Sync to trunk.

resources/recipes/expansion_spanish.recipe

@@ -1,59 +1,79 @@
 #!/usr/bin/env  python
-# -*- coding: utf-8 -*-
-
 __license__   = 'GPL v3'
-__copyright__ = '2009, Darko Miletic <darko.miletic at gmail.com>'
+__author__    = 'Gerardo Diez'
+__copyright__ = 'Gerardo Diez<gerardo.diez.garcia@gmail.com>'
+description   = 'Main daily newspaper from Spain - v1.00 (05, Enero 2011)'
+__docformat__ = 'restructuredtext en'
+
 '''
-www.expansion.com
+expansion.es
 '''
+from calibre.web.feeds.recipes import BasicNewsRecipe
+class Publico(BasicNewsRecipe):
+    title               =u'Expansion.com'
+    __author__      ='Gerardo Diez'
+    publisher       =u'Unidad Editorial Información Económica, S.L.'
+    category                ='finances, catalunya'
+    oldest_article      =1
+    max_articles_per_feed   =100
+    simultaneous_downloads  =10
+    cover_url       =u'http://estaticos01.expansion.com/iconos/v2.x/v2.0/cabeceras/logo_expansion.png'
+    timefmt         ='[%A, %d %B, %Y]'
+    encoding        ='latin'
+    language        ='es'
+    remove_javascript   =True
+    no_stylesheets      =True
+    keep_only_tags      =dict(name='div', attrs={'class':['noticia primer_elemento']})
+    remove_tags         =[
+                dict(name='div', attrs={'class':['compartir', 'metadata_desarrollo_noticia', 'relacionadas', 'mas_info','publicidad publicidad_textlink', 'ampliarfoto']}),
+                dict(name='ul', attrs={'class':['bolos_desarrollo_noticia']}),
+                dict(name='span', attrs={'class':['comentarios']}),
+                dict(name='p', attrs={'class':['cintillo_comentarios', 'cintillo_comentarios formulario']}),
+                dict(name='div', attrs={'id':['comentarios_lectores_listado']})
+                            ]
+    feeds               =[
+                (u'Portada', u'http://estaticos.expansion.com/rss/portada.xml'),
+                (u'Portada: Bolsas', u'http://estaticos.expansion.com/rss/mercados.xml'),
+                (u'Divisas', u'http://estaticos.expansion.com/rss/mercadosdivisas.xml'),
+                (u'Euribor', u'http://estaticos.expansion.com/rss/mercadoseuribor.xml'),
+                (u'Materias Primas', u'http://estaticos.expansion.com/rss/mercadosmateriasprimas.xml'),
+                (u'Renta Fija', u'http://estaticos.expansion.com/rss/mercadosrentafija.xml'),
 
-from calibre.web.feeds.news import BasicNewsRecipe
+                (u'Portada: Mi Dinero', u'http://estaticos.expansion.com/rss/midinero.xml'),
-from calibre.ebooks.BeautifulSoup import Tag
+                (u'Hipotecas', u'http://estaticos.expansion.com/rss/midinerohipotecas.xml'),
+                (u'Créditos', u'http://estaticos.expansion.com/rss/midinerocreditos.xml'),
+                (u'Pensiones', u'http://estaticos.expansion.com/rss/midineropensiones.xml'),
+                (u'Fondos de Inversión', u'http://estaticos.expansion.com/rss/midinerofondos.xml'),
+                (u'Motor', u'http://estaticos.expansion.com/rss/midineromotor.xml'),
 
-class Expansion(BasicNewsRecipe):
+                (u'Portada: Empresas', u'http://estaticos.expansion.com/rss/empresas.xml'),
-    title                 = 'Diario Expansion'
+                (u'Banca', u'http://estaticos.expansion.com/rss/empresasbanca.xml'),
-    __author__            = 'Darko Miletic'
+                (u'TMT', u'http://estaticos.expansion.com/rss/empresastmt.xml'),
-    description           = 'Lider de informacion de mercados, economica y politica'
+                (u'Energía', u'http://estaticos.expansion.com/rss/empresasenergia.xml'),
-    publisher             = 'expansion.com'
+                (u'Inmobiliario y Construcción', u'http://estaticos.expansion.com/rss/empresasinmobiliario.xml'),
-    category              = 'news, politics, Spain'
+                (u'Transporte y Turismo', u'http://estaticos.expansion.com/rss/empresastransporte.xml'),
-    oldest_article        = 2
+                (u'Automoción e Industria', u'http://estaticos.expansion.com/rss/empresasauto-industria.xml'),
-    max_articles_per_feed = 100
+                (u'Distribución', u'http://estaticos.expansion.com/rss/empresasdistribucion.xml'),
-    no_stylesheets        = True
+                (u'Deporte y Negocio', u' http://estaticos.expansion.com/rss/empresasdeporte.xml'),
-    use_embedded_content  = False
+                (u'Mi Negocio', u'http://estaticos.expansion.com/rss/empresasminegocio.xml'),
-    delay                 = 1
+                (u'Interiores', u'http://estaticos.expansion.com/rss/empresasinteriores.xml'),
-    encoding              = 'iso-8859-15'
+                (u'Digitech', u'http://estaticos.expansion.com/rss/empresasdigitech.xml'),
-    language = 'es'
 
-    direction             = 'ltr'
+                (u'Portada: Economía y Política', u'http://estaticos.expansion.com/rss/economiapolitica.xml'),
+                (u'Política', u'http://estaticos.expansion.com/rss/economia.xml'),
+                (u'Portada: Sociedad', u'http://estaticos.expansion.com/rss/entorno.xml'),
 
-    html2lrf_options = [
+                (u'Portada: Opinión', u'http://estaticos.expansion.com/rss/opinion.xml'),
-                          '--comment'  , description
+                (u'Llaves y editoriales', u'http://estaticos.expansion.com/rss/opinioneditorialyllaves.xml'),
-                        , '--category' , category
+                (u'Tribunas', u'http://estaticos.expansion.com/rss/opiniontribunas.xml'),
-                        , '--publisher', publisher
+
+                (u'Portada: Jurídico', u'http://estaticos.expansion.com/rss/juridico.xml'),
+                (u'Entrevistas', u'http://estaticos.expansion.com/rss/juridicoentrevistas.xml'),
+                (u'Opinión', u'http://estaticos.expansion.com/rss/juridicoopinion.xml'),
+                (u'Sentencias', u'http://estaticos.expansion.com/rss/juridicosentencias.xml'),
+
+                (u'Mujer', u'http://estaticos.expansion.com/rss/mujer-empresa.xml'),
+                (u'Catalu&ntilde;a', u'http://estaticos.expansion.com/rss/catalunya.xml'),
+                (u'Función pública', u'http://estaticos.expansion.com/rss/funcion-publica.xml')
                 ]
 
-    html2epub_options  = 'publisher="' + publisher + '"\ncomments="' + description + '"\ntags="' + category + '"'
-
-    feeds              = [
-                            (u'Ultimas noticias', u'http://rss.expansion.com/rss/descarga.htm?data2=178')
-                           ,(u'Temas del dia'   , u'http://rss.expansion.com/rss/descarga.htm?data2=178')
-                         ]
-
-
-    keep_only_tags = [dict(name='div', attrs={'id':'principal'})]
-
-    remove_tags        = [
-                             dict(name=['object','link','script'])
-                            ,dict(name='div', attrs={'class':['utilidades','tit_relacionadas']})
-                         ]
-
-    remove_tags_after = [dict(name='div', attrs={'class':'tit_relacionadas'})]
-
-    def preprocess_html(self, soup):
-        soup.html['dir' ] = self.direction
-        mcharset = Tag(soup,'meta',[("http-equiv","Content-Type"),("content","text/html; charset=utf-8")])
-        soup.head.insert(0,mcharset)
-        for item in soup.findAll(style=True):
-            del item['style']
-        return soup
-

resources/recipes/msnbc.recipe

@@ -1,10 +1,9 @@
 __license__   = 'GPL v3'
-__copyright__ = '2010, Darko Miletic <darko.miletic at gmail.com>'
+__copyright__ = '2010-2011, Darko Miletic <darko.miletic at gmail.com>'
 '''
 msnbc.msn.com
 '''
 
-import re
 from calibre.web.feeds.recipes import BasicNewsRecipe
 
 class MsNBC(BasicNewsRecipe):
@@ -19,7 +18,16 @@ class MsNBC(BasicNewsRecipe):
     publisher              = 'msnbc.com'
     category               = 'news, USA, world'
     language               = 'en'
-    extra_css              = ' body{ font-family: sans-serif } .head{font-family: serif; font-size: xx-large; font-weight: bold; color: #CC0000} .abstract{font-weight: bold} .source{font-size: small} .updateTime{font-size: small} '
+    extra_css              = """
+                                body{ font-family: Georgia,Times,serif }
+                                .hide{display: none}
+                                .caption{font-family: Arial,sans-serif; font-size: x-small}
+                                .entry-summary{font-family: Arial,sans-serif}
+                                .copyright{font-size: 0.95em; font-style: italic}
+                                .source-org{font-size: small; font-family: Arial,sans-serif}
+                                img{display: block; margin-bottom: 0.5em}
+                                span.byline{display: none}
+                            """
 
     conversion_options = {
                              'comments' : description
@@ -28,14 +36,20 @@ class MsNBC(BasicNewsRecipe):
                             ,'publisher': publisher
                          }
 
-    preprocess_regexps = [
+    remove_tags_before = dict(name='h1', attrs={'id':'headline'})
-        (re.compile(r'</style></head>', re.DOTALL|re.IGNORECASE),lambda match: '</style>')
+    remove_tags_after = dict(name='span', attrs={'class':['copyright','Linear copyright']})
-       ,(re.compile(r'<div class="head">', re.DOTALL|re.IGNORECASE),lambda match: '</head><body><div class="head">'),
+    keep_only_tags=[
+                      dict(attrs={'id':['headline','deck','byline','source','intelliTXT']})
+                     ,dict(attrs={'class':['gl_headline','articleText','drawer-content Linear','v-center3','byline','textBodyBlack']})
+                   ]
+    remove_attributes=['property','lang','rel','xmlns:fb','xmlns:v','xmlns:dc','xmlns:dcmitype','xmlns:og','xmlns:media','xmlns:vcard','typeof','itemscope','itemtype','itemprop','about','type','size','width','height','onreadystatechange','data','border','hspace','vspace']
+
+    remove_tags      = [
+                          dict(name=['iframe','object','link','embed','meta','table'])
+                         ,dict(name='span', attrs={'class':['copyright','Linear copyright']})
+                         ,dict(name='div', attrs={'class':'social'})
                        ]
 
-    remove_tags_before = dict(name='div', attrs={'class':'head'})
-    remove_tags_after = dict(name='div', attrs={'class':'copyright'})
-    remove_tags      = [dict(name=['iframe','object','link','script','form'])]
 
     feeds = [
                (u'US News'       , u'http://rss.msnbc.msn.com/id/3032524/device/rss/rss.xml'      )
@@ -48,11 +62,26 @@ class MsNBC(BasicNewsRecipe):
               ,(u'Tech & Science', u'http://rss.msnbc.msn.com/id/3032117/device/rss/rss.xml'      )
             ]
 
-    def print_version(self, url):
-        return url + 'print/1/displaymode/1098/'
-
     def preprocess_html(self, soup):
-        for item in soup.head.findAll('div'):
+        for item in soup.body.findAll('html'):
+            item.name='div'
+        for item in soup.body.findAll('div'):
+            if item.has_key('id') and item['id'].startswith('vine-'):
                item.extract()
+            if item.has_key('class') and ( item['class'].startswith('ad') or item['class'].startswith('vine')):
+               item.extract()
+        for item in soup.body.findAll('img'):
+            if not item.has_key('alt'):
+               item['alt'] = 'image'
+        for item in soup.body.findAll('ol'):
+            if item.has_key('class') and item['class'].startswith('grid'):
+               item.extract()
+        for item in soup.body.findAll('span'):
+            if ( item.has_key('id') and item['id'].startswith('byLine') and item.string is None) or ( item.has_key('class') and item['class'].startswith('inline') ):
+               item.extract()
+        for alink in soup.findAll('a'):
+            if alink.string is not None:
+               tstr = alink.string
+               alink.replaceWith(tstr)
         return soup
 

resources/recipes/nytimes.recipe

@@ -685,3 +685,28 @@ class NYTimes(BasicNewsRecipe):
             divTag.replaceWith(tag)
 
         return soup
+
+    def populate_article_metadata(self, article, soup, first):
+        shortparagraph = ""
+        try:
+            if len(article.text_summary.strip()) == 0:
+                articlebodies = soup.findAll('div',attrs={'class':'articleBody'})
+                if articlebodies:
+                    for articlebody in articlebodies:
+                        if articlebody:
+                            paras = articlebody.findAll('p')
+                            for p in paras:
+                                refparagraph = self.massageNCXText(self.tag_to_string(p,use_alt=False)).strip()
+                                #account for blank paragraphs and short paragraphs by appending them to longer ones
+                                if len(refparagraph) > 0:
+                                    if len(refparagraph) > 70: #approximately one line of text
+                                        article.summary = article.text_summary = shortparagraph + refparagraph
+                                        return
+                                    else:
+                                        shortparagraph = refparagraph + " "
+                                        if shortparagraph.strip().find(" ") == -1 and not shortparagraph.strip().endswith(":"):
+                                            shortparagraph = shortparagraph + "- "
+        except:
+            self.log("Error creating article descriptions")
+            return
+

resources/recipes/nytimes_sub.recipe

@@ -685,4 +685,27 @@ class NYTimes(BasicNewsRecipe):
             divTag.replaceWith(tag)
 
         return soup
+    def populate_article_metadata(self, article, soup, first):
+        shortparagraph = ""
+        try:
+            if len(article.text_summary.strip()) == 0:
+                articlebodies = soup.findAll('div',attrs={'class':'articleBody'})
+                if articlebodies:
+                    for articlebody in articlebodies:
+                        if articlebody:
+                            paras = articlebody.findAll('p')
+                            for p in paras:
+                                refparagraph = self.massageNCXText(self.tag_to_string(p,use_alt=False)).strip()
+                                #account for blank paragraphs and short paragraphs by appending them to longer ones
+                                if len(refparagraph) > 0:
+                                    if len(refparagraph) > 70: #approximately one line of text
+                                        article.summary = article.text_summary = shortparagraph + refparagraph
+                                        return
+                                    else:
+                                        shortparagraph = refparagraph + " "
+                                        if shortparagraph.strip().find(" ") == -1 and not shortparagraph.strip().endswith(":"):
+                                            shortparagraph = shortparagraph + "- "
+        except:
+            self.log("Error creating article descriptions")
+            return
 

resources/recipes/technology_review.recipe

@@ -35,7 +35,6 @@ class TechnologyReview(BasicNewsRecipe):
     def get_article_url(self, article):
         return article.get('guid', article.get('id', None))
 
-
     def print_version(self, url):
         baseurl='http://www.technologyreview.com/printer_friendly_article.aspx?id='
         split1 = string.split(url,"/")
@@ -43,3 +42,25 @@ class TechnologyReview(BasicNewsRecipe):
         split2= string.split(xxx,"/")
         s =  baseurl + split2[0]
         return s
+
+
+    def postprocess_html(self,soup, True):
+        #remove picture
+        headerhtml = soup.find(True, {'class':'header'})
+        headerhtml.replaceWith("")
+
+        #remove close button
+        closehtml = soup.find(True, {'class':'close'})
+        closehtml.replaceWith("")
+
+        #remove banner advertisement
+        bannerhtml = soup.find(True, {'class':'bannerad'})
+        bannerhtml.replaceWith("")
+
+        #thanks kiklop74!  This code removes all links from the text
+        for alink in soup.findAll('a'):
+            if alink.string is not None:
+               tstr = alink.string
+               alink.replaceWith(tstr)
+
+        return soup

resources/recipes/tyzden.recipe

@@ -28,7 +28,7 @@ class TyzdenRecipe(BasicNewsRecipe):
     if (weeknum > 1):
         weeknum -= 1
 
-    title = u'.tyzden ' + str(weeknum) + '/' + str(year)
+    title = u'tyzden'
 
     base_url_path = 'http://www.tyzden.sk/casopis/' + str(year) + '/' + str(weeknum)
     base_url = base_url_path + '.html'

resources/recipes/wired_daily.recipe

@@ -2,8 +2,10 @@
 __license__   = 'GPL v3'
 __docformat__ = 'restructuredtext en'
 
+import re
 
 from calibre.web.feeds.news import BasicNewsRecipe
+from calibre.ebooks.chardet import xml_to_unicode
 
 class Wired_Daily(BasicNewsRecipe):
 
@@ -15,30 +17,43 @@ class Wired_Daily(BasicNewsRecipe):
 
     no_stylesheets = True
 
+    preprocess_regexps = [(re.compile(r'<head.*</head>', re.DOTALL), lambda m:
+        '<head></head>')]
+
     remove_tags_before = dict(name='div', id='content')
-    remove_tags = [dict(id=['social_tools', 'outerWrapper', 'sidebar',
+    remove_tags = [dict(id=['header', 'commenting_module', 'post_nav',
-        'footer', 'advertisement', 'blog_subscription_unit',
+        'social_tools', 'sidebar', 'footer', 'social_wishlist', 'pgwidget',
-        'brightcove_component']),
+        'outerWrapper', 'inf_widget']),
-        {'class':'entryActions'},
+        {'class':['entryActions', 'advertisement', 'entryTags']},
-        dict(name=['noscript', 'script'])]
+        dict(name=['noscript', 'script']),
+        dict(name='h4', attrs={'class':re.compile(r'rat\d+')}),
+        {'class':lambda x: x and x.startswith('contentjump')},
+        dict(name='li', attrs={'class':['entryCategories', 'entryEdit']})]
+
 
     feeds = [
         ('Top News', 'http://feeds.wired.com/wired/index'),
-        ('Culture', 'http://feeds.wired.com/wired/culture'),
+        ('Product Reviews',
-        ('Software', 'http://feeds.wired.com/wired/software'),
+            'http://www.wired.com/reviews/feeds/latestProductsRss'),
-        ('Mac', 'http://feeds.feedburner.com/cultofmac/bFow'),
+        ('Autopia', 'http://www.wired.com/autopia/feed/'),
-        ('Gadgets', 'http://feeds.wired.com/wired/gadgets'),
+        ('Danger Room', 'http://www.wired.com/dangerroom/feed/'),
-        ('Cars', 'http://feeds.wired.com/wired/cars'),
+        ('Epicenter', 'http://www.wired.com/epicenter/feed/'),
-        ('Entertainment', 'http://feeds.wired.com/wired/entertainment'),
+        ('Gadget Lab', 'http://www.wired.com/gadgetlab/feed/'),
-        ('Gaming', 'http://feeds.wired.com/wired/gaming'),
+        ('Geek Dad', 'http://www.wired.com/geekdad/feed/'),
-        ('Science', 'http://feeds.wired.com/wired/science'),
+        ('Playbook', 'http://www.wired.com/playbook/feed/'),
-        ('Med Tech', 'http://feeds.wired.com/wired/medtech'),
+        ('Rawfile', 'http://www.wired.com/rawfile/feed/'),
-        ('Politics', 'http://feeds.wired.com/wired/politics'),
+        ('This Day in Tech', 'http://www.wired.com/thisdayintech/feed/'),
-        ('Tech Biz', 'http://feeds.wired.com/wired/techbiz'),
+        ('Threat Level', 'http://www.wired.com/threatlevel/feed/'),
-        ('Commentary', 'http://feeds.wired.com/wired/commentary'),
+        ('Underwire', 'http://www.wired.com/underwire/feed/'),
+        ('Web Monkey', 'http://www.webmonkey.com/feed/'),
+        ('Science', 'http://www.wired.com/wiredscience/feed/'),
         ]
 
+    def populate_article_metadata(self, article, soup, first):
+        if article.text_summary:
+            article.text_summary = xml_to_unicode(article.text_summary,
+                    resolve_entities=True)[0]
+
     def print_version(self, url):
-        return url.replace('http://www.wired.com/', 'http://www.wired.com/print/')
+        return url + '/all/1'
-
 

resources/recipes/zerohedge.recipe

@@ -0,0 +1,33 @@
+__license__   = 'GPL v3'
+__copyright__ = '2011, Darko Miletic <darko.miletic at gmail.com>'
+'''
+www.zerohedge.com
+'''
+
+from calibre.web.feeds.recipes import BasicNewsRecipe
+
+class ZeroHedge(BasicNewsRecipe):
+    title                  = 'Zero Hedge'
+    __author__             = 'Darko Miletic'
+    description            = 'On a long enough timeline the survival rate for everyone drops to zero'
+    oldest_article         = 10
+    max_articles_per_feed  = 100
+    no_stylesheets         = True
+    use_embedded_content   = True
+    encoding               = 'utf8'
+    publisher              = 'zero hedge'
+    category               = 'news, USA, world, economy, politics'
+    language               = 'en'
+    masthead_url           = 'http://www.zerohedge.com/themes/newsflash/logo.png'
+    publication_type       = 'blog'
+    extra_css              = 'body{ font-family: sans-serif }'
+
+    conversion_options = {
+                             'comments' : description
+                            ,'tags'     : category
+                            ,'language' : language
+                            ,'publisher': publisher
+                         }
+
+
+    feeds = [(u'Articles', u'http://feeds.feedburner.com/zerohedge/feed')]

resources/templates/rtf.xsl

@@ -287,7 +287,7 @@
                 <xsl:value-of select="count(preceding::rtf:footnote) + 1"/>
                 <xsl:text>]</xsl:text>
             </xsl:when>
-            <xsl:when test="(@superscript = 'true')">
+            <xsl:when test="(@superscript)">
                 <xsl:element name="sup">
                     <xsl:element name="span">
                         <xsl:attribute name="class">
@@ -297,7 +297,7 @@
                     </xsl:element>
                 </xsl:element>
             </xsl:when>
-            <xsl:when test="(@underscript = 'true')">
+            <xsl:when test="(@underscript or @subscript)">
                 <xsl:element name="sub">
                     <xsl:element name="span">
                         <xsl:attribute name="class">

src/calibre/__init__.py

@@ -459,6 +459,18 @@ def force_unicode(obj, enc=preferred_encoding):
                         obj = obj.decode('utf-8')
     return obj
 
+def as_unicode(obj, enc=preferred_encoding):
+    if not isbytestring(obj):
+        try:
+            obj = unicode(obj)
+        except:
+            try:
+                obj = str(obj)
+            except:
+                obj = repr(obj)
+    return force_unicode(obj, enc=enc)
+
+
 
 def human_readable(size):
     """ Convert a size in bytes into a human readable form """

src/calibre/devices/nook/driver.py

@@ -91,3 +91,19 @@ class NOOK_COLOR(NOOK):
 
     EBOOK_DIR_MAIN = 'My Files/Books'
 
+    '''
+    def create_upload_path(self, path, mdata, fname, create_dirs=True):
+        filepath = NOOK.create_upload_path(self, path, mdata, fname,
+                create_dirs=create_dirs)
+        edm = self.EBOOK_DIR_MAIN.replace('/', os.sep)
+        npath = os.path.join(edm, _('News')) + os.sep
+        if npath in filepath:
+            filepath = filepath.replace(npath, os.sep.join('My Files',
+                'Magazines')+os.sep)
+            filedir = os.path.dirname(filepath)
+            if create_dirs and not os.path.exists(filedir):
+                os.makedirs(filedir)
+
+        return filepath
+    '''
+

src/calibre/ebooks/conversion/plumber.py

@@ -88,6 +88,7 @@ class Plumber(object):
         self.ui_reporter = report_progress
         self.abort_after_input_dump = abort_after_input_dump
 
+        # Pipeline options {{{
         # Initialize the conversion options that are independent of input and
         # output formats. The input and output plugins can still disable these
         # options via recommendations.
@@ -527,6 +528,7 @@ OptionRecommendation(name='timestamp',
     help=_('Set the book timestamp (used by the date column in calibre).')),
 
 ]
+        # }}}
 
         input_fmt = os.path.splitext(self.input)[1]
         if not input_fmt:

src/calibre/ebooks/fb2/fb2ml.py

@@ -16,7 +16,6 @@ import uuid
 
 from lxml import etree
 
-from calibre import guess_type
 from calibre import prepare_string_for_xml
 from calibre.constants import __appname__, __version__
 from calibre.ebooks.oeb.base import XHTML, XHTML_NS, barename, namespace

src/calibre/ebooks/metadata/rtf.py

@@ -10,7 +10,8 @@ from calibre.ebooks.metadata import MetaInformation, string_to_authors
 title_pat    = re.compile(r'\{\\info.*?\{\\title(.*?)(?<!\\)\}', re.DOTALL)
 author_pat   = re.compile(r'\{\\info.*?\{\\author(.*?)(?<!\\)\}', re.DOTALL)
 comment_pat  = re.compile(r'\{\\info.*?\{\\subject(.*?)(?<!\\)\}', re.DOTALL)
-category_pat = re.compile(r'\{\\info.*?\{\\category(.*?)(?<!\\)\}', re.DOTALL)
+tags_pat = re.compile(r'\{\\info.*?\{\\category(.*?)(?<!\\)\}', re.DOTALL)
+publisher_pat = re.compile(r'\{\\info.*?\{\\manager(.*?)(?<!\\)\}', re.DOTALL)
 
 def get_document_info(stream):
     """
@@ -82,61 +83,73 @@ def decode(raw, codec):
 
 def get_metadata(stream):
     """ Return metadata as a L{MetaInfo} object """
-    title, author, comment, category = None, None, None, None
     stream.seek(0)
     if stream.read(5) != r'{\rtf':
-        return MetaInformation(None, None)
+        return MetaInformation(_('Unknown'))
     block = get_document_info(stream)[0]
     if not block:
-        return MetaInformation(None, None)
+        return MetaInformation(_('Unknown'))
 
     stream.seek(0)
     cpg = detect_codepage(stream)
     stream.seek(0)
 
     title_match = title_pat.search(block)
-    if title_match:
+    if title_match is not None:
         title = decode(title_match.group(1).strip(), cpg)
+    else:
+        title = _('Unknown')
     author_match = author_pat.search(block)
-    if author_match:
+    if author_match is not None:
         author = decode(author_match.group(1).strip(), cpg)
-    comment_match = comment_pat.search(block)
+    else:
-    if comment_match:
+        author = None
-        comment = decode(comment_match.group(1).strip(), cpg)
+    mi = MetaInformation(title)
-    category_match = category_pat.search(block)
-    if category_match:
-        category = decode(category_match.group(1).strip(), cpg)
-    mi = MetaInformation(title, author)
     if author:
         mi.authors = string_to_authors(author)
+
+    comment_match = comment_pat.search(block)
+    if comment_match is not None:
+        comment = decode(comment_match.group(1).strip(), cpg)
         mi.comments = comment
-    mi.category = category
+    tags_match = tags_pat.search(block)
+    if tags_match is not None:
+        tags = decode(tags_match.group(1).strip(), cpg)
+        mi.tags = tags
+    publisher_match = publisher_pat.search(block)
+    if publisher_match is not None:
+        publisher = decode(publisher_match.group(1).strip(), cpg)
+        mi.publisher = publisher
+
     return mi
 
-
 def create_metadata(stream, options):
-    md = r'{\info'
+    md = [r'{\info']
     if options.title:
         title = options.title.encode('ascii', 'ignore')
-        md += r'{\title %s}'%(title,)
+        md.append(r'{\title %s}'%(title,))
     if options.authors:
         au = options.authors
         if not isinstance(au, basestring):
             au = u', '.join(au)
         author = au.encode('ascii', 'ignore')
-        md += r'{\author %s}'%(author,)
+        md.append(r'{\author %s}'%(author,))
-    if options.get('category', None):
-        category = options.category.encode('ascii', 'ignore')
-        md += r'{\category %s}'%(category,)
     comp = options.comment if hasattr(options, 'comment') else options.comments
     if comp:
         comment = comp.encode('ascii', 'ignore')
-        md += r'{\subject %s}'%(comment,)
+        md.append(r'{\subject %s}'%(comment,))
-    if len(md) > 6:
+    if options.publisher:
-        md += '}'
+        publisher = options.publisher.encode('ascii', 'ignore')
+        md.append(r'{\manager %s}'%(publisher,))
+    if options.tags:
+        tags = u', '.join(options.tags)
+        tags = tags.encode('ascii', 'ignore')
+        md.append(r'{\category %s}'%(tags,))
+    if len(md) > 1:
+        md.append('}')
         stream.seek(0)
         src   = stream.read()
-        ans = src[:6] + md + src[6:]
+        ans = src[:6] + u''.join(md) + src[6:]
         stream.seek(0)
         stream.write(ans)
 
@@ -156,7 +169,7 @@ def set_metadata(stream, options):
 
         base_pat = r'\{\\name(.*?)(?<!\\)\}'
         title = options.title
-        if title != None:
+        if title is not None:
             title = title.encode('ascii', 'replace')
             pat = re.compile(base_pat.replace('name', 'title'), re.DOTALL)
             if pat.search(src):
@@ -164,7 +177,7 @@ def set_metadata(stream, options):
             else:
                 src = add_metadata_item(src, 'title', title)
         comment = options.comments
-        if comment != None:
+        if comment is not None:
             comment = comment.encode('ascii', 'replace')
             pat = re.compile(base_pat.replace('name', 'subject'), re.DOTALL)
             if pat.search(src):
@@ -172,7 +185,7 @@ def set_metadata(stream, options):
             else:
                 src = add_metadata_item(src, 'subject', comment)
         author = options.authors
-        if author != None:
+        if author is not None:
             author =  ', '.join(author)
             author = author.encode('ascii', 'ignore')
             pat = re.compile(base_pat.replace('name', 'author'), re.DOTALL)
@@ -180,14 +193,23 @@ def set_metadata(stream, options):
                 src = pat.sub(r'{\\author ' + author + r'}', src)
             else:
                 src = add_metadata_item(src, 'author', author)
-        category = options.get('category', None)
+        tags = options.tags
-        if category != None:
+        if tags is not None:
-            category = category.encode('ascii', 'replace')
+            tags =  ', '.join(tags)
+            tags = tags.encode('ascii', 'replace')
             pat = re.compile(base_pat.replace('name', 'category'), re.DOTALL)
             if pat.search(src):
-                src = pat.sub(r'{\\category ' + category + r'}', src)
+                src = pat.sub(r'{\\category ' + tags + r'}', src)
             else:
-                src = add_metadata_item(src, 'category', category)
+                src = add_metadata_item(src, 'category', tags)
+        publisher = options.publisher
+        if publisher is not None:
+            publisher = publisher.encode('ascii', 'replace')
+            pat = re.compile(base_pat.replace('name', 'manager'), re.DOTALL)
+            if pat.search(src):
+                src = pat.sub(r'{\\manager ' + publisher + r'}', src)
+            else:
+                src = add_metadata_item(src, 'manager', publisher)
         stream.seek(pos + olen)
         after = stream.read()
         stream.seek(pos)

src/calibre/ebooks/rtf/input.py

@@ -77,7 +77,15 @@ class RTFInput(InputFormatPlugin):
 
     def generate_xml(self, stream):
         from calibre.ebooks.rtf2xml.ParseRtf import ParseRtf
-        ofile = 'out.xml'
+        ofile = 'dataxml.xml'
+        run_lev, debug_dir = 1, None
+        if getattr(self.opts, 'debug_pipeline', None) is not None:
+            try:
+                os.mkdir(debug_dir)
+                debug_dir = 'rtfdebug'
+                run_lev = 4
+            except:
+                pass
         parser = ParseRtf(
             in_file    = stream,
             out_file   = ofile,
@@ -115,43 +123,45 @@ class RTFInput(InputFormatPlugin):
 
             # Write or do not write paragraphs. Default is 0.
             empty_paragraphs = 1,
+
+            #debug
+            deb_dir = debug_dir,
+            run_level = run_lev,
         )
         parser.parse_rtf()
-        ans = open('out.xml').read()
+        with open(ofile, 'rb') as f:
-        os.remove('out.xml')
+            return f.read()
-        return ans
 
     def extract_images(self, picts):
+        import imghdr
         self.log('Extracting images...')
 
+        with open(picts, 'rb') as f:
+            raw = f.read()
+        picts = filter(len, re.findall(r'\{\\pict([^}]+)\}', raw))
+        hex = re.compile(r'[^a-fA-F0-9]')
+        encs = [hex.sub('', pict) for pict in picts]
+
         count = 0
-        raw = open(picts, 'rb').read()
-        starts = []
-        for match in re.finditer(r'\{\\pict([^}]+)\}', raw):
-            starts.append(match.start(1))
-
         imap = {}
-
+        for enc in encs:
-        for start in starts:
-            pos, bc = start, 1
-            while bc > 0:
-                if raw[pos] == '}': bc -= 1
-                elif raw[pos] == '{': bc += 1
-                pos += 1
-            pict = raw[start:pos+1]
-            enc = re.sub(r'[^a-zA-Z0-9]', '', pict)
             if len(enc) % 2 == 1:
                 enc = enc[:-1]
             data = enc.decode('hex')
+            fmt = imghdr.what(None, data)
+            if fmt is None:
+                fmt = 'wmf'
             count += 1
-            name = (('%4d'%count).replace(' ', '0'))+'.wmf'
+            name = '%04d.%s' % (count, fmt)
-            open(name, 'wb').write(data)
+            with open(name, 'wb') as f:
+                f.write(data)
             imap[count] = name
             #open(name+'.hex', 'wb').write(enc)
         return self.convert_images(imap)
 
     def convert_images(self, imap):
-        for count, val in imap.items():
+        self.default_img = None
+        for count, val in imap.iteritems():
             try:
                 imap[count] = self.convert_image(val)
             except:
@@ -159,11 +169,35 @@ class RTFInput(InputFormatPlugin):
         return imap
 
     def convert_image(self, name):
-        from calibre.utils.magick import Image
+        if not name.endswith('.wmf'):
-        img = Image()
+            return name
-        img.open(name)
+        try:
+            return self.rasterize_wmf(name)
+        except:
+            self.log.exception('Failed to convert WMF image %r'%name)
+        return self.replace_wmf(name)
+
+    def replace_wmf(self, name):
+        from calibre.ebooks import calibre_cover
+        if self.default_img is None:
+            self.default_img = calibre_cover('Conversion of WMF images is not supported',
+            'Use Microsoft Word or OpenOffice to save this RTF file'
+            ' as HTML and convert that in calibre.', title_size=36,
+            author_size=20)
         name = name.replace('.wmf', '.jpg')
-        img.save(name)
+        with open(name, 'wb') as f:
+            f.write(self.default_img)
+        return name
+
+    def rasterize_wmf(self, name):
+        raise ValueError('Conversion of WMF images not supported')
+        from calibre.utils.wmf import extract_raster_image
+        with open(name, 'rb') as f:
+            data = f.read()
+        data = extract_raster_image(data)
+        name = name.replace('.wmf', '.jpg')
+        with open(name, 'wb') as f:
+            f.write(data)
         return name
 
 
@@ -192,27 +226,27 @@ class RTFInput(InputFormatPlugin):
         css += '\n'+'\n'.join(font_size_classes)
         css += '\n' +'\n'.join(color_classes)
 
-        for cls, val in border_styles.items():
+        for cls, val in border_styles.iteritems():
             css += '\n\n.%s {\n%s\n}'%(cls, val)
 
         with open('styles.css', 'ab') as f:
             f.write(css)
 
-    def preprocess(self, fname):
+    # def preprocess(self, fname):
-        self.log('\tPreprocessing to convert unicode characters')
+        # self.log('\tPreprocessing to convert unicode characters')
-        try:
+        # try:
-            data = open(fname, 'rb').read()
+            # data = open(fname, 'rb').read()
-            from calibre.ebooks.rtf.preprocess import RtfTokenizer, RtfTokenParser
+            # from calibre.ebooks.rtf.preprocess import RtfTokenizer, RtfTokenParser
-            tokenizer = RtfTokenizer(data)
+            # tokenizer = RtfTokenizer(data)
-            tokens = RtfTokenParser(tokenizer.tokens)
+            # tokens = RtfTokenParser(tokenizer.tokens)
-            data = tokens.toRTF()
+            # data = tokens.toRTF()
-            fname = 'preprocessed.rtf'
+            # fname = 'preprocessed.rtf'
-            with open(fname, 'wb') as f:
+            # with open(fname, 'wb') as f:
-                f.write(data)
+                # f.write(data)
-        except:
+        # except:
-            self.log.exception(
+            # self.log.exception(
-            'Failed to preprocess RTF to convert unicode sequences, ignoring...')
+            # 'Failed to preprocess RTF to convert unicode sequences, ignoring...')
-        return fname
+        # return fname
 
     def convert_borders(self, doc):
         border_styles = []
@@ -249,17 +283,14 @@ class RTFInput(InputFormatPlugin):
         self.log = log
         self.log('Converting RTF to XML...')
         #Name of the preprocesssed RTF file
-        fname = self.preprocess(stream.name)
+        # fname = self.preprocess(stream.name)
         try:
-            xml = self.generate_xml(fname)
+            xml = self.generate_xml(stream.name)
         except RtfInvalidCodeException, e:
+            raise
             raise ValueError(_('This RTF file has a feature calibre does not '
             'support. Convert it to HTML first and then try it.\n%s')%e)
 
-        '''dataxml = open('dataxml.xml', 'w')
-        dataxml.write(xml)
-        dataxml.close'''
-
         d = glob.glob(os.path.join('*_rtf_pict_dir', 'picts.rtf'))
         if d:
             imap = {}

src/calibre/ebooks/rtf2xml/ParseRtf.py

@@ -17,7 +17,8 @@
 #########################################################################
 # $Revision: 1.41 $
 # $Date: 2006/03/24 23:50:07 $
-import sys,os
+import sys, os
+
 from calibre.ebooks.rtf2xml import headings_to_sections, \
     line_endings, footnote, fields_small, default_encoding, \
     make_lists, preamble_div, header, colors, group_borders, \
@@ -90,7 +91,6 @@ class ParseRtf:
                 out_file = '',
                 out_dir = None,
                 dtd = '',
-                #debug = 0, #why? calibre
                 deb_dir = None,
                 convert_symbol = None,
                 convert_wingdings = None,
@@ -107,6 +107,7 @@ class ParseRtf:
                 no_dtd = 0,
                 char_data = '',
                 ):
+
         """
         Requires:
         'file' --file to parse
@@ -119,12 +120,11 @@ class ParseRtf:
             script tries to output to directory where is script is exectued.)
             'deb_dir' --debug directory. If a debug_dir is provided, the script
             will copy each run through as a file to examine in the debug_dir
-            'perl_script'--use perl to make tokens. This runs just a bit faster.
-            (I will probably phase this out.)
             'check_brackets' -- make sure the brackets match up after each run
             through a file. Only for debugging.
         Returns: Nothing
         """
+
         self.__file = in_file
         self.__out_file = out_file
         self.__out_dir = out_dir
@@ -132,7 +132,7 @@ class ParseRtf:
         self.__dtd_path = dtd
         self.__check_file(in_file,"file_to_parse")
         self.__char_data = char_data
-        self.__debug_dir = deb_dir #self.__debug_dir = debug calibre
+        self.__debug_dir = deb_dir
         self.__check_dir(self.__temp_dir)
         self.__copy = self.__check_dir(self.__debug_dir)
         self.__convert_caps = convert_caps
@@ -155,25 +155,24 @@ class ParseRtf:
         if hasattr(the_file, 'read'): return
         if the_file == None:
             if type == "file_to_parse":
-                message = "You must provide a file for the script to work"
+                msg = "\nYou must provide a file for the script to work"
-            msg = message
             raise RtfInvalidCodeException, msg
         elif os.path.exists(the_file):
             pass # do nothing
         else:
-            message = "The file '%s' cannot be found" % the_file
+            msg = "\nThe file '%s' cannot be found" % the_file
-            msg = message
             raise RtfInvalidCodeException, msg
+
     def __check_dir(self, the_dir):
         """Check to see if directory exists"""
         if not the_dir :
             return
         dir_exists = os.path.isdir(the_dir)
         if not dir_exists:
-            message = "%s is not a directory" % the_dir
+            msg = "\n%s is not a directory" % the_dir
-            msg = message
             raise RtfInvalidCodeException, msg
         return 1
+
     def parse_rtf(self):
         """
         Parse the file by calling on other classes.
@@ -194,13 +193,14 @@ class ParseRtf:
             copy_obj.set_dir(self.__debug_dir)
             copy_obj.remove_files()
             copy_obj.copy_file(self.__temp_file, "original_file")
-        # new as of 2005-08-02. Do I want this?
+        # Function to check if bracket are well handled
         if self.__debug_dir or self.__run_level > 2:
             self.__check_brack_obj = check_brackets.CheckBrackets\
             (file = self.__temp_file,
                 bug_handler = RtfInvalidCodeException,
                     )
-        # convert Macintosh line endings to Unix line endings
+        #convert Macintosh and Windows line endings to Unix line endings
+        #why do this if you don't wb after?
         line_obj = line_endings.FixLineEndings(
                 in_file = self.__temp_file,
                 bug_handler = RtfInvalidCodeException,
@@ -208,13 +208,13 @@ class ParseRtf:
                 run_level = self.__run_level,
                 replace_illegals = self.__replace_illegals,
                 )
-        return_value = line_obj.fix_endings()
+        return_value = line_obj.fix_endings() #calibre return what?
         self.__return_code(return_value)
         tokenize_obj = tokenize.Tokenize(
                 bug_handler = RtfInvalidCodeException,
                 in_file = self.__temp_file,
                 copy = self.__copy,
-                run_level = self.__run_level,)
+                run_level = self.__run_level)
         tokenize_obj.tokenize()
         process_tokens_obj = process_tokens.ProcessTokens(
             in_file = self.__temp_file,
@@ -230,11 +230,24 @@ class ParseRtf:
                 os.remove(self.__temp_file)
             except OSError:
                 pass
+            #Check to see if the file is correctly encoded
+            encode_obj = default_encoding.DefaultEncoding(
+            in_file = self.__temp_file,
+            run_level = self.__run_level,
+            bug_handler = RtfInvalidCodeException,
+            check_raw = True,
+            )
+            platform, code_page, default_font_num = encode_obj.find_default_encoding()
             check_encoding_obj = check_encoding.CheckEncoding(
                     bug_handler = RtfInvalidCodeException,
                         )
-            check_encoding_obj.check_encoding(self.__file)
+            enc = encode_obj.get_codepage()
-            sys.stderr.write('File "%s" does not appear to be RTF.\n' % self.__file if isinstance(self.__file, str) else self.__file.encode('utf-8'))
+            if enc != 'mac_roman':
+                enc = 'cp' + enc
+            if check_encoding_obj.check_encoding(self.__file, enc):
+                file_name = self.__file if isinstance(self.__file, str) \
+                                    else self.__file.encode('utf-8')
+                msg = 'File %s does not appear to be correctly encoded.\n' % file_name
                 raise InvalidRtfException, msg
         delete_info_obj = delete_info.DeleteInfo(
             in_file = self.__temp_file,
@@ -508,6 +521,7 @@ class ParseRtf:
                 indent = self.__indent,
                 run_level = self.__run_level,
                 no_dtd = self.__no_dtd,
+                encoding = encode_obj.get_codepage(),
                 bug_handler = RtfInvalidCodeException,
                 )
         tags_obj.convert_to_tags()
@@ -520,35 +534,28 @@ class ParseRtf:
         output_obj.output()
         os.remove(self.__temp_file)
         return self.__exit_level
+
     def __bracket_match(self, file_name):
         if self.__run_level > 2:
             good_br, msg =  self.__check_brack_obj.check_brackets()
             if good_br:
                 pass
-                # sys.stderr.write( msg + ' in ' + file_name + "\n")
+                #sys.stderr.write( msg + ' in ' + file_name + "\n")
             else:
-                msg += msg +  " in file '" + file_name + "'\n"
+                msg = '%s in file %s\n' % (msg, file_name)
                 raise RtfInvalidCodeException, msg
+
     def __return_code(self, num):
       if num == None:
           return
       if int(num) > self.__exit_level:
           self.__exit_level = num
+
     def __make_temp_file(self,file):
         """Make a temporary file to parse"""
         write_file="rtf_write_file"
         read_obj = file if hasattr(file, 'read') else open(file,'r')
-        write_obj = open(write_file, 'w')
+        with open(write_file, 'wb') as write_obj:
-        line = "dummy"
+            for line in read_obj:
-        while line:
+                write_obj.write(line)
-            line = read_obj.read(1000)
-            write_obj.write(line )
-        write_obj.close()
         return write_file
-    """
-mi<tg<open______<style-sheet\n
-mi<tg<close_____<style-sheet\n
-mi<tg<open-att__<footnote<num>1\n
-mi<tg<empty-att_<page-definition<margin>33\n
-mi<tg<empty_____<para\n
-"""

src/calibre/ebooks/rtf2xml/check_brackets.py

@@ -24,38 +24,38 @@ class CheckBrackets:
         self.__ob_count = 0
         self.__cb_count = 0
         self.__open_bracket_num = []
+
     def open_brack(self, line):
         num = line[-5:-1]
         self.__open_bracket_num.append(num)
         self.__bracket_count += 1
+
     def close_brack(self, line):
         num = line[-5:-1]
-        ##self.__open_bracket_num.append(num)
         try:
             last_num = self.__open_bracket_num.pop()
         except:
-            return 0
+            return False
         if num != last_num:
-            return 0
+            return False
         self.__bracket_count -= 1
-        return 1
+        return True
+
     def check_brackets(self):
-        read_obj = open(self.__file, 'r')
-        line = 'dummy'
         line_count = 0
-        while line:
+        with open(self.__file, 'r') as read_obj:
+            for line in read_obj:
                 line_count += 1
-            line = read_obj.readline()
                 self.__token_info = line[:16]
                 if self.__token_info == 'ob<nu<open-brack':
                     self.open_brack(line)
                 if self.__token_info == 'cb<nu<clos-brack':
-                right_count = self.close_brack(line)
+                    if not self.close_brack(line):
-                if not right_count:
+                        return (False, "closed bracket doesn't match, line %s" % line_count)
-                    return (0, "closed bracket doesn't match, line %s" % line_count)
+
-        read_obj.close()
         if self.__bracket_count != 0:
-            msg = 'At end of file open and closed brackets don\'t match\n'
+            msg = ('At end of file open and closed brackets don\'t match\n' \
-            msg = msg + 'total number of brackets is %s' % self.__bracket_count
+                        'total number of brackets is %s') % self.__bracket_count
-            return (0, msg)
+            return (False, msg)
-        return (1, "brackets match!")
+        return (True, "Brackets match!")
+

src/calibre/ebooks/rtf2xml/check_encoding.py

@@ -1,8 +1,11 @@
 #!/usr/bin/env python
 import sys
+
 class CheckEncoding:
+
     def __init__(self, bug_handler):
         self.__bug_handler = bug_handler
+
     def __get_position_error(self, line, encoding, line_num):
         char_position = 0
         for char in line:
@@ -12,21 +15,23 @@ class CheckEncoding:
             except UnicodeError, msg:
                 sys.stderr.write('line: %s char: %s\n' %  (line_num, char_position))
                 sys.stderr.write(str(msg) + '\n')
-    def check_encoding(self, path, encoding='us-ascii'):
+
-        read_obj = open(path, 'r')
+    def check_encoding(self, path, encoding='us-ascii', verbose=True):
-        line_to_read = 1
         line_num = 0
-        while line_to_read:
+        with open(path, 'r') as read_obj:
+            for line in read_obj:
                 line_num += 1
-            line_to_read = read_obj.readline()
-            line = line_to_read
                 try:
                     line.decode(encoding)
                 except UnicodeError:
+                    if verbose:
                         if len(line) < 1000:
                             self.__get_position_error(line, encoding, line_num)
                         else:
-                    sys.stderr.write('line: %d has bad encoding\n'%line_num)
+                            sys.stderr.write('line: %d has bad encoding\n' % line_num)
+                    return True
+        return False
+
 if __name__ == '__main__':
     check_encoding_obj = CheckEncoding()
     check_encoding_obj.check_encoding(sys.argv[1])

src/calibre/ebooks/rtf2xml/combine_borders.py

@@ -16,7 +16,9 @@
 #                                                                       #
 #########################################################################
 import os, tempfile
+
 from calibre.ebooks.rtf2xml import copy
+
 class CombineBorders:
     """Combine borders in RTF tokens to make later processing easier"""
     def __init__(self,
@@ -32,19 +34,21 @@ class CombineBorders:
         self.__state = 'default'
         self.__bord_pos = 'default'
         self.__bord_att = []
+
     def found_bd(self, line):
         #cw<bd<bor-t-r-vi
         self.__state = 'border'
         self.__bord_pos = line[6:16]
+
     def __default_func(self, line):
         #cw<bd<bor-t-r-vi
         if self.__first_five == 'cw<bd':
             self.found_bd(line)
             return ''
         return line
+
     def end_border(self, line, write_obj):
-        joiner = "|"
+        border_string = "|".join(self.__bord_att)
-        border_string = joiner.join(self.__bord_att)
         self.__bord_att = []
         write_obj.write('cw<bd<%s<nu<%s\n' % (self.__bord_pos,
                                                 border_string))
@@ -54,6 +58,7 @@ class CombineBorders:
             self. found_bd(line)
         else:
             write_obj.write(line)
+
     def add_to_border_desc(self, line):
         #cw<bt<bdr-hair__<nu<true
         #cw<bt<bdr-linew<nu<0.50
@@ -65,26 +70,22 @@ class CombineBorders:
         else:
             num = ':' + num
         self.__bord_att.append(border_desc + num)
+
     def __border_func(self, line, write_obj):
         if self.__first_five != 'cw<bt':
             self.end_border(line, write_obj)
         else:
             self.add_to_border_desc(line)
+
     def combine_borders(self):
-        read_obj = open(self.__file, 'r')
+        with open(self.__file, 'r') as read_obj:
-        write_obj = open(self.__write_to, 'w')
+            with open(self.__write_to, 'w') as write_obj:
-        line_to_read = 'dummy'
+                for line in read_obj:
-        while line_to_read:
-            line_to_read = read_obj.readline()
-            line = line_to_read
                     self.__first_five = line[0:5]
                     if self.__state == 'border':
                         self.__border_func(line, write_obj)
                     else:
-                to_print = self.__default_func(line)
+                        write_obj.write(self.__default_func(line))
-                write_obj.write(to_print)
-        read_obj.close()
-        write_obj.close()
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "combine_borders.data")

src/calibre/ebooks/rtf2xml/convert_to_tags.py

@@ -1,6 +1,9 @@
-import os, tempfile
+import os, tempfile, sys
-from calibre.ebooks.rtf2xml import copy
+
+from calibre.ebooks.rtf2xml import copy, check_encoding
+
 public_dtd = 'rtf2xml1.0.dtd'
+
 class ConvertToTags:
     """
     Convert file to XML
@@ -10,6 +13,7 @@ class ConvertToTags:
             bug_handler,
             dtd_path,
             no_dtd,
+            encoding,
             indent = None,
             copy = None,
             run_level = 1,
@@ -29,9 +33,14 @@ class ConvertToTags:
         self.__copy = copy
         self.__dtd_path = dtd_path
         self.__no_dtd = no_dtd
+        if encoding != 'mac_roman':
+            self.__encoding = 'cp' + encoding
+        else:
+            self.__encoding = 'mac_roman'
         self.__indent = indent
         self.__run_level = run_level
         self.__write_to = tempfile.mktemp()
+
     def __initiate_values(self):
         """
         Set values, including those for the dictionary.
@@ -61,6 +70,7 @@ class ConvertToTags:
         'tx<ut<__________'  :   self.__text_func,
         'mi<tg<empty_____'  :   self.__empty_func,
         }
+
     def __open_func(self, line):
         """
         Print the opening tag and newlines when needed.
@@ -73,6 +83,7 @@ class ConvertToTags:
         if info in self.__two_new_line:
             self.__write_extra_new_line()
         self.__write_obj.write('<%s>' % info)
+
     def __empty_func(self, line):
         """
         Print out empty tag and newlines when needed.
@@ -85,10 +96,11 @@ class ConvertToTags:
             self.__write_new_line()
         if info in self.__two_new_line:
             self.__write_extra_new_line()
+
     def __open_att_func(self, line):
         """
         Process lines for open tags that have attributes.
-        The important infor is between [17:-1]. Take this info and split it
+        The important info is between [17:-1]. Take this info and split it
         with the delimeter '<'. The first token in this group is the element
         name. The rest are attributes, separated fromt their values by '>'. So
         read each token one at a time, and split them by '>'.
@@ -119,6 +131,7 @@ class ConvertToTags:
             self.__write_new_line()
         if element_name in self.__two_new_line:
             self.__write_extra_new_line()
+
     def __empty_att_func(self, line):
         """
         Same as the __open_att_func, except a '/' is placed at the end of the tag.
@@ -143,6 +156,7 @@ class ConvertToTags:
             self.__write_new_line()
         if element_name in self.__two_new_line:
             self.__write_extra_new_line()
+
     def __close_func(self, line):
         """
         Print out the closed tag and new lines, if appropriate.
@@ -156,6 +170,7 @@ class ConvertToTags:
             self.__write_new_line()
         if info in self.__two_new_line:
             self.__write_extra_new_line()
+
     def __text_func(self, line):
         """
         Simply print out the information between [17:-1]
@@ -163,6 +178,7 @@ class ConvertToTags:
         #tx<nu<__________<Normal;
         # change this!
         self.__write_obj.write(line[17:-1])
+
     def __write_extra_new_line(self):
         """
         Print out extra new lines if the new lines have not exceeded two. If
@@ -172,8 +188,10 @@ class ConvertToTags:
             return
         if self.__new_line < 2:
             self.__write_obj.write('\n')
+
     def __default_func(self, line):
         pass
+
     def __write_new_line(self):
         """
         Print out a new line if a new line has not already been printed out.
@@ -183,11 +201,23 @@ class ConvertToTags:
         if not self.__new_line:
             self.__write_obj.write('\n')
             self.__new_line += 1
+
     def __write_dec(self):
         """
         Write the XML declaration at the top of the document.
         """
+        #keep maximum compatibility with previous version
+        check_encoding_obj = check_encoding.CheckEncoding(
+                    bug_handler=self.__bug_handler)
+
+        if not check_encoding_obj.check_encoding(self.__file, verbose=False):
             self.__write_obj.write('<?xml version="1.0" encoding="US-ASCII" ?>')
+        elif not check_encoding_obj.check_encoding(self.__file, self.__encoding):
+            self.__write_obj.write('<?xml version="1.0" encoding="%s" ?>' % self.__encoding)
+        else:
+            self.__write_obj.write('<?xml version="1.0" encoding="US-ASCII" ?>')
+            sys.stderr.write('Bad RTF encoding, revert to US-ASCII chars and'
+                    ' hope for the best')
         self.__new_line = 0
         self.__write_new_line()
         if self.__no_dtd:
@@ -207,6 +237,7 @@ class ConvertToTags:
             )
         self.__new_line = 0
         self.__write_new_line()
+
     def convert_to_tags(self):
         """
         Read in the file one line at a time. Get the important info, between
@@ -222,18 +253,14 @@ class ConvertToTags:
             an empty tag function.
             """
         self.__initiate_values()
-        read_obj = open(self.__file, 'r')
         self.__write_obj = open(self.__write_to, 'w')
         self.__write_dec()
-        line_to_read = 1
+        with open(self.__file, 'r') as read_obj:
-        while line_to_read:
+            for line in read_obj:
-            line_to_read = read_obj.readline()
-            line = line_to_read
                 self.__token_info = line[:16]
                 action = self.__state_dict.get(self.__token_info)
-            if action != None:
+                if action is not None:
                     action(line)
-        read_obj.close()
         self.__write_obj.close()
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:

src/calibre/ebooks/rtf2xml/copy.py

@@ -23,6 +23,7 @@ class Copy:
     def __init__(self, bug_handler, file = None, deb_dir = None, ):
         self.__file = file
         self.__bug_handler = bug_handler
+
     def set_dir(self, deb_dir):
         """Set the temporary directory to write files to"""
         if deb_dir is None:
@@ -33,19 +34,11 @@ class Copy:
             message = "%(deb_dir)s is not a directory" % vars()
             raise self.__bug_handler , message
         Copy.__dir = deb_dir
+
     def remove_files(self ):
         """Remove files from directory"""
         self.__remove_the_files(Copy.__dir)
-        """
+
-        list_of_files = os.listdir(Copy.__dir)
-        list_of_files = os.listdir(the_dir)
-        for file in list_of_files:
-            rem_file = os.path.join(Copy.__dir,file)
-            if os.path.isdir(rem_file):
-                self.remove_files(rem_file)
-            else:
-                os.remove(rem_file)
-        """
     def __remove_the_files(self, the_dir):
         """Remove files from directory"""
         list_of_files = os.listdir(the_dir)
@@ -58,6 +51,7 @@ class Copy:
                     os.remove(rem_file)
                 except OSError:
                     pass
+
     def copy_file(self, file, new_file):
         """
         Copy the file to a new name

src/calibre/ebooks/rtf2xml/default_encoding.py

@@ -1,61 +1,142 @@
 #########################################################################
 #                                                                       #
-#                                                                       #
 #   copyright 2002 Paul Henry Tremblay                                  #
 #                                                                       #
-#   This program is distributed in the hope that it will be useful,     #
-#   but WITHOUT ANY WARRANTY; without even the implied warranty of      #
-#   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU    #
-#   General Public License for more details.                            #
-#                                                                       #
-#   You should have received a copy of the GNU General Public License   #
-#   along with this program; if not, write to the Free Software         #
-#   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA            #
-#   02111-1307 USA                                                      #
-#                                                                       #
-#                                                                       #
 #########################################################################
+
+'''
+Codepages as to RTF 1.9.1:
+    437	United States IBM
+    708	Arabic (ASMO 708)
+    709	Arabic (ASMO 449+, BCON V4)
+    710	Arabic (transparent Arabic)
+    711	Arabic (Nafitha Enhanced)
+    720	Arabic (transparent ASMO)
+    819	Windows 3.1 (United States and Western Europe)
+    850	IBM multilingual
+    852	Eastern European
+    860	Portuguese
+    862	Hebrew
+    863	French Canadian
+    864	Arabic
+    865	Norwegian
+    866	Soviet Union
+    874	Thai
+    932	Japanese
+    936	Simplified Chinese
+    949	Korean
+    950	Traditional Chinese
+    1250	Eastern European
+    1251	Cyrillic
+    1252	Western European
+    1253	Greek
+    1254	Turkish
+    1255	Hebrew
+    1256	Arabic
+    1257	Baltic
+    1258	Vietnamese
+    1361	Johab
+    10000	MAC Roman
+    10001	MAC Japan
+    10004	MAC Arabic
+    10005	MAC Hebrew
+    10006	MAC Greek
+    10007	MAC Cyrillic
+    10029	MAC Latin2
+    10081	MAC Turkish
+    57002	Devanagari
+    57003	Bengali
+    57004	Tamil
+    57005	Telugu
+    57006	Assamese
+    57007	Oriya
+    57008	Kannada
+    57009	Malayalam
+    57010	Gujarati
+    57011	Punjabi
+'''
+import re
+
 class DefaultEncoding:
     """
     Find the default encoding for the doc
     """
-    def __init__(self, in_file, bug_handler, run_level = 1,):
+    def __init__(self, in_file, bug_handler, run_level = 1, check_raw = False):
-        """
-        Required:
-            'file'
-        Returns:
-            nothing
-            """
         self.__file = in_file
         self.__bug_handler = bug_handler
+        self.__platform = 'Windows'
+        self.__default_num = 'not-defined'
+        self.__code_page = '1252'
+        self.__datafetched = False
+        self.__fetchraw = check_raw
+
     def find_default_encoding(self):
-        platform = 'Windows'
+        if not self.__datafetched:
-        default_num = 'not-defined'
+            self._encoding()
-        code_page = 'ansicpg1252'
+            self.__datafetched = True
-        read_obj = open(self.__file, 'r')
+        if self.__platform == 'Macintosh':
-        line_to_read = 1
+            code_page = self.__code_page
-        while line_to_read:
+        else:
-            line_to_read = read_obj.readline()
+            code_page = 'ansicpg' + self.__code_page
-            line = line_to_read
+        return self.__platform, code_page, self.__default_num
+
+    def get_codepage(self):
+        if not self.__datafetched:
+            self._encoding()
+            self.__datafetched = True
+        return self.__code_page
+
+    def get_platform(self):
+        if not self.__datafetched:
+            self._encoding()
+            self.__datafetched = True
+        return self.__platform
+
+    def _encoding(self):
+        with open(self.__file, 'r') as read_obj:
+            if not self.__fetchraw:
+                for line in read_obj:
                     self.__token_info = line[:16]
                     if self.__token_info == 'mi<mk<rtfhed-end':
                         break
                     if self.__token_info == 'cw<ri<ansi-codpg':
                         #cw<ri<ansi-codpg<nu<10000
-                num = line[20:-1]
+                        self.__code_page = line[20:-1] if int(line[20:-1]) \
-                if not num:
+                                            else '1252'
-                    num = '1252'
-                code_page = 'ansicpg' + num
                     if self.__token_info == 'cw<ri<macintosh_':
-                platform = 'Macintosh'
+                        self.__platform = 'Macintosh'
+                        self.__code_page = 'mac_roman'
+                    elif self.__token_info == 'cw<ri<pc________':
+                        self.__platform = 'IBMPC'
+                        self.__code_page = '437'
+                    elif self.__token_info == 'cw<ri<pca_______':
+                        self.__platform = 'OS/2'
+                        self.__code_page = '850'
                     if self.__token_info == 'cw<ri<deflt-font':
-                default_num = line[20:-1]
+                        self.__default_num = line[20:-1]
                         #cw<ri<deflt-font<nu<0
-            #action = self.__state_dict.get(self.__state)
+            else:
-            #if action == None:
+                fenc = re.compile(r'\\(mac|pc|ansi|pca)[\\ \{\}\t\n]+')
-                #print self.__state
+                fenccp = re.compile(r'\\ansicpg(\d+)[\\ \{\}\t\n]+')
-            #action(line)
+                for line in read_obj:
-        read_obj.close()
+                    if fenccp.search(line):
-        if platform == 'Macintosh':
+                        cp = fenccp.search(line).group(1)
-            code_page = 'mac_roman'
+                        if not int(cp):
-        return platform, code_page, default_num
+                            self.__code_page = cp
+                        break
+                    if fenc.search(line):
+                        enc = fenc.search(line).group(1)
+                        if enc == 'mac':
+                            self.__code_page = 'mac_roman'
+                        elif enc == 'pc':
+                            self.__code_page = '437'
+                        elif enc == 'pca':
+                            self.__code_page = '850'
+
+# if __name__ == '__main__':
+    # encode_obj = DefaultEncoding(
+            # in_file = sys.argv[1],
+            # bug_handler = Exception,
+            # check_raw = True,
+            # )
+    # print encode_obj.get_codepage()

src/calibre/ebooks/rtf2xml/delete_info.py

@@ -16,7 +16,9 @@
 #                                                                       #
 #########################################################################
 import sys, os, tempfile
+
 from calibre.ebooks.rtf2xml import copy
+
 class DeleteInfo:
     """Delelet unecessary destination groups"""
     def __init__(self,
@@ -29,17 +31,18 @@ class DeleteInfo:
         self.__bug_handler = bug_handler
         self.__copy = copy
         self.__write_to = tempfile.mktemp()
-        self.__bracket_count=0
+        self.__bracket_count= 0
         self.__ob_count = 0
         self.__cb_count = 0
-        self.__after_asterisk = 0
+        # self.__after_asterisk = False
-        self.__delete = 0
+        # self.__delete = 0
         self.__initiate_allow()
         self.__ob = 0
-        self.__write_cb = 0
+        self.__write_cb = False
         self.__run_level = run_level
-        self.__found_delete = 0
+        self.__found_delete = False
-        self.__list = 0
+        # self.__list = False
+
     def __initiate_allow(self):
         """
         Initiate a list of destination groups which should be printed out.
@@ -69,6 +72,7 @@ class DeleteInfo:
             'delete'            : self.__delete_func,
             'list'              : self.__list_func,
         }
+
     def __default_func(self,line):
         """Handle lines when in no special state. Look for an asterisk to
         begin a special state. Otherwise, print out line."""
@@ -81,27 +85,29 @@ class DeleteInfo:
             if self.__ob:
                 self.__write_obj.write(self.__ob)
             self.__ob = line
-            return 0
+            return False
         else:
             # write previous bracket, since didn't fine asterisk
             if self.__ob:
                 self.__write_obj.write(self.__ob)
                 self.__ob = 0
-            return 1
+            return True
+
     def __delete_func(self,line):
         """Handle lines when in delete state. Don't print out lines
         unless the state has ended."""
         if self.__delete_count == self.__cb_count:
             self.__state = 'default'
             if self.__write_cb:
-                self.__write_cb = 0
+                self.__write_cb = True
-                return 1
+                return True
-            return 0
+            return False
+
     def __asterisk_func(self,line):
         """
         Determine whether to delete info in group
         Note on self.__cb flag.
-        If you find that you are in a delete group, and the preivous
+        If you find that you are in a delete group, and the previous
         token in not an open bracket (self.__ob = 0), that means
         that the delete group is nested inside another acceptable
         detination group. In this case, you have alrady written
@@ -110,21 +116,21 @@ class DeleteInfo:
         """
         # Test for {\*}, in which case don't enter
         # delete state
-        self.__after_asterisk = 0 # only enter this function once
+        # self.__after_asterisk = False # only enter this function once
-        self.__found_delete = 1
+        self.__found_delete = True
         if self.__token_info == 'cb<nu<clos-brack':
             if self.__delete_count == self.__cb_count:
                 self.__state = 'default'
                 self.__ob = 0
                 # changed this because haven't printed out start
-                return 0
+                return False
             else:
                 # not sure what happens here!
                 # believe I have a '{\*}
                 if self.__run_level > 3:
                     msg = 'flag problem\n'
                     raise self.__bug_handler, msg
-                return 1
+                return True
         elif self.__token_info in self.__allowable :
             if self.__ob:
                 self.__write_obj.write(self.__ob)
@@ -132,61 +138,61 @@ class DeleteInfo:
                 self.__state = 'default'
             else:
                 pass
-            return 1
+            return True
         elif self.__token_info == 'cw<ls<list______':
             self.__ob = 0
             self.__found_list_func(line)
         elif self.__token_info in self.__not_allowable:
             if not self.__ob:
-                self.__write_cb = 1
+                self.__write_cb = True
             self.__ob = 0
             self.__state = 'delete'
             self.__cb_count = 0
-            return 0
+            return False
         else:
             if self.__run_level > 5:
-                msg = 'After an asterisk, and found neither an allowable or non-allowble token\n'
+                msg = ('After an asterisk, and found neither an allowable or non-allowable token\n\
-                msg += 'token is "%s"\n' % self.__token_info
+                            token is "%s"\n') % self.__token_info
-                raise self.__bug_handler
+                raise self.__bug_handler, msg
             if not self.__ob:
-                self.__write_cb = 1
+                self.__write_cb = True
             self.__ob = 0
             self.__state = 'delete'
             self.__cb_count = 0
-            return 0
+            return False
+
     def __found_list_func(self, line):
         """
         print out control words in this group
         """
         self.__state = 'list'
+
     def __list_func(self, line):
         """
         Check to see if the group has ended.
-        Return 1 for all control words.
+        Return True for all control words.
-        Return 0 otherwise.
+        Return False otherwise.
         """
         if self.__delete_count == self.__cb_count and self.__token_info ==\
             'cb<nu<clos-brack':
             self.__state = 'default'
             if self.__write_cb:
-                self.__write_cb = 0
+                self.__write_cb = False
-                return 1
+                return True
-            return 0
+            return False
         elif line[0:2] == 'cw':
-            return 1
+            return True
         else:
-            return 0
+            return False
+
     def delete_info(self):
         """Main method for handling other methods. Read one line in at
-        a time, and determine wheter to print the line based on the state."""
+        a time, and determine whether to print the line based on the state."""
-        line_to_read = 'dummy'
+        with open(self.__file, 'r') as read_obj:
-        read_obj = open(self.__file, 'r')
+            with open(self.__write_to, 'w') as self.__write_obj:
-        self.__write_obj = open(self.__write_to, 'w')
+                for line in read_obj:
-        while line_to_read:
                     #ob<nu<open-brack<0001
-            to_print =1
+                    to_print = True
-            line_to_read = read_obj.readline()
-            line = line_to_read
                     self.__token_info = line[:16]
                     if self.__token_info == 'ob<nu<open-brack':
                         self.__ob_count = line[-5:-1]
@@ -194,23 +200,19 @@ class DeleteInfo:
                         self.__cb_count = line[-5:-1]
                     action = self.__state_dict.get(self.__state)
                     if not action:
-                sys.stderr.write('No action in dictionary state is "%s" \n'
+                        sys.stderr.write(_('No action in dictionary state is "%s" \n')
                                 % self.__state)
                     to_print = action(line)
-            """
+                    # if self.__after_asterisk:
-            if self.__after_asterisk:
+                        # to_print = self.__asterisk_func(line)
-                to_print = self.__asterisk_func(line)
+                    # elif self.__list:
-            elif self.__list:
+                        # self.__in_list_func(line)
-                self.__in_list_func(line)
+                    # elif self.__delete:
-            elif self.__delete:
+                        # to_print = self.__delete_func(line)
-                to_print = self.__delete_func(line)
+                    # else:
-            else:
+                        # to_print = self.__default_func(line)
-                to_print = self.__default_func(line)
-            """
                     if to_print:
                         self.__write_obj.write(line)
-        self.__write_obj.close()
-        read_obj.close()
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "delete_info.data")

src/calibre/ebooks/rtf2xml/footnote.py

@@ -16,7 +16,9 @@
 #                                                                       #
 #########################################################################
 import os, tempfile
+
 from calibre.ebooks.rtf2xml import copy
+
 class Footnote:
     """
     Two public methods are available. The first separates all of the
@@ -35,6 +37,7 @@ class Footnote:
         self.__copy = copy
         self.__write_to = tempfile.mktemp()
         self.__found_a_footnote = 0
+
     def __first_line_func(self, line):
         """
         Print the tag info for footnotes.  Check whether footnote is an
@@ -47,6 +50,7 @@ class Footnote:
             self.__write_to_foot_obj.write(
             'mi<tg<open-att__<footnote<num>%s\n' % self.__footnote_count)
         self.__first_line = 0
+
     def __in_footnote_func(self, line):
         """Handle all tokens that are part of footnote"""
         if self.__first_line:
@@ -68,6 +72,7 @@ class Footnote:
             'mi<mk<footnt-clo\n')
         else:
             self.__write_to_foot_obj.write(line)
+
     def __found_footnote(self, line):
         """ Found a footnote"""
         self.__found_a_footnote = 1
@@ -81,6 +86,7 @@ class Footnote:
         'mi<mk<footnt-ind<%04d\n' % self.__footnote_count)
         self.__write_to_foot_obj.write(
         'mi<mk<footnt-ope<%04d\n' % self.__footnote_count)
+
     def __default_sep(self, line):
         """Handle all tokens that are not footnote tokens"""
         if self.__token_info == 'cw<nt<footnote__':
@@ -91,6 +97,7 @@ class Footnote:
             self.__write_obj.write(
                 'tx<nu<__________<%s\n' % num
             )
+
     def __initiate_sep_values(self):
         """
         initiate counters for separate_footnotes method.
@@ -102,6 +109,7 @@ class Footnote:
         self.__in_footnote = 0
         self.__first_line = 0 #have not processed the first line of footnote
         self.__footnote_count = 0
+
     def separate_footnotes(self):
         """
         Separate all the footnotes in an RTF file and put them at the bottom,
@@ -111,14 +119,11 @@ class Footnote:
         bottom of the main file.
         """
         self.__initiate_sep_values()
-        read_obj = open(self.__file)
-        self.__write_obj = open(self.__write_to, 'w')
         self.__footnote_holder = tempfile.mktemp()
-        self.__write_to_foot_obj = open(self.__footnote_holder, 'w')
+        with open(self.__file) as read_obj:
-        line_to_read = 1
+            with open(self.__write_to, 'w') as self.__write_obj:
-        while line_to_read:
+                with open(self.__footnote_holder, 'w') as self.__write_to_foot_obj:
-            line_to_read = read_obj.readline()
+                    for line in read_obj:
-            line = line_to_read
                         self.__token_info = line[:16]
                         # keep track of opening and closing brackets
                         if self.__token_info == 'ob<nu<open-brack':
@@ -131,11 +136,8 @@ class Footnote:
                         # not in the middle of footnote text
                         else:
                             self.__default_sep(line)
-        self.__write_obj.close()
+        with open(self.__footnote_holder, 'r') as read_obj:
-        read_obj.close()
+            with open(self.__write_to, 'a') as write_obj:
-        self.__write_to_foot_obj.close()
-        read_obj = open(self.__footnote_holder, 'r')
-        write_obj = open(self.__write_to, 'a')
                 write_obj.write(
                     'mi<mk<sect-close\n'
                     'mi<mk<body-close\n'
@@ -143,26 +145,24 @@ class Footnote:
                     'mi<tg<close_____<body\n'
                     'mi<tg<close_____<doc\n'
                     'mi<mk<footnt-beg\n')
-        line = 1
+                for line in read_obj:
-        while line:
-            line = read_obj.readline()
                     write_obj.write(line)
                 write_obj.write(
                 'mi<mk<footnt-end\n')
-        read_obj.close()
-        write_obj.close()
         os.remove(self.__footnote_holder)
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "footnote_separate.data")
         copy_obj.rename(self.__write_to, self.__file)
         os.remove(self.__write_to)
+
     def update_info(self, file, copy):
         """
         Unused method
         """
         self.__file = file
         self.__copy = copy
+
     def __get_foot_body_func(self, line):
         """
         Process lines in main body and look for beginning of footnotes.
@@ -172,6 +172,7 @@ class Footnote:
             self.__state = 'foot'
         else:
             self.__write_obj.write(line)
+
     def __get_foot_foot_func(self, line):
         """
         Copy footnotes from bottom of file to a separate, temporary file.
@@ -180,6 +181,7 @@ class Footnote:
             self.__state = 'body'
         else:
             self.__write_to_foot_obj.write(line)
+
     def __get_footnotes(self):
         """
         Private method to remove footnotes from main file.  Read one line from
@@ -188,21 +190,16 @@ class Footnote:
         These two functions do the work of separating the footnotes form the
         body.
         """
-        read_obj = open(self.__file)
+        with open(self.__file) as read_obj:
-        self.__write_obj = open(self.__write_to, 'w')
+            with open(self.__write_to, 'w') as self.__write_obj:
-            # self.__write_to = "footnote_info.data"
+                with open(self.__footnote_holder, 'w') as self.__write_to_foot_obj:
-        self.__write_to_foot_obj = open(self.__footnote_holder, 'w')
+                    for line in read_obj:
-        line = 1
-        while line:
-            line = read_obj.readline()
                         self.__token_info = line[:16]
                         if self.__state == 'body':
                             self.__get_foot_body_func(line)
                         elif self.__state == 'foot':
                             self.__get_foot_foot_func(line)
-        read_obj.close()
+
-        self.__write_obj.close()
-        self.__write_to_foot_obj.close()
     def __get_foot_from_temp(self, num):
         """
         Private method for joining footnotes to body. This method reads from
@@ -213,9 +210,7 @@ class Footnote:
         look_for = 'mi<mk<footnt-ope<' + num + '\n'
         found_foot = 0
         string_to_return = ''
-        line = 1
+        for line in self.__read_from_foot_obj:
-        while line:
-            line = self.__read_from_foot_obj.readline()
             if found_foot:
                 if line == 'mi<mk<footnt-clo\n':
                     return string_to_return
@@ -223,6 +218,7 @@ class Footnote:
             else:
                 if line == look_for:
                     found_foot = 1
+
     def __join_from_temp(self):
         """
         Private method for rejoining footnotes to body.  Read from the
@@ -232,16 +228,14 @@ class Footnote:
         print out to the third file.
         If no footnote marker is found, simply print out the token (line).
         """
-        self.__read_from_foot_obj = open(self.__footnote_holder, 'r')
+        with open(self.__footnote_holder, 'r') as self.__read_from_foot_obj:
-        read_obj = open(self.__write_to, 'r')
+            with open(self.__write_to, 'r') as read_obj:
-        self.__write_obj = open(self.__write_to2, 'w')
+                with open(self.__write_to2, 'w') as self.__write_obj:
-        line = 1
+                    for line in read_obj:
-        while line:
-            line = read_obj.readline()
                         if line[:16] == 'mi<mk<footnt-ind':
                             line = self.__get_foot_from_temp(line[17:-1])
                         self.__write_obj.write(line)
-        read_obj.close()
+
     def join_footnotes(self):
         """
         Join the footnotes from the bottom of the file and put them in their
@@ -258,8 +252,8 @@ class Footnote:
         self.__state = 'body'
         self.__get_footnotes()
         self.__join_from_temp()
-        self.__write_obj.close()
+        # self.__write_obj.close()
-        self.__read_from_foot_obj.close()
+        # self.__read_from_foot_obj.close()
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to2, "footnote_joined.data")

src/calibre/ebooks/rtf2xml/get_char_map.py

@@ -43,16 +43,18 @@ class GetCharMap:
     def get_char_map(self, map):
         if map == 'ansicpg0':
             map = 'ansicpg1250'
-        found_map = 0
+        if map in ('ansicpg10000', '10000'):
+            map = 'mac_roman'
+        found_map = False
         map_dict = {}
         self.__char_file.seek(0)
-        for line in self.__char_file.readlines():
+        for line in self.__char_file:
             if not line.strip(): continue
             begin_element = '<%s>' % map;
             end_element = '</%s>' % map
             if not found_map:
                 if begin_element in line:
-                    found_map = 1
+                    found_map = True
             else:
                 if end_element in line:
                     break
@@ -62,8 +64,7 @@ class GetCharMap:
 
 
         if not found_map:
-            msg = 'no map found\n'
+            msg = 'no map found\nmap is "%s"\n'%(map,)
-            msg += 'map is "%s"\n'%(map,)
             raise self.__bug_handler, msg
         return map_dict
 

src/calibre/ebooks/rtf2xml/hex_2_utf8.py

@@ -57,7 +57,7 @@ class Hex2Utf8:
         """
         self.__file = in_file
         self.__copy = copy
-        if area_to_convert != 'preamble' and area_to_convert != 'body':
+        if area_to_convert not in ('preamble', 'body'):
             msg = (
             'Developer error! Wrong flag.\n'
             'in module "hex_2_utf8.py\n'
@@ -79,7 +79,8 @@ class Hex2Utf8:
         self.__write_to = tempfile.mktemp()
         self.__bug_handler = bug_handler
         self.__invalid_rtf_handler = invalid_rtf_handler
-    def update_values(  self,
+
+    def update_values(self,
                         file,
                         area_to_convert,
                         char_file,
@@ -132,6 +133,7 @@ class Hex2Utf8:
         # self.__convert_symbol = 0
         # self.__convert_wingdings = 0
         # self.__convert_zapf = 0
+
     def __initiate_values(self):
         """
         Required:
@@ -191,6 +193,7 @@ class Hex2Utf8:
             'body'          :       self.__body_func,
             'mi<mk<body-open_'  :   self.__found_body_func,
             'tx<hx<__________'  :   self.__hex_text_func,
+            # 'tx<nu<__________'  :   self.__text_func,
             }
         self.__body_state_dict = {
             'preamble'      :       self.__preamble_for_body_func,
@@ -209,6 +212,7 @@ class Hex2Utf8:
         }
         self.__caps_list = ['false']
         self.__font_list = ['not-defined']
+
     def __hex_text_func(self, line):
         """
         Required:
@@ -218,12 +222,12 @@ class Hex2Utf8:
             token is in the dictionary, then check if the value starts with a
             "&". If it does, then tag the result as utf text. Otherwise, tag it
             as normal text.
-            If the nex_num is not in the dictionary, then a mistake has been
+            If the hex_num is not in the dictionary, then a mistake has been
             made.
             """
         hex_num = line[17:-1]
         converted = self.__current_dict.get(hex_num)
-        if converted != None:
+        if converted is not None:
             # tag as utf-8
             if converted[0:1] == "&":
                 font = self.__current_dict_name
@@ -263,42 +267,43 @@ class Hex2Utf8:
                     # msg += 'dictionary is %s\n' % self.__current_dict_name
                     msg = 'Character "&#x%s;" does not appear to be valid (or is a control character)\n' % token
                     raise self.__bug_handler, msg
+
     def __found_body_func(self, line):
         self.__state = 'body'
         self.__write_obj.write(line)
+
     def __body_func(self, line):
         """
         When parsing preamble
         """
         self.__write_obj.write(line)
+
     def __preamble_func(self, line):
         action = self.__preamble_state_dict.get(self.__token_info)
-        if action != None:
+        if action is not None:
             action(line)
         else:
             self.__write_obj.write(line)
+
     def __convert_preamble(self):
         self.__state = 'preamble'
-        read_obj = open(self.__file, 'r')
         self.__write_obj = open(self.__write_to, 'w')
-        line_to_read = 1
+        with open(self.__file, 'r') as read_obj:
-        while line_to_read:
+           for line in read_obj:
-            line_to_read = read_obj.readline()
-            line = line_to_read
                 self.__token_info = line[:16]
                 action = self.__preamble_state_dict.get(self.__state)
-            if action == None:
+                if action is None:
-                sys.stderr.write('error no state found in hex_2_utf8',
+                    sys.stderr.write(_('error no state found in hex_2_utf8'),
                     self.__state
                     )
                 action(line)
-        read_obj.close()
         self.__write_obj.close()
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "preamble_utf_convert.data")
         copy_obj.rename(self.__write_to, self.__file)
         os.remove(self.__write_to)
+
     def __preamble_for_body_func(self, line):
         """
         Required:
@@ -311,6 +316,7 @@ class Hex2Utf8:
         if self.__token_info == 'mi<mk<body-open_':
             self.__found_body_func(line)
         self.__write_obj.write(line)
+
     def __body_for_body_func(self, line):
         """
         Required:
@@ -321,10 +327,11 @@ class Hex2Utf8:
             Used when parsing the body.
         """
         action = self.__in_body_dict.get(self.__token_info)
-        if action != None:
+        if action is not None:
             action(line)
         else:
             self.__write_obj.write(line)
+
     def __start_font_func(self, line):
         """
         Required:
@@ -348,6 +355,7 @@ class Hex2Utf8:
         else:
             self.__current_dict_name = 'default'
             self.__current_dict = self.__def_dict
+
     def __end_font_func(self, line):
         """
         Required:
@@ -376,6 +384,7 @@ class Hex2Utf8:
         else:
             self.__current_dict_name = 'default'
             self.__current_dict = self.__def_dict
+
     def __start_special_font_func_old(self, line):
         """
         Required:
@@ -398,6 +407,7 @@ class Hex2Utf8:
             self.__current_dict.append(self.__dingbats_dict)
             self.__special_fonts_found += 1
             self.__current_dict_name = 'Zapf Dingbats'
+
     def __end_special_font_func(self, line):
         """
         Required:
@@ -416,6 +426,7 @@ class Hex2Utf8:
             self.__current_dict.pop()
             self.__special_fonts_found -= 1
             self.__dict_name = 'default'
+
     def __start_caps_func_old(self, line):
         """
         Required:
@@ -427,6 +438,7 @@ class Hex2Utf8:
             self.__in_caps to 1
         """
         self.__in_caps = 1
+
     def __start_caps_func(self, line):
         """
         Required:
@@ -440,6 +452,7 @@ class Hex2Utf8:
         self.__in_caps = 1
         value = line[17:-1]
         self.__caps_list.append(value)
+
     def __end_caps_func(self, line):
         """
         Required:
@@ -455,7 +468,8 @@ class Hex2Utf8:
         else:
             sys.stderr.write('Module is hex_2_utf8\n')
             sys.stderr.write('method is __end_caps_func\n')
-            sys.stderr.write('caps list should be more than one?\n')
+            sys.stderr.write('caps list should be more than one?\n') #self.__in_caps not set
+
     def __text_func(self, line):
         """
         Required:
@@ -466,9 +480,8 @@ class Hex2Utf8:
             if in caps, convert. Otherwise, print out.
         """
         text = line[17:-1]
-        if self.__current_dict_name == 'Symbol'\
+        # print line
-          or self.__current_dict_name == 'Wingdings'\
+        if self.__current_dict_name in ('Symbol', 'Wingdings', 'Zapf Dingbats'):
-          or self.__current_dict_name == 'Zapf Dingbats':
             the_string = ''
             for letter in text:
                 hex_num = hex(ord(letter))
@@ -477,21 +490,21 @@ class Hex2Utf8:
                 hex_num = hex_num[2:]
                 hex_num = '\'%s' % hex_num
                 converted = self.__current_dict.get(hex_num)
-                if converted == None:
+                if converted is None:
                     sys.stderr.write('module is hex_2_ut8\n')
                     sys.stderr.write('method is __text_func\n')
                     sys.stderr.write('no hex value for "%s"\n' % hex_num)
                 else:
                     the_string += converted
             self.__write_obj.write('tx<nu<__________<%s\n' % the_string)
+            # print the_string
         else:
             if self.__caps_list[-1] == 'true' \
                 and self.__convert_caps\
-                and self.__current_dict_name != 'Symbol'\
+                and self.__current_dict_name not in ('Symbol', 'Wingdings', 'Zapf Dingbats'):
-                and self.__current_dict_name != 'Wingdings'\
-                and self.__current_dict_name != 'Zapf Dingbats':
                 text = text.upper()
             self.__write_obj.write('tx<nu<__________<%s\n' % text)
+
     def __utf_to_caps_func(self, line):
         """
         Required:
@@ -506,6 +519,7 @@ class Hex2Utf8:
             # utf_text = utf_text.upper()
             utf_text = self.__utf_token_to_caps_func(utf_text)
         self.__write_obj.write('tx<ut<__________<%s\n' % utf_text)
+
     def __utf_token_to_caps_func(self, char_entity):
         """
         Required:
@@ -530,28 +544,26 @@ class Hex2Utf8:
             return char_entity
         else:
             return converted
+
     def __convert_body(self):
         self.__state = 'body'
-        read_obj = open(self.__file, 'r')
+        with open(self.__file, 'r') as read_obj:
             self.__write_obj = open(self.__write_to, 'w')
-        line_to_read = 1
+            for line in read_obj:
-        while line_to_read:
-            line_to_read = read_obj.readline()
-            line = line_to_read
                 self.__token_info = line[:16]
                 action = self.__body_state_dict.get(self.__state)
-            if action == None:
+                if action is None:
                     sys.stderr.write('error no state found in hex_2_utf8',
                     self.__state
                     )
                 action(line)
-        read_obj.close()
         self.__write_obj.close()
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "body_utf_convert.data")
         copy_obj.rename(self.__write_to, self.__file)
         os.remove(self.__write_to)
+
     def convert_hex_2_utf8(self):
         self.__initiate_values()
         if self.__area_to_convert == 'preamble':

src/calibre/ebooks/rtf2xml/inline.py

@@ -1,5 +1,7 @@
 import sys, os, tempfile
+
 from calibre.ebooks.rtf2xml import copy
+
 """
 States.
 1. default
@@ -36,6 +38,7 @@ class Inline:
         self.__copy = copy
         self.__run_level = run_level
         self.__write_to = tempfile.mktemp()
+
     def __initiate_values(self):
         """
         Initiate all values.
@@ -51,7 +54,6 @@ class Inline:
             'tx<ut<__________'  :       self.__found_text_func,
             'mi<mk<inline-fld'  :       self.__found_text_func,
             'text'              :       self.__found_text_func,
-            'cw<nu<hard-lineb'  :       self.__found_text_func, #calibre
             'cb<nu<clos-brack'  :       self.__close_bracket_func,
             'mi<mk<par-end___'  :       self.__end_para_func,
             'mi<mk<footnt-ope'  :       self.__end_para_func,
@@ -63,7 +65,6 @@ class Inline:
             'tx<hx<__________'  :       self.__found_text_func,
             'tx<ut<__________'  :       self.__found_text_func,
             'text'              :       self.__found_text_func,
-            'cw<nu<hard-lineb'  :       self.__found_text_func, #calibre
             'mi<mk<inline-fld'  :       self.__found_text_func,
             'ob<nu<open-brack':         self.__found_open_bracket_func,
             'mi<mk<par-end___'  :       self.__end_para_func,
@@ -110,6 +111,7 @@ class Inline:
             'underlined'    :   'underlined',
         }
         self.__caps_list = ['false']
+
     def __set_list_func(self, line):
         """
         Requires:
@@ -128,6 +130,7 @@ class Inline:
                 self.__place = 'in_list'
                 self.__inline_list = self.__list_inline_list
                 self.__groups_in_waiting = self.__groups_in_waiting_list
+
     def __default_func(self, line):
         """
         Requires:
@@ -140,8 +143,8 @@ class Inline:
         action = self.__default_dict.get(self.__token_info)
         if action:
             action(line)
-        if self.__token_info != 'cw<nu<hard-lineb': #calibre
         self.__write_obj.write(line)
+
     def __found_open_bracket_func(self, line):
         """
         Requires:
@@ -156,6 +159,7 @@ class Inline:
         self.__groups_in_waiting[0] += 1
         self.__inline_list.append({})
         self.__inline_list[-1]['contains_inline'] = 0
+
     def __after_open_bracket_func(self, line):
         """
         Requires:
@@ -176,6 +180,7 @@ class Inline:
                 self.__state = 'default' #  a non control word?
                 action(line)
         self.__write_obj.write(line)
+
     def __handle_control_word(self, line):
         """
         Required:
@@ -206,6 +211,7 @@ class Inline:
                 elif char_value == 'Zapf Dingbats':
                     self.__write_obj.write('mi<mk<font-dingb\n')
             """
+
     def __close_bracket_func(self, line):
         """
         Requires:
@@ -244,6 +250,7 @@ class Inline:
         self.__inline_list.pop()
         if self.__groups_in_waiting[0] != 0:
             self.__groups_in_waiting[0] -= 1
+
     def __found_text_func(self, line):
         """
         Required:
@@ -257,7 +264,6 @@ class Inline:
                 Text can mark the start of a paragraph.
                 If already in a paragraph, check to see if any groups are waiting
                 to be added. If so, use another method to write these groups.
-            3. If not check if hardline break, then write
         """
         if self.__place == 'in_list':
             self.__write_inline()
@@ -265,10 +271,7 @@ class Inline:
             if not self.__in_para:
                 self.__in_para = 1
                 self.__start_para_func(line)
-            else:
+            elif self.__groups_in_waiting[0] != 0:
-                if self.__token_info == 'cw<nu<hard-lineb': #calibre
-                    self.__write_obj.write('mi<tg<empty_____<hardline-break\n')
-                if self.__groups_in_waiting[0] != 0:
                     self.__write_inline()
 
     def __write_inline(self):
@@ -314,6 +317,7 @@ class Inline:
                             self.__write_obj.write('<%s>%s' % (the_key, the_dict[the_key]))
                     self.__write_obj.write('\n')
         self.__groups_in_waiting[0] = 0
+
     def __end_para_func(self, line):
         """
         Requires:
@@ -342,6 +346,7 @@ class Inline:
                     self.__write_obj.write('mi<mk<caps-end__\n')
                 self.__write_obj.write('mi<tg<close_____<inline\n')
         self.__in_para = 0
+
     def __start_para_func(self, line):
         """
         Requires:
@@ -369,12 +374,14 @@ class Inline:
                         self.__write_obj.write('<%s>%s' % (the_key, the_dict[the_key]))
                 self.__write_obj.write('\n')
         self.__groups_in_waiting[0] = 0
+
     def __found_field_func(self, line):
         """
         Just a default function to make sure I don't prematurely exit
         default state
         """
         pass
+
     def form_tags(self):
         """
         Requires:
@@ -386,12 +393,9 @@ class Inline:
             the state.
         """
         self.__initiate_values()
-        read_obj = open(self.__file, 'r')
+        with open(self.__file, 'r') as read_obj:
-        self.__write_obj = open(self.__write_to, 'w')
+            with open(self.__write_to, 'w') as self.__write_obj:
-        line_to_read = 1
+                for line in read_obj:
-        while line_to_read:
-            line_to_read = read_obj.readline()
-            line = line_to_read
                     token = line[0:-1]
                     self.__token_info = ''
                     if token == 'tx<mc<__________<rdblquote'\
@@ -406,12 +410,10 @@ class Inline:
                         self.__token_info = line[:16]
                     self.__set_list_func(line)
                     action = self.__state_dict.get(self.__state)
-            if action == None:
+                    if action is None:
                         sys.stderr.write('No matching state in module inline_for_lists.py\n')
                         sys.stderr.write(self.__state + '\n')
                     action(line)
-        read_obj.close()
-        self.__write_obj.close()
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "inline.data")

src/calibre/ebooks/rtf2xml/line_endings.py

@@ -15,8 +15,11 @@
 #                                                                       #
 #                                                                       #
 #########################################################################
-import os, tempfile, re
+import os, tempfile
+
 from calibre.ebooks.rtf2xml import copy
+from calibre.utils.cleantext import clean_ascii_chars
+
 class FixLineEndings:
     """Fix line endings"""
     def __init__(self,
@@ -32,34 +35,21 @@ class FixLineEndings:
         self.__run_level = run_level
         self.__write_to = tempfile.mktemp()
         self.__replace_illegals = replace_illegals
+
     def fix_endings(self):
-        ##tempFileName = tempfile.mktemp()
+        #read
-        illegal_regx = re.compile( '\x00|\x01|\x02|\x03|\x04|\x05|\x06|\x07|\x08|\x0B|\x0E|\x0F|\x10|\x11|\x12|\x13')
+        with open(self.__file, 'r') as read_obj:
-        #nums = [0, 1, 2, 3, 4, 5, 6, 7, 8,  11,  14, 15, 16, 17, 18, 19]
+            input_file = read_obj.read()
-        """
+        #calibre go from win and mac to unix
-read_obj = open(self.__file, 'r')
+        input_file = input_file.replace ('\r\n', '\n')
-line = read_obj.read(1000)
+        input_file = input_file.replace ('\r', '\n')
-regexp = re.compile(r"\r")
+        #remove ASCII invalid chars : 0 to 8 and 11-14 to 24-26-27
-macintosh = regexp.search(line)
-read_obj.close()
-        """
-        # always check since I have to get rid of illegal characters
-        macintosh = 1
-        if macintosh:
-            line = 1
-            read_obj = open(self.__file, 'r')
-            write_obj = open(self.__write_to, 'w')
-            while line:
-                line = read_obj.read(1000)
-                # line = re.sub(regexp,"\n",line)
-                line = line.replace ('\r', '\n')
         if self.__replace_illegals:
-                    line = re.sub(illegal_regx, '', line)
+            input_file = clean_ascii_chars(input_file)
-                    # for num in nums:
+        #write
-                        # line = line.replace(chr(num), '')
+        with open(self.__write_to, 'wb') as write_obj:
-                write_obj.write(line )
+            write_obj.write(input_file)
-            read_obj.close()
+        #copy
-            write_obj.close()
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "line_endings.data")

src/calibre/ebooks/rtf2xml/pict.py

@@ -16,7 +16,9 @@
 #                                                                       #
 #########################################################################
 import sys, os, tempfile
+
 from calibre.ebooks.rtf2xml import copy
+
 class Pict:
     """Process graphic information"""
     def __init__(self,
@@ -36,13 +38,11 @@ class Pict:
         self.__ob_count = 0
         self.__cb_count = 0
         self.__pict_count = 0
-        self.__in_pict = 0
+        self.__in_pict = False
-        self.__already_found_pict = 0
+        self.__already_found_pict = False
         self.__orig_file = orig_file
         self.__initiate_pict_dict()
         self.__out_file = out_file
-        # this is left over
-        self.__no_ask = 1
 
     def __initiate_pict_dict(self):
         self.__pict_dict = {
@@ -71,26 +71,15 @@ class Pict:
                 self.__out_file))
         else:
             dir_name = os.path.dirname(self.__orig_file)
-        # self.__output_to_file_func()
         self.__dir_name = base_name + "_rtf_pict_dir/"
         self.__dir_name = os.path.join(dir_name, self.__dir_name)
         if not os.path.isdir(self.__dir_name):
             try:
                 os.mkdir(self.__dir_name)
             except OSError, msg:
-                msg = str(msg)
+                msg = "%sCouldn't make directory '%s':\n" % (str(msg), self.__dir_name)
-                msg += "Couldn't make directory '%s':\n" % (self.__dir_name)
                 raise self.__bug_handler
         else:
-            if self.__no_ask:
-                user_response = 'r'
-            else:
-                msg = 'Do you want to remove all files in %s?\n' % self.__dir_name
-                msg += 'Type "r" to remove.\n'
-                msg +=  'Type any other key to keep files in place.\n'
-                sys.stderr.write(msg)
-                user_response = raw_input()
-            if user_response == 'r':
             if self.__run_level > 1:
                 sys.stderr.write('Removing files from old pict directory...\n')
             all_files = os.listdir(self.__dir_name)
@@ -107,21 +96,18 @@ class Pict:
         """Create a file for all the pict data to be written to.
         """
         self.__pict_file = os.path.join(self.__dir_name, 'picts.rtf')
-        write_pic_obj = open(self.__pict_file, 'w')
-        write_pic_obj.close()
         self.__write_pic_obj = open(self.__pict_file, 'a')
 
     def __in_pict_func(self, line):
         if self.__cb_count == self.__pict_br_count:
-            self.__in_pict = 0
+            self.__in_pict = False
             self.__write_pic_obj.write("}\n")
-            return 1
+            return True
         else:
             action = self.__pict_dict.get(self.__token_info)
             if action:
-                line = action(line)
+                self.__write_pic_obj.write(action(line))
-                self.__write_pic_obj.write(line)
+            return False
-            return 0
 
     def __default(self, line, write_obj):
         """Determine if each token marks the beginning of pict data.
@@ -142,32 +128,27 @@ class Pict:
             write_obj.write('mi<mk<pict-end__\n')
             if not self.__already_found_pict:
                 self.__create_pict_file()
-                self.__already_found_pict=1;
+                self.__already_found_pict=True;
                 self.__print_rtf_header()
             self.__in_pict = 1
             self.__pict_br_count = self.__ob_count
             self.__cb_count = 0
             self.__write_pic_obj.write("{\\pict\n")
-            return 0
+            return False
-        return 1
+        return True
 
     def __print_rtf_header(self):
         """Print to pict file the necessary RTF data for the file to be
         recognized as an RTF file.
         """
-        self.__write_pic_obj.write("{\\rtf1 \n")
+        self.__write_pic_obj.write("{\\rtf1 \n{\\fonttbl\\f0\\null;} \n")
-        self.__write_pic_obj.write("{\\fonttbl\\f0\\null;} \n")
+        self.__write_pic_obj.write("{\\colortbl\\red255\\green255\\blue255;} \n\\pard \n")
-        self.__write_pic_obj.write("{\\colortbl\\red255\\green255\\blue255;} \n")
-        self.__write_pic_obj.write("\\pard \n")
 
     def process_pict(self):
         self.__make_dir()
-        read_obj = open(self.__file)
+        with open(self.__file) as read_obj:
-        write_obj = open(self.__write_to, 'w')
+            with open(self.__write_to, 'w') as write_obj:
-        line_to_read = 'dummy'
+                for line in read_obj:
-        while line_to_read:
-            line_to_read = read_obj.readline()
-            line = line_to_read
                     self.__token_info = line[:16]
                     if self.__token_info == 'ob<nu<open-brack':
                         self.__ob_count = line[-5:-1]
@@ -184,11 +165,13 @@ class Pict:
                 if self.__already_found_pict:
                     self.__write_pic_obj.write("}\n")
                     self.__write_pic_obj.close()
-        read_obj.close()
-        write_obj.close()
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "pict.data")
+            try:
+                copy_obj.copy_file(self.__pict_file, "pict.rtf")
+            except:
+                pass
         copy_obj.rename(self.__write_to, self.__file)
         os.remove(self.__write_to)
         if self.__pict_count == 0:

src/calibre/ebooks/rtf2xml/process_tokens.py

@@ -16,7 +16,9 @@
 #                                                                       #
 #########################################################################
 import os, re, tempfile
+
 from calibre.ebooks.rtf2xml import copy, check_brackets
+
 class ProcessTokens:
     """
     Process each token on a line and add information that will be useful for
@@ -41,9 +43,11 @@ class ProcessTokens:
         self.__bracket_count=0
         self.__exception_handler = exception_handler
         self.__bug_handler = bug_handler
+
     def compile_expressions(self):
         self.__num_exp = re.compile(r"([a-zA-Z]+)(.*)")
         self.__utf_exp = re.compile(r'(&.*?;)')
+
     def initiate_token_dict(self):
         self.__return_code = 0
         self.dict_token={
@@ -66,6 +70,7 @@ class ProcessTokens:
         ';'                  :	('mc', ';', self.ms_sub_func),
         # this must be wrong
         '-'                  :	('mc', '-', self.ms_sub_func),
+        'line'               :  ('mi', 'hardline-break', self.hardline_func), #calibre
         # misc => ml
         '*'                  :	('ml', 'asterisk__', self.default_func),
         ':'                  :	('ml', 'colon_____', self.default_func),
@@ -73,7 +78,6 @@ class ProcessTokens:
         'backslash'          :	('nu', '\\', self.text_func),
         'ob'                 :	('nu', '{', self.text_func),
         'cb'                 :	('nu', '}', self.text_func),
-        'line'               :  ('nu', 'hard-lineb', self.default_func), #calibre
         #'line'               :  ('nu', ' ', self.text_func), calibre
         # paragraph formatting => pf
         'page'               :  ('pf', 'page-break', self.default_func),
@@ -159,6 +163,8 @@ class ProcessTokens:
         'rtf'                :	('ri', 'rtf_______', self.default_func),
         'deff'               :	('ri', 'deflt-font', self.default_func),
         'mac'                :	('ri', 'macintosh_', self.default_func),
+        'pc'                 :	('ri', 'pc________', self.default_func),
+        'pca'                :	('ri', 'pca_______', self.default_func),
         'ansi'               :	('ri', 'ansi______', self.default_func),
         'ansicpg'            :	('ri', 'ansi-codpg', self.default_func),
         # notes => nt
@@ -595,30 +601,37 @@ class ProcessTokens:
         num = num[1:] # chop off leading 0, which I added
         num = num.upper() # the mappings store hex in caps
         return 'tx<hx<__________<\'%s\n' % num # add an ' for the mappings
+
     def ms_sub_func(self, pre, token, num):
         return 'tx<mc<__________<%s\n' % token
+
+    def hardline_func(self, pre, token, num):
+        return 'mi<tg<empty_____<%s\n' % token
+
     def default_func(self, pre, token, num):
-        if num == None:
+        if num is None:
             num = 'true'
         return 'cw<%s<%s<nu<%s\n' % (pre, token, num)
+
     def __list_type_func(self, pre, token, num):
         type = 'arabic'
-        if num == None:
+        if num is None:
             type = 'Arabic'
         else:
             try:
                 num = int(num)
             except ValueError:
                 if self.__run_level > 3:
-                    msg = 'number "%s" cannot be converted to integer\n' % num
+                    msg = 'Number "%s" cannot be converted to integer\n' % num
                     raise self.__bug_handler, msg
             type = self.__number_type_dict.get(num)
-            if type == None:
+            if type is None:
                 if self.__run_level > 3:
                     msg = 'No type for "%s" in self.__number_type_dict\n'
                     raise self.__bug_handler
                 type = 'Arabic'
         return 'cw<%s<%s<nu<%s\n' % (pre, token, type)
+
     def __language_func(self, pre, token, num):
         lang_name = self.__language_dict.get(int(re.search('[0-9]+', num).group()))
         if not lang_name:
@@ -627,31 +640,36 @@ class ProcessTokens:
                 msg = 'No entry for number "%s"' % num
                 raise self.__bug_handler, msg
         return 'cw<%s<%s<nu<%s\n' % (pre, token, lang_name)
+
     def two_part_func(self, pre, token, num):
         list = token.split("<")
         token = list[0]
         num = list[1]
         return 'cw<%s<%s<nu<%s\n' % (pre, token, num)
         ##return 'cw<nu<nu<nu<%s>num<%s\n' % (token, num)
+
     def divide_by_2(self, pre, token, num):
         num = self.divide_num(num, 2)
         return 'cw<%s<%s<nu<%s\n' % (pre, token, num)
         ##return 'cw<nu<nu<nu<%s>%s<%s\n' % (token, num, token)
+
     def divide_by_20(self, pre, token, num):
         num = self.divide_num(num, 20)
         return 'cw<%s<%s<nu<%s\n' % (pre, token, num)
         ##return 'cw<nu<nu<nu<%s>%s<%s\n' % (token, num, token)
+
     def text_func(self, pre, token, num=None):
         return 'tx<nu<__________<%s\n' % token
+
     def ob_func(self, pre, token, num=None):
         self.__bracket_count += 1
-        ##return 'ob<%04d\n' % self.__bracket_count
         return 'ob<nu<open-brack<%04d\n' % self.__bracket_count
+
     def cb_func(self, pre, token, num=None):
-        ##line = 'cb<%04d\n' % self.__bracket_count
         line = 'cb<nu<clos-brack<%04d\n' % self.__bracket_count
         self.__bracket_count -= 1
         return line
+
     def color_func(self, pre, token, num):
         third_field = 'nu'
         if num[-1] == ';':
@@ -662,6 +680,7 @@ class ProcessTokens:
             num = "0" + num
         return 'cw<%s<%s<%s<%s\n' % (pre, token, third_field, num)
         ##return 'cw<cl<%s<nu<nu<%s>%s<%s\n' % (third_field, token, num, token)
+
     def bool_st_func(self, pre, token, num):
         if num is None or num == '' or num == '1':
             return 'cw<%s<%s<nu<true\n' % (pre, token)
@@ -670,24 +689,23 @@ class ProcessTokens:
             return 'cw<%s<%s<nu<false\n' % (pre, token)
                 ##return 'cw<nu<nu<nu<%s>false<%s\n' % (token, token)
         else:
-            msg = 'boolean should have some value module process tokens\n'
+            msg = "boolean should have some value module process tokens\ntoken is %s\n'%s'\n" % (token, num)
-            msg += 'token is ' + token + "\n"
-            msg += "'" + num + "'" + "\n"
             raise self.__bug_handler, msg
+
     def __no_sup_sub_func(self, pre, token, num):
         the_string = 'cw<ci<subscript_<nu<false\n'
         the_string += 'cw<ci<superscrip<nu<false\n'
         return the_string
+
     def divide_num(self, numerator, denominator):
         try:
-            numerator = float(re.search('[0-9.]+', numerator).group())            
+            #calibre why ignore negative number? Wrong in case of \fi
+            numerator = float(re.search('[0-9.\-]+', numerator).group())
         except TypeError, msg:
             if self.__run_level > 3:
-                msg = 'no number to process?\n'
+                msg = ('No number to process?\nthis indicates that the token \(\\li\) \
-                msg += 'this indicates that the token '
+                should have a number and does not\nnumerator is \
-                msg += ' \(\\li\) should have a number and does not\n'
+                "%s"\ndenominator is "%s"\n') % (numerator, denominator)
-                msg += 'numerator is "%s"\n' % numerator
-                msg += 'denominator is "%s"\n' % denominator
                 raise self.__bug_handler, msg
             if 5 > self.__return_code:
                 self.__return_code = 5
@@ -698,9 +716,10 @@ class ProcessTokens:
         if string_num[-2:] == ".0":
             string_num = string_num[:-2]
         return string_num
+
     def split_let_num(self, token):
         match_obj = re.search(self.__num_exp,token)
-        if match_obj != None:
+        if match_obj is not None:
             first = match_obj.group(1)
             second = match_obj.group(2)
             if not second:
@@ -714,6 +733,7 @@ class ProcessTokens:
                 raise self.__bug_handler
             return token, 0
         return first, second
+
     def convert_to_hex(self,number):
         """Convert a string to uppercase hexidecimal"""
         num = int(number)
@@ -722,6 +742,7 @@ class ProcessTokens:
             return hex_num
         except:
             raise self.__bug_handler
+
     def process_cw(self, token):
         """Change the value of the control word by determining what dictionary
         it belongs to"""
@@ -737,69 +758,41 @@ class ProcessTokens:
         pre, token, action = self.dict_token.get(token, (None, None, None))
         if action:
             return action(pre, token, num)
-    # unused function
+
-    def initiate_token_actions(self):
-        self.action_for_token={
-        '{'     :   self.ob_func,
-        '}'     :   self.cb_func,
-        '\\'    :   self.process_cw,
-        }
-    # unused function
-    def evaluate_token(self,token):
-        """Evaluate tokens. Return a value if the token is not a
-        control word. Otherwise, pass token onto another method
-        for further evaluation."""
-        token, action = self.dict_token.get(token[0:1])
-        if action:
-            line = action(token)
-            return line
-        else :
-            return  'tx<nu<nu<nu<nu<%s\n' % token
     def __check_brackets(self, in_file):
         self.__check_brack_obj = check_brackets.CheckBrackets\
             (file = in_file)
         good_br =  self.__check_brack_obj.check_brackets()[0]
         if not good_br:
             return 1
+
     def process_tokens(self):
         """Main method for handling other methods. """
-        first_token = 0
-        second_token = 0
-        read_obj = open(self.__file, 'r')
-        write_obj = open(self.__write_to, 'w')
-        line_to_read = "dummy"
         line_count = 0
-        while line_to_read:
+        with open(self.__file, 'r') as read_obj:
-            line_to_read = read_obj.readline()
+            with open(self.__write_to, 'wb') as write_obj:
-            token = line_to_read
+                for line in read_obj:
-            token = token.replace("\n","")
+                    token = line.replace("\n","")
-            if not token:
-                continue
                     line_count += 1
+                    if line_count == 1 and token != '\\{':
+                            msg = 'Invalid RTF: document doesn\'t start with {\n'
+                            raise self.__exception_handler, msg
+                    elif line_count == 2 and token[0:4] != '\\rtf':
+                            msg = 'Invalid RTF: document doesn\'t start with \\rtf \n'
+                            raise self.__exception_handler, msg
+
+                    the_index = token.find('\\ ')
+                    if token is not None and  the_index > -1:
+                        msg = 'Invalid RTF: token "\\ " not valid.\n'
+                        raise self.__exception_handler, msg
+                    elif token[:1] == "\\":
                         try:
                             token.decode('us-ascii')
                         except UnicodeError, msg:
-                msg = str(msg)
+                            msg = 'Invalid RTF: Tokens not ascii encoded.\n%s' % str(msg)
-                msg += 'Invalid RTF: File not ascii encoded.\n'
                             raise self.__exception_handler, msg
-            if not first_token:
-                if token != '\\{':
-                    msg = 'Invalid RTF: document doesn\'t start with {\n'
-                    raise self.__exception_handler, msg
-                first_token = 1
-            elif first_token and not second_token:
-                if token[0:4] != '\\rtf':
-                    msg ='Invalid RTF: document doesn\'t start with \\rtf \n'
-                    raise self.__exception_handler, msg
-                second_token = 1
-            ##token = self.evaluate_token(token)
-            the_index = token.find('\\ ')
-            if token != None and  the_index > -1:
-                msg ='Invalid RTF: token "\\ " not valid. \n'
-                raise self.__exception_handler, msg
-            elif token[0:1] == "\\":
                         line = self.process_cw(token)
-                if line != None:
+                        if line is not None:
                             write_obj.write(line)
                     else:
                         fields = re.split(self.__utf_exp, token)
@@ -810,16 +803,17 @@ class ProcessTokens:
                                 write_obj.write('tx<ut<__________<%s\n' % field)
                             else:
                                 write_obj.write('tx<nu<__________<%s\n' % field)
-        read_obj.close()
+
-        write_obj.close()
         if not line_count:
-            msg ='Invalid RTF: file appears to be empty. \n'
+            msg = 'Invalid RTF: file appears to be empty.\n'
             raise self.__exception_handler, msg
+
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "processed_tokens.data")
         copy_obj.rename(self.__write_to, self.__file)
         os.remove(self.__write_to)
+
         bad_brackets = self.__check_brackets(self.__file)
         if bad_brackets:
             msg = 'Invalid RTF: document does not have matching brackets.\n'

src/calibre/ebooks/rtf2xml/replace_illegals.py

@@ -16,7 +16,10 @@
 #                                                                       #
 #########################################################################
 import os, tempfile
+
 from calibre.ebooks.rtf2xml import copy
+from calibre.utils.cleantext import clean_ascii_chars
+
 class ReplaceIllegals:
     """
     reaplace illegal lower ascii characters
@@ -30,21 +33,14 @@ class ReplaceIllegals:
         self.__copy = copy
         self.__run_level = run_level
         self.__write_to = tempfile.mktemp()
+
     def replace_illegals(self):
         """
         """
-        nums = [0, 1, 2, 3, 4, 5, 6, 7, 8,  11,  13, 14, 15, 16, 17, 18, 19]
+        with open(self.__file, 'r') as read_obj:
-        read_obj = open(self.__file, 'r')
+            with open(self.__write_to, 'w') as write_obj:
-        write_obj = open(self.__write_to, 'w')
+                for line in read_obj:
-        line_to_read = 1
+                    write_obj.write(clean_ascii_chars(line))
-        while line_to_read:
-            line_to_read = read_obj.readline()
-            line = line_to_read
-            for num in nums:
-                line = line.replace(chr(num), '')
-            write_obj.write(line)
-        read_obj.close()
-        write_obj.close()
         copy_obj = copy.Copy()
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "replace_illegals.data")

src/calibre/ebooks/rtf2xml/tokenize.py

@@ -16,7 +16,10 @@
 #                                                                       #
 #########################################################################
 import os, re, tempfile
+
 from calibre.ebooks.rtf2xml import copy
+from calibre.utils.mreplace import MReplace
+
 class Tokenize:
     """Tokenize RTF into one line per field. Each line will contain information useful for the rest of the script"""
     def __init__(self,
@@ -28,89 +31,175 @@ class Tokenize:
         self.__file = in_file
         self.__bug_handler = bug_handler
         self.__copy = copy
-        self.__special_tokens = [ '_', '~', "'", '{', '}' ]
         self.__write_to = tempfile.mktemp()
-    def __from_ms_to_utf8(self,match_obj):
+        self.__compile_expressions()
+        #variables
+        self.__uc_char = 0
+        self.__uc_bin = False
+        self.__uc_value = [1]
+
+    def __reini_utf8_counters(self):
+        self.__uc_char = 0
+        self.__uc_bin = False
+
+    def __remove_uc_chars(self, startchar, token):
+        for i in xrange(startchar, len(token)):
+            if token[i] == " ":
+                continue
+            elif self.__uc_char:
+                self.__uc_char -= 1
+            else:
+                return token[i:]
+        #if only " " and char to skip
+        return ''
+
+    def __unicode_process(self, token):
+        #change scope in
+        if token == '\{':
+            self.__uc_value.append(self.__uc_value[-1])
+            #basic error handling
+            self.__reini_utf8_counters()
+            return token
+        #change scope out
+        elif token == '\}':
+            self.__uc_value.pop()
+            self.__reini_utf8_counters()
+            return token
+        #add a uc control
+        elif token[:3] == '\uc':
+            self.__uc_value[-1] = int(token[3:])
+            self.__reini_utf8_counters()
+            return token
+        #bin data to slip
+        elif self.__uc_bin:
+            self.__uc_bin = False
+            return ''
+        #uc char to remove
+        elif self.__uc_char:
+            #handle \bin tag in case of uc char to skip
+            if token[:4] == '\bin':
+                self.__uc_char -=1
+                self.__uc_bin = True
+                return ''
+            elif token[:1] == "\\" :
+                self.__uc_char -=1
+                return ''
+            else:
+                return self.__remove_uc_chars(0, token)
+        #go for real \u token
+        match_obj = self.__utf_exp.match(token)
+        if match_obj is not None:
+            self.__reini_utf8_counters()
+            #get value and handle negative case
             uni_char = int(match_obj.group(1))
+            uni_len = len(match_obj.group(1)) + 2
             if uni_char < 0:
                 uni_char += 65536
-        return   '&#x' + str('%X' % uni_char) + ';'
+            uni_char = unichr(uni_char).encode('ascii', 'xmlcharrefreplace')
-    def __neg_unicode_func(self, match_obj):
+            self.__uc_char = self.__uc_value[-1]
-        neg_uni_char = int(match_obj.group(1)) * -1
+            #there is only an unicode char
-        # sys.stderr.write(str( neg_uni_char))
+            if len(token)<= uni_len:
-        uni_char = neg_uni_char + 65536
+                return uni_char
-        return   '&#x' + str('%X' % uni_char) + ';'
+            #an unicode char and something else
-    def __sub_line_reg(self,line):
+            #must be after as it is splited on \
-        line = line.replace("\\\\", "\\backslash ")
+            #necessary? maybe for \bin?
-        line = line.replace("\\~", "\\~ ")
+            elif not self.__uc_char:
-        line = line.replace("\\;", "\\; ")
+                return uni_char + token[uni_len:]
-        line = line.replace("&", "&amp;")
+            #if not uc0 and chars
-        line = line.replace("<", "&lt;")
-        line = line.replace(">", "&gt;")
-        line = line.replace("\\~", "\\~ ")
-        line = line.replace("\\_", "\\_ ")
-        line = line.replace("\\:", "\\: ")
-        line = line.replace("\\-", "\\- ")
-        # turn into a generic token to eliminate special
-        # cases and make processing easier
-        line = line.replace("\\{", "\\ob ")
-        # turn into a generic token to eliminate special
-        # cases and make processing easier
-        line = line.replace("\\}", "\\cb ")
-        # put a backslash in front of to eliminate special cases and
-        # make processing easier
-        line = line.replace("{", "\\{")
-        # put a backslash in front of to eliminate special cases and
-        # make processing easier
-        line = line.replace("}", "\\}")
-        line = re.sub(self.__utf_exp, self.__from_ms_to_utf8, line)
-        # line = re.sub( self.__neg_utf_exp, self.__neg_unicode_func, line)
-        line = re.sub(self.__ms_hex_exp, "\\mshex0\g<1> ", line)
-        ##line = line.replace("\\backslash", "\\\\")
-        # this is for older RTF
-        line = re.sub(self.__par_exp, '\\par ', line)
-        return line
-    def __compile_expressions(self):
-        self.__ms_hex_exp = re.compile(r"\\\'(..)")
-        self.__utf_exp = re.compile(r"\\u(-?\d{3,6}) {0,1}")
-        self.__splitexp = re.compile(r"(\\[\\{}]|{|}|\\[^\s\\{}&]+(?:\s)?)")
-        self.__par_exp = re.compile(r'\\$')
-        self.__mixed_exp = re.compile(r"(\\[a-zA-Z]+\d+)(\D+)")
-        ##self.num_exp = re.compile(r"(\*|:|[a-zA-Z]+)(.*)")
-    def __create_tokens(self):
-        self.__compile_expressions()
-        read_obj = open(self.__file, 'r')
-        write_obj = open(self.__write_to, 'w')
-        line_to_read = "dummy"
-        while line_to_read:
-            line_to_read = read_obj.readline()
-            line = line_to_read
-            line = line.replace("\n", "")
-            line =  self.__sub_line_reg(line)
-            tokens = re.split(self.__splitexp, line)
-            ##print tokens
-            for token in tokens:
-                if token != "":
-                    write_obj.write(token + "\n")
-                    """
-                    match_obj = re.search(self.__mixed_exp, token)
-                    if match_obj != None:
-                        first = match_obj.group(1)
-                        second = match_obj.group(2)
-                        write_obj.write(first + "\n")
-                        write_obj.write(second + "\n")
             else:
-                        write_obj.write(token + "\n")
+                return uni_char + self.__remove_uc_chars(uni_len, token)
-                    """
+        #default
-        read_obj.close()
+        return token
-        write_obj.close()
+
+    def __sub_reg_split(self,input_file):
+        input_file = self.__replace_spchar.mreplace(input_file)
+        input_file = self.__ms_hex_exp.sub("\\mshex0\g<1> ", input_file)
+        input_file = self.__utf_ud.sub("\\{\\uc0 \g<1>\\}", input_file)
+        #remove \n in bin data
+        input_file = self.__bin_exp.sub(lambda x: \
+                                        x.group().replace('\n', '') + '\n', input_file)
+        #split
+        tokens = re.split(self.__splitexp, input_file)
+        #remove empty tokens and \n
+        return filter(lambda x: len(x) > 0 and x != '\n', tokens)
+        #input_file = re.sub(self.__utf_exp, self.__from_ms_to_utf8, input_file)
+        # line = re.sub( self.__neg_utf_exp, self.__neg_unicode_func, line)
+        # this is for older RTF
+        #line = re.sub(self.__par_exp, '\\par ', line)
+        #return filter(lambda x: len(x) > 0, \
+            #(self.__remove_line.sub('', x) for x in tokens))
+
+    def __compile_expressions(self):
+        SIMPLE_RPL = {
+            "\\\\": "\\backslash ",
+            "\\~": "\\~ ",
+            "\\;": "\\; ",
+            "&": "&amp;",
+            "<": "&lt;",
+            ">": "&gt;",
+            "\\~": "\\~ ",
+            "\\_": "\\_ ",
+            "\\:": "\\: ",
+            "\\-": "\\- ",
+            # turn into a generic token to eliminate special
+            # cases and make processing easier
+            "\\{": "\\ob ",
+            # turn into a generic token to eliminate special
+            # cases and make processing easier
+            "\\}": "\\cb ",
+            # put a backslash in front of to eliminate special cases and
+            # make processing easier
+            "{": "\\{",
+            # put a backslash in front of to eliminate special cases and
+            # make processing easier
+            "}": "\\}",
+            # this is for older RTF
+            r'\\$': '\\par ',
+            }
+        self.__replace_spchar = MReplace(SIMPLE_RPL)
+        #add ;? in case of char following \u
+        self.__ms_hex_exp = re.compile(r"\\\'([0-9a-fA-F]{2})") #r"\\\'(..)"
+        self.__utf_exp = re.compile(r"\\u(-?\d{3,6}) ?")
+        self.__bin_exp = re.compile(r"(?:\\bin(-?\d{0,10})[\n ]+)[01\n]+")
+        #manage upr/ud situations
+        self.__utf_ud = re.compile(r"\\{[\n ]?\\upr[\n ]?(?:\\{.*?\\})[\n ]?" + \
+                       r"\\{[\n ]?\\*[\n ]?\\ud[\n ]?(\\{.*?\\})[\n ]?\\}[\n ]?\\}")
+        #add \n in split for whole file reading
+        #why keep backslash whereas \is replaced before?
+        #remove \n from endline char
+        self.__splitexp = re.compile(r"(\\[{}]|\n|\\[^\s\\{}&]+(?:[ \t\r\f\v])?)")
+        #self.__bin_exp = re.compile(r"\\bin(-?\d{1,8}) {0,1}")
+        #self.__utf_exp = re.compile(r"^\\u(-?\d{3,6})")
+        #self.__splitexp = re.compile(r"(\\[\\{}]|{|}|\n|\\[^\s\\{}&]+(?:\s)?)")
+        #self.__par_exp = re.compile(r'\\$')
+        #self.__remove_line = re.compile(r'\n+')
+        #self.__mixed_exp = re.compile(r"(\\[a-zA-Z]+\d+)(\D+)")
+        ##self.num_exp = re.compile(r"(\*|:|[a-zA-Z]+)(.*)")
+
     def tokenize(self):
-        """Main class for handling other methods. Reads in one line \
+        """Main class for handling other methods. Reads the file \
-        at a time, usues method self.sub_line to make basic substitutions,\
+        , uses method self.sub_reg to make basic substitutions,\
-        uses ? to process tokens"""
+        and process tokens by itself"""
-        self.__create_tokens()
+        #read
+        with open(self.__file, 'r') as read_obj:
+            input_file = read_obj.read()
+        
+        #process simple replacements and split giving us a correct list
+        #remove '' and \n in the process
+        tokens = self.__sub_reg_split(input_file)
+        #correct unicode
+        tokens = map(self.__unicode_process, tokens)
+        #remove empty items created by removing \uc
+        tokens = filter(lambda x: len(x) > 0, tokens)
+        
+        #write
+        with open(self.__write_to, 'wb') as write_obj:
+            write_obj.write('\n'.join(tokens))
+        #Move and copy
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "tokenize.data")
         copy_obj.rename(self.__write_to, self.__file)
         os.remove(self.__write_to)
+        
+        #self.__special_tokens = [ '_', '~', "'", '{', '}' ]

src/calibre/ebooks/textile/__init__.py

@@ -1,3 +1,6 @@
 from functions import textile, textile_restricted, Textile
 
+if False:
+    textile, textile_restricted, Textile
+
 __all__ = ['textile', 'textile_restricted']

src/calibre/ebooks/textile/functions.py

@@ -425,7 +425,7 @@ class Textile(object):
         text = text.split('\n\n')
 
         tag = 'p'
-        atts = cite = graf = ext = ''
+        atts = cite = graf = ext = c1 = ''
 
         out = []
 

src/calibre/ebooks/txt/processor.py

@@ -1,4 +1,8 @@
 # -*- coding: utf-8 -*-
+__license__   = 'GPL v3'
+__copyright__ = '2009, John Schember <john@nachtimwald.com>'
+__docformat__ = 'restructuredtext en'
+
 
 '''
 Read content from txt file.
@@ -7,15 +11,10 @@ Read content from txt file.
 import os, re
 
 from calibre import prepare_string_for_xml, isbytestring
-from calibre.ebooks.markdown import markdown
-from calibre.ebooks.textile import textile
 from calibre.ebooks.metadata.opf2 import OPFCreator
 from calibre.ebooks.txt.heuristicprocessor import TXTHeuristicProcessor
 from calibre.ebooks.conversion.preprocess import DocAnalysis
-
+from calibre.utils.cleantext import clean_ascii_chars
-__license__   = 'GPL v3'
-__copyright__ = '2009, John Schember <john@nachtimwald.com>'
-__docformat__ = 'restructuredtext en'
 
 HTML_TEMPLATE = u'<html><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"/><title>%s</title></head><body>\n%s\n</body></html>'
 
@@ -35,9 +34,7 @@ def clean_txt(txt):
     # Remove excessive line breaks.
     txt = re.sub('\n{3,}', '\n\n', txt)
     #remove ASCII invalid chars : 0 to 8 and 11-14 to 24
-    chars = list(range(8)) + [0x0B, 0x0E, 0x0F] + list(range(0x10, 0x19))
+    txt = clean_ascii_chars(txt)
-    illegal_chars = re.compile(u'|'.join(map(unichr, chars)))
-    txt = illegal_chars.sub('', txt)
 
     return txt
 
@@ -75,6 +72,7 @@ def convert_heuristic(txt, title='', epub_split_size_kb=0):
     return tp.convert(txt, title, epub_split_size_kb)
 
 def convert_markdown(txt, title='', disable_toc=False):
+    from calibre.ebooks.markdown import markdown
     md = markdown.Markdown(
           extensions=['footnotes', 'tables', 'toc'],
           extension_configs={"toc": {"disable_toc": disable_toc}},
@@ -82,6 +80,7 @@ def convert_markdown(txt, title='', disable_toc=False):
     return HTML_TEMPLATE % (title, md.convert(txt))
 
 def convert_textile(txt, title=''):
+    from calibre.ebooks.textile import textile
     html = textile(txt, encoding='utf-8')
     return HTML_TEMPLATE % (title, html)
 

src/calibre/gui2/__init__.py

@@ -269,10 +269,14 @@ def question_dialog(parent, title, msg, det_msg='', show_copy_button=True,
 
     return d.exec_() == yes_button
 
-def info_dialog(parent, title, msg, det_msg='', show=False):
+def info_dialog(parent, title, msg, det_msg='', show=False,
+        show_copy_button=True):
     d = MessageBox(QMessageBox.Information, title, msg, QMessageBox.Ok,
                     parent, det_msg)
     d.setIconPixmap(QPixmap(I('dialog_information.png')))
+    if not show_copy_button:
+        d.cb.setVisible(False)
+
     if show:
         return d.exec_()
     return d

src/calibre/gui2/catalog/catalog_bibtex.py

@@ -27,14 +27,17 @@ class PluginWidget(QWidget, Ui_Form):
     def __init__(self, parent=None):
         QWidget.__init__(self, parent)
         self.setupUi(self)
-        from calibre.library.catalog import FIELDS
-        self.all_fields = []
-        for x in FIELDS :
-            if x != 'all':
-                self.all_fields.append(x)
-                QListWidgetItem(x, self.db_fields)
 
     def initialize(self, name, db): #not working properly to update
+        from calibre.library.catalog import FIELDS
+
+        self.all_fields = [x for x in FIELDS if x != 'all']
+        #add custom columns
+        self.all_fields.extend([x for x in sorted(db.custom_field_keys())])
+        #populate
+        for x in self.all_fields:
+            QListWidgetItem(x, self.db_fields)
+
         self.name = name
         fields = gprefs.get(name+'_db_fields', self.all_fields)
         # Restore the activated db_fields from last use

src/calibre/gui2/dialogs/drm_error.py

@@ -0,0 +1,21 @@
+#!/usr/bin/env python
+# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
+
+__license__   = 'GPL v3'
+__copyright__ = '2011, Kovid Goyal <kovid@kovidgoyal.net>'
+__docformat__ = 'restructuredtext en'
+
+
+from PyQt4.Qt import QDialog
+from calibre.gui2.dialogs.drm_error_ui import Ui_Dialog
+
+class DRMErrorMessage(QDialog, Ui_Dialog):
+
+    def __init__(self, parent=None, title=None):
+        QDialog.__init__(self, parent)
+        self.setupUi(self)
+        if title is not None:
+            t = unicode(self.msg.text())
+            self.msg.setText('<h2>%s</h2>%s'%(title, t))
+        self.resize(self.sizeHint())
+

src/calibre/gui2/dialogs/drm_error.ui

@@ -0,0 +1,102 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<ui version="4.0">
+ <class>Dialog</class>
+ <widget class="QDialog" name="Dialog">
+  <property name="geometry">
+   <rect>
+    <x>0</x>
+    <y>0</y>
+    <width>417</width>
+    <height>235</height>
+   </rect>
+  </property>
+  <property name="windowTitle">
+   <string>This book is DRMed</string>
+  </property>
+  <layout class="QGridLayout" name="gridLayout">
+   <item row="0" column="0">
+    <widget class="QLabel" name="label">
+     <property name="sizePolicy">
+      <sizepolicy hsizetype="Preferred" vsizetype="Preferred">
+       <horstretch>0</horstretch>
+       <verstretch>0</verstretch>
+      </sizepolicy>
+     </property>
+     <property name="maximumSize">
+      <size>
+       <width>132</width>
+       <height>16777215</height>
+      </size>
+     </property>
+     <property name="text">
+      <string/>
+     </property>
+     <property name="pixmap">
+      <pixmap resource="../../../../resources/images.qrc">:/images/document-encrypt.png</pixmap>
+     </property>
+    </widget>
+   </item>
+   <item row="0" column="1">
+    <widget class="QLabel" name="msg">
+     <property name="text">
+      <string>&lt;p&gt;This book is locked by &lt;b&gt;DRM&lt;/b&gt;. To learn more about DRM and why you cannot read or convert this book in calibre, 
+&lt;a href=&quot;http://bugs.calibre-ebook.com/wiki/DRM&quot;&gt;click here&lt;/a&gt;.</string>
+     </property>
+     <property name="wordWrap">
+      <bool>true</bool>
+     </property>
+     <property name="openExternalLinks">
+      <bool>true</bool>
+     </property>
+    </widget>
+   </item>
+   <item row="1" column="0" colspan="2">
+    <widget class="QDialogButtonBox" name="buttonBox">
+     <property name="orientation">
+      <enum>Qt::Horizontal</enum>
+     </property>
+     <property name="standardButtons">
+      <set>QDialogButtonBox::Close</set>
+     </property>
+    </widget>
+   </item>
+  </layout>
+ </widget>
+ <resources>
+  <include location="../../../../resources/images.qrc"/>
+ </resources>
+ <connections>
+  <connection>
+   <sender>buttonBox</sender>
+   <signal>accepted()</signal>
+   <receiver>Dialog</receiver>
+   <slot>accept()</slot>
+   <hints>
+    <hint type="sourcelabel">
+     <x>248</x>
+     <y>254</y>
+    </hint>
+    <hint type="destinationlabel">
+     <x>157</x>
+     <y>274</y>
+    </hint>
+   </hints>
+  </connection>
+  <connection>
+   <sender>buttonBox</sender>
+   <signal>rejected()</signal>
+   <receiver>Dialog</receiver>
+   <slot>reject()</slot>
+   <hints>
+    <hint type="sourcelabel">
+     <x>316</x>
+     <y>260</y>
+    </hint>
+    <hint type="destinationlabel">
+     <x>286</x>
+     <y>274</y>
+    </hint>
+   </hints>
+  </connection>
+ </connections>
+</ui>

src/calibre/gui2/dialogs/metadata_bulk.py

@@ -15,7 +15,7 @@ from calibre.ebooks.metadata import string_to_authors, authors_to_string
 from calibre.ebooks.metadata.book.base import composite_formatter
 from calibre.ebooks.metadata.meta import get_metadata
 from calibre.gui2.custom_column_widgets import populate_metadata_page
-from calibre.gui2 import error_dialog
+from calibre.gui2 import error_dialog, ResizableDialog
 from calibre.gui2.progress_indicator import ProgressIndicator
 from calibre.utils.config import dynamic
 from calibre.utils.titlecase import titlecase
@@ -49,7 +49,7 @@ def get_cover_data(path):
 
 
 
-class MyBlockingBusy(QDialog):
+class MyBlockingBusy(QDialog): # {{{
 
     do_one_signal = pyqtSignal()
 
@@ -241,8 +241,9 @@ class MyBlockingBusy(QDialog):
         self.current_index += 1
         self.do_one_signal.emit()
 
+    # }}}
 
-class MetadataBulkDialog(QDialog, Ui_MetadataBulkDialog):
+class MetadataBulkDialog(ResizableDialog, Ui_MetadataBulkDialog):
 
     s_r_functions = {       ''              : lambda x: x,
                             _('Lower Case') : lambda x: icu_lower(x),
@@ -261,9 +262,8 @@ class MetadataBulkDialog(QDialog, Ui_MetadataBulkDialog):
                         ]
 
     def __init__(self, window, rows, model, tab):
-        QDialog.__init__(self, window)
+        ResizableDialog.__init__(self, window)
         Ui_MetadataBulkDialog.__init__(self)
-        self.setupUi(self)
         self.model = model
         self.db = model.db
         self.ids = [self.db.id(r) for r in rows]

src/calibre/gui2/dialogs/metadata_bulk.ui

@@ -6,8 +6,8 @@
    <rect>
     <x>0</x>
     <y>0</y>
-    <width>752</width>
+    <width>850</width>
-    <height>633</height>
+    <height>650</height>
    </rect>
   </property>
   <property name="windowTitle">
@@ -17,8 +17,8 @@
    <iconset resource="../../../../resources/images.qrc">
     <normaloff>:/images/edit_input.png</normaloff>:/images/edit_input.png</iconset>
   </property>
-  <layout class="QVBoxLayout" name="verticalLayout">
+  <layout class="QGridLayout" name="gridLayout_2">
-   <item>
+   <item row="0" column="0">
     <widget class="QLabel" name="box_title">
      <property name="text">
       <string/>
@@ -28,11 +28,27 @@
      </property>
     </widget>
    </item>
-   <item>
+   <item row="1" column="0">
-    <layout class="QVBoxLayout">
+    <widget class="QScrollArea" name="scrollArea">
-     <property name="spacing">
+     <property name="frameShape">
-      <number>6</number>
+      <enum>QFrame::NoFrame</enum>
      </property>
+     <property name="lineWidth">
+      <number>0</number>
+     </property>
+     <property name="widgetResizable">
+      <bool>true</bool>
+     </property>
+     <widget class="QWidget" name="scrollAreaWidgetContents">
+      <property name="geometry">
+       <rect>
+        <x>0</x>
+        <y>0</y>
+        <width>842</width>
+        <height>589</height>
+       </rect>
+      </property>
+      <layout class="QVBoxLayout" name="verticalLayout_2">
        <property name="margin">
         <number>0</number>
        </property>
@@ -784,8 +800,8 @@ not multiple and the destination field is multiple</string>
                <rect>
                 <x>0</x>
                 <y>0</y>
-              <width>726</width>
+                <width>197</width>
-              <height>334</height>
+                <height>60</height>
                </rect>
               </property>
               <layout class="QGridLayout" name="testgrid">
@@ -838,8 +854,10 @@ not multiple and the destination field is multiple</string>
         </widget>
        </item>
       </layout>
+     </widget>
+    </widget>
    </item>
-   <item>
+   <item row="2" column="0">
     <widget class="QDialogButtonBox" name="button_box">
      <property name="orientation">
       <enum>Qt::Horizontal</enum>
@@ -893,7 +911,6 @@ not multiple and the destination field is multiple</string>
   <tabstop>swap_title_and_author</tabstop>
   <tabstop>change_title_to_title_case</tabstop>
   <tabstop>button_box</tabstop>
-  <tabstop>central_widget</tabstop>
   <tabstop>search_field</tabstop>
   <tabstop>search_mode</tabstop>
   <tabstop>s_r_template</tabstop>

src/calibre/gui2/dialogs/metadata_single.py

@@ -823,7 +823,7 @@ class MetadataSingleDialog(ResizableDialog, Ui_MetadataSingleDialog):
                                 if book.series_index is not None:
                                     self.series_index.setValue(book.series_index)
                         if book.has_cover:
-                            if d.opt_auto_download_cover.isChecked() and book.has_cover:
+                            if d.opt_auto_download_cover.isChecked():
                                 self.fetch_cover()
                             else:
                                 self.fetch_cover_button.setFocus(Qt.OtherFocusReason)

src/calibre/gui2/dialogs/user_profiles.py

@@ -4,7 +4,7 @@ __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
 import time, os
 
 from PyQt4.Qt import SIGNAL, QUrl, QAbstractListModel, Qt, \
-        QVariant, QInputDialog
+        QVariant
 
 from calibre.web.feeds.recipes import compile_recipe
 from calibre.web.feeds.news import AutomaticNewsRecipe
@@ -256,15 +256,52 @@ class %(classname)s(%(base_class)s):
 
     def add_builtin_recipe(self):
         from calibre.web.feeds.recipes.collection import \
-            get_builtin_recipe_by_title, get_builtin_recipe_titles
+            get_builtin_recipe_collection, get_builtin_recipe_by_id
-        items = sorted(get_builtin_recipe_titles(), key=sort_key)
+        from PyQt4.Qt import QDialog, QVBoxLayout, QListWidgetItem, \
+                QListWidget, QDialogButtonBox, QSize
 
+        d = QDialog(self)
+        d.l = QVBoxLayout()
+        d.setLayout(d.l)
+        d.list = QListWidget(d)
+        d.list.doubleClicked.connect(lambda x: d.accept())
+        d.l.addWidget(d.list)
+        d.bb = QDialogButtonBox(QDialogButtonBox.Ok|QDialogButtonBox.Cancel,
+                Qt.Horizontal, d)
+        d.bb.accepted.connect(d.accept)
+        d.bb.rejected.connect(d.reject)
+        d.l.addWidget(d.bb)
+        d.setWindowTitle(_('Choose builtin recipe'))
+        items = []
+        for r in get_builtin_recipe_collection():
+            id_ = r.get('id', '')
+            title = r.get('title', '')
+            lang = r.get('language', '')
+            if id_ and title:
+                items.append((title + ' [%s]'%lang, id_))
+
+        items.sort(key=lambda x:sort_key(x[0]))
+        for title, id_ in items:
+            item = QListWidgetItem(title)
+            item.setData(Qt.UserRole, id_)
+            d.list.addItem(item)
+
+        d.resize(QSize(450, 400))
+        ret = d.exec_()
+        d.list.doubleClicked.disconnect()
+        if ret != d.Accepted:
+            return
+
+        items = list(d.list.selectedItems())
+        if not items:
+            return
+        item = items[-1]
+        id_ = unicode(item.data(Qt.UserRole).toString())
+        title = unicode(item.data(Qt.DisplayRole).toString()).rpartition(' [')[0]
+        profile = get_builtin_recipe_by_id(id_)
+        if profile is None:
+            raise Exception('Something weird happened')
 
-        title, ok = QInputDialog.getItem(self, _('Pick recipe'), _('Pick the recipe to customize'),
-                                     items, 0, False)
-        if ok:
-            title = unicode(title)
-            profile = get_builtin_recipe_by_title(title)
         if self._model.has_title(title):
             if question_dialog(self, _('Replace recipe?'),
                 _('A custom recipe named %s already exists. Do you want to '

src/calibre/gui2/layout.py

@@ -8,9 +8,9 @@ __docformat__ = 'restructuredtext en'
 from functools import partial
 
 from PyQt4.Qt import QIcon, Qt, QWidget, QToolBar, QSize, \
-    pyqtSignal, QToolButton, QPushButton, \
+    pyqtSignal, QToolButton, QMenu, QCheckBox, \
-    QObject, QVBoxLayout, QSizePolicy, QLabel, QHBoxLayout, QActionGroup, \
+    QObject, QVBoxLayout, QSizePolicy, QLabel, QHBoxLayout, QActionGroup
-    QMenu
+
 
 from calibre.constants import __appname__
 from calibre.gui2.search_box import SearchBox2, SavedSearchBox
@@ -178,7 +178,9 @@ class SearchBar(QWidget): # {{{
         x.setToolTip(_("<p>Search the list of books by title, author, publisher, tags, comments, etc.<br><br>Words separated by spaces are ANDed"))
         l.addWidget(x)
 
-        self.search_button = QPushButton(_('&Go!'))
+        self.search_button = QToolButton()
+        self.search_button.setToolButtonStyle(Qt.ToolButtonTextOnly)
+        self.search_button.setText(_('&Go!'))
         l.addWidget(self.search_button)
         self.search_button.setSizePolicy(QSizePolicy.Minimum,
                 QSizePolicy.Minimum)
@@ -192,6 +194,12 @@ class SearchBar(QWidget): # {{{
         l.addWidget(x)
         x.setToolTip(_("Reset Quick Search"))
 
+        x = parent.search_highlight_only = QCheckBox()
+        x.setText(_('&Highlight'))
+        x.setToolTip(_('Highlight matched books in the book list, instead '
+            'of restricting the book list to the matches.'))
+        l.addWidget(x)
+
         x = parent.saved_search = SavedSearchBox(self)
         x.setMaximumSize(QSize(150, 16777215))
         x.setMinimumContentsLength(15)

src/calibre/gui2/library/models.py

@@ -10,7 +10,7 @@ from contextlib import closing
 from operator import attrgetter
 
 from PyQt4.Qt import QAbstractTableModel, Qt, pyqtSignal, QIcon, QImage, \
-        QModelIndex, QVariant, QDate
+        QModelIndex, QVariant, QDate, QColor
 
 from calibre.gui2 import NONE, config, UNDEFINED_QDATE
 from calibre.utils.pyparsing import ParseException
@@ -93,6 +93,9 @@ class BooksModel(QAbstractTableModel): # {{{
         self.bool_no_icon = QIcon(I('list_remove.png'))
         self.bool_blank_icon = QIcon(I('blank.png'))
         self.device_connected = False
+        self.rows_matching = set()
+        self.lowest_row_matching = None
+        self.highlight_only = False
         self.read_config()
 
     def change_alignment(self, colname, alignment):
@@ -229,8 +232,26 @@ class BooksModel(QAbstractTableModel): # {{{
             self.endInsertRows()
             self.count_changed()
 
+    def set_highlight_only(self, toWhat):
+        self.highlight_only = toWhat
+        if self.last_search:
+            self.research()
+
     def search(self, text, reset=True):
         try:
+            if self.highlight_only:
+                self.db.search('')
+                if not text:
+                    self.rows_matching = set()
+                    self.lowest_row_matching = None
+                else:
+                    self.rows_matching = self.db.search(text, return_matches=True)
+                    if self.rows_matching:
+                        self.lowest_row_matching = self.db.row(self.rows_matching[0])
+                    self.rows_matching = set(self.rows_matching)
+            else:
+                self.rows_matching = set()
+                self.lowest_row_matching = None
                 self.db.search(text)
         except ParseException as e:
             self.searched.emit(e.msg)
@@ -337,8 +358,9 @@ class BooksModel(QAbstractTableModel): # {{{
             name, val = mi.format_field(key)
             if mi.metadata_for_field(key)['datatype'] == 'comments':
                 name += ':html'
-            if val:
+            if val and name not in data:
                 data[name] = val
+
         return data
 
 
@@ -651,6 +673,9 @@ class BooksModel(QAbstractTableModel): # {{{
             return NONE
         if role in (Qt.DisplayRole, Qt.EditRole):
             return self.column_to_dc_map[col](index.row())
+        elif role == Qt.BackgroundColorRole:
+            if self.id(index) in self.rows_matching:
+                return QColor('lightgreen')
         elif role == Qt.DecorationRole:
             if self.column_to_dc_decorator_map[col] is not None:
                 return self.column_to_dc_decorator_map[index.column()](index.row())

src/calibre/gui2/library/views.py

@@ -680,8 +680,14 @@ class BooksView(QTableView): # {{{
     def set_editable(self, editable, supports_backloading):
         self._model.set_editable(editable)
 
+    def search_proxy(self, txt):
+        self._model.search(txt)
+        if self._model.lowest_row_matching is not None:
+            self.select_rows([self._model.lowest_row_matching], using_ids=False)
+        self.setFocus(Qt.OtherFocusReason)
+
     def connect_to_search_box(self, sb, search_done):
-        sb.search.connect(self._model.search)
+        sb.search.connect(self.search_proxy)
         self._search_done = search_done
         self._model.searched.connect(self.search_done)
 

src/calibre/gui2/preferences/plugins.py

@@ -15,7 +15,8 @@ from calibre.gui2.preferences.plugins_ui import Ui_Form
 from calibre.customize.ui import initialized_plugins, is_disabled, enable_plugin, \
                                  disable_plugin, plugin_customization, add_plugin, \
                                  remove_plugin
-from calibre.gui2 import NONE, error_dialog, info_dialog, choose_files
+from calibre.gui2 import NONE, error_dialog, info_dialog, choose_files, \
+        question_dialog
 
 class PluginModel(QAbstractItemModel): # {{{
 
@@ -76,6 +77,16 @@ class PluginModel(QAbstractItemModel): # {{{
                     return self.index(j, 0, parent)
         return QModelIndex()
 
+    def plugin_to_index_by_properties(self, plugin):
+        for i, category in enumerate(self.categories):
+            parent = self.index(i, 0, QModelIndex())
+            for j, p in enumerate(self._data[category]):
+                if plugin.name == p.name and plugin.type == p.type and \
+                        plugin.author == p.author and plugin.version == p.version:
+                    return self.index(j, 0, parent)
+        return QModelIndex()
+
+
     def refresh_plugin(self, plugin, rescan=False):
         if rescan:
             self.populate()
@@ -132,7 +143,6 @@ class ConfigWidget(ConfigWidgetBase, Ui_Form):
         self.toggle_plugin_button.clicked.connect(self.toggle_plugin)
         self.customize_plugin_button.clicked.connect(self.customize_plugin)
         self.remove_plugin_button.clicked.connect(self.remove_plugin)
-        self.button_plugin_browse.clicked.connect(self.find_plugin)
         self.button_plugin_add.clicked.connect(self.add_plugin)
 
     def toggle_plugin(self, *args):
@@ -149,23 +159,39 @@ class ConfigWidget(ConfigWidgetBase, Ui_Form):
         self.modify_plugin(op='remove')
 
     def add_plugin(self):
-        path = unicode(self.plugin_path.text())
+        path = choose_files(self, 'add a plugin dialog', _('Add plugin'),
+                filters=[(_('Plugins'), ['zip'])], all_files=False,
+                    select_only_single_file=True)
+        if not path:
+            return
+        path = path[0]
         if path and  os.access(path, os.R_OK) and path.lower().endswith('.zip'):
-            add_plugin(path)
+            if not question_dialog(self, _('Are you sure?'), '<p>' + \
+                    _('Installing plugins is a <b>security risk</b>. '
+                    'Plugins can contain a virus/malware. '
+                        'Only install it if you got it from a trusted source.'
+                        ' Are you sure you want to proceed?'),
+                    show_copy_button=False):
+                return
+            plugin = add_plugin(path)
             self._plugin_model.populate()
             self._plugin_model.reset()
             self.changed_signal.emit()
-            self.plugin_path.setText('')
+            info_dialog(self, _('Success'),
+                    _('Plugin <b>{0}</b> successfully installed under <b>'
+                        ' {1} plugins</b>. You may have to restart calibre '
+                        'for the plugin to take effect.').format(plugin.name, plugin.type),
+                    show=True, show_copy_button=False)
+            idx = self._plugin_model.plugin_to_index_by_properties(plugin)
+            if idx.isValid():
+                self.plugin_view.scrollTo(idx,
+                        self.plugin_view.PositionAtCenter)
+                self.plugin_view.scrollTo(idx,
+                        self.plugin_view.PositionAtCenter)
         else:
             error_dialog(self, _('No valid plugin path'),
                          _('%s is not a valid plugin path')%path).exec_()
 
-    def find_plugin(self):
-        path = choose_files(self, 'choose plugin dialog', _('Choose plugin'),
-                            filters=[('Plugins', ['zip'])], all_files=False,
-                            select_only_single_file=True)
-        if path:
-            self.plugin_path.setText(path[0])
 
     def modify_plugin(self, op=''):
         index = self.plugin_view.currentIndex()
@@ -191,10 +217,13 @@ class ConfigWidget(ConfigWidgetBase, Ui_Form):
                 if plugin.do_user_config():
                     self._plugin_model.refresh_plugin(plugin)
             elif op == 'remove':
+                msg = _('Plugin {0} successfully removed').format(plugin.name)
                 if remove_plugin(plugin):
                     self._plugin_model.populate()
                     self._plugin_model.reset()
                     self.changed_signal.emit()
+                    info_dialog(self, _('Success'), msg, show=True,
+                            show_copy_button=False)
                 else:
                     error_dialog(self, _('Cannot remove builtin plugin'),
                          plugin.name + _(' cannot be removed. It is a '

src/calibre/gui2/preferences/plugins.ui

@@ -72,67 +72,17 @@
     </layout>
    </item>
    <item>
-    <widget class="QGroupBox" name="groupBox_4">
+    <widget class="QPushButton" name="button_plugin_add">
-     <property name="title">
-      <string>Add new plugin</string>
-     </property>
-     <layout class="QVBoxLayout" name="verticalLayout_5">
-      <item>
-       <layout class="QHBoxLayout" name="horizontalLayout_5">
-        <item>
-         <widget class="QLabel" name="label_14">
      <property name="text">
-           <string>Plugin &amp;file:</string>
+      <string>&amp;Add a new plugin</string>
-          </property>
-          <property name="buddy">
-           <cstring>plugin_path</cstring>
-          </property>
-         </widget>
-        </item>
-        <item>
-         <widget class="QLineEdit" name="plugin_path"/>
-        </item>
-        <item>
-         <widget class="QToolButton" name="button_plugin_browse">
-          <property name="text">
-           <string>...</string>
      </property>
      <property name="icon">
       <iconset resource="../../../../resources/images.qrc">
-            <normaloff>:/images/document_open.png</normaloff>:/images/document_open.png</iconset>
+       <normaloff>:/images/plugins.png</normaloff>:/images/plugins.png</iconset>
      </property>
     </widget>
    </item>
   </layout>
-      </item>
-      <item>
-       <layout class="QHBoxLayout" name="horizontalLayout_4">
-        <item>
-         <spacer name="horizontalSpacer_2">
-          <property name="orientation">
-           <enum>Qt::Horizontal</enum>
-          </property>
-          <property name="sizeHint" stdset="0">
-           <size>
-            <width>40</width>
-            <height>20</height>
-           </size>
-          </property>
-         </spacer>
-        </item>
-        <item>
-         <widget class="QPushButton" name="button_plugin_add">
-          <property name="text">
-           <string>&amp;Add</string>
-          </property>
-         </widget>
-        </item>
-       </layout>
-      </item>
-     </layout>
-    </widget>
-   </item>
-  </layout>
  </widget>
  <resources>
   <include location="../../../../resources/images.qrc"/>

src/calibre/gui2/preferences/toolbar.py

@@ -37,7 +37,10 @@ class BaseModel(QAbstractListModel):
                     dont_remove_from=set(['toolbar-device']))
         if name is None:
             return FakeAction('--- '+_('Separator')+' ---', None)
+        try:
             return gui.iactions[name]
+        except:
+            return None
 
     def rowCount(self, parent):
         return len(self._data)
@@ -125,6 +128,7 @@ class CurrentModel(BaseModel):
         self.gprefs_name = 'action-layout-'+key
         current = gprefs[self.gprefs_name]
         self._data = [self.name_to_action(x, gui) for x in current]
+        self._data = [x for x in self._data if x is not None]
         self.key = key
         self.gui = gui
 

src/calibre/gui2/search_box.py

@@ -16,6 +16,7 @@ from calibre.gui2 import config
 from calibre.gui2.dialogs.confirm_delete import confirm
 from calibre.gui2.dialogs.saved_search_editor import SavedSearchEditor
 from calibre.gui2.dialogs.search import SearchDialog
+from calibre.utils.config import dynamic
 from calibre.utils.search_query_parser import saved_searches
 from calibre.utils.icu import sort_key
 
@@ -375,6 +376,9 @@ class SearchBoxMixin(object): # {{{
             unicode(self.search.toolTip())))
         self.advanced_search_button.setStatusTip(self.advanced_search_button.toolTip())
         self.clear_button.setStatusTip(self.clear_button.toolTip())
+        self.search_highlight_only.stateChanged.connect(self.highlight_only_changed)
+        self.search_highlight_only.setChecked(
+                            dynamic.get('search_highlight_only', False))
 
     def focus_search_box(self, *args):
         self.search.setFocus(Qt.OtherFocusReason)
@@ -401,6 +405,11 @@ class SearchBoxMixin(object): # {{{
     def focus_to_library(self):
         self.current_view().setFocus(Qt.OtherFocusReason)
 
+    def highlight_only_changed(self, toWhat):
+        dynamic.set('search_highlight_only', toWhat)
+        self.current_view().model().set_highlight_only(toWhat)
+        self.focus_to_library()
+
     # }}}
 
 class SavedSearchBoxMixin(object): # {{{

src/calibre/gui2/shortcuts.py

@@ -150,7 +150,7 @@ class Delegate(QStyledItemDelegate):
         custom = []
         if editor.custom.isChecked():
             for x in ('1', '2'):
-                sc = getattr(editor, 'shortcut'+x)
+                sc = getattr(editor, 'shortcut'+x, None)
                 if sc is not None:
                     custom.append(sc)
 
@@ -266,6 +266,11 @@ class ShortcutConfig(QWidget):
         self.view.scrollTo(index)
 
 
+    @property
+    def is_editing(self):
+        return self.view.state() == self.view.EditingState
+
+
 if __name__ == '__main__':
     from calibre.gui2 import is_ok_to_use_qt
     from calibre.gui2.viewer.keys import SHORTCUTS

src/calibre/gui2/ui.py

@@ -19,7 +19,7 @@ from PyQt4.Qt import Qt, SIGNAL, QTimer, \
                      QMessageBox, QHelpEvent
 
 from calibre import  prints
-from calibre.constants import __appname__, isosx, DEBUG
+from calibre.constants import __appname__, isosx
 from calibre.ptempfile import PersistentTemporaryFile
 from calibre.utils.config import prefs, dynamic
 from calibre.utils.ipc.server import Server
@@ -103,7 +103,15 @@ class Main(MainWindow, MainWindowMixin, DeviceMixin, EmailMixin, # {{{
         self.gui_debug = gui_debug
         acmap = OrderedDict()
         for action in interface_actions():
+            try:
                 ac = action.load_actual_plugin(self)
+            except:
+                # Ignore errors in loading user supplied plugins
+                import traceback
+                traceback.print_exc()
+                if ac.plugin_path is None:
+                    raise
+
             ac.plugin_path = action.plugin_path
             ac.interface_action_base_plugin = action
             if ac.name in acmap:
@@ -460,12 +468,8 @@ class Main(MainWindow, MainWindowMixin, DeviceMixin, EmailMixin, # {{{
         try:
             if 'calibre.ebooks.DRMError' in job.details:
                 if not minz:
-                    d = error_dialog(self, _('Conversion Error'),
+                    from calibre.gui2.dialogs.drm_error import DRMErrorMessage
-                        _('<p>Could not convert: %s<p>It is a '
+                    d = DRMErrorMessage(self, job.description.split(':')[-1])
-                        '<a href="%s">DRM</a>ed book. You must first remove the '
-                        'DRM using third party tools.')%\
-                            (job.description.split(':')[-1],
-                                'http://bugs.calibre-ebook.com/wiki/DRM'))
                     d.setModal(False)
                     d.show()
                     self._modeless_dialogs.append(d)
@@ -582,9 +586,6 @@ class Main(MainWindow, MainWindowMixin, DeviceMixin, EmailMixin, # {{{
             # Goes here, because if cf is valid, db is valid.
             db.prefs['field_metadata'] = db.field_metadata.all_metadata()
             db.commit_dirty_cache()
-            if DEBUG and db.gm_count > 0:
-                print 'get_metadata cache: {0:d} calls, {1:4.2f}% misses'.format(
-                        db.gm_count, (db.gm_missed*100.0)/db.gm_count)
         for action in self.iactions.values():
             if not action.shutting_down():
                 return

src/calibre/gui2/viewer/documentview.py

@@ -120,6 +120,13 @@ class ConfigDialog(QDialog, Ui_Dialog):
 
 
     def accept(self, *args):
+        if self.shortcut_config.is_editing:
+            from calibre.gui2 import info_dialog
+            info_dialog(self, _('Still editing'),
+                    _('You are in the middle of editing a keyboard shortcut'
+                        ' first complete that, by clicking outside the '
+                        ' shortcut editing box.'), show=True)
+            return
         c = config()
         c.set('serif_family', unicode(self.serif_family.currentFont().family()))
         c.set('sans_family', unicode(self.sans_family.currentFont().family()))
@@ -279,7 +286,7 @@ class Document(QWebPage): # {{{
 
     @pyqtSignature("")
     def init_hyphenate(self):
-        if self.hyphenate:
+        if self.hyphenate and getattr(self, 'loaded_lang', ''):
             self.javascript('do_hyphenation("%s")'%self.loaded_lang)
 
     def after_load(self):

src/calibre/gui2/viewer/main.py

@@ -26,6 +26,7 @@ from calibre.gui2.search_box import SearchBox2
 from calibre.ebooks.metadata import MetaInformation
 from calibre.customize.ui import available_input_formats
 from calibre.gui2.viewer.dictionary import Lookup
+from calibre import as_unicode
 
 class TOCItem(QStandardItem):
 
@@ -626,13 +627,12 @@ class EbookViewer(MainWindow, Ui_EbookViewer):
             QApplication.processEvents()
         if worker.exception is not None:
             if isinstance(worker.exception, DRMError):
-                error_dialog(self, _('DRM Error'),
+                from calibre.gui2.dialogs.drm_error import DRMErrorMessage
-                        _('<p>This book is protected by <a href="%s">DRM</a>')
+                DRMErrorMessage(self).exec_()
-                        %'http://wiki.mobileread.com/wiki/DRM').exec_()
             else:
                 r = getattr(worker.exception, 'reason', worker.exception)
                 error_dialog(self, _('Could not open ebook'),
-                        unicode(r), det_msg=worker.traceback, show=True)
+                        as_unicode(r), det_msg=worker.traceback, show=True)
             self.close_progress_indicator()
         else:
             self.metadata.show_opf(self.iterator.opf, os.path.splitext(pathtoebook)[1][1:])

src/calibre/library/caches.py

@@ -411,7 +411,8 @@ class ResultCache(SearchQueryParser): # {{{
             if isinstance(location, list):
                 if allow_recursion:
                     for loc in location:
-                        matches |= self.get_matches(loc, query, allow_recursion=False)
+                        matches |= self.get_matches(loc, query, candidates,
+                                                    allow_recursion=False)
                     return matches
                 raise ParseException(query, len(query), 'Recursive query group detected', self)
 
@@ -419,11 +420,11 @@ class ResultCache(SearchQueryParser): # {{{
                 fm = self.field_metadata[location]
                 # take care of dates special case
                 if fm['datatype'] == 'datetime':
-                    return self.get_dates_matches(location, query.lower())
+                    return self.get_dates_matches(location, query.lower(), candidates)
 
                 # take care of numbers special case
                 if fm['datatype'] in ('rating', 'int', 'float'):
-                    return self.get_numeric_matches(location, query.lower())
+                    return self.get_numeric_matches(location, query.lower(), candidates)
 
                 # take care of the 'count' operator for is_multiples
                 if fm['is_multiple'] and \
@@ -431,7 +432,8 @@ class ResultCache(SearchQueryParser): # {{{
                         query[1:1] in '=<>!':
                     vf = lambda item, loc=fm['rec_index'], ms=fm['is_multiple']:\
                             len(item[loc].split(ms)) if item[loc] is not None else 0
-                    return self.get_numeric_matches(location, query[1:], val_func=vf)
+                    return self.get_numeric_matches(location, query[1:],
+                                                    candidates, val_func=vf)
 
             # everything else, or 'all' matches
             matchkind = CONTAINS_MATCH
@@ -598,7 +600,6 @@ class ResultCache(SearchQueryParser): # {{{
 
     def set(self, row, col, val, row_is_id=False):
         id = row if row_is_id else self._map_filtered[row]
-        self._data[id][self.FIELD_MAP['all_metadata']] = None
         self._data[id][col] = val
 
     def get(self, row, col, row_is_id=False):
@@ -629,7 +630,6 @@ class ResultCache(SearchQueryParser): # {{{
                 self._data[id] = CacheRow(db, self.composites,
                         db.conn.get('SELECT * from meta2 WHERE id=?', (id,))[0])
                 self._data[id].append(db.book_on_device_string(id))
-                self._data[id].append(None)
             except IndexError:
                 return None
         try:
@@ -646,7 +646,6 @@ class ResultCache(SearchQueryParser): # {{{
             self._data[id] = CacheRow(db, self.composites,
                         db.conn.get('SELECT * from meta2 WHERE id=?', (id,))[0])
             self._data[id].append(db.book_on_device_string(id))
-            self._data[id].append(None)
         self._map[0:0] = ids
         self._map_filtered[0:0] = ids
 
@@ -671,7 +670,6 @@ class ResultCache(SearchQueryParser): # {{{
         for item in self._data:
             if item is not None:
                 item.append(db.book_on_device_string(item[0]))
-                item.append(None)
         self._map = [i[0] for i in self._data if i is not None]
         if field is not None:
             self.sort(field, ascending)

src/calibre/library/catalog.py

@@ -1524,17 +1524,30 @@ class EPUB_MOBI(CatalogPlugin):
                     this_title['formats'] = formats
 
                 # Add user notes to be displayed in header
-                # Special case handling for datetime fields
+                # Special case handling for datetime fields and lists
                 if self.opts.header_note_source_field:
                     field_md = self.__db.metadata_for_field(self.opts.header_note_source_field)
                     notes = self.__db.get_field(record['id'],
                                         self.opts.header_note_source_field,
                                         index_is_id=True)
-                    if notes and field_md['datatype'] == 'datetime':
-                        # Reformat date fields to match UI presentation: dd MMM YYYY
-                        notes = format_date(notes,'dd MMM yyyy')
-
                     if notes:
+                        if field_md['datatype'] == 'text':
+                            if isinstance(notes,list):
+                                notes = ' · '.join(notes)
+                        elif field_md['datatype'] == 'datetime':
+                            notes = format_date(notes,'dd MMM yyyy')
+                        elif field_md['datatype'] == 'composite':
+                            m = re.match(r'\[(.+)\]$', notes)
+                            if m is not None:
+                                # Sniff for special pseudo-list string "[<item, item>]"
+                                bracketed_content = m.group(1)
+                                if ',' in bracketed_content:
+                                    # Recast the comma-separated items as a list
+                                    items = bracketed_content.split(',')
+                                    items = [i.strip() for i in items]
+                                    notes = ' &middot; '.join(items)
+                                else:
+                                    notes = bracketed_content
                         this_title['notes'] = {'source':field_md['name'],
                                                    'content':notes}
 

src/calibre/library/database2.py

@@ -298,10 +298,8 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
                             base,
                             prefer_custom=True)
 
-        self.FIELD_MAP['ondevice'] = base+1
+        self.FIELD_MAP['ondevice'] = base = base+1
-        self.field_metadata.set_field_record_index('ondevice', base+1, prefer_custom=False)
+        self.field_metadata.set_field_record_index('ondevice', base, prefer_custom=False)
-        self.FIELD_MAP['all_metadata'] = base+2
-        self.field_metadata.set_field_record_index('all_metadata', base+2, prefer_custom=False)
 
         script = '''
         DROP VIEW IF EXISTS meta2;
@@ -343,10 +341,6 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
         self.has_id  = self.data.has_id
         self.count   = self.data.count
 
-        # Count times get_metadata is called, and how many times in the cache
-        self.gm_count  = 0
-        self.gm_missed = 0
-
         for prop in ('author_sort', 'authors', 'comment', 'comments', 'isbn',
                      'publisher', 'rating', 'series', 'series_index', 'tags',
                      'title', 'timestamp', 'uuid', 'pubdate', 'ondevice'):
@@ -690,19 +684,7 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
         '''
         row = self.data._data[idx] if index_is_id else self.data[idx]
         fm = self.FIELD_MAP
-
-        self.gm_count += 1
-        mi = row[self.FIELD_MAP['all_metadata']]
-        if mi is not None:
-            if get_cover:
-                # Always get the cover, because the value can be wrong if the
-                # original mi was from the OPF
-                mi.cover = self.cover(idx, index_is_id=index_is_id, as_path=True)
-            return mi
-
-        self.gm_missed += 1
         mi = Metadata(None)
-        self.data.set(idx, fm['all_metadata'], mi, row_is_id = index_is_id)
 
         aut_list = row[fm['au_map']]
         aut_list = [p.split(':::') for p in aut_list.split(':#:')]
@@ -724,6 +706,8 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
         formats = row[fm['formats']]
         if not formats:
             formats = None
+        else:
+            formats = formats.split(',')
         mi.formats = formats
         tags = row[fm['tags']]
         if tags:
@@ -1387,7 +1371,8 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
             if r is not None:
                 if (now - r[self.FIELD_MAP['timestamp']]) > delta:
                     tags = r[self.FIELD_MAP['tags']]
-                    if tags and tag in tags.lower():
+                    if tags and tag in [x.strip() for x in
+                            tags.lower().split(',')]:
                         yield r[self.FIELD_MAP['id']]
 
     def get_next_series_num_for(self, series):

src/calibre/library/field_metadata.py

@@ -162,15 +162,6 @@ class FieldMetadata(dict):
                            'search_terms':['tags', 'tag'],
                            'is_custom':False,
                            'is_category':True}),
-            ('all_metadata',{'table':None,
-                             'column':None,
-                             'datatype':None,
-                             'is_multiple':None,
-                             'kind':'field',
-                             'name':None,
-                             'search_terms':[],
-                             'is_custom':False,
-                             'is_category':False}),
             ('author_sort',{'table':None,
                             'column':None,
                             'datatype':'text',

src/calibre/trac/bzr_commit_plugin.py

@@ -110,6 +110,7 @@ class cmd_commit(_cmd_commit):
             suffix = 'The fix will be in the next release.'
         action = action+'ed'
         msg = '%s in branch %s. %s'%(action, nick, suffix)
+        msg = msg.replace('Fixesed', 'Fixed')
         server = xmlrpclib.ServerProxy(url)
         server.ticket.update(int(bug), msg,
                              {'status':'closed', 'resolution':'fixed'},

src/calibre/utils/cleantext.py

@@ -3,7 +3,7 @@ __license__ = 'GPL 3'
 __copyright__ = '2010, sengian <sengian1@gmail.com>'
 __docformat__ = 'restructuredtext en'
 
-import re
+import re, htmlentitydefs
 
 _ascii_pat = None
 
@@ -21,3 +21,32 @@ def clean_ascii_chars(txt, charlist=None):
         pat = re.compile(u'|'.join(map(unichr, charlist)))
     return pat.sub('', txt)
 
+##
+# Fredrik Lundh: http://effbot.org/zone/re-sub.htm#unescape-html
+# Removes HTML or XML character references and entities from a text string.
+#
+# @param text The HTML (or XML) source text.
+# @return The plain text, as a Unicode string, if necessary.
+
+def unescape(text, rm=False, rchar=u''):
+    def fixup(m, rm=rm, rchar=rchar):
+        text = m.group(0)
+        if text[:2] == "&#":
+            # character reference
+            try:
+                if text[:3] == "&#x":
+                    return unichr(int(text[3:-1], 16))
+                else:
+                    return unichr(int(text[2:-1]))
+            except ValueError:
+                pass
+        else:
+            # named entity
+            try:
+                text = unichr(htmlentitydefs.name2codepoint[text[1:-1]])
+            except KeyError:
+                pass
+        if rm:
+            return rchar #replace by char
+        return text # leave as is
+    return re.sub("&#?\w+;", fixup, text)

src/calibre/utils/formatter.py

@@ -18,6 +18,24 @@ class _Parser(object):
     LEX_NUM = 4
     LEX_EOF = 5
 
+    def _python(self, func):
+        locals = {}
+        exec func in locals
+        if 'evaluate' not in locals:
+            self.error('no evaluate function in python')
+        try:
+            result = locals['evaluate'](self.parent.kwargs)
+            if isinstance(result, (float, int)):
+                result = unicode(result)
+            elif isinstance(result, list):
+                result = ','.join(result)
+            elif isinstance(result, str):
+                result = unicode(result)
+            return result
+        except Exception as e:
+            self.error('python function threw exception: ' + e.msg)
+
+
     def _strcmp(self, x, y, lt, eq, gt):
         v = strcmp(x, y)
         if v < 0:
@@ -79,6 +97,7 @@ class _Parser(object):
             'field'    : (1, lambda s, x: s.parent.get_value(x, [], s.parent.kwargs)),
             'multiply' : (2, partial(_math, op='*')),
             'print'    : (-1, _print),
+            'python'   : (1, _python),
             'strcat'   : (-1, _concat),
             'strcmp'   : (5, _strcmp),
             'substr'   : (3, lambda s, x, y, z: x[int(y): len(x) if int(z) == 0 else int(z)]),
@@ -362,7 +381,7 @@ class TemplateFormatter(string.Formatter):
                 (r'\'.*?((?<!\\)\')',   lambda x,t: (3, t[1:-1])),
                 (r'\n#.*?(?=\n)',       None),
                 (r'\s',                 None)
-        ])
+        ], flags=re.DOTALL)
 
     def _eval_program(self, val, prog):
         # keep a cache of the lex'ed program under the theory that re-lexing

src/calibre/utils/magick/draw.py

@@ -92,6 +92,9 @@ def identify_data(data):
     or raises an Exception if data is not an image.
     '''
     img = Image()
+    if hasattr(img, 'identify'):
+        img.identify(data)
+    else:
         img.load(data)
     width, height = img.size
     fmt = img.format

src/calibre/utils/magick/magick.c

@@ -456,6 +456,26 @@ magick_Image_load(magick_Image *self, PyObject *args, PyObject *kwargs) {
 
 // }}}
 
+// Image.identify {{{
+static PyObject *
+magick_Image_identify(magick_Image *self, PyObject *args, PyObject *kwargs) {
+    const char *data;
+	Py_ssize_t dlen;
+    MagickBooleanType res;
+    
+    NULL_CHECK(NULL)
+    if (!PyArg_ParseTuple(args, "s#", &data, &dlen)) return NULL;
+
+    res = MagickPingImageBlob(self->wand, data, dlen);
+
+    if (!res)
+        return magick_set_exception(self->wand);
+
+    Py_RETURN_NONE;
+}
+
+// }}}
+
 // Image.open {{{
 static PyObject *
 magick_Image_read(magick_Image *self, PyObject *args, PyObject *kwargs) {
@@ -993,6 +1013,10 @@ static PyMethodDef magick_Image_methods[] = {
     {"destroy", (PyCFunction)magick_Image_destroy, METH_VARARGS,
     "Destroy the underlying ImageMagick Wand. WARNING: After using this method, all methods on this object will raise an exception."},
 
+    {"identify", (PyCFunction)magick_Image_identify, METH_VARARGS,
+     "Identify an image from a byte buffer (string)"
+    },
+
     {"load", (PyCFunction)magick_Image_load, METH_VARARGS,
      "Load an image from a byte buffer (string)"
     },

src/calibre/utils/wmf/__init__.py

@@ -5,5 +5,52 @@ __license__   = 'GPL v3'
 __copyright__ = '2011, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
+import glob
+
+from calibre.constants import plugins, iswindows, filesystem_encoding
+from calibre.ptempfile import TemporaryDirectory
+from calibre import CurrentDir
+from calibre.utils.magick import Image, PixelWand
+
+class Unavailable(Exception):
+    pass
+
+class NoRaster(Exception):
+    pass
+
+def extract_raster_image(wmf_data):
+    try:
+        wmf, wmf_err = plugins['wmf']
+    except KeyError:
+        raise Unavailable('libwmf not available on this platform')
+    if wmf_err:
+        raise Unavailable(wmf_err)
+
+    if iswindows:
+        import sys, os
+        appdir = sys.app_dir
+        if isinstance(appdir, unicode):
+            appdir = appdir.encode(filesystem_encoding)
+        fdir = os.path.join(appdir, 'wmffonts')
+        wmf.set_font_dir(fdir)
+
+    data = ''
+
+    with TemporaryDirectory('wmf2png') as tdir:
+        with CurrentDir(tdir):
+            wmf.render(wmf_data)
+
+            images = list(sorted(glob.glob('*.png')))
+            if not images:
+                raise NoRaster('No raster images in WMF')
+            data = open(images[0], 'rb').read()
+
+    im = Image()
+    im.load(data)
+    pw = PixelWand()
+    pw.color = '#ffffff'
+    im.rotate(pw, 180)
+
+    return im.export('png')
 
 

src/calibre/utils/wmf/wmf.c

@@ -4,6 +4,7 @@
 
 #include <libwmf/api.h>
 #include <libwmf/svg.h>
+//#include <libwmf/gd.h>
 
 typedef struct {
     char *data;
@@ -13,7 +14,7 @@ typedef struct {
 
 //This code is taken mostly from the Abiword wmf plugin
 
-
+// Buffer read {{{
 // returns unsigned char cast to int, or EOF
 static int wmf_WMF_read(void * context) {
     char c;
@@ -22,11 +23,11 @@ static int wmf_WMF_read(void * context) {
 	if (info->pos == info->len)
 		return EOF;
 
-	c = info->data[pos];
+	c = info->data[info->pos];
 
 	info->pos++;
 
-	return (int)c;
+	return (int)((unsigned char)c);
 }
 
 // returns (-1) on error, else 0
@@ -44,8 +45,17 @@ static long wmf_WMF_tell(void * context) {
 
 	return (long) info->pos;
 }
+// }}}
 
 
+char _png_name_buf[100];
+char *wmf_png_name(void *ctxt) {
+    int *num = (int*)ctxt;
+    *num = *num + 1;
+    snprintf(_png_name_buf, 90, "%04d.png", *num);
+    return _png_name_buf;
+}
+
 #define CLEANUP if(API) { if (stream) wmf_free(API, stream); wmf_api_destroy(API); };
 
 static PyObject *
@@ -66,9 +76,9 @@ wmf_render(PyObject *self, PyObject *args) {
 
 	unsigned int max_width  = 1600;
 	unsigned int max_height = 1200;
-	unsigned long max_flags = 0;
 
 	static const char* Default_Description = "wmf2svg";
+    int fname_counter = 0;
 
 	wmf_error_t err;
 
@@ -125,6 +135,8 @@ wmf_render(PyObject *self, PyObject *args) {
 	ddata->Description = (char *)Default_Description;
 
 	ddata->bbox = bbox;
+    ddata->image.context = (void *)&fname_counter;
+    ddata->image.name = wmf_png_name;
 
 	wmf_display_size(API, &disp_width, &disp_height, 96, 96);
 
@@ -156,9 +168,9 @@ wmf_render(PyObject *self, PyObject *args) {
 		ddata->height = (unsigned int) ceil ((double) wmf_height);
 	}
 
-	ddata->flags |= WMF_SVG_INLINE_IMAGES;
+    // Needs GD
-
+	//ddata->flags |= WMF_SVG_INLINE_IMAGES;
-	ddata->flags |= WMF_GD_OUTPUT_MEMORY | WMF_GD_OWN_BUFFER;
+	//ddata->flags |= WMF_GD_OUTPUT_MEMORY | WMF_GD_OWN_BUFFER;
 
     err = wmf_play(API, 0, &(bbox));
 
@@ -178,11 +190,32 @@ wmf_render(PyObject *self, PyObject *args) {
     return ans;
 }
 
+#ifdef _WIN32
+void set_libwmf_fontdir(const char *);
+
+static PyObject *
+wmf_setfontdir(PyObject *self, PyObject *args) {
+    char *path;
+    if (!PyArg_ParseTuple(args, "s", &path))
+        return NULL;
+    set_libwmf_fontdir(path);
+
+    Py_RETURN_NONE;
+}
+#endif
+
+
+
 
 static PyMethodDef wmf_methods[] = {
     {"render", wmf_render, METH_VARARGS,
-        "render(path) -> Render wmf as svg."
+        "render(data) -> Render wmf as svg."
     },
+#ifdef _WIN32
+    {"set_font_dir", wmf_setfontdir, METH_VARARGS,
+        "set_font_dir(path) -> Set the path to the fonts dir on windows, must be called at least once before using render()"
+    },
+#endif
 
     {NULL}  /* Sentinel */
 };

src/calibre/utils/zipfile.py

@@ -982,9 +982,12 @@ class ZipFile:
             zef_file.read(fheader[_FH_EXTRA_FIELD_LENGTH])
 
         if fname != zinfo.orig_filename:
-            raise BadZipfile, \
+            print ('WARNING: Header (%r) and directory (%r) filenames do not'
-                      'File name in directory "%s" and header "%s" differ.' % (
+                    ' match inside ZipFile')%(fname, zinfo.orig_filename)
-                          zinfo.orig_filename, fname)
+            print 'Using directory filename %r'%zinfo.orig_filename
+            #raise BadZipfile, \
+            #          'File name in directory "%r" and header "%r" differ.' % (
+            #              zinfo.orig_filename, fname)
 
         # check for encrypted flag & handle password
         is_encrypted = zinfo.flag_bits & 0x1

src/calibre/web/feeds/recipes/collection.py

@@ -108,7 +108,6 @@ def download_builtin_recipe(urn):
     br = browser()
     return br.open_novisit('http://status.calibre-ebook.com/recipe/'+urn).read()
 
-
 def get_builtin_recipe_by_title(title, log=None, download_recipe=False):
     for x in get_builtin_recipe_collection():
         if x.get('title') == title:
@@ -127,6 +126,24 @@ def get_builtin_recipe_by_title(title, log=None, download_recipe=False):
                         'Failed to download recipe, using builtin version')
             return P('recipes/%s.recipe'%urn, data=True)
 
+def get_builtin_recipe_by_id(id_, log=None, download_recipe=False):
+    for x in get_builtin_recipe_collection():
+        if x.get('id') == id_:
+            urn = x.get('id')[8:]
+            if download_recipe:
+                try:
+                    if log is not None:
+                        log('Trying to get latest version of recipe:', urn)
+                    return download_builtin_recipe(urn)
+                except:
+                    if log is None:
+                        import traceback
+                        traceback.print_exc()
+                    else:
+                        log.exception(
+                        'Failed to download recipe, using builtin version')
+            return P('recipes/%s.recipe'%urn, data=True)
+
 class SchedulerConfig(object):
 
     def __init__(self):