Merge from trunk

resources/recipes/adevarul.recipe

@@ -0,0 +1,50 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+adevarul.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class Adevarul(BasicNewsRecipe):
+    title                 = u'Adev\u0103rul'
+    language              = 'ro'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'\u0218tiri din Rom\u00e2nia'
+    publisher             = 'Adevarul'
+    category              = 'Ziare,Stiri,Romania'
+    oldest_article        = 5
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    encoding              = 'utf-8'
+    remove_javascript     = True
+    cover_url         = 'http://upload.wikimedia.org/wikipedia/en/d/d6/Logo_noul_adevarul.png'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [  dict(name='div', attrs={'class':'article_header'})
+                       ,dict(name='div', attrs={'class':'bd'})
+                     ]
+
+
+    remove_tags = [  dict(name='div', attrs={'class':'bb-wg-article_related_attachements'})
+                    ,dict(name='div', attrs={'class':'bb-md bb-md-article_comments'})
+                ,dict(name='form', attrs={'id':'bb-comment-create-form'})
+              ]
+
+    remove_tags_after = [ dict(name='form', attrs={'id':'bb-comment-create-form'}) ]
+
+    feeds = [ (u'\u0218tiri', u'http://www.adevarul.ro/rss/latest') ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)
+

resources/recipes/capital.recipe

@@ -0,0 +1,44 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+capital.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class Capital(BasicNewsRecipe):
+    title                 = 'Capital'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'\u0218tiri din Rom\u00e2nia'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare,Stiri,Romania'
+    encoding              = 'utf-8'
+    remove_javascript     = True
+    publisher             = 'Capital'
+    cover_url         = 'http://www.mediapress.ro/imagini/sigla-capital-s16.gif'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [  dict(name='div', attrs={'class':'single one_article'})
+                     ]
+
+    remove_tags = [  dict(name='div', attrs={'class':'single_details'})
+                   , dict(name='div', attrs={'class':'tx-addoceansbanners-pi1'})
+          ]
+
+    feeds = [(u'\u0218tiri', u'http://www.capital.ro/rss.html') ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)

resources/recipes/catavencu.recipe

@@ -0,0 +1,53 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+catavencu.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class Catavencu(BasicNewsRecipe):
+    title                 = u'Academia Ca\u0163avencu'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = 'Tagma cum laude'
+    publisher             = 'Catavencu'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare'
+    encoding              = 'utf-8'
+    cover_url         = 'http://upload.wikimedia.org/wikipedia/en/1/1e/Academia_Catavencu.jpg'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [
+            dict(name='ul', attrs={'class':'articles'})
+                     ]
+
+    remove_tags = [
+             dict(name='div', attrs={'class':['tools']})
+           , dict(name='div', attrs={'class':['share']})
+           , dict(name='div', attrs={'class':['category']})
+           , dict(name='div', attrs={'id':['comments']})
+                  ]
+
+    remove_tags_after = [
+              dict(name='div', attrs={'id':'comments'})
+            ]
+
+    feeds          = [
+            (u'Feeds', u'http://catavencu.ro/feed/rss')
+                 ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)

resources/recipes/gandul.recipe

@@ -0,0 +1,47 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+gandul.info
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class Gandul(BasicNewsRecipe):
+    title                 = u'G\u00E2ndul'
+    __author__            = u'Silviu Cotoar\u0103'
+    publisher             = 'Gandul'
+    description           = 'Cotidian Online'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare,Stiri,Romania'
+    encoding              = 'utf-8'
+    cover_url         = 'http://storage0.dms.mpinteractiv.ro/media/1/1/1706/1064063/1/logo.jpg?width=400'
+
+    conversion_options = {
+                'comments'    : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [
+                    dict(name='div', attrs={'class':'article'})
+             ]
+
+    remove_tags = [
+             dict(name='a', attrs={'class':'photo'})
+           , dict(name='div', attrs={'class':'ad'})
+                  ]
+
+    feeds  = [
+        (u'\u0218tiri', u'http://www.gandul.info/rss-stiri-prima-pagina.xml')
+         ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)

resources/recipes/hotnews.recipe

@@ -0,0 +1,46 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+hotnews.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class Hotnews(BasicNewsRecipe):
+    title                 = 'Hotnews'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'\u0218tiri din Rom\u00e2nia'
+    publisher             = 'Hotnews'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare,Stiri,Romania'
+    encoding              = 'utf-8'
+    cover_url         = 'http://www.hotnews.ro/images/new/logo.gif'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [
+            dict(name='h1', attrs={'class':'title'})
+                   ,dict(name='div', attrs={'id':'articleContent'})
+                     ]
+
+    feeds          = [   (u'\u0218tiri', u'http://www.hotnews.ro/rss/actualitate')
+            ,(u'English', u'http://www.hotnews.ro/rss/english')
+                 ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)
+
+
+

resources/recipes/jurnalulnational.recipe

@@ -0,0 +1,54 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+jurnalul.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class JurnalulNational(BasicNewsRecipe):
+    title                 = u'Jurnalul Na\u0163ional'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'\u0218tiri din Rom\u00e2nia'
+    publisher             = 'Jurnalul National'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare,Stiri,Romania'
+    encoding              = 'utf-8'
+    cover_url         = 'http://www.jurnalul.ro/images/sigla.png'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [
+            dict(name='h1', attrs={'class':'h3 art_title'})
+                   ,dict(name='div', attrs={'class':'only_text'})
+                     ]
+
+    feeds          = [
+             (u'\u0218tiri', u'http://www.jurnalul.ro/rss/stiri-3028.html')
+                ,(u'Special', u'http://www.jurnalul.ro/rss/special-3001.html')
+                    ,(u'Sport', u'http://www.jurnalul.ro/rss/sport-3035.html')
+            ,(u'Bani Afaceri', u'http://www.jurnalul.ro/rss/bani-afaceri-3006.html')
+            ,(u'Viata Sanatoasa', u'http://www.jurnalul.ro/rss/viata-sanatoasa-3010.html')
+            ,(u'Stiinta Tehnica', u'http://www.jurnalul.ro/rss/stiinta-tehnica-3019.html')
+            ,(u'Timp Liber', u'http://www.jurnalul.ro/rss/timp-liber-3022.html')
+            ,(u'Fun', u'http://www.jurnalul.ro/rss/fun-3038.html')
+            ,(u'Acum 20 de ani', u'http://www.jurnalul.ro/rss/acum-20-de-ani-3073.html')
+                 ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)
+
+
+

resources/recipes/mediafax.recipe

@@ -0,0 +1,52 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+mediafax.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class Mediafax(BasicNewsRecipe):
+    title                 = 'Mediafax'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'\u0218tiri din Rom\u00e2nia'
+    publisher             = 'Mediafax'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare,Stiri,Romania'
+    encoding              = 'utf-8'
+    cover_url         = 'http://storage0.dms.mpinteractiv.ro/media/1/1/1706/4134575/2/logo-mediafax-mass-media-news.jpg?width=400'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+
+                         }
+
+    keep_only_tags = [
+            dict(name='div', attrs={'class':'news tabs-container'})
+                     ]
+
+    remove_tags = [
+            dict(name='ul', attrs={'class':['CategoryNews']})
+           ,dict(name='div', attrs={'class':['read']})
+                  ]
+
+    remove_tags_after = [ dict(name='div', attrs={'class':'cmsItemViews'}) ]
+
+
+    feeds       = [
+            (u'Actualitate', u'http://www.mediafax.ro/rss/')
+              ]
+
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)

resources/recipes/moneyro.recipe

@@ -0,0 +1,54 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+money.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class MoneyRo(BasicNewsRecipe):
+    title                 = 'Money Ro'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'\u0218tiri din Rom\u00e2nia'
+    publisher             = 'MoneyRo'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare,Stiri,Romania'
+    encoding              = 'utf-8'
+    remove_javascript     = True
+    cover_url         = 'http://assets.moneyweb.ro/images/logo_money.jpg'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [   dict(name='div', attrs={'id':'titluArticol'})
+               , dict(name='img', attrs={'id':'objImage'})
+               , dict(name='div', attrs={'class':'leftColumnArticle'})
+                     ]
+
+    remove_tags_after = [ dict(name='div', attrs={'id':'articleTags'}) ]
+
+    remove_tags = [  dict(name='div', attrs={'id':'ads'})
+           , dict(name='div', attrs={'id':'aus'})
+           , dict(name='div', attrs={'id':'bb-comment-create-form'})
+           , dict(name='div', attrs={'id':'articleTags'})
+           , dict(name='div', attrs={'class':'breadcrumb'})
+          ]
+
+    feeds = [(u'\u0218tiri', u'http://moneyro.feedsportal.com/c/32533/fe.ed/rss.money.ro/stiri.xml') ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)
+
+
+

resources/recipes/prosport.recipe

@@ -0,0 +1,49 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+prosport.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class Prosport(BasicNewsRecipe):
+    title                 = 'Prosport'
+    __author__            = u'Silviu Cotoar\u0103'
+    publisher             = 'Prosport'
+    description           = u'\u0218tiri Sportive din Rom\u00e2nia'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare,Stiri,Romania,Sport'
+    encoding              = 'utf-8'
+    cover_url         = 'http://storage0.dms.mpinteractiv.ro/media/401/581/7946/3688311/1/logo-pro.jpg?width=610'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [
+            dict(name='h1', attrs={'class':'a-title'})
+                   ,dict(name='div', attrs={'class':'a-entry'})
+                     ]
+
+    remove_tags = [  dict(name='div', attrs={'class':'utils'})
+            ,dict(name='div', attrs={'class':'g-slide'})
+              ]
+
+
+    feeds          = [ (u'\u0218tiri', u'http://www.prosport.ro/rss.xml')]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)
+
+
+

resources/recipes/realitatea.recipe

@@ -0,0 +1,45 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+realitatea.net
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class Realitatea(BasicNewsRecipe):
+    title                 = 'Realitatea'
+    __author__            = u'Silviu Cotoar\u0103'
+    publisher             = 'Realitatea'
+    description           = u'\u0218tiri din Rom\u00e2nia'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare,Stiri,Romania'
+    encoding              = 'utf-8'
+    cover_url         = 'http://assets.realitatea.ro/images/logo.jpg'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [
+            dict(name='div', attrs={'class':'articleTitle '})
+                   ,dict(name='div', attrs={'class':'articleBody'})
+                     ]
+
+    remove_tags = [  dict(name='div', attrs={'id':'aus'}) ]
+    feeds          = [ (u'\u0218tiri', u'http://realitatea.feedsportal.com/c/32533/fe.ed/rss.realitatea.net/stiri.xml') ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)
+
+
+

resources/recipes/standardmoney.recipe

@@ -0,0 +1,46 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+standard.money.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class StandardMoneyRo(BasicNewsRecipe):
+    title                 = 'Standard Money Ro'
+    __author__            = u'Silviu Cotoar\u0103'
+    publisher             = 'Standard Money'
+    description           = 'Portal de Business'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare,Stiri,Romania'
+    encoding              = 'utf-8'
+    cover_url         = 'http://assets.standard.ro/wp-content/themes/standard/images/standard-logo.gif'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [
+             dict(name='h1', attrs={'class':'post-title'})
+                   , dict(name='div', attrs={'class':'content_post'})
+                     ]
+
+    feeds          = [
+            (u'Actualitate', u'http://standard.money.ro/feed')
+                 ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)
+
+
+

resources/recipes/ziarulfinanciar.recipe

@@ -0,0 +1,45 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = u'2011, Silviu Cotoar\u0103'
+'''
+zf.ro
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class ZiarulFinanciar(BasicNewsRecipe):
+    title                 = 'Ziarul Financiar'
+    __author__            = u'Silviu Cotoar\u0103'
+    description           = u'\u0218tiri din Business'
+    publisher             = 'Ziarul Financiar'
+    oldest_article        = 5
+    language              = 'ro'
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    use_embedded_content  = False
+    category              = 'Ziare,Stiri,Romania'
+    encoding              = 'utf-8'
+    cover_url         = 'http://storage0.dms.mpinteractiv.ro/media/1/1/1706/7462721/1/ziarul-financiar-big.jpg?width=400'
+
+    conversion_options = {
+                             'comments'   : description
+                            ,'tags'       : category
+                            ,'language'   : language
+                ,'publisher'  : publisher
+                         }
+
+    keep_only_tags = [
+            dict(name='div', attrs={'class':'article'})
+                     ]
+
+    feeds          = [
+            (u'\u0218tiri', u'http://www.zf.ro/rss/zf-24/')
+                 ]
+
+    def preprocess_html(self, soup):
+        return self.adeify_images(soup)
+
+
+

src/calibre/customize/builtins.py

@@ -2,11 +2,13 @@ import os.path
 __license__   = 'GPL v3'
 __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
 
-import textwrap, os, glob, functools
+import textwrap, os, glob, functools, re
+from calibre import guess_type
 from calibre.customize import FileTypePlugin, MetadataReaderPlugin, \
     MetadataWriterPlugin, PreferencesPlugin, InterfaceActionBase
 from calibre.constants import numeric_version
 from calibre.ebooks.metadata.archive import ArchiveExtract, get_cbz_metadata
+from calibre.ebooks.oeb.base import OEB_IMAGES
 
 # To archive plugins {{{
 class HTML2ZIP(FileTypePlugin):
@@ -82,6 +84,66 @@ class PML2PMLZ(FileTypePlugin):
 
         return of.name
 
+class TXT2TXTZ(FileTypePlugin):
+    name = 'TXT to TXTZ'
+    author = 'John Schember'
+    description = _('Create a TXTZ archive when a TXT file is imported '
+        'containing Markdown or Textile references to images. The referenced '
+        'images as well as the TXT file are added to the archive.')
+    version = numeric_version
+    file_types = set(['txt'])
+    supported_platforms = ['windows', 'osx', 'linux']
+    on_import = True
+    
+    def _get_image_references(self, txt, base_dir):
+        images = []
+        
+        # Textile
+        for m in re.finditer(ur'(?mu)(?:[\[{])?\!(?:\. )?(?P<path>[^\s(!]+)\s?(?:\(([^\)]+)\))?\!(?::(\S+))?(?:[\]}]|(?=\s|$))', txt):
+            path = m.group('path')
+            if path and not os.path.isabs(path) and guess_type(path)[0] in OEB_IMAGES and os.path.exists(os.path.join(base_dir, path)):
+                images.append(path)
+                
+        # Markdown inline        
+        for m in re.finditer(ur'(?mu)\!\[([^\]\[]*(\[[^\]\[]*(\[[^\]\[]*(\[[^\]\[]*(\[[^\]\[]*(\[[^\]\[]*(\[[^\]\[]*\])*[^\]\[]*\])*[^\]\[]*\])*[^\]\[]*\])*[^\]\[]*\])*[^\]\[]*\])*[^\]\[]*)\]\s*\((?P<path>[^\)]*)\)', txt):
+            path = m.group('path')
+            if path and not os.path.isabs(path) and guess_type(path)[0] in OEB_IMAGES and os.path.exists(os.path.join(base_dir, path)):
+                images.append(path)
+        
+        # Markdown reference
+        refs = {}
+        for m in re.finditer(ur'(?mu)^(\ ?\ ?\ ?)\[(?P<id>[^\]]*)\]:\s*(?P<path>[^\s]*)$', txt):
+            if m.group('id') and m.group('path'):
+                refs[m.group('id')] = m.group('path')
+        for m in re.finditer(ur'(?mu)\!\[([^\]\[]*(\[[^\]\[]*(\[[^\]\[]*(\[[^\]\[]*(\[[^\]\[]*(\[[^\]\[]*(\[[^\]\[]*\])*[^\]\[]*\])*[^\]\[]*\])*[^\]\[]*\])*[^\]\[]*\])*[^\]\[]*\])*[^\]\[]*)\]\s*\[(?P<id>[^\]]*)\]', txt):
+            path = refs.get(m.group('id'), None)
+            if path and not os.path.isabs(path) and guess_type(path)[0] in OEB_IMAGES and os.path.exists(os.path.join(base_dir, path)):
+                images.append(path)
+
+        # Remove duplicates
+        return list(set(images))
+    
+    def run(self, path_to_ebook):
+        with open(path_to_ebook, 'rb') as ebf:
+            txt = ebf.read()
+        base_dir = os.path.dirname(path_to_ebook)
+        images = self._get_image_references(txt, base_dir)
+        
+        if images:
+            # Create TXTZ and put file plus images inside of it.
+            import zipfile
+            of = self.temporary_file('_plugin_txt2txtz.txtz')
+            txtz = zipfile.ZipFile(of.name, 'w')
+            txtz.write(path_to_ebook, os.path.basename(path_to_ebook), zipfile.ZIP_DEFLATED)
+            for image in images:
+                txtz.write(os.path.join(base_dir, image), image)
+            txtz.close()
+
+            return of.name
+        else:
+            # No images so just import the TXT file.
+            return path_to_ebook
+
 # }}}
 
 # Metadata reader plugins {{{
@@ -516,7 +578,7 @@ from calibre.library.catalog import CSV_XML, EPUB_MOBI, BIBTEX
 from calibre.ebooks.epub.fix.unmanifested import Unmanifested
 from calibre.ebooks.epub.fix.epubcheck import Epubcheck
 
-plugins = [HTML2ZIP, PML2PMLZ, ArchiveExtract, GoogleBooks, ISBNDB, Amazon,
+plugins = [HTML2ZIP, PML2PMLZ, TXT2TXTZ, ArchiveExtract, GoogleBooks, ISBNDB, Amazon,
         KentDistrictLibrary, DoubanBooks, NiceBooks, CSV_XML, EPUB_MOBI, BIBTEX, Unmanifested,
         Epubcheck, OpenLibraryCovers, AmazonCovers, DoubanCovers,
         NiceBooksCovers]

src/calibre/gui2/library/models.py

@@ -825,7 +825,7 @@ class BooksModel(QAbstractTableModel): # {{{
                     return False
                 val = int(value.toInt()[0]) if column == 'rating' else \
                       value.toDate() if column in ('timestamp', 'pubdate') else \
-                      unicode(value.toString())
+                      unicode(value.toString()).strip()
                 id = self.db.id(row)
                 books_to_refresh = set([id])
                 if column == 'rating':

src/calibre/gui2/preferences/search.py

@@ -5,7 +5,7 @@ __license__   = 'GPL v3'
 __copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-from PyQt4.Qt import QApplication, QIcon
+from PyQt4.Qt import QApplication
 
 from calibre.gui2.preferences import ConfigWidgetBase, test_widget, \
         CommaSeparatedList
@@ -30,23 +30,31 @@ class ConfigWidget(ConfigWidgetBase, Ui_Form):
         self.opt_limit_search_columns_to.update_items_cache(fl)
         self.clear_history_button.clicked.connect(self.clear_histories)
 
-        self.gst_explanation.setText(_(
+        self.gst_explanation.setText('<p>' + _(
-    "Grouped search terms are search names that permit a query to automatically "
+    "<b>Grouped search terms</b> are search names that permit a query to automatically "
     "search across more than one column. For example, if you create a grouped "
-    "search term 'myseries' with the value 'series, #myseries, #myseries2', "
+    "search term <code>allseries</code> with the value "
-    "the query 'myseries:adhoc' will find the string 'adhoc' in any of the "
+    "<code>series, #myseries, #myseries2</code>, then "
-    "columns 'series', '#myseries', and '#myseries2'. Enter the name of the "
+    "the query <code>allseries:adhoc</code> will find 'adhoc' in any of the "
+    "columns 'series', '#myseries', and '#myseries2'.<p> Enter the name of the "
     "grouped search term in the drop-down box, enter the list of columns "
     "to search in the value box, then push the Save button. "
-    "Notes: You cannot create a search term that is a duplicate of an existing "
+    "<p>Note: Search terms are forced to lower case; 'MySearch' "
-    "term or user category. Search terms are forced to lower case; 'MySearch' "
+    "and 'mysearch' are the same term."
-    "and 'mysearch' are the same term."))
+    "<p>You can have your grouped search term show up as a user category in "
-
+    " the Tag Browser. Just add the grouped search term name to the user "
+    "category box. You can add multiple terms separated by commas. "
+    "The user category will be "
+    "populated with all the items in the categories included in the grouped "
+    "search term. <p>This permits you to see easily all the category items that "
+    "are in the fields contained in the grouped search term. Using the above "
+    "'myseries' example, the automatically-generated user category would contain "
+    "all the series mentioned in 'series', '#myseries1', and '#myseries2'. This "
+    "can be useful to check for duplicates or to find which column contains "
+    "a particular item."))
         self.gst = db.prefs.get('grouped_search_terms', {})
         self.orig_gst_keys = self.gst.keys()
 
-        del_icon = QIcon(I('trash.png'))
-        self.gst_delete_button.setIcon(del_icon)
         fl = []
         for f in db.all_field_keys():
             fm = db.metadata_for_field(f)
@@ -67,18 +75,7 @@ class ConfigWidget(ConfigWidgetBase, Ui_Form):
         self.gst_delete_button.clicked.connect(self.gst_delete_clicked)
         self.gst_changed = False
 
-        self.muc_explanation.setText(_(
+        if db.prefs.get('grouped_search_make_user_categories', None) is None:
-    "Add a grouped search term name to this box to automatically generate "
-    "a user category with the name of the search term. The user category will be "
-    "populated with all the items in the categories included in the grouped "
-    "search term. This permits you to see easily all the category items that "
-    "are in the fields contained in the grouped search term. Using the above "
-    "'myseries' example, the automatically-generated user category would contain "
-    "all the series mentioned in 'series', '#myseries1', and '#myseries2'. This "
-    "can be useful to check for duplications or to find which column contains "
-    "a particular item."))
-
-        if not db.prefs.get('grouped_search_make_user_categories', None):
             db.prefs.set('grouped_search_make_user_categories', [])
         r('grouped_search_make_user_categories', db.prefs, setting=CommaSeparatedList)
         self.muc_changed = False

src/calibre/gui2/preferences/search.ui

@@ -7,7 +7,7 @@
     <x>0</x>
     <y>0</y>
     <width>670</width>
-    <height>391</height>
+    <height>556</height>
    </rect>
   </property>
   <property name="windowTitle">
@@ -92,26 +92,16 @@
      <property name="title">
       <string>Grouped Search Terms</string>
      </property>
-     <layout class="QVBoxLayout">
+     <layout class="QGridLayout" name="gridLayout_3">
-      <item>
+      <item row="0" column="0">
-       <widget class="QLabel" name="gst_explanation">
-        <property name="text">
-         <string>TextLabel</string>
-        </property>
-        <property name="wordWrap">
-         <bool>true</bool>
-        </property>
-        <property name="textInteractionFlags">
-         <set>Qt::LinksAccessibleByMouse|Qt::TextSelectableByKeyboard|Qt::TextSelectableByMouse</set>
-        </property>
-       </widget>
-      </item>
-      <item>
        <layout class="QHBoxLayout" name="l12">
         <item>
          <widget class="QLabel" name="la10">
           <property name="text">
-           <string>Grouped search terms:</string>
+           <string>&amp;Grouped search terms:</string>
+          </property>
+          <property name="buddy">
+           <cstring>gst_names</cstring>
           </property>
          </widget>
         </item>
@@ -133,6 +123,10 @@
           <property name="text">
            <string>...</string>
           </property>
+          <property name="icon">
+           <iconset resource="../../../../resources/images.qrc">
+            <normaloff>:/images/trash.png</normaloff>:/images/trash.png</iconset>
+          </property>
          </widget>
         </item>
         <item>
@@ -146,31 +140,31 @@ changing the name then pressing Save. You can change the value
 of a search term by changing the value box then pressing Save.</string>
           </property>
           <property name="text">
-           <string>Save</string>
+           <string>&amp;Save</string>
           </property>
          </widget>
         </item>
        </layout>
       </item>
-      <item>
+      <item row="0" column="1" rowspan="3">
-       <widget class="QLabel" name="muc_explanation">
+       <widget class="QTextBrowser" name="gst_explanation">
-        <property name="text">
+        <property name="sizePolicy">
-         <string>TextLabel</string>
+         <sizepolicy hsizetype="Expanding" vsizetype="Expanding">
-        </property>
+          <horstretch>0</horstretch>
-        <property name="wordWrap">
+          <verstretch>100</verstretch>
-         <bool>true</bool>
+         </sizepolicy>
-        </property>
-        <property name="textInteractionFlags">
-         <set>Qt::LinksAccessibleByMouse|Qt::TextSelectableByKeyboard|Qt::TextSelectableByMouse</set>
         </property>
        </widget>
       </item>
-      <item>
+      <item row="1" column="0">
        <layout class="QHBoxLayout">
         <item>
          <widget class="QLabel" name="l11">
           <property name="text">
-           <string>Make user categories:</string>
+           <string>Make &amp;user categories:</string>
+          </property>
+          <property name="buddy">
+           <cstring>opt_grouped_search_make_user_categories</cstring>
           </property>
          </widget>
         </item>
@@ -179,24 +173,24 @@ of a search term by changing the value box then pressing Save.</string>
         </item>
        </layout>
       </item>
-     </layout>
+      <item row="2" column="0">
-    </widget>
-   </item>
-   <item row="5" column="0">
        <spacer name="verticalSpacer">
         <property name="orientation">
          <enum>Qt::Vertical</enum>
         </property>
         <property name="sizeHint" stdset="0">
          <size>
-       <width>0</width>
+          <width>20</width>
-       <height>0</height>
+          <height>40</height>
          </size>
         </property>
        </spacer>
       </item>
      </layout>
     </widget>
+   </item>
+  </layout>
+ </widget>
  <customwidgets>
   <customwidget>
    <class>MultiCompleteLineEdit</class>
@@ -204,6 +198,8 @@ of a search term by changing the value box then pressing Save.</string>
    <header>calibre/gui2/complete.h</header>
   </customwidget>
  </customwidgets>
- <resources/>
+ <resources>
+  <include location="../../../../resources/images.qrc"/>
+ </resources>
  <connections/>
 </ui>

src/calibre/library/database2.py

@@ -191,8 +191,7 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
         # migrate grouped_search_terms
         if self.prefs.get('grouped_search_terms', None) is None:
             try:
-                ogst = tweaks['grouped_search_terms']
+                ogst = tweaks.get('grouped_search_terms', {})
-                print 'migrating'
                 ngst = {}
                 for t in ogst:
                     ngst[icu_lower(t)] = ogst[t]