes(BasicNewsRecipe):
         from contextlib import closing
         import copy
         from calibre.ebooks.chardet import xml_to_unicode
+        print("ARTICLE_TO_SOUP "+url_or_raw)
         if re.match(r'\w+://', url_or_raw):
             br = self.clone_browser(self.browser)
             open_func = getattr(br, 'open_novisit', br.open)
@@ -489,6 +486,67 @@ class NYTimes(BasicNewsRecipe):
                             description=description, author=author,
                             content=''))
 
+    def get_popular_articles(self,ans):
+        if self.getPopularArticles:
+            popular_articles = {}
+            key_list = []
+
+            def handleh3(h3tag):
+                try:
+                    url = h3tag.a['href']
+                except:
+                    return ('','','','')
+                url = re.sub(r'\?.*', '', url)
+                if self.exclude_url(url):
+                    return ('','','','')
+                url += '?pagewanted=all'
+                title = self.tag_to_string(h3tag.a,False)
+                h6tag = h3tag.findNextSibling('h6')
+                if h6tag is not None:
+                    author = self.tag_to_string(h6tag,False)
+                else:
+                    author = ''
+                ptag = h3tag.findNextSibling('p')
+                if ptag is not None:
+                    desc = self.tag_to_string(ptag,False)
+                else:
+                    desc = ''
+                return(title,url,author,desc)
+
+
+            have_emailed = False
+            emailed_soup = self.index_to_soup('http://www.nytimes.com/most-popular-emailed?period='+self.popularPeriod)
+            for h3tag in emailed_soup.findAll('h3'):
+                (title,url,author,desc) = handleh3(h3tag)
+                if url=='':
+                    continue
+                if not have_emailed:
+                    key_list.append('Most E-Mailed')
+                    popular_articles['Most E-Mailed'] = []
+                    have_emailed = True
+                popular_articles['Most E-Mailed'].append(
+                    dict(title=title, url=url, date=strftime('%a, %d %b'),
+                        description=desc, author=author,
+                        content=''))
+            have_viewed = False
+            viewed_soup = self.index_to_soup('http://www.nytimes.com/most-popular-viewed?period='+self.popularPeriod)
+            for h3tag in viewed_soup.findAll('h3'):
+                (title,url,author,desc) = handleh3(h3tag)
+                if url=='':
+                    continue
+                if not have_viewed:
+                    key_list.append('Most Viewed')
+                    popular_articles['Most Viewed'] = []
+                    have_viewed = True
+                popular_articles['Most Viewed'].append(
+                    dict(title=title, url=url, date=strftime('%a, %d %b'),
+                        description=desc, author=author,
+                        content=''))
+            viewed_ans = [(k, popular_articles[k]) for k in key_list if popular_articles.has_key(k)]
+            for x in viewed_ans:
+                ans.append(x)
+        return ans
+
     def get_tech_feeds(self,ans):
         if self.getTechBlogs:
             tech_articles = {}
@@ -550,7 +608,7 @@ class NYTimes(BasicNewsRecipe):
                         self.handle_article(lidiv)
 
         self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
-        return self.filter_ans(self.get_tech_feeds(self.ans))
+        return self.filter_ans(self.get_tech_feeds(self.get_popular_articles(self.ans)))
 
 
     def parse_todays_index(self):
@@ -583,7 +641,7 @@ class NYTimes(BasicNewsRecipe):
                         self.handle_article(lidiv)
 
         self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
-        return self.filter_ans(self.get_tech_feeds(self.ans))
+        return self.filter_ans(self.get_tech_feeds(self.get_popular_articles(self.ans)))
 
     def parse_headline_index(self):
 
@@ -657,7 +715,7 @@ class NYTimes(BasicNewsRecipe):
                         self.articles[section_name].append(dict(title=title, url=url, date=pubdate, description=description, author=author, content=''))
 
         self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
-        return self.filter_ans(self.get_tech_feeds(self.ans))
+        return self.filter_ans(self.get_tech_feeds(self.get_popular_articles(self.ans)))
 
     def parse_index(self):
         if self.headlinesOnly:
@@ -745,11 +803,12 @@ class NYTimes(BasicNewsRecipe):
 
 
     def preprocess_html(self, soup):
-        print("PREPROCESS TITLE="+self.tag_to_string(soup.title))
+        #print(strftime("%H:%M:%S")+" --  PREPROCESS TITLE="+self.tag_to_string(soup.title))
         skip_tag = soup.find(True, {'name':'skip'})
         if skip_tag is not None:
-            url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
+            #url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
-            url += '?pagewanted=all'
+            url = 'http://www.nytimes.com' + skip_tag.parent['href']
+            #url += '?pagewanted=all'
             self.log.warn("Skipping ad to article at '%s'" % url)
             sleep(5)
             soup = self.handle_tags(self.article_to_soup(url))
@@ -920,6 +979,7 @@ class NYTimes(BasicNewsRecipe):
             for aside in soup.findAll('div','aside'):
                 aside.extract()
             soup = self.strip_anchors(soup,True)
+            #print("RECURSIVE: "+self.tag_to_string(soup.title))
 
         if soup.find('div',attrs={'id':'blogcontent'}) is None:
             if first_fetch:
@@ -1005,7 +1065,7 @@ class NYTimes(BasicNewsRecipe):
                         if headline:
                                 tag = Tag(soup, "h2")
                                 tag['class'] = "headline"
-                                    tag.insert(0, self.fixChars(headline.renderContents()))
+                                tag.insert(0, self.fixChars(self.tag_to_string(headline,False)))
                                 soup.insert(0, tag)
                                 hrs = soup.findAll('hr')
                                 for hr in hrs:
@@ -1019,7 +1079,7 @@ class NYTimes(BasicNewsRecipe):
                 if bylineauthor:
                     tag = Tag(soup, "h6")
                     tag['class'] = "byline"
-                        tag.insert(0, self.fixChars(bylineauthor.renderContents()))
+                    tag.insert(0, self.fixChars(self.tag_to_string(bylineauthor,False)))
                     bylineauthor.replaceWith(tag)
             except:
                 self.log("ERROR:  fixing byline author format")
@@ -1030,7 +1090,7 @@ class NYTimes(BasicNewsRecipe):
                 if blogcredit:
                     tag = Tag(soup, "h6")
                     tag['class'] = "credit"
-                        tag.insert(0, self.fixChars(blogcredit.renderContents()))
+                    tag.insert(0, self.fixChars(self.tag_to_string(blogcredit,False)))
                     blogcredit.replaceWith(tag)
             except:
                 self.log("ERROR:  fixing credit format")
@@ -1084,7 +1144,7 @@ class NYTimes(BasicNewsRecipe):
                         divTag.replaceWith(tag)
             except:
                 self.log("ERROR:  Problem in Add class=authorId to <div> so we can format with CSS")
-
+        #print(strftime("%H:%M:%S")+" -- POSTPROCESS TITLE="+self.tag_to_string(soup.title))
         return soup
 
     def populate_article_metadata(self, article, soup, first):

recipes/todays_zaman.recipe

@@ -26,24 +26,28 @@ class TodaysZaman_en(BasicNewsRecipe):
    # remove_attributes = ['width','height']
 
     feeds          = [
-                      ( u'Home', u'http://www.todayszaman.com/rss?sectionId=0'),
+                      ( u'Home', u'http://www.todayszaman.com/0.rss'),
-                      ( u'News', u'http://www.todayszaman.com/rss?sectionId=100'),
+                      ( u'Sports', u'http://www.todayszaman.com/5.rss'),
-                      ( u'Business', u'http://www.todayszaman.com/rss?sectionId=105'),
+                      ( u'Columnists', u'http://www.todayszaman.com/6.rss'),
-                      ( u'Interviews', u'http://www.todayszaman.com/rss?sectionId=8'),
+                      ( u'Interviews', u'http://www.todayszaman.com/9.rss'),
-                      ( u'Columnists', u'http://www.todayszaman.com/rss?sectionId=6'),
+                      ( u'News', u'http://www.todayszaman.com/100.rss'),
-                      ( u'Op-Ed', u'http://www.todayszaman.com/rss?sectionId=109'),
+                      ( u'National', u'http://www.todayszaman.com/101.rss'),
-                      ( u'Arts & Culture', u'http://www.todayszaman.com/rss?sectionId=110'),
+                      ( u'Diplomacy', u'http://www.todayszaman.com/102.rss'),
-                      ( u'Expat Zone', u'http://www.todayszaman.com/rss?sectionId=132'),
+                      ( u'World', u'http://www.todayszaman.com/104.rss'),
-                      ( u'Sports', u'http://www.todayszaman.com/rss?sectionId=5'),
+                      ( u'Business', u'http://www.todayszaman.com/105.rss'),
-                      ( u'Features', u'http://www.todayszaman.com/rss?sectionId=116'),
+                      ( u'Op-Ed', u'http://www.todayszaman.com/109.rss'),
-                      ( u'Travel', u'http://www.todayszaman.com/rss?sectionId=117'),
+                      ( u'Arts & Culture', u'http://www.todayszaman.com/110.rss'),
-                      ( u'Leisure', u'http://www.todayszaman.com/rss?sectionId=118'),
+                      ( u'Features', u'http://www.todayszaman.com/116.rss'),
-                      ( u'Weird But True', u'http://www.todayszaman.com/rss?sectionId=134'),
+                      ( u'Travel', u'http://www.todayszaman.com/117.rss'),
-                      ( u'Life', u'http://www.todayszaman.com/rss?sectionId=133'),
+                      ( u'Food', u'http://www.todayszaman.com/124.rss'),
-                      ( u'Health', u'http://www.todayszaman.com/rss?sectionId=126'),
+                      ( u'Press Review', u'http://www.todayszaman.com/130.rss'),
-                      ( u'Press Review', u'http://www.todayszaman.com/rss?sectionId=130'),
+                      ( u'Expat Zone', u'http://www.todayszaman.com/132.rss'),
-                      ( u'Todays think tanks', u'http://www.todayszaman.com/rss?sectionId=159'),
+                      ( u'Life', u'http://www.todayszaman.com/133.rss'),
-
+                      ( u'Think Tanks', u'http://www.todayszaman.com/159.rss'),
+                      ( u'Almanac', u'http://www.todayszaman.com/161.rss'),
+                      ( u'Health', u'http://www.todayszaman.com/162.rss'),
+                      ( u'Fashion & Beauty', u'http://www.todayszaman.com/163.rss'),
+                      ( u'Science & Technology', u'http://www.todayszaman.com/349.rss'),
                      ]
 
     #def preprocess_html(self, soup):
@@ -51,3 +55,4 @@ class TodaysZaman_en(BasicNewsRecipe):
     #def print_version(self, url):       #there is a probem caused by table format
      #return url.replace('http://www.todayszaman.com/newsDetail_getNewsById.action?load=detay&', 'http://www.todayszaman.com/newsDetail_openPrintPage.action?')
 
+

setup/iso_639/ca.po

@@ -12,13 +12,13 @@ msgstr ""
 "Report-Msgid-Bugs-To: Debian iso-codes team <pkg-isocodes-"
 "devel@lists.alioth.debian.org>\n"
 "POT-Creation-Date: 2011-11-25 14:01+0000\n"
-"PO-Revision-Date: 2012-12-22 17:18+0000\n"
+"PO-Revision-Date: 2012-12-31 12:50+0000\n"
 "Last-Translator: Ferran Rius <frius64@hotmail.com>\n"
 "Language-Team: Catalan <linux@softcatala.org>\n"
 "MIME-Version: 1.0\n"
 "Content-Type: text/plain; charset=UTF-8\n"
 "Content-Transfer-Encoding: 8bit\n"
-"X-Launchpad-Export-Date: 2012-12-23 04:38+0000\n"
+"X-Launchpad-Export-Date: 2013-01-01 04:45+0000\n"
 "X-Generator: Launchpad (build 16378)\n"
 "Language: ca\n"
 
@@ -1744,7 +1744,7 @@ msgstr "Asu (Nigèria)"
 
 #. name for aun
 msgid "One; Molmo"
-msgstr "One; Molmo"
+msgstr "Oneià; Molmo"
 
 #. name for auo
 msgid "Auyokawa"
@@ -1964,7 +1964,7 @@ msgstr "Leyigha"
 
 #. name for ayk
 msgid "Akuku"
-msgstr "Akuku"
+msgstr "Okpe-Idesa-Akuku; Akuku"
 
 #. name for ayl
 msgid "Arabic; Libyan"
@@ -9984,7 +9984,7 @@ msgstr "Indri"
 
 #. name for ids
 msgid "Idesa"
-msgstr "Idesa"
+msgstr "Okpe-Idesa-Akuku; Idesa"
 
 #. name for idt
 msgid "Idaté"
@@ -19524,7 +19524,7 @@ msgstr ""
 
 #. name for obi
 msgid "Obispeño"
-msgstr ""
+msgstr "Obispeño"
 
 #. name for obk
 msgid "Bontok; Southern"
@@ -19532,7 +19532,7 @@ msgstr "Bontoc; meridional"
 
 #. name for obl
 msgid "Oblo"
-msgstr ""
+msgstr "Oblo"
 
 #. name for obm
 msgid "Moabite"
@@ -19552,11 +19552,11 @@ msgstr "Bretó; antic"
 
 #. name for obu
 msgid "Obulom"
-msgstr ""
+msgstr "Obulom"
 
 #. name for oca
 msgid "Ocaina"
-msgstr ""
+msgstr "Ocaina"
 
 #. name for och
 msgid "Chinese; Old"
@@ -19576,11 +19576,11 @@ msgstr "Matlazinca; Atzingo"
 
 #. name for oda
 msgid "Odut"
-msgstr ""
+msgstr "Odut"
 
 #. name for odk
 msgid "Od"
-msgstr ""
+msgstr "Od"
 
 #. name for odt
 msgid "Dutch; Old"
@@ -19588,11 +19588,11 @@ msgstr "Holandès; antic"
 
 #. name for odu
 msgid "Odual"
-msgstr ""
+msgstr "Odual"
 
 #. name for ofo
 msgid "Ofo"
-msgstr ""
+msgstr "Ofo"
 
 #. name for ofs
 msgid "Frisian; Old"
@@ -19604,11 +19604,11 @@ msgstr ""
 
 #. name for ogb
 msgid "Ogbia"
-msgstr ""
+msgstr "Ogbia"
 
 #. name for ogc
 msgid "Ogbah"
-msgstr ""
+msgstr "Ogbah"
 
 #. name for oge
 msgid "Georgian; Old"
@@ -19616,7 +19616,7 @@ msgstr ""
 
 #. name for ogg
 msgid "Ogbogolo"
-msgstr ""
+msgstr "Ogbogolo"
 
 #. name for ogo
 msgid "Khana"
@@ -19624,7 +19624,7 @@ msgstr ""
 
 #. name for ogu
 msgid "Ogbronuagum"
-msgstr ""
+msgstr "Ogbronuagum"
 
 #. name for oht
 msgid "Hittite; Old"
@@ -19636,27 +19636,27 @@ msgstr "Hongarès; antic"
 
 #. name for oia
 msgid "Oirata"
-msgstr ""
+msgstr "Oirata"
 
 #. name for oin
 msgid "One; Inebu"
-msgstr ""
+msgstr "Oneià; Inebu"
 
 #. name for ojb
 msgid "Ojibwa; Northwestern"
-msgstr ""
+msgstr "Ojibwa; Nordoccidental"
 
 #. name for ojc
 msgid "Ojibwa; Central"
-msgstr ""
+msgstr "Ojibwa; Central"
 
 #. name for ojg
 msgid "Ojibwa; Eastern"
-msgstr ""
+msgstr "Ojibwa; Oriental"
 
 #. name for oji
 msgid "Ojibwa"
-msgstr ""
+msgstr "Ojibwa; Occidental"
 
 #. name for ojp
 msgid "Japanese; Old"
@@ -19664,11 +19664,11 @@ msgstr "Japonès; antic"
 
 #. name for ojs
 msgid "Ojibwa; Severn"
-msgstr ""
+msgstr "Ojibwa; Severn"
 
 #. name for ojv
 msgid "Ontong Java"
-msgstr ""
+msgstr "Ontong Java"
 
 #. name for ojw
 msgid "Ojibwa; Western"
@@ -19676,19 +19676,19 @@ msgstr ""
 
 #. name for oka
 msgid "Okanagan"
-msgstr ""
+msgstr "Colville-Okanagà"
 
 #. name for okb
 msgid "Okobo"
-msgstr ""
+msgstr "Okobo"
 
 #. name for okd
 msgid "Okodia"
-msgstr ""
+msgstr "Okodia"
 
 #. name for oke
 msgid "Okpe (Southwestern Edo)"
-msgstr ""
+msgstr "Okpe"
 
 #. name for okh
 msgid "Koresh-e Rostam"
@@ -19696,15 +19696,15 @@ msgstr ""
 
 #. name for oki
 msgid "Okiek"
-msgstr ""
+msgstr "Okiek"
 
 #. name for okj
 msgid "Oko-Juwoi"
-msgstr ""
+msgstr "Oko-Juwoi"
 
 #. name for okk
 msgid "One; Kwamtim"
-msgstr ""
+msgstr "Oneià; Kwamtim"
 
 #. name for okl
 msgid "Kentish Sign Language; Old"
@@ -19716,7 +19716,7 @@ msgstr ""
 
 #. name for okn
 msgid "Oki-No-Erabu"
-msgstr ""
+msgstr "Oki-No-Erabu"
 
 #. name for oko
 msgid "Korean; Old (3rd-9th cent.)"
@@ -19728,19 +19728,19 @@ msgstr ""
 
 #. name for oks
 msgid "Oko-Eni-Osayen"
-msgstr ""
+msgstr "Oko-Eni-Osayen"
 
 #. name for oku
 msgid "Oku"
-msgstr ""
+msgstr "Oku"
 
 #. name for okv
 msgid "Orokaiva"
-msgstr ""
+msgstr "Orokaiwa"
 
 #. name for okx
 msgid "Okpe (Northwestern Edo)"
-msgstr ""
+msgstr "Okpe-Idesa-Akuku; Okpe"
 
 #. name for ola
 msgid "Walungge"
@@ -19752,11 +19752,11 @@ msgstr ""
 
 #. name for ole
 msgid "Olekha"
-msgstr ""
+msgstr "Olekha"
 
 #. name for olm
 msgid "Oloma"
-msgstr ""
+msgstr "Oloma"
 
 #. name for olo
 msgid "Livvi"
@@ -19768,7 +19768,7 @@ msgstr ""
 
 #. name for oma
 msgid "Omaha-Ponca"
-msgstr ""
+msgstr "Omaha-Ponca"
 
 #. name for omb
 msgid "Ambae; East"
@@ -19780,23 +19780,23 @@ msgstr ""
 
 #. name for ome
 msgid "Omejes"
-msgstr ""
+msgstr "Omejes"
 
 #. name for omg
 msgid "Omagua"
-msgstr ""
+msgstr "Omagua"
 
 #. name for omi
 msgid "Omi"
-msgstr ""
+msgstr "Omi"
 
 #. name for omk
 msgid "Omok"
-msgstr ""
+msgstr "Omok"
 
 #. name for oml
 msgid "Ombo"
-msgstr ""
+msgstr "Ombo"
 
 #. name for omn
 msgid "Minoan"
@@ -19816,11 +19816,11 @@ msgstr ""
 
 #. name for omt
 msgid "Omotik"
-msgstr ""
+msgstr "Omotik"
 
 #. name for omu
 msgid "Omurano"
-msgstr ""
+msgstr "Omurano"
 
 #. name for omw
 msgid "Tairora; South"
@@ -19832,7 +19832,7 @@ msgstr ""
 
 #. name for ona
 msgid "Ona"
-msgstr ""
+msgstr "Ona"
 
 #. name for onb
 msgid "Lingao"
@@ -19840,31 +19840,31 @@ msgstr ""
 
 #. name for one
 msgid "Oneida"
-msgstr ""
+msgstr "Oneida"
 
 #. name for ong
 msgid "Olo"
-msgstr ""
+msgstr "Olo"
 
 #. name for oni
 msgid "Onin"
-msgstr ""
+msgstr "Onin"
 
 #. name for onj
 msgid "Onjob"
-msgstr ""
+msgstr "Onjob"
 
 #. name for onk
 msgid "One; Kabore"
-msgstr ""
+msgstr "Oneià; Kabore"
 
 #. name for onn
 msgid "Onobasulu"
-msgstr ""
+msgstr "Onobasulu"
 
 #. name for ono
 msgid "Onondaga"
-msgstr ""
+msgstr "Onondaga"
 
 #. name for onp
 msgid "Sartang"
@@ -19872,15 +19872,15 @@ msgstr ""
 
 #. name for onr
 msgid "One; Northern"
-msgstr ""
+msgstr "Oneià; Septentrional"
 
 #. name for ons
 msgid "Ono"
-msgstr ""
+msgstr "Ono"
 
 #. name for ont
 msgid "Ontenu"
-msgstr ""
+msgstr "Ontenu"
 
 #. name for onu
 msgid "Unua"
@@ -19900,23 +19900,23 @@ msgstr ""
 
 #. name for oog
 msgid "Ong"
-msgstr ""
+msgstr "Ong"
 
 #. name for oon
 msgid "Önge"
-msgstr ""
+msgstr "Onge"
 
 #. name for oor
 msgid "Oorlams"
-msgstr ""
+msgstr "Oorlams"
 
 #. name for oos
 msgid "Ossetic; Old"
-msgstr ""
+msgstr "Osset"
 
 #. name for opa
 msgid "Okpamheri"
-msgstr ""
+msgstr "Okpamheri"
 
 #. name for opk
 msgid "Kopkaka"
@@ -19924,39 +19924,39 @@ msgstr ""
 
 #. name for opm
 msgid "Oksapmin"
-msgstr ""
+msgstr "Oksapmin"
 
 #. name for opo
 msgid "Opao"
-msgstr ""
+msgstr "Opao"
 
 #. name for opt
 msgid "Opata"
-msgstr ""
+msgstr "Opata"
 
 #. name for opy
 msgid "Ofayé"
-msgstr ""
+msgstr "Opaie"
 
 #. name for ora
 msgid "Oroha"
-msgstr ""
+msgstr "Oroha"
 
 #. name for orc
 msgid "Orma"
-msgstr ""
+msgstr "Orma"
 
 #. name for ore
 msgid "Orejón"
-msgstr ""
+msgstr "Orejon"
 
 #. name for org
 msgid "Oring"
-msgstr ""
+msgstr "Oring"
 
 #. name for orh
 msgid "Oroqen"
-msgstr ""
+msgstr "Orotxen"
 
 #. name for ori
 msgid "Oriya"
@@ -19968,19 +19968,19 @@ msgstr "Oromo"
 
 #. name for orn
 msgid "Orang Kanaq"
-msgstr ""
+msgstr "Orang; Kanaq"
 
 #. name for oro
 msgid "Orokolo"
-msgstr ""
+msgstr "Orocolo"
 
 #. name for orr
 msgid "Oruma"
-msgstr ""
+msgstr "Oruma"
 
 #. name for ors
 msgid "Orang Seletar"
-msgstr ""
+msgstr "Orang; Seletar"
 
 #. name for ort
 msgid "Oriya; Adivasi"
@@ -19988,7 +19988,7 @@ msgstr "Oriya; Adivasi"
 
 #. name for oru
 msgid "Ormuri"
-msgstr ""
+msgstr "Ormuri"
 
 #. name for orv
 msgid "Russian; Old"
@@ -19996,31 +19996,31 @@ msgstr "Rus; antic"
 
 #. name for orw
 msgid "Oro Win"
-msgstr ""
+msgstr "Oro Win"
 
 #. name for orx
 msgid "Oro"
-msgstr ""
+msgstr "Oro"
 
 #. name for orz
 msgid "Ormu"
-msgstr ""
+msgstr "Ormu"
 
 #. name for osa
 msgid "Osage"
-msgstr ""
+msgstr "Osage"
 
 #. name for osc
 msgid "Oscan"
-msgstr ""
+msgstr "Osc"
 
 #. name for osi
 msgid "Osing"
-msgstr ""
+msgstr "Osing"
 
 #. name for oso
 msgid "Ososo"
-msgstr ""
+msgstr "Ososo"
 
 #. name for osp
 msgid "Spanish; Old"
@@ -20028,15 +20028,15 @@ msgstr "Espanyol; antic"
 
 #. name for oss
 msgid "Ossetian"
-msgstr ""
+msgstr "Osset"
 
 #. name for ost
 msgid "Osatu"
-msgstr ""
+msgstr "Osatu"
 
 #. name for osu
 msgid "One; Southern"
-msgstr ""
+msgstr "One; Meridional"
 
 #. name for osx
 msgid "Saxon; Old"
@@ -20052,15 +20052,15 @@ msgstr ""
 
 #. name for otd
 msgid "Ot Danum"
-msgstr ""
+msgstr "Dohoi"
 
 #. name for ote
 msgid "Otomi; Mezquital"
-msgstr ""
+msgstr "Otomí; Mezquital"
 
 #. name for oti
 msgid "Oti"
-msgstr ""
+msgstr "Oti"
 
 #. name for otk
 msgid "Turkish; Old"
@@ -20068,43 +20068,43 @@ msgstr "Turc; antic"
 
 #. name for otl
 msgid "Otomi; Tilapa"
-msgstr ""
+msgstr "Otomí; Tilapa"
 
 #. name for otm
 msgid "Otomi; Eastern Highland"
-msgstr ""
+msgstr "Otomí; Oriental"
 
 #. name for otn
 msgid "Otomi; Tenango"
-msgstr ""
+msgstr "Otomí; Tenango"
 
 #. name for otq
 msgid "Otomi; Querétaro"
-msgstr ""
+msgstr "Otomí; Queretaro"
 
 #. name for otr
 msgid "Otoro"
-msgstr ""
+msgstr "Otoro"
 
 #. name for ots
 msgid "Otomi; Estado de México"
-msgstr ""
+msgstr "Otomí; Estat de Mèxic"
 
 #. name for ott
 msgid "Otomi; Temoaya"
-msgstr ""
+msgstr "Otomí; Temoaya"
 
 #. name for otu
 msgid "Otuke"
-msgstr ""
+msgstr "Otuke"
 
 #. name for otw
 msgid "Ottawa"
-msgstr ""
+msgstr "Ottawa"
 
 #. name for otx
 msgid "Otomi; Texcatepec"
-msgstr ""
+msgstr "Otomí; Texcatepec"
 
 #. name for oty
 msgid "Tamil; Old"
@@ -20112,7 +20112,7 @@ msgstr ""
 
 #. name for otz
 msgid "Otomi; Ixtenco"
-msgstr ""
+msgstr "Otomí; Ixtenc"
 
 #. name for oua
 msgid "Tagargrent"
@@ -20124,7 +20124,7 @@ msgstr ""
 
 #. name for oue
 msgid "Oune"
-msgstr ""
+msgstr "Oune"
 
 #. name for oui
 msgid "Uighur; Old"
@@ -20132,15 +20132,15 @@ msgstr ""
 
 #. name for oum
 msgid "Ouma"
-msgstr ""
+msgstr "Ouma"
 
 #. name for oun
 msgid "!O!ung"
-msgstr ""
+msgstr "Oung"
 
 #. name for owi
 msgid "Owiniga"
-msgstr ""
+msgstr "Owiniga"
 
 #. name for owl
 msgid "Welsh; Old"
@@ -20148,11 +20148,11 @@ msgstr "Gal·lès; antic"
 
 #. name for oyb
 msgid "Oy"
-msgstr ""
+msgstr "Oy"
 
 #. name for oyd
 msgid "Oyda"
-msgstr ""
+msgstr "Oyda"
 
 #. name for oym
 msgid "Wayampi"
@@ -20160,7 +20160,7 @@ msgstr ""
 
 #. name for oyy
 msgid "Oya'oya"
-msgstr ""
+msgstr "Oya'oya"
 
 #. name for ozm
 msgid "Koonzime"
@@ -20168,27 +20168,27 @@ msgstr ""
 
 #. name for pab
 msgid "Parecís"
-msgstr ""
+msgstr "Pareci"
 
 #. name for pac
 msgid "Pacoh"
-msgstr ""
+msgstr "Pacoh"
 
 #. name for pad
 msgid "Paumarí"
-msgstr ""
+msgstr "Paumarí"
 
 #. name for pae
 msgid "Pagibete"
-msgstr ""
+msgstr "Pagibete"
 
 #. name for paf
 msgid "Paranawát"
-msgstr ""
+msgstr "Paranawat"
 
 #. name for pag
 msgid "Pangasinan"
-msgstr ""
+msgstr "Pangasi"
 
 #. name for pah
 msgid "Tenharim"
@@ -20196,19 +20196,19 @@ msgstr ""
 
 #. name for pai
 msgid "Pe"
-msgstr ""
+msgstr "Pe"
 
 #. name for pak
 msgid "Parakanã"
-msgstr ""
+msgstr "Akwawa; Parakanà"
 
 #. name for pal
 msgid "Pahlavi"
-msgstr ""
+msgstr "Pahlavi"
 
 #. name for pam
 msgid "Pampanga"
-msgstr ""
+msgstr "Pampangà"
 
 #. name for pan
 msgid "Panjabi"
@@ -20220,63 +20220,63 @@ msgstr ""
 
 #. name for pap
 msgid "Papiamento"
-msgstr ""
+msgstr "Papiament"
 
 #. name for paq
 msgid "Parya"
-msgstr ""
+msgstr "Parya"
 
 #. name for par
 msgid "Panamint"
-msgstr ""
+msgstr "Panamint"
 
 #. name for pas
 msgid "Papasena"
-msgstr ""
+msgstr "Papasena"
 
 #. name for pat
 msgid "Papitalai"
-msgstr ""
+msgstr "Papitalai"
 
 #. name for pau
 msgid "Palauan"
-msgstr ""
+msgstr "Palavà"
 
 #. name for pav
 msgid "Pakaásnovos"
-msgstr ""
+msgstr "Pakaa Nova"
 
 #. name for paw
 msgid "Pawnee"
-msgstr ""
+msgstr "Pawnee"
 
 #. name for pax
 msgid "Pankararé"
-msgstr ""
+msgstr "Pankararé"
 
 #. name for pay
 msgid "Pech"
-msgstr ""
+msgstr "Pech"
 
 #. name for paz
 msgid "Pankararú"
-msgstr ""
+msgstr "Pankarurú"
 
 #. name for pbb
 msgid "Páez"
-msgstr ""
+msgstr "Páez"
 
 #. name for pbc
 msgid "Patamona"
-msgstr ""
+msgstr "Patamona"
 
 #. name for pbe
 msgid "Popoloca; Mezontla"
-msgstr ""
+msgstr "Popoloca; Mezontla"
 
 #. name for pbf
 msgid "Popoloca; Coyotepec"
-msgstr ""
+msgstr "Popoloca; Coyotepec"
 
 #. name for pbg
 msgid "Paraujano"
@@ -20288,7 +20288,7 @@ msgstr ""
 
 #. name for pbi
 msgid "Parkwa"
-msgstr ""
+msgstr "Parkwa"
 
 #. name for pbl
 msgid "Mak (Nigeria)"
@@ -20300,7 +20300,7 @@ msgstr ""
 
 #. name for pbo
 msgid "Papel"
-msgstr ""
+msgstr "Papel"
 
 #. name for pbp
 msgid "Badyara"
@@ -20336,7 +20336,7 @@ msgstr ""
 
 #. name for pca
 msgid "Popoloca; Santa Inés Ahuatempan"
-msgstr ""
+msgstr "Popoloca; Ahuatempan"
 
 #. name for pcb
 msgid "Pear"
@@ -20832,7 +20832,7 @@ msgstr "Senufo; Palaka"
 
 #. name for pls
 msgid "Popoloca; San Marcos Tlalcoyalco"
-msgstr ""
+msgstr "Popoloca; Tlalcoyalc"
 
 #. name for plt
 msgid "Malagasy; Plateau"
@@ -21040,7 +21040,7 @@ msgstr ""
 
 #. name for poe
 msgid "Popoloca; San Juan Atzingo"
-msgstr ""
+msgstr "Popoloca; Atzingo"
 
 #. name for pof
 msgid "Poke"
@@ -21104,7 +21104,7 @@ msgstr ""
 
 #. name for pow
 msgid "Popoloca; San Felipe Otlaltepec"
-msgstr ""
+msgstr "Popoloca; Otlaltepec"
 
 #. name for pox
 msgid "Polabian"
@@ -21160,7 +21160,7 @@ msgstr ""
 
 #. name for pps
 msgid "Popoloca; San Luís Temalacayuca"
-msgstr ""
+msgstr "Popoloca; Temalacayuca"
 
 #. name for ppt
 msgid "Pare"

setup/iso_639/es.po

@@ -9,13 +9,13 @@ msgstr ""
 "Project-Id-Version: calibre\n"
 "Report-Msgid-Bugs-To: FULL NAME <EMAIL@ADDRESS>\n"
 "POT-Creation-Date: 2011-11-25 14:01+0000\n"
-"PO-Revision-Date: 2012-12-24 08:05+0000\n"
+"PO-Revision-Date: 2012-12-28 09:13+0000\n"
-"Last-Translator: Adolfo Jayme Barrientos <fitoschido@gmail.com>\n"
+"Last-Translator: Jellby <Unknown>\n"
 "Language-Team: Español; Castellano <>\n"
 "MIME-Version: 1.0\n"
 "Content-Type: text/plain; charset=UTF-8\n"
 "Content-Transfer-Encoding: 8bit\n"
-"X-Launchpad-Export-Date: 2012-12-25 04:46+0000\n"
+"X-Launchpad-Export-Date: 2012-12-29 05:00+0000\n"
 "X-Generator: Launchpad (build 16378)\n"
 
 #. name for aaa
@@ -9584,7 +9584,7 @@ msgstr "Holikachuk"
 
 #. name for hoj
 msgid "Hadothi"
-msgstr "Hadothi"
+msgstr "Hadoti"
 
 #. name for hol
 msgid "Holu"
@@ -11796,7 +11796,7 @@ msgstr ""
 
 #. name for khq
 msgid "Songhay; Koyra Chiini"
-msgstr ""
+msgstr "Songhay koyra chiini"
 
 #. name for khr
 msgid "Kharia"

setup/translations.py

@@ -227,9 +227,22 @@ class GetTranslations(Translations): # {{{
                 ans.append(line.split()[-1])
         return ans
 
+    def resolve_conflicts(self):
+        conflict = False
+        for line in subprocess.check_output(['bzr', 'status']).splitlines():
+            if line == 'conflicts:':
+                conflict = True
+                break
+        if not conflict:
+            raise Exception('bzr merge failed and no conflicts found')
+        subprocess.check_call(['bzr', 'resolve', '--take-other'])
+
     def run(self, opts):
         if not self.modified_translations:
+            try:
                 subprocess.check_call(['bzr', 'merge', self.BRANCH])
+            except subprocess.CalledProcessError:
+                self.resolve_conflicts()
         self.check_for_errors()
 
         if self.modified_translations:

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, 12)
+numeric_version = (0, 9, 13)
 __version__   = u'.'.join(map(unicode, numeric_version))
 __author__    = u"Kovid Goyal <kovid@kovidgoyal.net>"
 

src/calibre/customize/builtins.py

@@ -1529,6 +1529,15 @@ class StoreNextoStore(StoreBase):
     formats = ['EPUB', 'MOBI', 'PDF']
     affiliate = True
 
+class StoreNookUKStore(StoreBase):
+    name = 'Nook UK'
+    author = 'John Schember'
+    description = u'Barnes & Noble S.à r.l, a subsidiary of Barnes & Noble, Inc., a leading retailer of content, digital media and educational products, is proud to bring the award-winning NOOK® reading experience and a leading digital bookstore to the UK.'
+    actual_plugin = 'calibre.gui2.store.stores.nook_uk_plugin:NookUKStore'
+
+    headquarters = 'UK'
+    formats = ['NOOK']
+
 class StoreOpenBooksStore(StoreBase):
     name = 'Open Books'
     description = u'Comprehensive listing of DRM free ebooks from a variety of sources provided by users of calibre.'
@@ -1660,7 +1669,7 @@ plugins += [
     StoreAmazonITKindleStore,
     StoreAmazonUKKindleStore,
     StoreBaenWebScriptionStore,
-    StoreBNStore, StoreSonyStore,
+    StoreBNStore,
     StoreBeWriteStore,
     StoreBiblioStore,
     StoreBookotekaStore,
@@ -1686,12 +1695,14 @@ plugins += [
     StoreMillsBoonUKStore,
     StoreMobileReadStore,
     StoreNextoStore,
+    StoreNookUKStore,
     StoreOpenBooksStore,
     StoreOzonRUStore,
     StorePragmaticBookshelfStore,
     StorePublioStore,
     StoreRW2010Store,
     StoreSmashwordsStore,
+    StoreSonyStore,
     StoreVirtualoStore,
     StoreWaterstonesUKStore,
     StoreWeightlessBooksStore,

src/calibre/devices/android/driver.py

@@ -48,6 +48,7 @@ class ANDROID(USBMS):
                        0x2910 : HTC_BCDS,
                        0xe77  : HTC_BCDS,
                        0xff9  : HTC_BCDS,
+                       0x0001 : [0x255],
             },
 
             # Eken
@@ -190,7 +191,7 @@ class ANDROID(USBMS):
             0x10a9 : { 0x6050 : [0x227] },
 
             # Prestigio
-            0x2207 : { 0 : [0x222] },
+            0x2207 : { 0 : [0x222], 0x10 : [0x222] },
 
             }
     EBOOK_DIR_MAIN = ['eBooks/import', 'wordplayer/calibretransfer', 'Books',
@@ -212,7 +213,8 @@ class ANDROID(USBMS):
             'VIZIO', 'GOOGLE', 'FREESCAL', 'KOBO_INC', 'LENOVO', 'ROCKCHIP',
             'POCKET', 'ONDA_MID', 'ZENITHIN', 'INGENIC', 'PMID701C', 'PD',
             'PMP5097C', 'MASS', 'NOVO7', 'ZEKI', 'COBY', 'SXZ', 'USB_2.0',
-            'COBY_MID', 'VS', 'AINOL', 'TOPWISE', 'PAD703', 'NEXT8D12']
+            'COBY_MID', 'VS', 'AINOL', 'TOPWISE', 'PAD703', 'NEXT8D12',
+            'MEDIATEK']
     WINDOWS_MAIN_MEM = ['ANDROID_PHONE', 'A855', 'A853', 'INC.NEXUS_ONE',
             '__UMS_COMPOSITE', '_MB200', 'MASS_STORAGE', '_-_CARD', 'SGH-I897',
             'GT-I9000', 'FILE-STOR_GADGET', 'SGH-T959_CARD', 'SGH-T959', 'SAMSUNG_ANDROID',
@@ -232,7 +234,7 @@ class ANDROID(USBMS):
             'THINKPAD_TABLET', 'SGH-T989', 'YP-G70', 'STORAGE_DEVICE',
             'ADVANCED', 'SGH-I727', 'USB_FLASH_DRIVER', 'ANDROID',
             'S5830I_CARD', 'MID7042', 'LINK-CREATE', '7035', 'VIEWPAD_7E',
-            'NOVO7', 'MB526', '_USB#WYK7MSF8KE', 'TABLET_PC', 'F']
+            'NOVO7', 'MB526', '_USB#WYK7MSF8KE', 'TABLET_PC', 'F', 'MT65XX_MS']
     WINDOWS_CARD_A_MEM = ['ANDROID_PHONE', 'GT-I9000_CARD', 'SGH-I897',
             'FILE-STOR_GADGET', 'SGH-T959_CARD', 'SGH-T959', 'SAMSUNG_ANDROID', 'GT-P1000_CARD',
             'A70S', 'A101IT', '7', 'INCREDIBLE', 'A7EB', 'SGH-T849_CARD',

src/calibre/devices/mtp/unix/libmtp.c

@@ -734,6 +734,7 @@ initlibmtp(void) {
     // who designs a library without anyway to control/redirect the debugging
     // output, and hardcoded paths that cannot be changed?
     int bak, new;
+    fprintf(stdout, "\n"); // This is needed, without it, for some odd reason the code below causes stdout to buffer all output after it is restored, rather than using line buffering, and setlinebuf does not work.
     fflush(stdout);
     bak = dup(STDOUT_FILENO);
     new = open("/dev/null", O_WRONLY);

src/calibre/ebooks/conversion/plugins/pdf_output.py

@@ -8,11 +8,13 @@ __docformat__ = 'restructuredtext en'
 Convert OEB ebook format to PDF.
 '''
 
-import glob
+import glob, os
-import os
 
-from calibre.customize.conversion import OutputFormatPlugin, \
+from PyQt4.Qt import QRawFont, QFont
-    OptionRecommendation
+
+from calibre.constants import iswindows
+from calibre.customize.conversion import (OutputFormatPlugin,
+    OptionRecommendation)
 from calibre.ptempfile import TemporaryDirectory
 
 UNITS = ['millimeter', 'centimeter', 'point', 'inch' , 'pica' , 'didot',
@@ -91,12 +93,14 @@ class PDFOutput(OutputFormatPlugin):
         OptionRecommendation(name='pdf_mono_font_size',
             recommended_value=16, help=_(
                 'The default font size for monospaced text')),
-        # OptionRecommendation(name='old_pdf_engine', recommended_value=False,
+        OptionRecommendation(name='pdf_mark_links', recommended_value=False,
-        #     help=_('Use the old, less capable engine to generate the PDF')),
+            help=_('Surround all links with a red box, useful for debugging.')),
-        # OptionRecommendation(name='uncompressed_pdf',
+        OptionRecommendation(name='old_pdf_engine', recommended_value=False,
-        #     recommended_value=False, help=_(
+            help=_('Use the old, less capable engine to generate the PDF')),
-        #         'Generate an uncompressed PDF, useful for debugging, '
+        OptionRecommendation(name='uncompressed_pdf',
-        #         'only works with the new PDF engine.')),
+            recommended_value=False, help=_(
+                'Generate an uncompressed PDF, useful for debugging, '
+                'only works with the new PDF engine.')),
         ])
 
     def convert(self, oeb_book, output_path, input_plugin, opts, log):
@@ -134,7 +138,7 @@ class PDFOutput(OutputFormatPlugin):
         '''
         from calibre.ebooks.oeb.base import urlnormalize
         from calibre.gui2 import must_use_qt
-        from calibre.utils.fonts.utils import get_font_names, remove_embed_restriction
+        from calibre.utils.fonts.utils import remove_embed_restriction
         from PyQt4.Qt import QFontDatabase, QByteArray
 
         # First find all @font-face rules and remove them, adding the embedded
@@ -164,11 +168,13 @@ class PDFOutput(OutputFormatPlugin):
                     except:
                         continue
                     must_use_qt()
-                    QFontDatabase.addApplicationFontFromData(QByteArray(raw))
+                    fid = QFontDatabase.addApplicationFontFromData(QByteArray(raw))
-                    try:
-                        family_name = get_font_names(raw)[0]
-                    except:
                     family_name = None
+                    if fid > -1:
+                        try:
+                            family_name = unicode(QFontDatabase.applicationFontFamilies(fid)[0])
+                        except (IndexError, KeyError):
+                            pass
                     if family_name:
                         family_map[icu_lower(font_family)] = family_name
 
@@ -177,6 +183,7 @@ class PDFOutput(OutputFormatPlugin):
 
         # Now map the font family name specified in the css to the actual
         # family name of the embedded font (they may be different in general).
+        font_warnings = set()
         for item in self.oeb.manifest:
             if not hasattr(item.data, 'cssRules'): continue
             for i, rule in enumerate(item.data.cssRules):
@@ -188,15 +195,30 @@ class PDFOutput(OutputFormatPlugin):
                     k = icu_lower(val[i].value)
                     if k in family_map:
                         val[i].value = family_map[k]
+                if iswindows:
+                    # On windows, Qt uses GDI which does not support OpenType
+                    # (CFF) fonts, so we need to nuke references to OpenType
+                    # fonts. Note that you could compile QT with configure
+                    # -directwrite, but that requires atleast Vista SP2
+                    for i in xrange(val.length):
+                        family = val[i].value
+                        if family:
+                            f = QRawFont.fromFont(QFont(family))
+                            if len(f.fontTable('head')) == 0:
+                                if family not in font_warnings:
+                                    self.log.warn('Ignoring unsupported font: %s'
+                                                %family)
+                                    font_warnings.add(family)
+                                # Either a bitmap or (more likely) a CFF font
+                                val[i].value = 'times'
 
     def convert_text(self, oeb_book):
-        from calibre.utils.config import tweaks
+        from calibre.ebooks.metadata.opf2 import OPF
-        if tweaks.get('new_pdf_engine', False):
+        if self.opts.old_pdf_engine:
-            from calibre.ebooks.pdf.render.from_html import PDFWriter
+            from calibre.ebooks.pdf.writer import PDFWriter
             PDFWriter
         else:
-            from calibre.ebooks.pdf.writer import PDFWriter
+            from calibre.ebooks.pdf.render.from_html import PDFWriter
-        from calibre.ebooks.metadata.opf2 import OPF
 
         self.log.debug('Serializing oeb input to disk for processing...')
         self.get_cover_data()
@@ -231,7 +253,15 @@ class PDFOutput(OutputFormatPlugin):
         out_stream.seek(0)
         out_stream.truncate()
         self.log.debug('Rendering pages to PDF...')
+        import time
+        st = time.time()
+        if False:
+            import cProfile
+            cProfile.runctx('writer.dump(items, out_stream, PDFMetadata(self.metadata))',
+                        globals(), locals(), '/tmp/profile')
+        else:
             writer.dump(items, out_stream, PDFMetadata(self.metadata))
+        self.log('Rendered PDF in %g seconds:'%(time.time()-st))
 
         if close:
             out_stream.close()

src/calibre/ebooks/html/input.py

@@ -17,7 +17,7 @@ from urllib import unquote
 
 from calibre.ebooks.chardet import detect_xml_encoding
 from calibre.constants import iswindows
-from calibre import unicode_path, as_unicode
+from calibre import unicode_path, as_unicode, replace_entities
 
 class Link(object):
     '''
@@ -147,6 +147,7 @@ class HTMLFile(object):
                 url = match.group(i)
                 if url:
                     break
+            url = replace_entities(url)
             try:
                 link = self.resolve(url)
             except ValueError:

src/calibre/ebooks/lrf/__init__.py

@@ -41,7 +41,6 @@ def find_custom_fonts(options, logger):
     if options.serif_family:
         f = family(options.serif_family)
         fonts['serif'] = font_scanner.legacy_fonts_for_family(f)
-        print (111111, fonts['serif'])
         if not fonts['serif']:
             logger.warn('Unable to find serif family %s'%f)
     if options.sans_family:

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

@@ -9,7 +9,10 @@ __docformat__ = 'restructuredtext en'
 
 import codecs, zlib
 from io import BytesIO
-from struct import pack
+
+from calibre.constants import plugins, ispy3
+
+pdf_float = plugins['speedup'][0].pdf_float
 
 EOL = b'\n'
 
@@ -51,15 +54,25 @@ PAPER_SIZES = {k:globals()[k.upper()] for k in ('a0 a1 a2 a3 a4 a5 a6 b0 b1 b2'
 
 # Basic PDF datatypes {{{
 
+ic = str if ispy3 else unicode
+icb = (lambda x: str(x).encode('ascii')) if ispy3 else bytes
+
+def fmtnum(o):
+    if isinstance(o, float):
+        return pdf_float(o)
+    return ic(o)
+
 def serialize(o, stream):
-    if hasattr(o, 'pdf_serialize'):
+    if isinstance(o, float):
+        stream.write_raw(pdf_float(o).encode('ascii'))
+    elif isinstance(o, (int, long)):
+        stream.write_raw(icb(o))
+    elif hasattr(o, 'pdf_serialize'):
         o.pdf_serialize(stream)
-    elif isinstance(o, bool):
-        stream.write(b'true' if o else b'false')
-    elif isinstance(o, (int, long, float)):
-        stream.write(type(u'')(o).encode('ascii'))
     elif o is None:
-        stream.write(b'null')
+        stream.write_raw(b'null')
+    elif isinstance(o, bool):
+        stream.write_raw(b'true' if o else b'false')
     else:
         raise ValueError('Unknown object: %r'%o)
 
@@ -85,19 +98,13 @@ class String(unicode):
             raw = codecs.BOM_UTF16_BE + s.encode('utf-16-be')
         stream.write(b'('+raw+b')')
 
-class GlyphIndex(int):
-
-    def pdf_serialize(self, stream):
-        byts = bytearray(pack(b'>H', self))
-        stream.write('<%s>'%''.join(map(
-            lambda x: bytes(hex(x)[2:]).rjust(2, b'0'), byts)))
-
 class Dictionary(dict):
 
     def pdf_serialize(self, stream):
         stream.write(b'<<' + EOL)
         sorted_keys = sorted(self.iterkeys(),
-                    key=lambda x:((' ' if x == 'Type' else '')+x))
+                             key=lambda x:({'Type':'1', 'Subtype':'2'}.get(
+                                 x, x)+x))
         for k in sorted_keys:
             serialize(Name(k), stream)
             stream.write(b' ')
@@ -161,6 +168,9 @@ class Stream(BytesIO):
         super(Stream, self).write(raw if isinstance(raw, bytes) else
                                   raw.encode('ascii'))
 
+    def write_raw(self, raw):
+        BytesIO.write(self, raw)
+
 class Reference(object):
 
     def __init__(self, num, obj):
@@ -169,5 +179,11 @@ class Reference(object):
     def pdf_serialize(self, stream):
         raw = '%d 0 R'%self.num
         stream.write(raw.encode('ascii'))
+
+    def __repr__(self):
+        return '%d 0 R'%self.num
+
+    def __str__(self):
+        return repr(self)
 # }}}
 

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

@@ -8,24 +8,27 @@ __copyright__ = '2012, Kovid Goyal <kovid at kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
 import sys, traceback
-from math import sqrt
 from collections import namedtuple
 from functools import wraps, partial
+from future_builtins import map
 
 import sip
-from PyQt4.Qt import (QPaintEngine, QPaintDevice, Qt, QApplication, QPainter,
+from PyQt4.Qt import (QPaintEngine, QPaintDevice, Qt, QTransform, QBrush)
-                      QTransform, QPainterPath, QImage, QByteArray, QBuffer,
-                      qRgba)
 
 from calibre.constants import plugins
-from calibre.ebooks.pdf.render.serialize import (Color, PDFStream, Path)
+from calibre.ebooks.pdf.render.serialize import (PDFStream, Path)
-from calibre.ebooks.pdf.render.common import inch, A4
+from calibre.ebooks.pdf.render.common import inch, A4, fmtnum
-from calibre.utils.fonts.sfnt.container import Sfnt
+from calibre.ebooks.pdf.render.graphics import convert_path, Graphics
+from calibre.utils.fonts.sfnt.container import Sfnt, UnsupportedFont
 from calibre.utils.fonts.sfnt.metrics import FontMetrics
 
 Point = namedtuple('Point', 'x y')
 ColorState = namedtuple('ColorState', 'color opacity do')
 
+def repr_transform(t):
+    vals = map(fmtnum, (t.m11(), t.m12(), t.m21(), t.m22(), t.dx(), t.dy()))
+    return '[%s]'%' '.join(vals)
+
 def store_error(func):
 
     @wraps(func)
@@ -38,146 +41,6 @@ def store_error(func):
 
     return errh
 
-class GraphicsState(object): # {{{
-
-    def __init__(self):
-        self.ops = {}
-        self.initial_state = {
-            'fill': ColorState(Color(0., 0., 0., 1.), 1.0, False),
-            'transform': QTransform(),
-            'dash': [],
-            'line_width': 0,
-            'stroke': ColorState(Color(0., 0., 0., 1.), 1.0, True),
-            'line_cap': 'flat',
-            'line_join': 'miter',
-            'clip': (Qt.NoClip, QPainterPath()),
-        }
-        self.current_state = self.initial_state.copy()
-
-    def reset(self):
-        self.current_state = self.initial_state.copy()
-
-    def update_color_state(self, which, color=None, opacity=None,
-                           brush_style=None, pen_style=None):
-        current = self.ops.get(which, self.current_state[which])
-        n = ColorState(*current)
-        if color is not None:
-            n = n._replace(color=Color(*color.getRgbF()))
-        if opacity is not None:
-            n = n._replace(opacity=opacity)
-        if opacity is not None:
-            opacity *= n.color.opacity
-        if brush_style is not None:
-            if which == 'fill':
-                do = (False if opacity == 0.0 or brush_style == Qt.NoBrush else
-                    True)
-            else:
-                do = (False if opacity == 0.0 or brush_style == Qt.NoBrush or
-                    pen_style == Qt.NoPen else True)
-            n = n._replace(do=do)
-        self.ops[which] = n
-
-    def read(self, state):
-        flags = state.state()
-
-        if flags & QPaintEngine.DirtyTransform:
-            self.ops['transform'] = state.transform()
-
-        # TODO: Add support for brush patterns
-        if flags & QPaintEngine.DirtyBrush:
-            brush = state.brush()
-            color = brush.color()
-            self.update_color_state('fill', color=color,
-                                    brush_style=brush.style())
-
-        if flags & QPaintEngine.DirtyPen:
-            pen = state.pen()
-            brush = pen.brush()
-            color = pen.color()
-            self.update_color_state('stroke', color, brush_style=brush.style(),
-                                    pen_style=pen.style())
-            ps = {Qt.DashLine:[3], Qt.DotLine:[1,2], Qt.DashDotLine:[3,2,1,2],
-                  Qt.DashDotDotLine:[3, 2, 1, 2, 1, 2]}.get(pen.style(), [])
-            self.ops['dash'] = ps
-            self.ops['line_width'] = pen.widthF()
-            self.ops['line_cap'] = {Qt.FlatCap:'flat', Qt.RoundCap:'round',
-                            Qt.SquareCap:'square'}.get(pen.capStyle(), 'flat')
-            self.ops['line_join'] = {Qt.MiterJoin:'miter', Qt.RoundJoin:'round',
-                            Qt.BevelJoin:'bevel'}.get(pen.joinStyle(), 'miter')
-
-        if flags & QPaintEngine.DirtyOpacity:
-            self.update_color_state('fill', opacity=state.opacity())
-            self.update_color_state('stroke', opacity=state.opacity())
-
-        if flags & QPaintEngine.DirtyClipPath or flags & QPaintEngine.DirtyClipRegion:
-            self.ops['clip'] = True
-
-    def __call__(self, engine):
-        if not self.ops:
-            return
-        pdf = engine.pdf
-        ops = self.ops
-        current_transform = self.current_state['transform']
-        transform_changed = 'transform' in ops and ops['transform'] != current_transform
-        reset_stack = transform_changed or 'clip' in ops
-
-        if reset_stack:
-            pdf.restore_stack()
-            pdf.save_stack()
-            # Since we have reset the stack we need to re-apply all previous
-            # operations, that are different from the default value (clip is
-            # handled separately).
-            for op in set(self.initial_state) - {'clip'}:
-                if op in ops: # These will be applied below
-                    self.current_state[op] = self.initial_state[op]
-                elif self.current_state[op] != self.initial_state[op]:
-                    self.apply(op, self.current_state[op], engine, pdf)
-
-        # Now apply the new operations
-        for op, val in ops.iteritems():
-            if op != 'clip' and self.current_state[op] != val:
-                self.apply(op, val, engine, pdf)
-                self.current_state[op] = val
-
-        if 'clip' in ops:
-            # Get the current clip
-            path = engine.painter().clipPath()
-            if not path.isEmpty():
-                engine.add_clip(path)
-        self.ops = {}
-
-    def apply(self, op, val, engine, pdf):
-        getattr(self, 'apply_'+op)(val, engine, pdf)
-
-    def apply_transform(self, val, engine, pdf):
-        if not val.isIdentity():
-            pdf.transform(val)
-
-    def apply_stroke(self, val, engine, pdf):
-        self.apply_color_state('stroke', val, engine, pdf)
-
-    def apply_fill(self, val, engine, pdf):
-        self.apply_color_state('fill', val, engine, pdf)
-
-    def apply_color_state(self, which, val, engine, pdf):
-        color = val.color._replace(opacity=val.opacity*val.color.opacity)
-        getattr(pdf, 'set_%s_color'%which)(color)
-        setattr(engine, 'do_%s'%which, val.do)
-
-    def apply_dash(self, val, engine, pdf):
-        pdf.set_dash(val)
-
-    def apply_line_width(self, val, engine, pdf):
-        pdf.set_line_width(val)
-
-    def apply_line_cap(self, val, engine, pdf):
-        pdf.set_line_cap(val)
-
-    def apply_line_join(self, val, engine, pdf):
-        pdf.set_line_join(val)
-
-# }}}
-
 class Font(FontMetrics):
 
     def __init__(self, sfnt):
@@ -186,12 +49,21 @@ class Font(FontMetrics):
 
 class PdfEngine(QPaintEngine):
 
+    FEATURES = QPaintEngine.AllFeatures & ~(
+        QPaintEngine.PorterDuff | QPaintEngine.PerspectiveTransform
+        | QPaintEngine.ObjectBoundingModeGradients
+        | QPaintEngine.LinearGradientFill
+        | QPaintEngine.RadialGradientFill
+        | QPaintEngine.ConicalGradientFill
+    )
+
     def __init__(self, file_object, page_width, page_height, left_margin,
                  top_margin, right_margin, bottom_margin, width, height,
-                 errors=print, debug=print, compress=True):
+                 errors=print, debug=print, compress=True,
-        QPaintEngine.__init__(self, self.features)
+                 mark_links=False):
+        QPaintEngine.__init__(self, self.FEATURES)
         self.file_object = file_object
-        self.compress = compress
+        self.compress, self.mark_links = compress, mark_links
         self.page_height, self.page_width = page_height, page_width
         self.left_margin, self.top_margin = left_margin, top_margin
         self.right_margin, self.bottom_margin = right_margin, bottom_margin
@@ -210,49 +82,48 @@ class PdfEngine(QPaintEngine):
                             self.bottom_margin) / self.pixel_height
 
         self.pdf_system = QTransform(sx, 0, 0, -sy, dx, dy)
-        self.do_stroke = True
+        self.graphics = Graphics()
-        self.do_fill = False
-        self.scale = sqrt(sy**2 + sx**2)
-        self.xscale, self.yscale = sx, sy
-        self.graphics_state = GraphicsState()
         self.errors_occurred = False
         self.errors, self.debug = errors, debug
         self.fonts = {}
-        i = QImage(1, 1, QImage.Format_ARGB32)
-        i.fill(qRgba(0, 0, 0, 255))
-        self.alpha_bit = i.constBits().asstring(4).find(b'\xff')
         self.current_page_num = 1
         self.current_page_inited = False
         self.qt_hack, err = plugins['qt_hack']
         if err:
             raise RuntimeError('Failed to load qt_hack with err: %s'%err)
 
-    def init_page(self):
+    def apply_graphics_state(self):
-        self.pdf.transform(self.pdf_system)
+        self.graphics(self.pdf_system, self.painter())
-        self.pdf.set_rgb_colorspace()
+
-        width = self.painter().pen().widthF() if self.isActive() else 0
+    def resolve_fill(self, rect):
-        self.pdf.set_line_width(width)
+        self.graphics.resolve_fill(rect, self.pdf_system,
-        self.do_stroke = True
+                                   self.painter().transform())
-        self.do_fill = False
-        self.graphics_state.reset()
-        self.pdf.save_stack()
-        self.current_page_inited = True
 
     @property
-    def features(self):
+    def do_fill(self):
-        return (QPaintEngine.Antialiasing | QPaintEngine.AlphaBlend |
+        return self.graphics.current_state.do_fill
-                QPaintEngine.ConstantOpacity | QPaintEngine.PainterPaths |
+
-                QPaintEngine.PaintOutsidePaintEvent |
+    @property
-                QPaintEngine.PrimitiveTransform)
+    def do_stroke(self):
+        return self.graphics.current_state.do_stroke
+
+    def init_page(self):
+        self.pdf.transform(self.pdf_system)
+        self.graphics.reset()
+        self.pdf.save_stack()
+        self.current_page_inited = True
 
     def begin(self, device):
         if not hasattr(self, 'pdf'):
             try:
                 self.pdf = PDFStream(self.file_object, (self.page_width,
-                                                        self.page_height),
+                        self.page_height), compress=self.compress,
-                                    compress=self.compress)
+                                     mark_links=self.mark_links,
+                                     debug=self.debug)
+                self.graphics.begin(self.pdf)
             except:
-                self.errors.append(traceback.format_exc())
+                self.errors(traceback.format_exc())
+                self.errors_occurred = True
                 return False
         return True
 
@@ -268,7 +139,8 @@ class PdfEngine(QPaintEngine):
             self.end_page()
             self.pdf.end()
         except:
-            self.errors.append(traceback.format_exc())
+            self.errors(traceback.format_exc())
+            self.errors_occurred = True
             return False
         finally:
             self.pdf = self.file_object = None
@@ -277,139 +149,63 @@ class PdfEngine(QPaintEngine):
     def type(self):
         return QPaintEngine.Pdf
 
+    def add_image(self, img, cache_key):
+        if img.isNull(): return
+        return self.pdf.add_image(img, cache_key)
+
+    @store_error
+    def drawTiledPixmap(self, rect, pixmap, point):
+        self.apply_graphics_state()
+        brush = QBrush(pixmap)
+        bl = rect.topLeft()
+        color, opacity, pattern, do_fill = self.graphics.convert_brush(
+            brush, bl-point, 1.0, self.pdf_system,
+            self.painter().transform())
+        self.pdf.save_stack()
+        self.pdf.apply_fill(color, pattern)
+        self.pdf.draw_rect(bl.x(), bl.y(), rect.width(), rect.height(),
+                            stroke=False, fill=True)
+        self.pdf.restore_stack()
+
     @store_error
     def drawPixmap(self, rect, pixmap, source_rect):
-        self.graphics_state(self)
+        self.apply_graphics_state()
         source_rect = source_rect.toRect()
         pixmap = (pixmap if source_rect == pixmap.rect() else
                   pixmap.copy(source_rect))
         image = pixmap.toImage()
         ref = self.add_image(image, pixmap.cacheKey())
         if ref is not None:
-            self.pdf.draw_image(rect.x(), rect.height()+rect.y(), rect.width(),
+            self.pdf.draw_image(rect.x(), rect.y(), rect.width(),
-                                -rect.height(), ref)
+                                rect.height(), ref)
 
     @store_error
     def drawImage(self, rect, image, source_rect, flags=Qt.AutoColor):
-        self.graphics_state(self)
+        self.apply_graphics_state()
         source_rect = source_rect.toRect()
         image = (image if source_rect == image.rect() else
                  image.copy(source_rect))
         ref = self.add_image(image, image.cacheKey())
         if ref is not None:
-            self.pdf.draw_image(rect.x(), rect.height()+rect.y(), rect.width(),
+            self.pdf.draw_image(rect.x(), rect.y(), rect.width(),
-                                -rect.height(), ref)
+                                rect.height(), ref)
-
-    def add_image(self, img, cache_key):
-        if img.isNull(): return
-        ref = self.pdf.get_image(cache_key)
-        if ref is not None:
-            return ref
-
-        fmt = img.format()
-        image = QImage(img)
-        if (image.depth() == 1 and img.colorTable().size() == 2 and
-            img.colorTable().at(0) == QColor(Qt.black).rgba() and
-            img.colorTable().at(1) == QColor(Qt.white).rgba()):
-            if fmt == QImage.Format_MonoLSB:
-                image = image.convertToFormat(QImage.Format_Mono)
-            fmt = QImage.Format_Mono
-        else:
-            if (fmt != QImage.Format_RGB32 and fmt != QImage.Format_ARGB32):
-                image = image.convertToFormat(QImage.Format_ARGB32)
-                fmt = QImage.Format_ARGB32
-
-        w = image.width()
-        h = image.height()
-        d = image.depth()
-
-        if fmt == QImage.Format_Mono:
-            bytes_per_line = (w + 7) >> 3
-            data = image.constBits().asstring(bytes_per_line * h)
-            return self.pdf.write_image(data, w, h, d, cache_key=cache_key)
-
-        ba = QByteArray()
-        buf = QBuffer(ba)
-        image.save(buf, 'jpeg', 94)
-        data = bytes(ba.data())
-        has_alpha = has_mask = False
-        soft_mask = mask = None
-
-        if fmt == QImage.Format_ARGB32:
-            tmask = image.constBits().asstring(4*w*h)[self.alpha_bit::4]
-            sdata = bytearray(tmask)
-            vals = set(sdata)
-            vals.discard(255)
-            has_mask = bool(vals)
-            vals.discard(0)
-            has_alpha = bool(vals)
-
-        if has_alpha:
-            soft_mask = self.pdf.write_image(tmask, w, h, 8)
-        elif has_mask:
-            # dither the soft mask to 1bit and add it. This also helps PDF
-            # viewers without transparency support
-            bytes_per_line = (w + 7) >> 3
-            mdata = bytearray(0 for i in xrange(bytes_per_line * h))
-            spos = mpos = 0
-            for y in xrange(h):
-                for x in xrange(w):
-                    if sdata[spos]:
-                        mdata[mpos + x>>3] |= (0x80 >> (x&7))
-                    spos += 1
-                mpos += bytes_per_line
-            mdata = bytes(mdata)
-            mask = self.pdf.write_image(mdata, w, h, 1)
-
-        return self.pdf.write_image(data, w, h, 32, mask=mask, dct=True,
-                                    soft_mask=soft_mask, cache_key=cache_key)
 
     @store_error
     def updateState(self, state):
-        self.graphics_state.read(state)
+        self.graphics.update_state(state, self.painter())
-
-    def convert_path(self, path):
-        p = Path()
-        i = 0
-        while i < path.elementCount():
-            elem = path.elementAt(i)
-            em = (elem.x, elem.y)
-            i += 1
-            if elem.isMoveTo():
-                p.move_to(*em)
-            elif elem.isLineTo():
-                p.line_to(*em)
-            elif elem.isCurveTo():
-                added = False
-                if path.elementCount() > i+1:
-                    c1, c2 = path.elementAt(i), path.elementAt(i+1)
-                    if (c1.type == path.CurveToDataElement and c2.type ==
-                        path.CurveToDataElement):
-                        i += 2
-                        p.curve_to(em[0], em[1], c1.x, c1.y, c2.x, c2.y)
-                        added = True
-                if not added:
-                    raise ValueError('Invalid curve to operation')
-        return p
 
     @store_error
     def drawPath(self, path):
-        self.graphics_state(self)
+        self.apply_graphics_state()
-        p = self.convert_path(path)
+        p = convert_path(path)
         fill_rule = {Qt.OddEvenFill:'evenodd',
                     Qt.WindingFill:'winding'}[path.fillRule()]
         self.pdf.draw_path(p, stroke=self.do_stroke,
                                 fill=self.do_fill, fill_rule=fill_rule)
 
-    def add_clip(self, path):
-        p = self.convert_path(path)
-        fill_rule = {Qt.OddEvenFill:'evenodd',
-                    Qt.WindingFill:'winding'}[path.fillRule()]
-        self.pdf.add_clip(p, fill_rule=fill_rule)
-
     @store_error
     def drawPoints(self, points):
-        self.graphics_state(self)
+        self.apply_graphics_state()
         p = Path()
         for point in points:
             p.move_to(point.x(), point.y())
@@ -418,15 +214,21 @@ class PdfEngine(QPaintEngine):
 
     @store_error
     def drawRects(self, rects):
-        self.graphics_state(self)
+        self.apply_graphics_state()
+        with self.graphics:
             for rect in rects:
+                self.resolve_fill(rect)
                 bl = rect.topLeft()
                 self.pdf.draw_rect(bl.x(), bl.y(), rect.width(), rect.height(),
                                 stroke=self.do_stroke, fill=self.do_fill)
 
     def create_sfnt(self, text_item):
         get_table = partial(self.qt_hack.get_sfnt_table, text_item)
+        try:
             ans = Font(Sfnt(get_table))
+        except UnsupportedFont as e:
+            raise UnsupportedFont('The font %s is not a valid sfnt. Error: %s'%(
+                text_item.font().family(), e))
         glyph_map = self.qt_hack.get_glyph_map(text_item)
         gm = {}
         for uc, glyph_id in enumerate(glyph_map):
@@ -438,7 +240,7 @@ class PdfEngine(QPaintEngine):
     @store_error
     def drawTextItem(self, point, text_item):
         # super(PdfEngine, self).drawTextItem(point, text_item)
-        self.graphics_state(self)
+        self.apply_graphics_state()
         gi = self.qt_hack.get_glyphs(point, text_item)
         if not gi.indices:
             sip.delete(gi)
@@ -453,23 +255,19 @@ class PdfEngine(QPaintEngine):
             except (KeyError, ValueError):
                 pass
         glyphs = []
-        pdf_pos = point
+        last_x = last_y = 0
-        first_baseline = None
         for i, pos in enumerate(gi.positions):
-            if first_baseline is None:
+            x, y = pos.x(), pos.y()
-                first_baseline = pos.y()
+            glyphs.append((x-last_x, last_y - y, gi.indices[i]))
-            glyph_pos = pos
+            last_x, last_y = x, y
-            delta = glyph_pos - pdf_pos
-            glyphs.append((delta.x(), pos.y()-first_baseline, gi.indices[i]))
-            pdf_pos = glyph_pos
 
-        self.pdf.draw_glyph_run([1, 0, 0, -1, point.x(),
+        self.pdf.draw_glyph_run([gi.stretch, 0, 0, -1, 0, 0], gi.size, metrics,
-            point.y()], gi.size, metrics, glyphs)
+                                glyphs)
         sip.delete(gi)
 
     @store_error
     def drawPolygon(self, points, mode):
-        self.graphics_state(self)
+        self.apply_graphics_state()
         if not points: return
         p = Path()
         p.move_to(points[0].x(), points[0].y())
@@ -484,20 +282,31 @@ class PdfEngine(QPaintEngine):
     def set_metadata(self, *args, **kwargs):
         self.pdf.set_metadata(*args, **kwargs)
 
-    def __enter__(self):
+    def add_outline(self, toc):
-        self.pdf.save_stack()
+        self.pdf.links.add_outline(toc)
-        self.saved_ps = (self.do_stroke, self.do_fill)
 
-    def __exit__(self, *args):
+    def add_links(self, current_item, start_page, links, anchors):
-        self.do_stroke, self.do_fill = self.saved_ps
+        for pos in anchors.itervalues():
-        self.pdf.restore_stack()
+            pos['left'], pos['top'] = self.pdf_system.map(pos['left'], pos['top'])
+        for link in links:
+            pos = link[1]
+            llx = pos['left']
+            lly = pos['top'] + pos['height']
+            urx = pos['left'] + pos['width']
+            ury = pos['top']
+            llx, lly = self.pdf_system.map(llx, lly)
+            urx, ury = self.pdf_system.map(urx, ury)
+            link[1] = pos['column'] + start_page
+            link.append((llx, lly, urx, ury))
+        self.pdf.links.add(current_item, start_page, links, anchors)
 
 class PdfDevice(QPaintDevice): # {{{
 
 
     def __init__(self, file_object, page_size=A4, left_margin=inch,
                  top_margin=inch, right_margin=inch, bottom_margin=inch,
-                 xdpi=1200, ydpi=1200, errors=print, debug=print, compress=True):
+                 xdpi=1200, ydpi=1200, errors=print, debug=print,
+                 compress=True, mark_links=False):
         QPaintDevice.__init__(self)
         self.xdpi, self.ydpi = xdpi, ydpi
         self.page_width, self.page_height = page_size
@@ -506,7 +315,10 @@ class PdfDevice(QPaintDevice): # {{{
         self.engine = PdfEngine(file_object, self.page_width, self.page_height,
                                 left_margin, top_margin, right_margin,
                                 bottom_margin, self.width(), self.height(),
-                                errors=errors, debug=debug, compress=compress)
+                                errors=errors, debug=debug, compress=compress,
+                                mark_links=mark_links)
+        self.add_outline = self.engine.add_outline
+        self.add_links = self.engine.add_links
 
     def paintEngine(self):
         return self.engine
@@ -553,59 +365,4 @@ class PdfDevice(QPaintDevice): # {{{
 
 # }}}
 
-if __name__ == '__main__':
-    from PyQt4.Qt import (QBrush, QColor, QPoint, QPixmap)
-    QBrush, QColor, QPoint, QPixmap
-    app = QApplication([])
-    p = QPainter()
-    with open('/tmp/painter.pdf', 'wb') as f:
-        dev = PdfDevice(f, compress=False)
-        p.begin(dev)
-        dev.init_page()
-        xmax, ymax = p.viewport().width(), p.viewport().height()
-        try:
-            p.drawRect(0, 0, xmax, ymax)
-            # p.drawPolyline(QPoint(0, 0), QPoint(xmax, 0), QPoint(xmax, ymax),
-            #                QPoint(0, ymax), QPoint(0, 0))
-            # pp = QPainterPath()
-            # pp.addRect(0, 0, xmax, ymax)
-            # p.drawPath(pp)
-            # p.save()
-            # for i in xrange(3):
-            #     col = [0, 0, 0, 200]
-            #     col[i] = 255
-            #     p.setOpacity(0.3)
-            #     p.setBrush(QBrush(QColor(*col)))
-            #     p.drawRect(0, 0, xmax/10, xmax/10)
-            #     p.translate(xmax/10, xmax/10)
-            #     p.scale(1, 1.5)
-            # p.restore()
-
-            # # p.scale(2, 2)
-            # # p.rotate(45)
-            # p.drawPixmap(0, 0, 2048, 2048, QPixmap(I('library.png')))
-            # p.drawRect(0, 0, 2048, 2048)
-
-            # p.save()
-            # p.drawLine(0, 0, 5000, 0)
-            # p.rotate(45)
-            # p.drawLine(0, 0, 5000, 0)
-            # p.restore()
-
-            f = p.font()
-            f.setPointSize(20)
-            # f.setLetterSpacing(f.PercentageSpacing, 200)
-            # f.setUnderline(True)
-            # f.setOverline(True)
-            # f.setStrikeOut(True)
-            f.setFamily('Calibri')
-            p.setFont(f)
-            # p.setPen(QColor(0, 0, 255))
-            # p.scale(2, 2)
-            # p.rotate(45)
-            p.drawText(QPoint(300, 300), 'Some—text not By’s ū --- Д AV ff ff')
-        finally:
-            p.end()
-        if dev.engine.errors_occurred:
-            raise SystemExit(1)
 

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

@@ -20,7 +20,6 @@ from calibre.ebooks.oeb.display.webview import load_html
 from calibre.ebooks.pdf.render.common import (inch, cm, mm, pica, cicero,
                                               didot, PAPER_SIZES)
 from calibre.ebooks.pdf.render.engine import PdfDevice
-from calibre.ebooks.pdf.render.links import Links
 
 def get_page_size(opts, for_comic=False): # {{{
     use_profile = not (opts.override_profile_size or
@@ -143,7 +142,6 @@ class PDFWriter(QObject):
             self.view.page().mainFrame().setScrollBarPolicy(x,
                     Qt.ScrollBarAlwaysOff)
         self.report_progress = lambda x, y: x
-        self.links = Links()
 
     def dump(self, items, out_stream, pdf_metadata):
         opts = self.opts
@@ -156,7 +154,8 @@ class PDFWriter(QObject):
                              top_margin=0, right_margin=mr, bottom_margin=0,
                              xdpi=xdpi, ydpi=ydpi, errors=self.log.error,
                              debug=self.log.debug, compress=not
-                             opts.uncompressed_pdf)
+                             opts.uncompressed_pdf,
+                             mark_links=opts.pdf_mark_links)
 
         self.page.setViewportSize(QSize(self.doc.width(), self.doc.height()))
         self.render_queue = items
@@ -187,7 +186,9 @@ class PDFWriter(QObject):
         QTimer.singleShot(0, self.render_book)
         self.loop.exec_()
 
-        # TODO: Outline and links
+        if self.toc is not None and len(self.toc) > 0:
+            self.doc.add_outline(self.toc)
+
         self.painter.end()
 
         if self.doc.errors_occurred:
@@ -261,8 +262,7 @@ class PDFWriter(QObject):
         amap = self.bridge_value
         if not isinstance(amap, dict):
             amap = {'links':[], 'anchors':{}} # Some javascript error occurred
-        self.links.add(self.current_item, self.current_page_num, amap['links'],
+        start_page = self.current_page_num
-                       amap['anchors'])
 
         mf = self.view.page().mainFrame()
         while True:
@@ -278,3 +278,6 @@ class PDFWriter(QObject):
             if self.doc.errors_occurred:
                 break
 
+        self.doc.add_links(self.current_item, start_page, amap['links'],
+                           amap['anchors'])
+

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

@@ -0,0 +1,470 @@
+#!/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__ = '2012, Kovid Goyal <kovid at kovidgoyal.net>'
+__docformat__ = 'restructuredtext en'
+
+from math import sqrt
+from collections import namedtuple
+
+from PyQt4.Qt import (
+    QBrush, QPen, Qt, QPointF, QTransform, QPainterPath, QPaintEngine, QImage)
+
+from calibre.ebooks.pdf.render.common import (
+    Name, Array, fmtnum, Stream, Dictionary)
+from calibre.ebooks.pdf.render.serialize import Path
+
+def convert_path(path): # {{{
+    p = Path()
+    i = 0
+    while i < path.elementCount():
+        elem = path.elementAt(i)
+        em = (elem.x, elem.y)
+        i += 1
+        if elem.isMoveTo():
+            p.move_to(*em)
+        elif elem.isLineTo():
+            p.line_to(*em)
+        elif elem.isCurveTo():
+            added = False
+            if path.elementCount() > i+1:
+                c1, c2 = path.elementAt(i), path.elementAt(i+1)
+                if (c1.type == path.CurveToDataElement and c2.type ==
+                    path.CurveToDataElement):
+                    i += 2
+                    p.curve_to(em[0], em[1], c1.x, c1.y, c2.x, c2.y)
+                    added = True
+            if not added:
+                raise ValueError('Invalid curve to operation')
+    return p
+# }}}
+
+Brush = namedtuple('Brush', 'origin brush color')
+
+class TilingPattern(Stream):
+
+    def __init__(self, cache_key, matrix, w=8, h=8, paint_type=2, compress=False):
+        Stream.__init__(self, compress=compress)
+        self.paint_type = paint_type
+        self.w, self.h = w, h
+        self.matrix = (matrix.m11(), matrix.m12(), matrix.m21(), matrix.m22(),
+                       matrix.dx(), matrix.dy())
+        self.resources = Dictionary()
+        self.cache_key = (self.__class__.__name__, cache_key, self.matrix)
+
+    def add_extra_keys(self, d):
+        d['Type'] = Name('Pattern')
+        d['PatternType'] = 1
+        d['PaintType'] = self.paint_type
+        d['TilingType'] = 1
+        d['BBox'] = Array([0, 0, self.w, self.h])
+        d['XStep'] = self.w
+        d['YStep'] = self.h
+        d['Matrix'] = Array(self.matrix)
+        d['Resources'] = self.resources
+
+class QtPattern(TilingPattern):
+
+    qt_patterns = ( # {{{
+        "0 J\n"
+        "6 w\n"
+        "[] 0 d\n"
+        "4 0 m\n"
+        "4 8 l\n"
+        "0 4 m\n"
+        "8 4 l\n"
+        "S\n", # Dense1Pattern
+
+        "0 J\n"
+        "2 w\n"
+        "[6 2] 1 d\n"
+        "0 0 m\n"
+        "0 8 l\n"
+        "8 0 m\n"
+        "8 8 l\n"
+        "S\n"
+        "[] 0 d\n"
+        "2 0 m\n"
+        "2 8 l\n"
+        "6 0 m\n"
+        "6 8 l\n"
+        "S\n"
+        "[6 2] -3 d\n"
+        "4 0 m\n"
+        "4 8 l\n"
+        "S\n", # Dense2Pattern
+
+        "0 J\n"
+        "2 w\n"
+        "[6 2] 1 d\n"
+        "0 0 m\n"
+        "0 8 l\n"
+        "8 0 m\n"
+        "8 8 l\n"
+        "S\n"
+        "[2 2] -1 d\n"
+        "2 0 m\n"
+        "2 8 l\n"
+        "6 0 m\n"
+        "6 8 l\n"
+        "S\n"
+        "[6 2] -3 d\n"
+        "4 0 m\n"
+        "4 8 l\n"
+        "S\n", # Dense3Pattern
+
+        "0 J\n"
+        "2 w\n"
+        "[2 2] 1 d\n"
+        "0 0 m\n"
+        "0 8 l\n"
+        "8 0 m\n"
+        "8 8 l\n"
+        "S\n"
+        "[2 2] -1 d\n"
+        "2 0 m\n"
+        "2 8 l\n"
+        "6 0 m\n"
+        "6 8 l\n"
+        "S\n"
+        "[2 2] 1 d\n"
+        "4 0 m\n"
+        "4 8 l\n"
+        "S\n", # Dense4Pattern
+
+        "0 J\n"
+        "2 w\n"
+        "[2 6] -1 d\n"
+        "0 0 m\n"
+        "0 8 l\n"
+        "8 0 m\n"
+        "8 8 l\n"
+        "S\n"
+        "[2 2] 1 d\n"
+        "2 0 m\n"
+        "2 8 l\n"
+        "6 0 m\n"
+        "6 8 l\n"
+        "S\n"
+        "[2 6] 3 d\n"
+        "4 0 m\n"
+        "4 8 l\n"
+        "S\n", # Dense5Pattern
+
+        "0 J\n"
+        "2 w\n"
+        "[2 6] -1 d\n"
+        "0 0 m\n"
+        "0 8 l\n"
+        "8 0 m\n"
+        "8 8 l\n"
+        "S\n"
+        "[2 6] 3 d\n"
+        "4 0 m\n"
+        "4 8 l\n"
+        "S\n", # Dense6Pattern
+
+        "0 J\n"
+        "2 w\n"
+        "[2 6] -1 d\n"
+        "0 0 m\n"
+        "0 8 l\n"
+        "8 0 m\n"
+        "8 8 l\n"
+        "S\n", # Dense7Pattern
+
+        "1 w\n"
+        "0 4 m\n"
+        "8 4 l\n"
+        "S\n", # HorPattern
+
+        "1 w\n"
+        "4 0 m\n"
+        "4 8 l\n"
+        "S\n", # VerPattern
+
+        "1 w\n"
+        "4 0 m\n"
+        "4 8 l\n"
+        "0 4 m\n"
+        "8 4 l\n"
+        "S\n", # CrossPattern
+
+        "1 w\n"
+        "-1 5 m\n"
+        "5 -1 l\n"
+        "3 9 m\n"
+        "9 3 l\n"
+        "S\n", # BDiagPattern
+
+        "1 w\n"
+        "-1 3 m\n"
+        "5 9 l\n"
+        "3 -1 m\n"
+        "9 5 l\n"
+        "S\n", # FDiagPattern
+
+        "1 w\n"
+        "-1 3 m\n"
+        "5 9 l\n"
+        "3 -1 m\n"
+        "9 5 l\n"
+        "-1 5 m\n"
+        "5 -1 l\n"
+        "3 9 m\n"
+        "9 3 l\n"
+        "S\n", # DiagCrossPattern
+    ) # }}}
+
+    def __init__(self, pattern_num, matrix):
+        super(QtPattern, self).__init__(pattern_num, matrix)
+        self.write(self.qt_patterns[pattern_num-2])
+
+class TexturePattern(TilingPattern):
+
+    def __init__(self, pixmap, matrix, pdf, clone=None):
+        if clone is None:
+            image = pixmap.toImage()
+            cache_key = pixmap.cacheKey()
+            imgref = pdf.add_image(image, cache_key)
+            paint_type = (2 if image.format() in {QImage.Format_MonoLSB,
+                                                QImage.Format_Mono} else 1)
+            super(TexturePattern, self).__init__(
+                cache_key, matrix, w=image.width(), h=image.height(),
+                paint_type=paint_type)
+            m = (self.w, 0, 0, -self.h, 0, self.h)
+            self.resources['XObject'] = Dictionary({'Texture':imgref})
+            self.write_line('%s cm /Texture Do'%(' '.join(map(fmtnum, m))))
+        else:
+            super(TexturePattern, self).__init__(
+                clone.cache_key[1], matrix, w=clone.w, h=clone.h,
+                paint_type=clone.paint_type)
+            self.resources['XObject'] = Dictionary(clone.resources['XObject'])
+            self.write(clone.getvalue())
+
+class GraphicsState(object):
+
+    FIELDS = ('fill', 'stroke', 'opacity', 'transform', 'brush_origin',
+                  'clip', 'do_fill', 'do_stroke')
+
+    def __init__(self):
+        self.fill = QBrush()
+        self.stroke = QPen()
+        self.opacity = 1.0
+        self.transform = QTransform()
+        self.brush_origin = QPointF()
+        self.clip = QPainterPath()
+        self.do_fill = False
+        self.do_stroke = True
+        self.qt_pattern_cache = {}
+
+    def __eq__(self, other):
+        for x in self.FIELDS:
+            if getattr(other, x) != getattr(self, x):
+                return False
+        return True
+
+    def copy(self):
+        ans = GraphicsState()
+        ans.fill = QBrush(self.fill)
+        ans.stroke = QPen(self.stroke)
+        ans.opacity = self.opacity
+        ans.transform = self.transform * QTransform()
+        ans.brush_origin = QPointF(self.brush_origin)
+        ans.clip = self.clip
+        ans.do_fill, ans.do_stroke = self.do_fill, self.do_stroke
+        return ans
+
+class Graphics(object):
+
+    def __init__(self):
+        self.base_state = GraphicsState()
+        self.current_state = GraphicsState()
+        self.pending_state = None
+
+    def begin(self, pdf):
+        self.pdf = pdf
+
+    def update_state(self, state, painter):
+        flags = state.state()
+        if self.pending_state is None:
+            self.pending_state = self.current_state.copy()
+
+        s = self.pending_state
+
+        if flags & QPaintEngine.DirtyTransform:
+            s.transform = state.transform()
+
+        if flags & QPaintEngine.DirtyBrushOrigin:
+            s.brush_origin = state.brushOrigin()
+
+        if flags & QPaintEngine.DirtyBrush:
+            s.fill = state.brush()
+
+        if flags & QPaintEngine.DirtyPen:
+            s.stroke = state.pen()
+
+        if flags & QPaintEngine.DirtyOpacity:
+            s.opacity = state.opacity()
+
+        if flags & QPaintEngine.DirtyClipPath or flags & QPaintEngine.DirtyClipRegion:
+            s.clip = painter.clipPath()
+
+    def reset(self):
+        self.current_state = GraphicsState()
+        self.pending_state = None
+
+    def __call__(self, pdf_system, painter):
+        # Apply the currently pending state to the PDF
+        if self.pending_state is None:
+            return
+
+        pdf_state = self.current_state
+        ps = self.pending_state
+        pdf = self.pdf
+
+        if (ps.transform != pdf_state.transform or ps.clip != pdf_state.clip):
+            pdf.restore_stack()
+            pdf.save_stack()
+            pdf_state = self.base_state
+
+        if (pdf_state.transform != ps.transform):
+            pdf.transform(ps.transform)
+
+        if (pdf_state.opacity != ps.opacity or pdf_state.stroke != ps.stroke):
+            self.apply_stroke(ps, pdf_system, painter)
+
+        if (pdf_state.opacity != ps.opacity or pdf_state.fill != ps.fill or
+            pdf_state.brush_origin != ps.brush_origin):
+            self.apply_fill(ps, pdf_system, painter)
+
+        if (pdf_state.clip != ps.clip):
+            p = convert_path(ps.clip)
+            fill_rule = {Qt.OddEvenFill:'evenodd',
+                        Qt.WindingFill:'winding'}[ps.clip.fillRule()]
+            pdf.add_clip(p, fill_rule=fill_rule)
+
+        self.current_state = self.pending_state
+        self.pending_state = None
+
+    def convert_brush(self, brush, brush_origin, global_opacity,
+                      pdf_system, qt_system):
+        # Convert a QBrush to PDF operators
+        style = brush.style()
+        pdf = self.pdf
+
+        pattern = color = pat = None
+        opacity = 1.0
+        do_fill = True
+
+        matrix = (QTransform.fromTranslate(brush_origin.x(), brush_origin.y())
+                  * pdf_system * qt_system.inverted()[0])
+        vals = list(brush.color().getRgbF())
+        self.brushobj = None
+
+        if style <= Qt.DiagCrossPattern:
+            opacity = global_opacity * vals[-1]
+            color = vals[:3]
+
+            if style > Qt.SolidPattern:
+                pat = QtPattern(style, matrix)
+                pattern = pdf.add_pattern(pat)
+
+            if opacity < 1e-4 or style == Qt.NoBrush:
+                do_fill = False
+
+        elif style == Qt.TexturePattern:
+            pat = TexturePattern(brush.texture(), matrix, pdf)
+            opacity = global_opacity
+            if pat.paint_type == 2:
+                opacity *= vals[-1]
+                color = vals[:3]
+            pattern = pdf.add_pattern(pat)
+
+            if opacity < 1e-4 or style == Qt.NoBrush:
+                do_fill = False
+
+        self.brushobj = Brush(brush_origin, pat, color)
+        # TODO: Add support for gradient fills
+        return color, opacity, pattern, do_fill
+
+    def apply_stroke(self, state, pdf_system, painter):
+        # TODO: Support miter limit by using QPainterPathStroker
+        pen = state.stroke
+        self.pending_state.do_stroke = True
+        pdf = self.pdf
+
+        # Width
+        w = pen.widthF()
+        if pen.isCosmetic():
+            t = painter.transform()
+            w /= sqrt(t.m11()**2 + t.m22()**2)
+        pdf.serialize(w)
+        pdf.current_page.write(' w ')
+
+        # Line cap
+        cap = {Qt.FlatCap:0, Qt.RoundCap:1, Qt.SquareCap:
+               2}.get(pen.capStyle(), 0)
+        pdf.current_page.write('%d J '%cap)
+
+        # Line join
+        join = {Qt.MiterJoin:0, Qt.RoundJoin:1,
+                Qt.BevelJoin:2}.get(pen.joinStyle(), 0)
+        pdf.current_page.write('%d j '%join)
+
+        # Dash pattern
+        ps = {Qt.DashLine:[3], Qt.DotLine:[1,2], Qt.DashDotLine:[3,2,1,2],
+              Qt.DashDotDotLine:[3, 2, 1, 2, 1, 2]}.get(pen.style(), [])
+        if ps:
+            pdf.serialize(Array(ps))
+            pdf.current_page.write(' 0 d ')
+
+        # Stroke fill
+        color, opacity, pattern, self.pending_state.do_stroke = self.convert_brush(
+            pen.brush(), state.brush_origin, state.opacity, pdf_system,
+            painter.transform())
+        self.pdf.apply_stroke(color, pattern, opacity)
+        if pen.style() == Qt.NoPen:
+            self.pending_state.do_stroke = False
+
+    def apply_fill(self, state, pdf_system, painter):
+        self.pending_state.do_fill = True
+        color, opacity, pattern, self.pending_state.do_fill = self.convert_brush(
+            state.fill, state.brush_origin, state.opacity, pdf_system,
+            painter.transform())
+        self.pdf.apply_fill(color, pattern, opacity)
+        self.last_fill = self.brushobj
+
+    def __enter__(self):
+        self.pdf.save_stack()
+
+    def __exit__(self, *args):
+        self.pdf.restore_stack()
+
+    def resolve_fill(self, rect, pdf_system, qt_system):
+        '''
+        Qt's paint system does not update brushOrigin when using
+        TexturePatterns and it also uses TexturePatterns to emulate gradients,
+        leading to brokenness. So this method allows the paint engine to update
+        the brush origin before painting an object. While not perfect, this is
+        better than nothing.
+        '''
+        if not hasattr(self, 'last_fill') or not self.current_state.do_fill:
+            return
+
+        if isinstance(self.last_fill.brush, TexturePattern):
+            tl = rect.topLeft()
+            if tl == self.last_fill.origin:
+                return
+
+            matrix = (QTransform.fromTranslate(tl.x(), tl.y())
+                * pdf_system * qt_system.inverted()[0])
+
+            pat = TexturePattern(None, matrix, self.pdf, clone=self.last_fill.brush)
+            pattern = self.pdf.add_pattern(pat)
+            self.pdf.apply_fill(self.last_fill.color, pattern)
+
+

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

@@ -8,25 +8,115 @@ __copyright__ = '2012, Kovid Goyal <kovid at kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
 import os
+from future_builtins import map
+from urlparse import urlparse, urlunparse
+from urllib2 import quote, unquote
 
-from calibre.ebooks.pdf.render.common import Array, Name
+from calibre.ebooks.pdf.render.common import Array, Name, Dictionary, String
 
 class Destination(Array):
 
-    def __init__(self, start_page, pos):
+    def __init__(self, start_page, pos, get_pageref):
         super(Destination, self).__init__(
-            [start_page + pos['column'], Name('FitH'), pos['y']])
+            [get_pageref(start_page + pos['column']), Name('XYZ'), pos['left'],
+             pos['top'], None]
+        )
 
 class Links(object):
 
-    def __init__(self):
+    def __init__(self, pdf, mark_links, page_size):
         self.anchors = {}
+        self.links = []
+        self.start = {'top':page_size[1], 'column':0, 'left':0}
+        self.pdf = pdf
+        self.mark_links = mark_links
 
     def add(self, base_path, start_page, links, anchors):
         path = os.path.normcase(os.path.abspath(base_path))
         self.anchors[path] = a = {}
-        a[None] = Destination(start_page, {'y':0, 'column':0})
+        a[None] = Destination(start_page, self.start, self.pdf.get_pageref)
         for anchor, pos in anchors.iteritems():
-            a[anchor] = Destination(start_page, pos)
+            a[anchor] = Destination(start_page, pos, self.pdf.get_pageref)
+        for link in links:
+            href, page, rect = link
+            p, frag = href.partition('#')[0::2]
+            link = ((path, p, frag or None), self.pdf.get_pageref(page).obj, Array(rect))
+            self.links.append(link)
+
+    def add_links(self):
+        for link in self.links:
+            path, href, frag = link[0]
+            page, rect = link[1:]
+            combined_path = os.path.abspath(os.path.join(os.path.dirname(path), *href.split('/')))
+            is_local = not href or combined_path in self.anchors
+            annot = Dictionary({
+                'Type':Name('Annot'), 'Subtype':Name('Link'),
+                'Rect':rect, 'Border':Array([0,0,0]),
+            })
+            if self.mark_links:
+                annot.update({'Border':Array([16, 16, 1]), 'C':Array([1.0, 0,
+                                                                      0])})
+            if is_local:
+                path = combined_path if href else path
+                annot['Dest'] = self.anchors[path][frag]
+            else:
+                url = href + (('#'+frag) if frag else '')
+                purl = urlparse(url)
+                if purl.scheme and purl.scheme != 'file':
+                    action = Dictionary({
+                        'Type':Name('Action'), 'S':Name('URI'),
+                    })
+                    parts = (x.encode('utf-8') if isinstance(x, type(u'')) else
+                             x for x in purl)
+                    url = urlunparse(map(quote, map(unquote,
+                                                    parts))).decode('ascii')
+                    action['URI'] = String(url)
+                    annot['A'] = action
+            if 'A' in annot or 'Dest' in annot:
+                if 'Annots' not in page:
+                    page['Annots'] = Array()
+                page['Annots'].append(self.pdf.objects.add(annot))
+            else:
+                self.pdf.debug('Could not find destination for link: %s in file %s'%
+                               (href, path))
+
+    def add_outline(self, toc):
+        parent = Dictionary({'Type':Name('Outlines')})
+        parentref = self.pdf.objects.add(parent)
+        self.process_children(toc, parentref, parent_is_root=True)
+        self.pdf.catalog.obj['Outlines'] = parentref
+
+    def process_children(self, toc, parentref, parent_is_root=False):
+        childrefs = []
+        for child in toc:
+            childref = self.process_toc_item(child, parentref)
+            if childref is None:
+                continue
+            if childrefs:
+                childrefs[-1].obj['Next'] = childref
+                childref.obj['Prev'] = childrefs[-1]
+            childrefs.append(childref)
+
+            if len(child) > 0:
+                self.process_children(child, childref)
+        if childrefs:
+            parentref.obj['First'] = childrefs[0]
+            parentref.obj['Last'] = childrefs[-1]
+            if not parent_is_root:
+                parentref.obj['Count'] = -len(childrefs)
+
+    def process_toc_item(self, toc, parentref):
+        path = toc.abspath or None
+        frag = toc.fragment or None
+        if path is None:
+            return
+        path = os.path.normcase(os.path.abspath(path))
+        if path not in self.anchors:
+            return None
+        a = self.anchors[path]
+        dest = a.get(frag, a[None])
+        item = Dictionary({'Parent':parentref, 'Dest':dest,
+                           'Title':String(toc.text or _('Unknown'))})
+        return self.pdf.objects.add(item)
 
 

src/calibre/ebooks/pdf/render/qt_hack.cpp

@@ -17,18 +17,25 @@ GlyphInfo* get_glyphs(QPointF &p, const QTextItem &text_item) {
     QFontEngine *fe = ti.fontEngine;
     qreal size = ti.fontEngine->fontDef.pixelSize;
 #ifdef Q_WS_WIN
-    if (ti.fontEngine->type() == QFontEngine::Win) {
+    if (false && ti.fontEngine->type() == QFontEngine::Win) {
+        // This is used in the Qt sourcecode, but it gives incorrect results,
+        // so I have disabled it. I dont understand how it works in qpdf.cpp
         QFontEngineWin *fe = static_cast<QFontEngineWin *>(ti.fontEngine);
+        // I think this should be tmHeight - tmInternalLeading, but pixelSize
+        // seems to work on windows as well, so leave it as pixelSize
         size = fe->tm.tmHeight;
     }
 #endif
+    int synthesized = ti.fontEngine->synthesized();
+    qreal stretch = synthesized & QFontEngine::SynthesizedStretch ? ti.fontEngine->fontDef.stretch/100. : 1.;
+
     QVarLengthArray<glyph_t> glyphs;
     QVarLengthArray<QFixedPoint> positions;
     QTransform m = QTransform::fromTranslate(p.x(), p.y());
     fe->getGlyphPositions(ti.glyphs, m, ti.flags, glyphs, positions);
     QVector<QPointF> points = QVector<QPointF>(positions.count());
     for (int i = 0; i < positions.count(); i++) {
-        points[i].setX(positions[i].x.toReal());
+        points[i].setX(positions[i].x.toReal()/stretch);
         points[i].setY(positions[i].y.toReal());
     }
 
@@ -38,10 +45,10 @@ GlyphInfo* get_glyphs(QPointF &p, const QTextItem &text_item) {
 
     const quint32 *tag = reinterpret_cast<const quint32 *>("name");
 
-    return new GlyphInfo(fe->getSfntTable(qToBigEndian(*tag)), size, points, indices);
+    return new GlyphInfo(fe->getSfntTable(qToBigEndian(*tag)), size, stretch, points, indices);
 }
 
-GlyphInfo::GlyphInfo(const QByteArray& name, qreal size, const QVector<QPointF> &positions, const QVector<unsigned int> &indices) :name(name), positions(positions), size(size), indices(indices) {
+GlyphInfo::GlyphInfo(const QByteArray& name, qreal size, qreal stretch, const QVector<QPointF> &positions, const QVector<unsigned int> &indices) :name(name), positions(positions), size(size), stretch(stretch), indices(indices) {
 }
 
 QByteArray get_sfnt_table(const QTextItem &text_item, const char* tag_name) {

src/calibre/ebooks/pdf/render/qt_hack.h

@@ -17,9 +17,10 @@ class GlyphInfo {
         QByteArray name;
         QVector<QPointF> positions;
         qreal size;
+        qreal stretch;
         QVector<unsigned int> indices;
 
-        GlyphInfo(const QByteArray &name, qreal size, const QVector<QPointF> &positions, const QVector<unsigned int> &indices);
+        GlyphInfo(const QByteArray &name, qreal size, qreal stretch, const QVector<QPointF> &positions, const QVector<unsigned int> &indices);
 
     private:
         GlyphInfo(const GlyphInfo&);

src/calibre/ebooks/pdf/render/qt_hack.sip

@@ -13,9 +13,10 @@ class GlyphInfo {
 public:
     QByteArray name;
     qreal size;
+    qreal stretch;
     QVector<QPointF> &positions;
     QVector<unsigned int> indices;
-    GlyphInfo(const QByteArray &name, qreal size, const QVector<QPointF> &positions, const QVector<unsigned int> &indices);
+    GlyphInfo(const QByteArray &name, qreal size, qreal stretch, const QVector<QPointF> &positions, const QVector<unsigned int> &indices);
 private:
     GlyphInfo(const GlyphInfo& g);
 

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

@@ -9,19 +9,18 @@ __docformat__ = 'restructuredtext en'
 
 import hashlib
 from future_builtins import map
-from itertools import izip
+
-from collections import namedtuple
+from PyQt4.Qt import QBuffer, QByteArray, QImage, Qt, QColor, qRgba
 
 from calibre.constants import (__appname__, __version__)
 from calibre.ebooks.pdf.render.common import (
     Reference, EOL, serialize, Stream, Dictionary, String, Name, Array,
-    GlyphIndex)
+    fmtnum)
 from calibre.ebooks.pdf.render.fonts import FontManager
+from calibre.ebooks.pdf.render.links import Links
 
 PDFVER = b'%PDF-1.3'
 
-Color = namedtuple('Color', 'red green blue opacity')
-
 class IndirectObjects(object):
 
     def __init__(self):
@@ -89,6 +88,7 @@ class Page(Stream):
         self.opacities = {}
         self.fonts = {}
         self.xobjects = {}
+        self.patterns = {}
 
     def set_opacity(self, opref):
         if opref not in self.opacities:
@@ -107,6 +107,11 @@ class Page(Stream):
             self.xobjects[imgref] = 'Image%d'%len(self.xobjects)
         return self.xobjects[imgref]
 
+    def add_pattern(self, patternref):
+        if patternref not in self.patterns:
+            self.patterns[patternref] = 'Pat%d'%len(self.patterns)
+        return self.patterns[patternref]
+
     def add_resources(self):
         r = Dictionary()
         if self.opacities:
@@ -124,6 +129,13 @@ class Page(Stream):
             for ref, name in self.xobjects.iteritems():
                 xobjects[name] = ref
             r['XObject'] = xobjects
+        if self.patterns:
+            r['ColorSpace'] = Dictionary({'PCSp':Array(
+                [Name('Pattern'), Name('DeviceRGB')])})
+            patterns = Dictionary()
+            for ref, name in self.patterns.iteritems():
+                patterns[name] = ref
+            r['Pattern'] = patterns
         if r:
             self.page_dict['Resources'] = r
 
@@ -153,54 +165,6 @@ class Path(object):
     def close(self):
         self.ops.append(('h',))
 
-class Text(object):
-
-    def __init__(self):
-        self.transform = self.default_transform = [1, 0, 0, 1, 0, 0]
-        self.font_name = 'Times-Roman'
-        self.font_path = None
-        self.horizontal_scale = self.default_horizontal_scale = 100
-        self.word_spacing = self.default_word_spacing = 0
-        self.char_space = self.default_char_space = 0
-        self.glyph_adjust = self.default_glyph_adjust = None
-        self.size = 12
-        self.text = ''
-
-    def set_transform(self, *args):
-        if len(args) == 1:
-            m = args[0]
-            vals = [m.m11(), m.m12(), m.m21(), m.m22(), m.dx(), m.dy()]
-        else:
-            vals = args
-        self.transform = vals
-
-    def pdf_serialize(self, stream, font_name):
-        if not self.text: return
-        stream.write_line('BT ')
-        serialize(Name(font_name), stream)
-        stream.write(' %g Tf '%self.size)
-        stream.write(' '.join(map(type(u''), self.transform)) + ' Tm ')
-        if self.horizontal_scale != self.default_horizontal_scale:
-            stream.write('%g Tz '%self.horizontal_scale)
-        if self.word_spacing != self.default_word_spacing:
-            stream.write('%g Tw '%self.word_spacing)
-        if self.char_space != self.default_char_space:
-            stream.write('%g Tc '%self.char_space)
-        stream.write_line()
-        if self.glyph_adjust is self.default_glyph_adjust:
-            serialize(String(self.text), stream)
-            stream.write(' Tj ')
-        else:
-            chars = Array()
-            frac, widths = self.glyph_adjust
-            for c, width in izip(self.text, widths):
-                chars.append(String(c))
-                chars.append(int(width * frac))
-            serialize(chars, stream)
-            stream.write(' TJ ')
-        stream.write_line('ET')
-
-
 class Catalog(Dictionary):
 
     def __init__(self, pagetree):
@@ -219,6 +183,9 @@ class PageTree(Dictionary):
         self['Kids'].append(pageref)
         self['Count'] += 1
 
+    def get_ref(self, num):
+        return self['Kids'][num-1]
+
 class HashingStream(object):
 
     def __init__(self, f):
@@ -228,7 +195,9 @@ class HashingStream(object):
         self.last_char = b''
 
     def write(self, raw):
-        raw = raw if isinstance(raw, bytes) else raw.encode('ascii')
+        self.write_raw(raw if isinstance(raw, bytes) else raw.encode('ascii'))
+
+    def write_raw(self, raw):
         self.f.write(raw)
         self.hashobj.update(raw)
         if raw:
@@ -277,7 +246,8 @@ class PDFStream(object):
         ( True,  True,  'evenodd')  : 'B*',
     }
 
-    def __init__(self, stream, page_size, compress=False):
+    def __init__(self, stream, page_size, compress=False, mark_links=False,
+                 debug=print):
         self.stream = HashingStream(stream)
         self.compress = compress
         self.write_line(PDFVER)
@@ -294,6 +264,12 @@ class PDFStream(object):
         self.stroke_opacities, self.fill_opacities = {}, {}
         self.font_manager = FontManager(self.objects, self.compress)
         self.image_cache = {}
+        self.pattern_cache = {}
+        self.debug = debug
+        self.links = Links(self, mark_links, page_size)
+        i = QImage(1, 1, QImage.Format_ARGB32)
+        i.fill(qRgba(0, 0, 0, 255))
+        self.alpha_bit = i.constBits().asstring(4).find(b'\xff')
 
     @property
     def page_tree(self):
@@ -303,6 +279,9 @@ class PDFStream(object):
     def catalog(self):
         return self.objects[1]
 
+    def get_pageref(self, pagenum):
+        return self.page_tree.obj.get_ref(pagenum)
+
     def set_metadata(self, title=None, author=None, tags=None):
         if title:
             self.info['Title'] = String(title)
@@ -321,12 +300,9 @@ class PDFStream(object):
             vals = [m.m11(), m.m12(), m.m21(), m.m22(), m.dx(), m.dy()]
         else:
             vals = args
-        cm = ' '.join(map(type(u''), vals))
+        cm = ' '.join(map(fmtnum, vals))
         self.current_page.write_line(cm + ' cm')
 
-    def set_rgb_colorspace(self):
-        self.current_page.write_line('/DeviceRGB CS /DeviceRGB cs')
-
     def save_stack(self):
         self.current_page.write_line('q')
 
@@ -337,7 +313,7 @@ class PDFStream(object):
         self.current_page.write_line('Q q')
 
     def draw_rect(self, x, y, width, height, stroke=True, fill=False):
-        self.current_page.write('%g %g %g %g re '%(x, y, width, height))
+        self.current_page.write('%s re '%' '.join(map(fmtnum, (x, y, width, height))))
         self.current_page.write_line(self.PATH_OPS[(stroke, fill, 'winding')])
 
     def write_path(self, path):
@@ -345,7 +321,8 @@ class PDFStream(object):
             if i != 0:
                 self.current_page.write_line()
             for x in op:
-                self.current_page.write(type(u'')(x) + ' ')
+                self.current_page.write(
+                (fmtnum(x) if isinstance(x, (int, long, float)) else x) + ' ')
 
     def draw_path(self, path, stroke=True, fill=False, fill_rule='winding'):
         if not path.ops: return
@@ -358,67 +335,38 @@ class PDFStream(object):
         op = 'W' if fill_rule == 'winding' else 'W*'
         self.current_page.write_line(op + ' ' + 'n')
 
-    def set_dash(self, array, phase=0):
+    def serialize(self, o):
-        array = Array(array)
+        serialize(o, self.current_page)
-        serialize(array, self.current_page)
-        self.current_page.write(b' ')
-        serialize(phase, self.current_page)
-        self.current_page.write_line(' d')
 
-    def set_line_width(self, width):
+    def set_stroke_opacity(self, opacity):
-        serialize(width, self.current_page)
-        self.current_page.write_line(' w')
-
-    def set_line_cap(self, style):
-        serialize({'flat':0, 'round':1, 'square':2}.get(style),
-                  self.current_page)
-        self.current_page.write_line(' J')
-
-    def set_line_join(self, style):
-        serialize({'miter':0, 'round':1, 'bevel':2}[style], self.current_page)
-        self.current_page.write_line(' j')
-
-    def set_stroke_color(self, color):
-        opacity = color.opacity
         if opacity not in self.stroke_opacities:
             op = Dictionary({'Type':Name('ExtGState'), 'CA': opacity})
             self.stroke_opacities[opacity] = self.objects.add(op)
         self.current_page.set_opacity(self.stroke_opacities[opacity])
-        self.current_page.write_line(' '.join(map(type(u''), color[:3])) + ' SC')
 
-    def set_fill_color(self, color):
+    def set_fill_opacity(self, opacity):
-        opacity = color.opacity
+        opacity = float(opacity)
         if opacity not in self.fill_opacities:
             op = Dictionary({'Type':Name('ExtGState'), 'ca': opacity})
             self.fill_opacities[opacity] = self.objects.add(op)
         self.current_page.set_opacity(self.fill_opacities[opacity])
-        self.current_page.write_line(' '.join(map(type(u''), color[:3])) + ' sc')
 
     def end_page(self):
         pageref = self.current_page.end(self.objects, self.stream)
         self.page_tree.obj.add_page(pageref)
         self.current_page = Page(self.page_tree, compress=self.compress)
 
-    def draw_text(self, text_object):
-        if text_object.font_path is None:
-            fontref = self.font_manager.add_standard_font(text_object.font_name)
-        else:
-            raise NotImplementedError()
-        name = self.current_page.add_font(fontref)
-        text_object.pdf_serialize(self.current_page, name)
-
     def draw_glyph_run(self, transform, size, font_metrics, glyphs):
         glyph_ids = {x[-1] for x in glyphs}
         fontref = self.font_manager.add_font(font_metrics, glyph_ids)
         name = self.current_page.add_font(fontref)
         self.current_page.write(b'BT ')
         serialize(Name(name), self.current_page)
-        self.current_page.write(' %g Tf '%size)
+        self.current_page.write(' %s Tf '%fmtnum(size))
-        self.current_page.write('%s Tm '%' '.join(map(type(u''), transform)))
+        self.current_page.write('%s Tm '%' '.join(map(fmtnum, transform)))
         for x, y, glyph_id in glyphs:
-            self.current_page.write('%g %g Td '%(x, y))
+            self.current_page.write_raw(('%s %s Td <%04X> Tj '%(
-            serialize(GlyphIndex(glyph_id), self.current_page)
+                fmtnum(x), fmtnum(y), glyph_id)).encode('ascii'))
-            self.current_page.write(' Tj ')
         self.current_page.write_line(b' ET')
 
     def get_image(self, cache_key):
@@ -431,17 +379,109 @@ class PDFStream(object):
         self.objects.commit(r, self.stream)
         return r
 
-    def draw_image(self, x, y, xscale, yscale, imgref):
+    def add_image(self, img, cache_key):
+        ref = self.get_image(cache_key)
+        if ref is not None:
+            return ref
+
+        fmt = img.format()
+        image = QImage(img)
+        if (image.depth() == 1 and img.colorTable().size() == 2 and
+            img.colorTable().at(0) == QColor(Qt.black).rgba() and
+            img.colorTable().at(1) == QColor(Qt.white).rgba()):
+            if fmt == QImage.Format_MonoLSB:
+                image = image.convertToFormat(QImage.Format_Mono)
+            fmt = QImage.Format_Mono
+        else:
+            if (fmt != QImage.Format_RGB32 and fmt != QImage.Format_ARGB32):
+                image = image.convertToFormat(QImage.Format_ARGB32)
+                fmt = QImage.Format_ARGB32
+
+        w = image.width()
+        h = image.height()
+        d = image.depth()
+
+        if fmt == QImage.Format_Mono:
+            bytes_per_line = (w + 7) >> 3
+            data = image.constBits().asstring(bytes_per_line * h)
+            return self.write_image(data, w, h, d, cache_key=cache_key)
+
+        ba = QByteArray()
+        buf = QBuffer(ba)
+        image.save(buf, 'jpeg', 94)
+        data = bytes(ba.data())
+        has_alpha = has_mask = False
+        soft_mask = mask = None
+
+        if fmt == QImage.Format_ARGB32:
+            tmask = image.constBits().asstring(4*w*h)[self.alpha_bit::4]
+            sdata = bytearray(tmask)
+            vals = set(sdata)
+            vals.discard(255)
+            has_mask = bool(vals)
+            vals.discard(0)
+            has_alpha = bool(vals)
+
+        if has_alpha:
+            soft_mask = self.write_image(tmask, w, h, 8)
+        elif has_mask:
+            # dither the soft mask to 1bit and add it. This also helps PDF
+            # viewers without transparency support
+            bytes_per_line = (w + 7) >> 3
+            mdata = bytearray(0 for i in xrange(bytes_per_line * h))
+            spos = mpos = 0
+            for y in xrange(h):
+                for x in xrange(w):
+                    if sdata[spos]:
+                        mdata[mpos + x>>3] |= (0x80 >> (x&7))
+                    spos += 1
+                mpos += bytes_per_line
+            mdata = bytes(mdata)
+            mask = self.write_image(mdata, w, h, 1)
+
+        return self.write_image(data, w, h, 32, mask=mask, dct=True,
+                                    soft_mask=soft_mask, cache_key=cache_key)
+
+    def add_pattern(self, pattern):
+        if pattern.cache_key not in self.pattern_cache:
+            self.pattern_cache[pattern.cache_key] = self.objects.add(pattern)
+        return self.current_page.add_pattern(self.pattern_cache[pattern.cache_key])
+
+    def draw_image(self, x, y, width, height, imgref):
         name = self.current_page.add_image(imgref)
-        self.current_page.write('q %g 0 0 %g %g %g cm '%(xscale, yscale, x, y))
+        self.current_page.write('q %s 0 0 %s %s %s cm '%(fmtnum(width),
+                            fmtnum(-height), fmtnum(x), fmtnum(y+height)))
         serialize(Name(name), self.current_page)
         self.current_page.write_line(' Do Q')
 
+    def apply_color_space(self, color, pattern, stroke=False):
+        wl = self.current_page.write_line
+        if color is not None and pattern is None:
+            wl(' '.join(map(fmtnum, color)) + (' RG' if stroke else ' rg'))
+        elif color is None and pattern is not None:
+            wl('/Pattern %s /%s %s'%('CS' if stroke else 'cs', pattern,
+                                     'SCN' if stroke else 'scn'))
+        elif color is not None and pattern is not None:
+            col = ' '.join(map(fmtnum, color))
+            wl('/PCSp %s %s /%s %s'%('CS' if stroke else 'cs', col, pattern,
+                                     'SCN' if stroke else 'scn'))
+
+    def apply_fill(self, color=None, pattern=None, opacity=None):
+        if opacity is not None:
+            self.set_fill_opacity(opacity)
+        self.apply_color_space(color, pattern)
+
+    def apply_stroke(self, color=None, pattern=None, opacity=None):
+        if opacity is not None:
+            self.set_stroke_opacity(opacity)
+        self.apply_color_space(color, pattern, stroke=True)
+
     def end(self):
         if self.current_page.getvalue():
             self.end_page()
         self.font_manager.embed_fonts()
         inforef = self.objects.add(self.info)
+        self.links.add_links()
         self.objects.pdf_serialize(self.stream)
         self.write_line()
         startxref = self.objects.write_xref(self.stream)

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

@@ -0,0 +1,132 @@
+#!/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__ = '2012, Kovid Goyal <kovid at kovidgoyal.net>'
+__docformat__ = 'restructuredtext en'
+
+import os
+
+from PyQt4.Qt import (QBrush, QColor, QPoint, QPixmap, QPainterPath, QRectF,
+                      QApplication, QPainter, Qt, QImage, QLinearGradient,
+                      QPointF, QPen)
+QBrush, QColor, QPoint, QPixmap, QPainterPath, QRectF, Qt, QPointF
+
+from calibre.ebooks.pdf.render.engine import PdfDevice
+
+def full(p, xmax, ymax):
+    p.drawRect(0, 0, xmax, ymax)
+    p.drawPolyline(QPoint(0, 0), QPoint(xmax, 0), QPoint(xmax, ymax),
+                    QPoint(0, ymax), QPoint(0, 0))
+    pp = QPainterPath()
+    pp.addRect(0, 0, xmax, ymax)
+    p.drawPath(pp)
+    p.save()
+    for i in xrange(3):
+        col = [0, 0, 0, 200]
+        col[i] = 255
+        p.setOpacity(0.3)
+        p.fillRect(0, 0, xmax/10, xmax/10, QBrush(QColor(*col)))
+        p.setOpacity(1)
+        p.drawRect(0, 0, xmax/10, xmax/10)
+        p.translate(xmax/10, xmax/10)
+        p.scale(1, 1.5)
+    p.restore()
+
+    # p.scale(2, 2)
+    # p.rotate(45)
+    p.drawPixmap(0, 0, xmax/4, xmax/4, QPixmap(I('library.png')))
+    p.drawRect(0, 0, xmax/4, xmax/4)
+
+    f = p.font()
+    f.setPointSize(20)
+    # f.setLetterSpacing(f.PercentageSpacing, 200)
+    f.setUnderline(True)
+    # f.setOverline(True)
+    # f.setStrikeOut(True)
+    f.setFamily('Calibri')
+    p.setFont(f)
+    # p.setPen(QColor(0, 0, 255))
+    # p.scale(2, 2)
+    # p.rotate(45)
+    p.drawText(QPoint(xmax/3.9, 30), 'Some—text not By’s ū --- Д AV ff ff')
+
+    b = QBrush(Qt.HorPattern)
+    b.setColor(QColor(Qt.blue))
+    pix = QPixmap(I('console.png'))
+    w = xmax/4
+    p.fillRect(0, ymax/3, w, w, b)
+    p.fillRect(xmax/3, ymax/3, w, w, QBrush(pix))
+    x, y = 2*xmax/3, ymax/3
+    p.drawTiledPixmap(QRectF(x, y, w, w), pix, QPointF(10, 10))
+
+    x, y = 1, ymax/1.9
+    g = QLinearGradient(QPointF(x, y), QPointF(x+w, y+w))
+    g.setColorAt(0, QColor('#00f'))
+    g.setColorAt(1, QColor('#fff'))
+    p.fillRect(x, y, w, w, QBrush(g))
+
+
+def run(dev, func):
+    p = QPainter(dev)
+    if isinstance(dev, PdfDevice):
+        dev.init_page()
+    xmax, ymax = p.viewport().width(), p.viewport().height()
+    try:
+        func(p, xmax, ymax)
+    finally:
+        p.end()
+    if isinstance(dev, PdfDevice):
+        if dev.engine.errors_occurred:
+            raise SystemExit(1)
+
+def brush(p, xmax, ymax):
+    x = xmax/3
+    y = 0
+    w = xmax/2
+    pix = QPixmap(I('console.png'))
+    p.fillRect(x, y, w, w, QBrush(pix))
+
+    p.fillRect(0, y+xmax/1.9, w, w, QBrush(pix))
+
+def pen(p, xmax, ymax):
+    pix = QPixmap(I('console.png'))
+    pen = QPen(QBrush(pix), 60)
+    p.setPen(pen)
+    p.drawRect(0, xmax/3, xmax/3, xmax/2)
+
+def text(p, xmax, ymax):
+    f = p.font()
+    f.setPixelSize(24)
+    f.setFamily('Candara')
+    p.setFont(f)
+    p.drawText(QPoint(0, 100),
+        'Test intra glyph spacing ffagain imceo')
+
+def main():
+    app = QApplication([])
+    app
+    tdir = os.path.abspath('.')
+    pdf = os.path.join(tdir, 'painter.pdf')
+    func = full
+    dpi = 100
+    with open(pdf, 'wb') as f:
+        dev = PdfDevice(f, xdpi=dpi, ydpi=dpi, compress=False)
+        img = QImage(dev.width(), dev.height(),
+                     QImage.Format_ARGB32_Premultiplied)
+        img.setDotsPerMeterX(dpi*39.37)
+        img.setDotsPerMeterY(dpi*39.37)
+        img.fill(Qt.white)
+        run(dev, func)
+    run(img, func)
+    path = os.path.join(tdir, 'painter.png')
+    img.save(path)
+    print ('PDF written to:', pdf)
+    print ('Image written to:', path)
+
+if __name__ == '__main__':
+    main()
+
+

src/calibre/gui2/convert/single.py

@@ -33,7 +33,10 @@ from calibre.utils.config import prefs
 from calibre.utils.logging import Log
 
 class NoSupportedInputFormats(Exception):
-    pass
+
+    def __init__(self, available_formats):
+        Exception.__init__(self)
+        self.available_formats = available_formats
 
 def sort_formats_by_preference(formats, prefs):
     uprefs = [x.upper() for x in prefs]
@@ -86,7 +89,7 @@ def get_supported_input_formats_for_book(db, book_id):
     input_formats = set([x.lower() for x in supported_input_formats()])
     input_formats = sorted(available_formats.intersection(input_formats))
     if not input_formats:
-        raise NoSupportedInputFormats
+        raise NoSupportedInputFormats(tuple(x for x in available_formats if x))
     return input_formats
 
 

src/calibre/gui2/custom_column_widgets.py

@@ -372,7 +372,7 @@ class Series(Base):
 
         self.widgets.append(QLabel('&'+self.col_metadata['name']+_(' index:'), parent))
         w = QDoubleSpinBox(parent)
-        w.setRange(-100., float(100000000))
+        w.setRange(-10000., float(100000000))
         w.setDecimals(2)
         w.setSingleStep(1)
         self.idx_widget=w

src/calibre/gui2/library/models.py

@@ -5,7 +5,7 @@ __license__   = 'GPL v3'
 __copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-import functools, re, os, traceback, errno
+import functools, re, os, traceback, errno, time
 from collections import defaultdict
 
 from PyQt4.Qt import (QAbstractTableModel, Qt, pyqtSignal, QIcon, QImage,
@@ -1419,7 +1419,11 @@ class DeviceBooksModel(BooksModel): # {{{
                 return QVariant(human_readable(size))
             elif cname == 'timestamp':
                 dt = self.db[self.map[row]].datetime
+                try:
                     dt = dt_factory(dt, assume_utc=True, as_utc=False)
+                except OverflowError:
+                    dt = dt_factory(time.gmtime(), assume_utc=True,
+                                    as_utc=False)
                 return QVariant(strftime(TIME_FMT, dt.timetuple()))
             elif cname == 'collections':
                 tags = self.db[self.map[row]].device_collections

src/calibre/gui2/metadata/basic_widgets.py

@@ -1094,6 +1094,9 @@ class RatingEdit(QSpinBox): # {{{
         db.set_rating(id_, 2*self.current_val, notify=False, commit=False)
         return True
 
+    def zero(self):
+        self.setValue(0)
+
 # }}}
 
 class TagsEdit(EditWithComplete): # {{{

src/calibre/gui2/metadata/single.py

@@ -181,6 +181,11 @@ class MetadataSingleDialogBase(ResizableDialog):
         self.basic_metadata_widgets.append(self.comments)
 
         self.rating = RatingEdit(self)
+        self.clear_ratings_button = QToolButton(self)
+        self.clear_ratings_button.setToolTip(_('Clear rating'))
+        self.clear_ratings_button.setIcon(QIcon(I('trash.png')))
+        self.clear_ratings_button.clicked.connect(self.rating.zero)
+
         self.basic_metadata_widgets.append(self.rating)
 
         self.tags = TagsEdit(self)
@@ -659,8 +664,9 @@ class MetadataSingleDialog(MetadataSingleDialogBase): # {{{
                 QSizePolicy.Expanding)
         l.addItem(self.tabs[0].spc_one, 1, 0, 1, 3)
         sto(self.cover.buttons[-1], self.rating)
-        create_row2(1, self.rating)
+        create_row2(1, self.rating, self.clear_ratings_button)
-        sto(self.rating, self.tags_editor_button)
+        sto(self.rating, self.clear_ratings_button)
+        sto(self.clear_ratings_button, self.tags_editor_button)
         sto(self.tags_editor_button, self.tags)
         create_row2(2, self.tags, self.clear_tags_button, front_button=self.tags_editor_button)
         sto(self.clear_tags_button, self.paste_isbn_button)
@@ -780,7 +786,7 @@ class MetadataSingleDialogAlt1(MetadataSingleDialogBase): # {{{
                    button=self.clear_series_button, icon='trash.png')
         create_row(5, self.series_index, self.tags)
         create_row(6, self.tags, self.rating, button=self.clear_tags_button)
-        create_row(7, self.rating, self.pubdate)
+        create_row(7, self.rating, self.pubdate, button=self.clear_ratings_button)
         create_row(8, self.pubdate, self.publisher,
                    button=self.pubdate.clear_button, icon='trash.png')
         create_row(9, self.publisher, self.languages)
@@ -917,7 +923,7 @@ class MetadataSingleDialogAlt2(MetadataSingleDialogBase): # {{{
                    button=self.clear_series_button, icon='trash.png')
         create_row(5, self.series_index, self.tags)
         create_row(6, self.tags, self.rating, button=self.clear_tags_button)
-        create_row(7, self.rating, self.pubdate)
+        create_row(7, self.rating, self.pubdate, button=self.clear_ratings_button)
         create_row(8, self.pubdate, self.publisher,
                    button=self.pubdate.clear_button, icon='trash.png')
         create_row(9, self.publisher, self.languages)

src/calibre/gui2/preferences/ignored_devices.py

@@ -7,8 +7,10 @@ __license__   = 'GPL v3'
 __copyright__ = '2012, Kovid Goyal <kovid at kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-from PyQt4.Qt import (QLabel, QVBoxLayout, QListWidget, QListWidgetItem, Qt)
+from PyQt4.Qt import (QLabel, QVBoxLayout, QListWidget, QListWidgetItem, Qt,
+                      QIcon)
 
+from calibre.customize.ui import enable_plugin
 from calibre.gui2.preferences import ConfigWidgetBase, test_widget
 
 class ConfigWidget(ConfigWidgetBase):
@@ -31,6 +33,18 @@ class ConfigWidget(ConfigWidgetBase):
         f.itemChanged.connect(self.changed_signal)
         f.itemDoubleClicked.connect(self.toggle_item)
 
+        self.la2 = la = QLabel(_(
+            'The list of device plugins you have disabled. Uncheck an entry '
+            'to enable the plugin. calibre cannot detect devices that are '
+            'managed by disabled plugins.'))
+        la.setWordWrap(True)
+        l.addWidget(la)
+
+        self.device_plugins = f = QListWidget(f)
+        l.addWidget(f)
+        f.itemChanged.connect(self.changed_signal)
+        f.itemDoubleClicked.connect(self.toggle_item)
+
     def toggle_item(self, item):
         item.setCheckState(Qt.Checked if item.checkState() == Qt.Unchecked else
                 Qt.Unchecked)
@@ -46,6 +60,17 @@ class ConfigWidget(ConfigWidgetBase):
                 item.setCheckState(Qt.Checked)
         self.devices.blockSignals(False)
 
+        self.device_plugins.blockSignals(True)
+        for dev in self.gui.device_manager.disabled_device_plugins:
+            n = dev.get_gui_name()
+            item = QListWidgetItem(n, self.device_plugins)
+            item.setData(Qt.UserRole, dev)
+            item.setFlags(Qt.ItemIsEnabled|Qt.ItemIsUserCheckable|Qt.ItemIsSelectable)
+            item.setCheckState(Qt.Checked)
+            item.setIcon(QIcon(I('plugins.png')))
+        self.device_plugins.sortItems()
+        self.device_plugins.blockSignals(False)
+
     def restore_defaults(self):
         if self.devices.count() > 0:
             self.devices.clear()
@@ -63,6 +88,12 @@ class ConfigWidget(ConfigWidgetBase):
         for dev, bl in devs.iteritems():
             dev.set_user_blacklisted_devices(bl)
 
+        for i in xrange(self.device_plugins.count()):
+            e = self.device_plugins.item(i)
+            dev = e.data(Qt.UserRole).toPyObject()
+            if e.checkState() == Qt.Unchecked:
+                enable_plugin(dev)
+
         return True # Restart required
 
 if __name__ == '__main__':

src/calibre/gui2/preferences/server.ui

@@ -273,7 +273,7 @@
     <widget class="QLabel" name="label_13">
      <property name="text">
       <string>&lt;p&gt;Remember to leave calibre running as the server only runs as long as calibre is running.
-&lt;p&gt;To connect to the calibre server from your device you should use a URL of the form &lt;b&gt;http://myhostname:8080&lt;/b&gt; as a new catalog in the Stanza reader on your iPhone. Here myhostname should be either the fully qualified hostname or the IP address of the computer calibre is running on.</string>
+&lt;p&gt;To connect to the calibre server from your device you should use a URL of the form &lt;b&gt;http://myhostname:8080&lt;/b&gt;. Here myhostname should be either the fully qualified hostname or the IP address of the computer calibre is running on. If you want to access the server from anywhere in the world, you will have to setup port forwarding for it on your router.</string>
      </property>
      <property name="wordWrap">
       <bool>true</bool>

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

@@ -6,6 +6,7 @@ __license__ = 'GPL 3'
 __copyright__ = '2011, John Schember <john@nachtimwald.com>'
 __docformat__ = 'restructuredtext en'
 
+import re
 import urllib
 from contextlib import closing
 
@@ -50,12 +51,17 @@ class BNStore(BasicStoreConfig, StorePlugin):
                 if not id:
                     continue
 
-                cover_url = ''.join(data.xpath('.//img[contains(@class, "product-image")]/@src'))
+                cover_url = ''
+                cover_id = ''.join(data.xpath('.//img[contains(@class, "product-image")]/@id'))
+                m = re.search(r"%s'.*?srcUrl: '(?P<iurl>.*?)'.*?}" % cover_id, raw)
+                if m:
+                    cover_url = m.group('iurl')
 
                 title = ''.join(data.xpath('descendant::p[@class="title"]//span[@class="name"]//text()')).strip()
-                if not title: continue
+                if not title:
+                    continue
 
-                author = ', '.join(data.xpath('.//ul[@class="contributors"]//a[@class="subtle"]//text()')).strip()
+                author = ', '.join(data.xpath('.//ul[contains(@class, "contributors")]//a[contains(@class, "subtle")]//text()')).strip()
                 price = ''.join(data.xpath('.//a[contains(@class, "bn-price")]//text()'))
 
                 counter -= 1

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

@@ -59,7 +59,7 @@ class GoogleBooksStore(BasicStoreConfig, StorePlugin):
         counter = max_results
         with closing(br.open(url, timeout=timeout)) as f:
             doc = html.fromstring(f.read())
-            for data in doc.xpath('//ol[@id="rso"]/li'):
+            for data in doc.xpath('//ol/li'):
                 if counter <= 0:
                     break
 
@@ -68,7 +68,7 @@ class GoogleBooksStore(BasicStoreConfig, StorePlugin):
                     continue
 
                 title = ''.join(data.xpath('.//h3/a//text()'))
-                authors = data.xpath('.//div[@class="f"]//a//text()')
+                authors = data.xpath('.//span[contains(@class, "f")]//a//text()')
                 while authors and authors[-1].strip().lower() in ('preview', 'read', 'more editions'):
                     authors = authors[:-1]
                 if not authors:

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

@@ -0,0 +1,75 @@
+# -*- coding: utf-8 -*-
+
+from __future__ import (unicode_literals, division, absolute_import, print_function)
+
+__license__ = 'GPL 3'
+__copyright__ = '2012, John Schember <john@nachtimwald.com>'
+__docformat__ = 'restructuredtext en'
+
+import re
+import urllib
+from contextlib import closing
+
+from lxml import html
+
+from PyQt4.Qt import QUrl
+
+from calibre import browser, url_slash_cleaner
+from calibre.gui2 import open_url
+from calibre.gui2.store import StorePlugin
+from calibre.gui2.store.basic_config import BasicStoreConfig
+from calibre.gui2.store.search_result import SearchResult
+from calibre.gui2.store.web_store_dialog import WebStoreDialog
+
+class NookUKStore(BasicStoreConfig, StorePlugin):
+
+    def open(self, parent=None, detail_item=None, external=False):
+        url = "http://uk.nook.com"
+
+        if external or self.config.get('open_external', False):
+            open_url(QUrl(url_slash_cleaner(detail_item if detail_item else url)))
+        else:
+            d = WebStoreDialog(self.gui, url, parent, detail_item)
+            d.setWindowTitle(self.name)
+            d.set_tags(self.config.get('tags', ''))
+            d.exec_()
+
+    def search(self, query, max_results=10, timeout=60):
+        url = u'http://uk.nook.com/s/%s?s%%5Bdref%%5D=1&s%%5Bkeyword%%5D=%s' % (query.replace(' ', '-'), urllib.quote(query))
+
+        br = browser()
+
+        counter = max_results
+        with closing(br.open(url, timeout=timeout)) as f:
+            raw = f.read()
+            doc = html.fromstring(raw)
+            for data in doc.xpath('//ul[contains(@class, "product_list")]/li'):
+                if counter <= 0:
+                    break
+
+                id = ''.join(data.xpath('.//span[contains(@class, "image")]/a/@href'))
+                if not id:
+                    continue
+
+                cover_url = ''.join(data.xpath('.//span[contains(@class, "image")]//img/@data-src'))
+
+                title = ''.join(data.xpath('.//div[contains(@class, "title")]//text()')).strip()
+                if not title:
+                    continue
+
+                author = ', '.join(data.xpath('.//div[contains(@class, "contributor")]//a/text()')).strip()
+                price = ''.join(data.xpath('.//div[contains(@class, "action")]//a//text()')).strip()
+                price = re.sub(r'[^\d.,£]', '', price);
+
+                counter -= 1
+
+                s = SearchResult()
+                s.cover_url = cover_url
+                s.title = title.strip()
+                s.author = author.strip()
+                s.price = price.strip()
+                s.detail_item = 'http://uk.nook.com/' + id.strip()
+                s.drm = SearchResult.DRM_UNKNOWN
+                s.formats = 'Nook'
+
+                yield s

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

@@ -76,7 +76,7 @@ class SmashwordsStore(BasicStoreConfig, StorePlugin):
                 
                 title = ''.join(data.xpath('//a[@class="bookTitle"]/text()'))
                 subnote = ''.join(data.xpath('//span[@class="subnote"]/text()'))
-                author = ''.join(data.xpath('//span[@class="subnote"]/a/text()'))
+                author = ''.join(data.xpath('//span[@class="subnote"]//a[1]//text()'))
                 if '$' in subnote:
                     price = subnote.partition('$')[2]
                     price = price.split(u'\xa0')[0]

src/calibre/gui2/tools.py

@@ -88,19 +88,34 @@ def convert_single_ebook(parent, db, book_ids, auto_conversion=False, # {{{
 
                 changed = True
                 d.break_cycles()
-        except NoSupportedInputFormats:
+        except NoSupportedInputFormats as nsif:
-            bad.append(book_id)
+            bad.append((book_id, nsif.available_formats))
 
     if bad and show_no_format_warning:
+        if len(bad) == 1 and not bad[0][1]:
+            title = db.title(bad[0][0], True)
+            warning_dialog(parent, _('Could not convert'), '<p>'+
+                _('Could not convert <b>%s</b> as it has no ebook files. If you '
+                  'think it should have files, but calibre is not finding '
+                  'them, that is most likely because you moved the book\'s '
+                  'files around outside of calibre. You will need to find those files '
+                  'and re-add them to calibre.')%title, show=True)
+        else:
             res = []
-        for id in bad:
+            for id, available_formats in bad:
                 title = db.title(id, True)
-            res.append('%s'%title)
+                if available_formats:
+                    msg = _('No supported formats (Available formats: %s)')%(
+                        ', '.join(available_formats))
+                else:
+                    msg = _('This book has no actual ebook files')
+                res.append('%s - %s'%(title, msg))
+
 
             msg = '%s' % '\n'.join(res)
             warning_dialog(parent, _('Could not convert some books'),
-            _('Could not convert %(num)d of %(tot)d books, because no suitable source'
+                _('Could not convert %(num)d of %(tot)d books, because no supported source'
-               ' format was found.') % dict(num=len(res), tot=total),
+                ' formats were found.') % dict(num=len(res), tot=total),
                 msg).exec_()
 
     return jobs, changed, bad

src/calibre/library/server/browse.py

@@ -441,6 +441,10 @@ class BrowseServer(object):
             cat_len = len(category)
             if not (len(ucat) > cat_len and ucat.startswith(category+'.')):
                 continue
+
+            if ucat in self.icon_map:
+                icon = '_'+quote(self.icon_map[ucat])
+            else:
                 icon = category_icon_map['user:']
             # we have a subcategory. Find any further dots (further subcats)
             cat_len += 1