Sync to trunk.

recipes/divahair.recipe

@@ -0,0 +1,53 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+divahair.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class DivaHair(BasicNewsRecipe):
+    title                 = u'Diva Hair'
+    language              = 'ro'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'Coafuri, frizuri, tunsori ..'
+    publisher             = u'Diva Hair'
+    category              = u'Ziare,Stiri,Coafuri,Femei'
+    oldest_article        = 5
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    encoding              = 'utf-8'
+    remove_javascript     = True
+    cover_url             = 'http://www.divahair.ro/imgs/logo.jpg'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                            ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [
+                         dict(name='td', attrs={'class':'spatiuart'})
+                        , dict(name='div', attrs={'class':'spatiuart'})
+                     ]
+
+
+    remove_tags = [
+                     dict(name='div', attrs={'class':'categorie'})
+                     , dict(name='div', attrs={'class':'gri gri2 detaliiart'})
+                     , dict(name='div', attrs={'class':'articol_box_bottom'})
+                  ]
+
+    remove_tags_after = [
+                             dict(name='div', attrs={'class':'articol_box_bottom'})
+                        ]
+
+    feeds = [ (u'\u0218tiri', u'http://www.divahair.ro/feed') ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)

recipes/mayra.recipe

@@ -0,0 +1,51 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+mayra.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class Mayra(BasicNewsRecipe):
+    title                 = u'Mayra'
+    language              = 'ro'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'Traieste urban, cool, sexy'
+    publisher             = 'Mayra'
+    category              = 'Ziare,Stiri,Reviste'
+    oldest_article        = 5
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    encoding              = 'utf-8'
+    remove_javascript     = True
+    cover_url             = 'http://img.konkurs.ro/img/concursuri-cu-premii/147/14672_front.jpg'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                            ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [
+                        dict(name='div', attrs={'id':'article_details'})
+                     ]
+
+    remove_tags = [
+                     dict(name='div', attrs={'id':'LikePluginPagelet'})
+                     , dict(name='p', attrs={'id':'tags'})
+                    , dict(name='span', attrs={'id':'tweet-button'})
+                  ]
+
+    remove_tags_after = [
+                             dict(name='div', attrs={'id':'LikePluginPagelet'})
+                        ]
+
+    feeds = [ (u'\u0218tiri', u'http://www.mayra.ro/rss') ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)

recipes/moldovaazi.recipe

@@ -0,0 +1,50 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+azi.md
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class MoldovaAzi(BasicNewsRecipe):
+    title                 = u'Moldova Azi'
+    language              = 'ro'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'Moldova pe internet'
+    publisher             = 'Moldova Azi'
+    category              = 'Ziare,Stiri,Moldova'
+    oldest_article        = 5
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    encoding              = 'utf-8'
+    remove_javascript     = True
+    cover_url             = 'http://www.azi.md/images/logo.gif'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                            ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [  dict(name='div', attrs={'id':'in'})
+                     ]
+
+
+    remove_tags = [
+                     dict(name='div', attrs={'class':'in-more-stories'})
+                  ]
+
+    remove_tags_after = [
+                              dict(name='div', attrs={'id':'comment_wrapper'})
+                            , dict(name='div', attrs={'class':'box-title4'})
+                        ]
+
+    feeds = [ (u'\u0218tiri', u'http://www.azi.md/ro/feeds/0/rss201') ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)

recipes/newsmoldova.recipe

@@ -0,0 +1,50 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+newsmoldova.md
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class NewsMoldova(BasicNewsRecipe):
+    title                 = u'Agen\u0163ia de \u015ftiri Moldova'
+    language              = 'ro'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'Agen\u0163ia de \u015ftiri Moldova'
+    publisher             = 'Moldova'
+    category              = 'Ziare,Stiri,Moldova'
+    oldest_article        = 5
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    encoding              = 'utf-8'
+    remove_javascript     = True
+    cover_url             = 'http://www.newsmoldova.md/i/logo_top_md.gif'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                            ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [  dict(name='div', attrs={'class':'main-article-index article'})
+                     ]
+
+
+    remove_tags = [
+                      dict(name='div', attrs={'id':'actions'})
+                     , dict(name='li', attrs={'class':'invisible'})
+                  ]
+
+    remove_tags_after = [
+                             dict(name='div', attrs={'id':'actions'})
+                        ]
+
+    feeds = [ (u'\u0218tiri', u'http://newsmoldova.md/export/rss2/archive/index.xml') ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)

setup/commands.py

@@ -11,7 +11,7 @@ __all__ = [
         'build', 'build_pdf2xml', 'server',
         'gui',
         'develop', 'install',
-        'resources',
+        'kakasi', 'resources',
         'check',
         'sdist',
         'manual', 'tag_release',
@@ -49,8 +49,9 @@ gui = GUI()
 from setup.check import Check
 check = Check()
 
-from setup.resources import Resources
+from setup.resources import Resources, Kakasi
 resources = Resources()
+kakasi = Kakasi()
 
 from setup.publish import Manual, TagRelease, Stage1, Stage2, \
         Stage3, Stage4, Publish

setup/installer/windows/__init__.py

@@ -32,6 +32,7 @@ class Win32(VMInstaller):
     FREEZE_TEMPLATE = 'python -OO setup.py {freeze_command} --no-ice'
     INSTALLER_EXT = 'msi'
     SHUTDOWN_CMD = ['shutdown.exe', '-s', '-f', '-t', '0']
+    BUILD_BUILD  =  ['python setup.py kakasi',] + VMInstaller.BUILD_BUILD
 
     def download_installer(self):
         installer = self.installer()

setup/installer/windows/freeze.py

@@ -14,7 +14,7 @@ from setup.build_environment import msvc, MT, RC
 from setup.installer.windows.wix import WixMixIn
 
 OPENSSL_DIR = r'Q:\openssl'
-QT_DIR = 'Q:\\Qt\\4.7.2'
+QT_DIR = 'Q:\\Qt\\4.7.3'
 QT_DLLS = ['Core', 'Gui', 'Network', 'Svg', 'WebKit', 'Xml', 'XmlPatterns']
 LIBUSB_DIR       = 'C:\\libusb'
 LIBUNRAR         = 'C:\\Program Files\\UnrarDLL\\unrar.dll'

setup/installer/windows/wix-template.xml

@@ -11,9 +11,6 @@
 						SummaryCodepage='1252' />
 
                 <Media Id="1" Cabinet="{app}.cab" CompressionLevel="{compression}" EmbedCab="yes" />
-                <!-- The following line is needed because of the patch to QtCore4.dll. You can remove this line
-                     after you update Qt beyond 4.7.2. 'emus' means re-install even if version is the same not just if it is older. -->
-                <Property Id='REINSTALLMODE' Value='emus'/>
 
                 <Upgrade Id="{upgrade_code}">
                     <UpgradeVersion Maximum="{version}"

setup/resources.py

@@ -9,7 +9,7 @@ __docformat__ = 'restructuredtext en'
 import os, cPickle, re, anydbm, shutil, marshal, zipfile, glob
 from zlib import compress
 
-from setup import Command, basenames, __appname__
+from setup import Command, basenames, __appname__, iswindows
 
 def get_opts_from_parser(parser):
     def do_opt(opt):
@@ -23,13 +23,119 @@ def get_opts_from_parser(parser):
         for o in g.option_list:
             for x in do_opt(o): yield x
 
-class Resources(Command):
+class Kakasi(Command):
 
-    description = 'Compile various needed calibre resources'
+    description = 'Compile resources for unihandecode'
 
     KAKASI_PATH = os.path.join(Command.SRC,  __appname__,
             'ebooks', 'unihandecode', 'pykakasi')
 
+    def run(self, opts):
+        self.records = {}
+        src = self.j(self.KAKASI_PATH, 'kakasidict.utf8')
+        dest = self.j(self.RESOURCES, 'localization',
+                'pykakasi','kanwadict2.db')
+        base = os.path.dirname(dest)
+        if not os.path.exists(base):
+            os.makedirs(base)
+
+        if self.newer(dest, src) or iswindows:
+            self.info('\tGenerating Kanwadict')
+
+            for line in open(src, "r"):
+                self.parsekdict(line)
+            self.kanwaout(dest)
+
+        src = self.j(self.KAKASI_PATH, 'itaijidict.utf8')
+        dest = self.j(self.RESOURCES, 'localization',
+                'pykakasi','itaijidict2.pickle')
+
+        if self.newer(dest, src) or iswindows:
+            self.info('\tGenerating Itaijidict')
+            self.mkitaiji(src, dest)
+
+        src = self.j(self.KAKASI_PATH, 'kanadict.utf8')
+        dest = self.j(self.RESOURCES, 'localization',
+                'pykakasi','kanadict2.pickle')
+
+        if self.newer(dest, src) or iswindows:
+            self.info('\tGenerating kanadict')
+            self.mkkanadict(src, dest)
+
+        return
+
+
+    def mkitaiji(self, src, dst):
+        dic = {}
+        for line in open(src, "r"):
+            line = line.decode("utf-8").strip()
+            if line.startswith(';;'): # skip comment
+                continue
+            if re.match(r"^$",line):
+                continue
+            pair = re.sub(r'\\u([0-9a-fA-F]{4})', lambda x:unichr(int(x.group(1),16)), line)
+            dic[pair[0]] = pair[1]
+        cPickle.dump(dic, open(dst, 'w'), protocol=-1) #pickle
+
+    def mkkanadict(self, src, dst):
+        dic = {}
+        for line in open(src, "r"):
+            line = line.decode("utf-8").strip()
+            if line.startswith(';;'): # skip comment
+                continue
+            if re.match(r"^$",line):
+                continue
+            (alpha, kana) = line.split(' ')
+            dic[kana] = alpha
+        cPickle.dump(dic, open(dst, 'w'), protocol=-1) #pickle
+
+    def parsekdict(self, line):
+        line = line.decode("utf-8").strip()
+        if line.startswith(';;'): # skip comment
+            return
+        (yomi, kanji) = line.split(' ')
+        if ord(yomi[-1:]) <= ord('z'):
+            tail = yomi[-1:]
+            yomi = yomi[:-1]
+        else:
+            tail = ''
+        self.updaterec(kanji, yomi, tail)
+
+    def updaterec(self, kanji, yomi, tail):
+        key = "%04x"%ord(kanji[0])
+        if key in self.records:
+            if kanji in self.records[key]:
+                rec = self.records[key][kanji]
+                rec.append((yomi,tail))
+                self.records[key].update( {kanji: rec} )
+            else:
+                self.records[key][kanji]=[(yomi, tail)]
+        else:
+            self.records[key] = {}
+            self.records[key][kanji]=[(yomi, tail)]
+
+    def kanwaout(self, out):
+        try:
+            # Needed as otherwise anydbm tries to create a gdbm db when the db
+            # created on Unix is found
+            os.remove(out)
+        except:
+            pass
+        dic = anydbm.open(out, 'n')
+        for (k, v) in self.records.iteritems():
+            dic[k] = compress(marshal.dumps(v))
+        dic.close()
+
+    def clean(self):
+        kakasi = self.j(self.RESOURCES, 'localization', 'pykakasi')
+        if os.path.exists(kakasi):
+            shutil.rmtree(kakasi)
+
+class Resources(Command):
+
+    description = 'Compile various needed calibre resources'
+    sub_commands = ['kakasi']
+
     def run(self, opts):
         scripts = {}
         for x in ('console', 'gui'):
@@ -117,108 +223,13 @@ class Resources(Command):
         import json
         json.dump(function_dict, open(dest, 'wb'), indent=4)
 
-        self.run_kakasi(opts)
-
-    def run_kakasi(self, opts):
-        self.records = {}
-        src = self.j(self.KAKASI_PATH, 'kakasidict.utf8')
-        dest = self.j(self.RESOURCES, 'localization',
-                'pykakasi','kanwadict2.db')
-        base = os.path.dirname(dest)
-        if not os.path.exists(base):
-            os.makedirs(base)
-
-        if self.newer(dest, src):
-            self.info('\tGenerating Kanwadict')
-
-            for line in open(src, "r"):
-                self.parsekdict(line)
-            self.kanwaout(dest)
-
-        src = self.j(self.KAKASI_PATH, 'itaijidict.utf8')
-        dest = self.j(self.RESOURCES, 'localization',
-                'pykakasi','itaijidict2.pickle')
-
-        if self.newer(dest, src):
-            self.info('\tGenerating Itaijidict')
-            self.mkitaiji(src, dest)
-
-        src = self.j(self.KAKASI_PATH, 'kanadict.utf8')
-        dest = self.j(self.RESOURCES, 'localization',
-                'pykakasi','kanadict2.pickle')
-
-        if self.newer(dest, src):
-            self.info('\tGenerating kanadict')
-            self.mkkanadict(src, dest)
-
-        return
-
-
-    def mkitaiji(self, src, dst):
-        dic = {}
-        for line in open(src, "r"):
-            line = line.decode("utf-8").strip()
-            if line.startswith(';;'): # skip comment
-                continue
-            if re.match(r"^$",line):
-                continue
-            pair = re.sub(r'\\u([0-9a-fA-F]{4})', lambda x:unichr(int(x.group(1),16)), line)
-            dic[pair[0]] = pair[1]
-        cPickle.dump(dic, open(dst, 'w'), protocol=-1) #pickle
-
-    def mkkanadict(self, src, dst):
-        dic = {}
-        for line in open(src, "r"):
-            line = line.decode("utf-8").strip()
-            if line.startswith(';;'): # skip comment
-                continue
-            if re.match(r"^$",line):
-                continue
-            (alpha, kana) = line.split(' ')
-            dic[kana] = alpha
-        cPickle.dump(dic, open(dst, 'w'), protocol=-1) #pickle
-
-    def parsekdict(self, line):
-        line = line.decode("utf-8").strip()
-        if line.startswith(';;'): # skip comment
-            return
-        (yomi, kanji) = line.split(' ')
-        if ord(yomi[-1:]) <= ord('z'):
-            tail = yomi[-1:]
-            yomi = yomi[:-1]
-        else:
-            tail = ''
-        self.updaterec(kanji, yomi, tail)
-
-    def updaterec(self, kanji, yomi, tail):
-            key = "%04x"%ord(kanji[0])
-            if key in self.records:
-                if kanji in self.records[key]:
-                    rec = self.records[key][kanji]
-                    rec.append((yomi,tail))
-                    self.records[key].update( {kanji: rec} )
-                else:
-                    self.records[key][kanji]=[(yomi, tail)]
-            else:
-                self.records[key] = {}
-                self.records[key][kanji]=[(yomi, tail)]
-
-    def kanwaout(self, out):
-        dic = anydbm.open(out, 'c')
-        for (k, v) in self.records.iteritems():
-            dic[k] = compress(marshal.dumps(v))
-        dic.close()
-
-
     def clean(self):
         for x in ('scripts', 'recipes', 'ebook-convert-complete'):
             x = self.j(self.RESOURCES, x+'.pickle')
             if os.path.exists(x):
                 os.remove(x)
-        kakasi = self.j(self.RESOURCES, 'localization', 'pykakasi')
-        if os.path.exists(kakasi):
-            shutil.rmtree(kakasi)
-
+        from setup.commands import kakasi
+        kakasi.clean()
 
 
 

src/calibre/ebooks/rtf/input.py

@@ -86,7 +86,7 @@ class RTFInput(InputFormatPlugin):
                 run_lev = 4
                 self.log('Running RTFParser in debug mode')
             except:
-                pass
+                self.log.warn('Impossible to run RTFParser in debug mode')
         parser = ParseRtf(
             in_file    = stream,
             out_file   = ofile,

src/calibre/ebooks/rtf2xml/process_tokens.py

@@ -197,8 +197,8 @@ class ProcessTokens:
         # character info => ci
         'b'                  :	('ci', 'bold______', self.bool_st_func),
         'blue'               :	('ci', 'blue______', self.color_func),
-        'caps'               :  ('ci', 'caps______', self.bool_st_func),
-        'cf'                 :	('ci', 'font-color', self.default_func),
+        'caps'               :	('ci', 'caps______', self.bool_st_func),
+        'cf'                 :	('ci', 'font-color', self.colorz_func),
         'chftn'              :	('ci', 'footnot-mk', self.bool_st_func),
         'dn'                 :	('ci', 'font-down_', self.divide_by_2),
         'embo'               :	('ci', 'emboss____', self.bool_st_func),
@@ -624,6 +624,11 @@ class ProcessTokens:
             num = 'true'
         return 'cw<%s<%s<nu<%s\n' % (pre, token, num)
 
+    def colorz_func(self, pre, token, num):
+        if num is None:
+            num = '0'
+        return 'cw<%s<%s<nu<%s\n' % (pre, token, num)
+
     def __list_type_func(self, pre, token, num):
         type = 'arabic'
         if num is None:

src/calibre/gui2/actions/store.py

@@ -27,9 +27,9 @@ class StoreAction(InterfaceAction):
     def load_menu(self):
         self.store_menu.clear()
         self.store_menu.addAction(_('Search for ebooks'), self.search)
-        self.store_menu.addAction(_('Search by this author'), self.search_author)
-        self.store_menu.addAction(_('Search by this title'), self.search_title)
-        self.store_menu.addAction(_('Search by this author and title'), self.search_author_title)
+        self.store_menu.addAction(_('Search for this author'), self.search_author)
+        self.store_menu.addAction(_('Search for this title'), self.search_title)
+        self.store_menu.addAction(_('Search for this book'), self.search_author_title)
         self.store_menu.addSeparator()
         self.store_list_menu = self.store_menu.addMenu(_('Stores'))
         for n, p in sorted(self.gui.istores.items(), key=lambda x: x[0].lower()):

src/calibre/gui2/preferences/plugins.py

@@ -75,6 +75,8 @@ class PluginModel(QAbstractItemModel, SearchQueryParser): # {{{
 
     def find(self, query):
         query = query.strip()
+        if not query:
+            return QModelIndex()
         matches = self.parse(query)
         if not matches:
             return QModelIndex()
@@ -87,6 +89,8 @@ class PluginModel(QAbstractItemModel, SearchQueryParser): # {{{
 
     def find_next(self, idx, query, backwards=False):
         query = query.strip()
+        if not query:
+            return idx
         matches = self.parse(query)
         if not matches:
             return idx

src/calibre/gui2/wizard/__init__.py

@@ -435,7 +435,7 @@ class DevicePage(QWizardPage, DeviceUI):
         self.registerField("device", self.device_view)
 
     def initializePage(self):
-        self.label.setText(_('Choose you e-book device. If your device is'
+        self.label.setText(_('Choose your e-book device. If your device is'
             ' not in the list, choose a "%s" device.')%Device.manufacturer)
         self.man_model = ManufacturerModel()
         self.manufacturer_view.setModel(self.man_model)

src/odf/odf2xhtml.py

@@ -1415,18 +1415,34 @@ ol, ul { padding-left: 2em; }
         self.writedata()
         c = attrs.get( (TEXTNS,'style-name'), None)
         htmlattrs = {}
+        # Changed by Kovid to handle inline special styles defined on <text:span> tags.
+        # Apparently LibreOffice does this.
+        special = 'span'
         if c:
             c = c.replace(".","_")
             special = special_styles.get("S-"+c)
-            if special is None and self.generate_css:
-                htmlattrs['class'] = "S-%s" % c
-        self.opentag('span', htmlattrs)
+            if special is None:
+                special = 'span'
+                if self.generate_css:
+                    htmlattrs['class'] = "S-%s" % c
+
+        self.opentag(special, htmlattrs)
         self.purgedata()
 
     def e_text_span(self, tag, attrs):
         """ End the <text:span> """
         self.writedata()
-        self.closetag('span', False)
+        c = attrs.get( (TEXTNS,'style-name'), None)
+        # Changed by Kovid to handle inline special styles defined on <text:span> tags.
+        # Apparently LibreOffice does this.
+        special = 'span'
+        if c:
+            c = c.replace(".","_")
+            special = special_styles.get("S-"+c)
+            if special is None:
+                special = 'span'
+
+        self.closetag(special, False)
         self.purgedata()
 
     def s_text_tab(self, tag, attrs):