Various fixes for calibre to run on python 2.6

src/calibre/debug.py

@@ -22,7 +22,7 @@ Run an embedded python interpreter.
     )
     parser.add_option('-c', '--command', help='Run python code.', default=None)
     parser.add_option('--migrate', action='store_true', default=False, 
-                      help='Migrate old database. Needs two arguments. Path to library1.db and path to new library folder.', default=False)
+                      help='Migrate old database. Needs two arguments. Path to library1.db and path to new library folder.')
     return parser
 
 def update_zipfile(zipfile, mod, path):

src/calibre/ebooks/lrf/web/__init__.py

@@ -1,37 +1,6 @@
 __license__   = 'GPL v3'
 __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
 
-from calibre.ebooks.lrf.web.profiles.nytimes       import NYTimes
-from calibre.ebooks.lrf.web.profiles.bbc           import BBC
-from calibre.ebooks.lrf.web.profiles.newsweek      import Newsweek
-from calibre.ebooks.lrf.web.profiles.economist     import Economist
-from calibre.ebooks.lrf.web.profiles.newyorkreview import NewYorkReviewOfBooks
-from calibre.ebooks.lrf.web.profiles.spiegelde     import SpiegelOnline
-from calibre.ebooks.lrf.web.profiles.zeitde        import ZeitNachrichten
-from calibre.ebooks.lrf.web.profiles.faznet        import FazNet
-from calibre.ebooks.lrf.web.profiles.wsj           import WallStreetJournal
-from calibre.ebooks.lrf.web.profiles.barrons       import Barrons
-from calibre.ebooks.lrf.web.profiles.portfolio     import Portfolio
-from calibre.ebooks.lrf.web.profiles.cnn           import CNN
-from calibre.ebooks.lrf.web.profiles.chr_mon       import ChristianScienceMonitor
-from calibre.ebooks.lrf.web.profiles.jpost         import JerusalemPost
-from calibre.ebooks.lrf.web.profiles.reuters       import Reuters
-from calibre.ebooks.lrf.web.profiles.atlantic      import Atlantic 
-from calibre.ebooks.lrf.web.profiles.ap            import AssociatedPress 
-from calibre.ebooks.lrf.web.profiles.newyorker     import NewYorker 
-from calibre.ebooks.lrf.web.profiles.jutarnji      import Jutarnji
-from calibre.ebooks.lrf.web.profiles.usatoday      import USAToday
-from calibre.ebooks.lrf.web.profiles.upi           import UnitedPressInternational 
-from calibre.ebooks.lrf.web.profiles.wash_post     import WashingtonPost 
-from calibre.ebooks.lrf.web.profiles.nasa          import NASA 
 
-
+builtin_profiles   = []
-builtin_profiles   = [Atlantic, AssociatedPress, Barrons, BBC, 
+available_profiles = [i.__module__.rpartition('.')[2] for i in builtin_profiles]
-                      ChristianScienceMonitor, CNN, Economist, FazNet, 
-                      JerusalemPost, Jutarnji, NASA, Newsweek, NewYorker, 
-                      NewYorkReviewOfBooks, NYTimes, UnitedPressInternational, USAToday,  
-                      Portfolio, Reuters, SpiegelOnline, WallStreetJournal, 
-                      WashingtonPost, ZeitNachrichten,   
-                     ]
-
-available_profiles = [i.__module__.rpartition('.')[2] for i in builtin_profiles]

src/calibre/ebooks/lrf/web/profiles/nasa.py

@@ -61,7 +61,7 @@ class NASA(DefaultProfile):
 		(re.compile(r'<!-- Top Header starts -->.*?<!---->', re.IGNORECASE | re.DOTALL), lambda match : '<New Stuff>'),
 		
 		## This removes the "download image" of various sizes from the Image of the day.
-		(re.compile(r'<div id="download_image_box_print">.*?<div id="caption_region_print">', re.IGNORECASE | re.DOTALL), lambda match : '<New Stuff>'),
+		(re.compile(r'(?is)<div id="download_image_box_print">.*?<div id="caption_region_print">'), lambda match : '<New Stuff>'),
 
 
 		]

src/calibre/library/database2.py

@@ -304,10 +304,11 @@ class ResultCache(SearchQueryParser):
     
 class Tag(unicode):
     
-    def __init__(self, name):
+    def __new__(cls, *args):
-        unicode.__init__(self, name)
+        obj = super(Tag, cls).__new__(cls, *args)
-        self.count = 0
+        obj.count = 0
-        self.state = 0
+        obj.state = 0
+        return obj
         
     def as_string(self):
         return u'[%d] %s'%(self.count, self)

src/calibre/manual/faq.rst

@@ -102,7 +102,7 @@ Device Integration
 
 What devices does |app| support?
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-At the moment |app| has full support for the SONY PRS500 and PRS505. However, using the :guilabel:`Save to disk` function you can use it with any ebook reader that exports itself as a USB disk.
+At the moment |app| has full support for the SONY PRS 500/505/700 as well as the iPhone. In addition, using the :guilabel:`Save to disk` function you can use it with any ebook reader that exports itself as a USB disk.
 
 I used |app| to transfer some books to my reader, and now the SONY software hangs every time I connect the reader?
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -117,6 +117,17 @@ Can I use the collections feature of the SONY reader?
 |app| has full support for collections. When you add tags to a book's metadata, those tags are turned into collections when you upload the book to the SONY reader. Also, the series information is automatically
 turned into a collection on the reader. Note that the PRS-500 does not support collections for books stored on the SD card. The PRS-505 does. 
 
+How do I use |app| with my iPhone?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+First install the Stanza reader on your iPhone from http://www.lexcycle.com . Then,
+  * Set the output format for calibre to EPUB (this can be done in the configuration dialog accessed by the little hammer icon next to the search bar)
+  * Convert the books you want to read on your iPhone to EPUB format by selecting them and clicking the Convert button.
+  * Turn on the Content Server in the configurations dialog and leave |app| running.
+  * In the Stanza reader on your iPhone, add a new catalog. The URL of the catalog is of the form 
+    ``http://10.34.56.89:8080/stanza``,  where you should replace the IP address ``10.34.56.89`` 
+    with the IP address of your computer. Stanza will the use the |app| content server to access all the
+    EPUB books in your |app| database.
+
 Library Management
 ------------------
 

src/calibre/path.py

@@ -29,7 +29,7 @@ Date:    9 Mar 2007
 
 from __future__ import generators
 
-import sys, warnings, os, fnmatch, glob, shutil, codecs, md5
+import sys, warnings, os, fnmatch, glob, shutil, codecs, hashlib
 
 __version__ = '2.2'
 __all__ = ['path']
@@ -767,7 +767,7 @@ class path(_base):
         """
         f = self.open('rb')
         try:
-            m = md5.new()
+            m = hashlib.md5()
             while True:
                 d = f.read(8192)
                 if not d:

src/calibre/web/feeds/news.py

@@ -335,9 +335,9 @@ class BasicNewsRecipe(object, LoggingInterface):
         It can be used to do arbitrarily powerful post-processing on the :term:`HTML`.
         It should return `soup` after processing it. 
         
-        :param soup: A `BeautifulSoup <http://www.crummy.com/software/BeautifulSoup/documentation.html>`_ 
+        :param soup: A `BeautifulSoup <http://www.crummy.com/software/BeautifulSoup/documentation.html>`_  instance containing the downloaded :term:`HTML`.
-        instance containing the downloaded :term:`HTML`.
         :param first_fetch: True if this is the first page of an article.
+
         '''
         return soup
     

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

@@ -5,18 +5,18 @@ __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
 Builtin recipes.
 '''
 recipe_modules = [
-           'newsweek', 'atlantic', 'economist', 'portfolio', 
+           'newsweek', 'atlantic', 'economist', 'portfolio',
            'nytimes', 'usatoday', 'outlook_india', 'bbc', 'greader', 'wsj',
            'wired', 'globe_and_mail', 'smh', 'espn', 'business_week',
            'ars_technica', 'upi', 'new_yorker', 'irish_times', 'iht', 'lanacion',
            'discover_magazine', 'scientific_american', 'new_york_review_of_books',
-           'daily_telegraph', 'guardian', 'el_pais', 'new_scientist', 'b92', 
+           'daily_telegraph', 'guardian', 'el_pais', 'new_scientist', 'b92',
            'politika', 'moscow_times', 'latimes', 'japan_times', 'san_fran_chronicle',
            'demorgen_be', 'de_standaard', 'ap', 'barrons', 'chr_mon', 'cnn', 'faznet',
            'jpost', 'jutarnji', 'nasa', 'reuters', 'spiegelde', 'wash_post', 'zeitde',
            'blic', 'novosti', 'danas', 'vreme', 'times_online', 'the_scotsman',
            'nytimes_sub', 'security_watch', 'cyberpresse', 'st_petersburg_times',
-           'clarin', 'financial_times', 'heise'
+           'clarin', 'financial_times', 'heise', 'le_monde'
           ]
 
 import re, imp, inspect, time, os
@@ -58,7 +58,7 @@ def compile_recipe(src):
     Compile the code in src and return the first object that is a recipe or profile.
     @param src: Python source code
     @type src: string
-    @return: Recipe/Profile class or None, if no such class was found in C{src} 
+    @return: Recipe/Profile class or None, if no such class was found in C{src}
     '''
     global _tdir, _crep
     if _tdir is None or not os.path.exists(_tdir):
@@ -77,13 +77,13 @@ def compile_recipe(src):
     f.close()
     module = imp.find_module(temp.namebase, [temp.dirname()])
     module = imp.load_module(temp.namebase, *module)
-    classes = inspect.getmembers(module, 
+    classes = inspect.getmembers(module,
             lambda x : inspect.isclass(x) and \
                 issubclass(x, (DefaultProfile, BasicNewsRecipe)) and \
                 x not in basic_recipes)
     if not classes:
         return None
-    
+
     return classes[0][1]
 
 
@@ -92,7 +92,7 @@ def get_builtin_recipe(title):
     Return a builtin recipe/profile class whose title == C{title} or None if no such
     recipe exists. Also returns a flag that is True iff the found recipe is really
     an old-style Profile.
-    
+
     @type title: string
     @rtype: class or None, boolean
     '''
@@ -117,9 +117,9 @@ class BasicUserRecipe%d(AutomaticNewsRecipe):
     oldest_article = %d
     max_articles_per_feed = %d
     summary_length = %d
-    
+
     feeds = %s
-    
+
-'''%(int(time.time()), repr(profile.title), profile.oldest_article, 
+'''%(int(time.time()), repr(profile.title), profile.oldest_article,
     profile.max_articles_per_feed, profile.summary_length, repr(profile.feeds))
-    
+

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

@@ -53,13 +53,12 @@ class NASA(BasicNewsRecipe):
 		(r'<a.*?onclick.*?>.*?(<img .*?>)', lambda match: match.group(1),),
 		
 		## This removes header and footer information from each print version.
-       	(re.compile(r'<!-- Top Header starts -->.*?<!-- Body starts -->', re.IGNORECASE | re.DOTALL), lambda match : '<New Stuff>'),
+       	(r'<!-- Top Header starts -->.*?<!-- Body starts -->', lambda match : '<New Stuff>'),
-		(re.compile(r'<hr align="center" width="200"><p align="center">.*?<!-- Press Release standard text ends -->', re.IGNORECASE | re.DOTALL), lambda match : '<New Stuff>'),
+		(r'<hr align="center" width="200"><p align="center">.*?<!-- Press Release standard text ends -->', lambda match : '<New Stuff>'),
-		(re.compile(r'<!-- Top Header starts -->.*?<!---->', re.IGNORECASE | re.DOTALL), lambda match : '<New Stuff>'),
+		(r'<!-- Top Header starts -->.*?<!---->', lambda match : '<New Stuff>'),
 		
 		## This removes the "download image" of various sizes from the Image of the day.
-		(re.compile(r'<div id="download_image_box_print">.*?<div id="caption_region_print">', re.IGNORECASE | re.DOTALL), lambda match : '<New Stuff>'),
+		(r'<div id="download_image_box_print">.*?<div id="caption_region_print">', lambda match : '<New Stuff>'),
-
 
 		]
 	]

src/pyPdf/pdf.py

@@ -49,7 +49,7 @@ import utils
 import warnings
 from generic import *
 from utils import readNonWhitespace, readUntilWhitespace, ConvertFunctionsToVirtualList
-from sets import ImmutableSet
+
 
 ##
 # This class supports writing PDF files out, given pages produced by another
@@ -986,8 +986,8 @@ class PageObject(DictionaryObject):
 
         # Combine /ProcSet sets.
         newResources[NameObject("/ProcSet")] = ArrayObject(
-            ImmutableSet(originalResources.get("/ProcSet", ArrayObject()).getObject()).union(
+            frozenset(originalResources.get("/ProcSet", ArrayObject()).getObject()).union(
-                ImmutableSet(page2Resources.get("/ProcSet", ArrayObject()).getObject())
+                frozenset(page2Resources.get("/ProcSet", ArrayObject()).getObject())
             )
         )
 

todo-2.6

@@ -0,0 +1,9 @@
+
+* Refactor web.fetch.simple to use per connection timeouts via the timeout kwarg for mechanize.open
+
+* Refactor IPC code to use communication logic from multiprocessing
+
+* Use multiprocessing for cpu_count instead of QThread
+
+* Windows build:
+  * Compile all dependencies with MSVC 2008 since this is what python now uses