Merge remote-tracking branch 'origin/master' into calibre_cover_generate

calibre-plugin/plugin-defaults.ini

@@ -1471,6 +1471,8 @@ extra_titlepage_entries:readings,awards
 awards_label:Awards
 readings_label:Readings
 
+## some sites include images that we don't ever want becoming the
+## cover image.  This lets you exclude them.
 cover_exclusion_regexp:art/.*Awards.jpg
 
 [voracity2.e-fic.com]
@@ -1667,6 +1669,10 @@ dateUpdated_format:%%Y-%%m-%%d %%H:%%M:%%S
 ## to on, but can be switched off if it is found to cause problems.
 fix_fimf_blockquotes:true
 
+## some sites include images that we don't ever want becoming the
+## cover image.  This lets you exclude them.
+cover_exclusion_regexp:/images/emoticons/
+
 ## Site dedicated to these categories/characters/ships
 extracategories:My Little Pony: Friendship is Magic
 

calibre-plugin/translations/es.po

@@ -5,12 +5,13 @@
 # Adolfo Jayme Barrientos, 2014
 # dario hereñu <magallania@gmail.com>, 2015
 # Jellby <jellby@yahoo.com>, 2014-2015
+# JimmXinu <retiefjimm@gmail.com>, 2015
 msgid ""
 msgstr ""
 "Project-Id-Version: calibre-plugins\n"
 "POT-Creation-Date: 2015-04-11 11:17+Central Daylight Time\n"
-"PO-Revision-Date: 2015-04-17 13:02+0000\n"
-"Last-Translator: Jellby <jellby@yahoo.com>\n"
+"PO-Revision-Date: 2015-04-22 22:08+0000\n"
+"Last-Translator: JimmXinu <retiefjimm@gmail.com>\n"
 "Language-Team: Spanish (http://www.transifex.com/projects/p/calibre-plugins/language/es/)\n"
 "MIME-Version: 1.0\n"
 "Content-Type: text/plain; charset=UTF-8\n"
@@ -558,7 +559,7 @@ msgid ""
 "The %(pini)s parameter %(gcset)s allows you to choose a GC setting based on "
 "metadata rather than site, but it's much more complex.<br \\>%(gcset)s is "
 "ignored when this is off."
-msgstr "El parámetro %(gcset)s de %(pini) le permite elegir una configuración de GC según los metadatos en vez del sitio, pero es mucho más complejo.<br>%(gcset)s no se tiene en cuenta si esta opción está desactivada."
+msgstr "El parámetro %(gcset)s de %(pini)s le permite elegir una configuración de GC según los metadatos en vez del sitio, pero es mucho más complejo.<br>%(gcset)s no se tiene en cuenta si esta opción está desactivada."
 
 #: config.py:811
 msgid "Use calibre's Polish feature to inject/update the cover"
@@ -794,7 +795,7 @@ msgid ""
 "The %(pini)s parameter %(ccset)s allows you to set custom columns to site "
 "specific values that aren't common to all sites.<br />%(ccset)s is ignored "
 "when this is off."
-msgstr "El parámetro %(ccset)s de %(pini) le permite asignar a las columnas personalizadas valores específicos para cada sitio.<br>%(ccset)s no se tiene en cuenta si esta opción está desactivada."
+msgstr "El parámetro %(ccset)s de %(pini)s le permite asignar a las columnas personalizadas valores específicos para cada sitio.<br>%(ccset)s no se tiene en cuenta si esta opción está desactivada."
 
 #: config.py:1065
 msgid "Special column:"

fanficfare/adapters/adapter_fimfictionnet.py

@@ -176,7 +176,10 @@ class FimFictionNetSiteAdapter(BaseSiteAdapter):
             if coverurl.startswith('//'): # fix for img urls missing 'http:'
                 coverurl = "http:"+coverurl
             if get_cover:
-                self.setCoverImage(self.url,coverurl)
+                # try setting from href, if fails, try using the img src
+                if self.setCoverImage(self.url,coverurl)[0] == "failedtoload":
+                    coverurl = storyImage.find('img')['src']
+                    self.setCoverImage(self.url,coverurl)
 
             coverSource = storyImage.find('a', {'class':'source'})
             if coverSource:

fanficfare/adapters/adapter_storiesonlinenet.py

@@ -21,7 +21,7 @@ logger = logging.getLogger(__name__)
 import re
 import urllib2
 
-from .. import BeautifulSoup as bs
+#from .. import BeautifulSoup as bs
 from ..htmlcleanup import stripHTML
 from .. import exceptions as exceptions
 
@@ -144,7 +144,7 @@ class StoriesOnlineNetAdapter(BaseSiteAdapter):
             raise exceptions.FailedToDownload(self.getSiteDomain() +" says: Error! The story you're trying to access is being filtered by your choice of contents filtering.")
             
         # use BeautifulSoup HTML parser to make everything easier to find.
-        soup = bs.BeautifulSoup(data)
+        soup = self.make_soup(data)
         #print data
 
         # Now go hunting for all the meta data and the chapter list.
@@ -178,7 +178,7 @@ class StoriesOnlineNetAdapter(BaseSiteAdapter):
         page=0
         i=0
         while i == 0:
-            asoup = bs.BeautifulSoup(self._fetchUrl(self.story.getList('authorUrl')[0]+"/"+str(page)))
+            asoup = self.make_soup(self._fetchUrl(self.story.getList('authorUrl')[0]+"/"+str(page)))
 
             a = asoup.findAll('td', {'class' : 'lc2'})
             for lc2 in a:
@@ -212,7 +212,7 @@ class StoriesOnlineNetAdapter(BaseSiteAdapter):
                 self.story.setMetadata('seriesUrl',seriesUrl)
                 series_name = stripHTML(a)
                 logger.debug("Series name= %s" % series_name)
-                series_soup = bs.BeautifulSoup(self._fetchUrl(seriesUrl))
+                series_soup = self.make_soup(self._fetchUrl(seriesUrl))
                 if series_soup:
                     logger.debug("Retrieving Series - looking for name")
                     series_name = series_soup.find('span', {'id' : 'ptitle'}).text.partition(' — ')[0]
@@ -221,7 +221,7 @@ class StoriesOnlineNetAdapter(BaseSiteAdapter):
                 desc = lc4.contents[2]
                 # Check if series is in a universe
                 universe_url = self.story.getList('authorUrl')[0]  + "&type=uni"
-                universes_soup = bs.BeautifulSoup(self._fetchUrl(universe_url) )
+                universes_soup = self.make_soup(self._fetchUrl(universe_url) )
                 logger.debug("Universe url='{0}'".format(universe_url))
                 if universes_soup:
                     universes = universes_soup.findAll('div', {'class' : 'ser-box'})
@@ -255,7 +255,7 @@ class StoriesOnlineNetAdapter(BaseSiteAdapter):
                 universe_name = stripHTML(a)
                 universeUrl = 'http://'+self.host+a['href']
                 logger.debug("Retrieving Universe - about to get page - universeUrl='{0}".format(universeUrl))
-                universe_soup = bs.BeautifulSoup(self._fetchUrl(universeUrl))
+                universe_soup = self.make_soup(self._fetchUrl(universeUrl))
                 logger.debug("Retrieving Universe - have page")
                 if universe_soup:
                     logger.debug("Retrieving Universe - looking for name")
@@ -318,10 +318,12 @@ class StoriesOnlineNetAdapter(BaseSiteAdapter):
 
         logger.debug('Getting chapter text from: %s' % url)
 
-        soup = bs.BeautifulSoup(self._fetchUrl(url),
-                                     selfClosingTags=('br','hr')) # otherwise soup eats the br/hr tags.
+        soup = self.make_soup(self._fetchUrl(url))
         
-        div = soup.find('article', {'id' : 'story'})
+        div = soup.find('div', {'id' : 'story'})
+        if not div:
+            logger.debug("div id=story not found, try article")
+            div = soup.find('article', {'id' : 'story'})
         
         # some big chapters are split over several pages
         pager = div.find('span', {'class' : 'pager'})
@@ -340,10 +342,12 @@ class StoriesOnlineNetAdapter(BaseSiteAdapter):
 #            logger.debug(div)
                         
             for ur in urls:
-                soup = bs.BeautifulSoup(self._fetchUrl("http://"+self.getSiteDomain()+ur['href']),
-                                     selfClosingTags=('br','hr')) # otherwise soup eats the br/hr tags.
+                soup = self.make_soup(self._fetchUrl("http://"+self.getSiteDomain()+ur['href']))
         
-                div1 = soup.find('article', {'id' : 'story'})
+                div1 = soup.find('div', {'id' : 'story'})
+                if not div1:
+                    logger.debug("div id=story not found, try article")
+                    div1 = soup.find('article', {'id' : 'story'})
                 
                 # Find the "Continues" marker on the current page and remove everything after that. 
                 continues = div.find('span', {'class' : 'conTag'})
@@ -368,7 +372,7 @@ class StoriesOnlineNetAdapter(BaseSiteAdapter):
                 if endPager != None:
                     b = endPager.nextSibling
                     while endPager != None:
-                        logger.debug("removing end: {0}".format(endPager))
+#                        logger.debug("removing end: {0}".format(endPager))
                         b = endPager.nextSibling
                         endPager.extract()
                         endPager = b
@@ -401,21 +405,35 @@ class StoriesOnlineNetAdapter(BaseSiteAdapter):
 
         # For a story or the last chapter, remove voting form and the in library box
         a = div.find('div', {'id' : 'vote-form'})
+        if a != None:
+            a.extract()
+        a = div.find('div', {'id' : 'top-header'})
         if a != None:
             a.extract()
         a = div.find('div', {'id' : 'b-man-div'})
         if a != None:
             a.extract()
 
-        # Kill the "The End" header and everything after it.
-        a = div.find(['h2', 'h3'], {'class' : 'end'})
-        logger.debug("Chapter end= '{0}'".format(a))
+        # Kill the vote form and everything after it.
+        a = div.find('div', {'class' : 'vform'})
+#        logger.debug("Chapter end= '{0}'".format(a))
         while a != None:
             b = a.nextSibling
             a.extract()
             a=b
 
+        # Kill the vote form and everything after it.
+        a = div.find('h3', {'class' : 'end'})
+#        logger.debug("Chapter end= '{0}'".format(a))
+        while a != None:
+            b = a.nextSibling
+            a.extract()
+            a=b
 
+        foot = div.find('footer')
+        if foot != None:
+            foot.extract()
+            
         if None == div:
             raise exceptions.FailedToDownload("Error downloading Chapter: %s!  Missing required element!" % url)
     

fanficfare/adapters/base_adapter.py

@@ -481,8 +481,10 @@ class BaseSiteAdapter(Configurable):
 
     def setCoverImage(self,storyurl,imgurl):
         if self.getConfig('include_images'):
-            self.story.addImgUrl(storyurl,imgurl,self._fetchUrlRaw,cover=True,
-                                 coverexclusion=self.getConfig('cover_exclusion_regexp'))
+            return self.story.addImgUrl(storyurl,imgurl,self._fetchUrlRaw,cover=True,
+                                        coverexclusion=self.getConfig('cover_exclusion_regexp'))
+        else:
+            return (None,None)
 
     # bs3 & bs4 are different here.
     # will move to a bs3 vs bs4 block if there's lots of changes.

fanficfare/cli.py

@@ -144,14 +144,20 @@ def main(argv=None, parser=None, passed_defaultsini=None, passed_personalini=Non
     else:
         urls = args
 
-    for url in urls:
-        try:
-            do_download(url,
-                        options,
-                        passed_defaultsini,
-                        passed_personalini)
-        except Exception, e:
-            print "URL(%s) Failed: Exception (%s). Run URL individually for more detail."%(url,e)
+    if len(urls) > 1:
+        for url in urls:
+            try:
+                do_download(url,
+                            options,
+                            passed_defaultsini,
+                            passed_personalini)
+            except Exception, e:
+                print "URL(%s) Failed: Exception (%s). Run URL individually for more detail."%(url,e)
+    else:
+        do_download(urls[0],
+                    options,
+                    passed_defaultsini,
+                    passed_personalini)
 
 # make rest a function and loop on it.
 def do_download(arg,

fanficfare/defaults.ini

@@ -1656,6 +1656,10 @@ dateUpdated_format:%%Y-%%m-%%d %%H:%%M:%%S
 ## to on, but can be switched off if it is found to cause problems.
 fix_fimf_blockquotes:true
 
+## some sites include images that we don't ever want becoming the
+## cover image.  This lets you exclude them.
+cover_exclusion_regexp:/images/emoticons/
+
 ## Site dedicated to these categories/characters/ships
 extracategories:My Little Pony: Friendship is Magic
 

fanficfare/exceptions.py

@@ -24,6 +24,13 @@ class FailedToDownload(Exception):
     def __str__(self):
         return self.error
 
+class RejectImage(Exception):
+    def __init__(self,error):
+        self.error=error
+
+    def __str__(self):
+        return self.error
+
 class InvalidStoryURL(Exception):
     def __init__(self,url,domain,example):
         self.url=url

fanficfare/story.py

@@ -44,6 +44,8 @@ imagetypes = {
 
 try:
     from calibre.utils.magick import Image
+    from StringIO import StringIO
+    from gif import GifInfo, CHECK_IS_ANIMATED
     convtype = {'jpg':'JPG', 'png':'PNG'}
 
     def convert_image(url,data,sizes,grayscale,
@@ -55,6 +57,10 @@ try:
         owidth, oheight = img.size
         nwidth, nheight = sizes
         scaled, nwidth, nheight = fit_image(owidth, oheight, nwidth, nheight)
+
+        if normalize_format_name(img.format)=="gif" and GifInfo(StringIO(data),CHECK_IS_ANIMATED).frameCount > 1:
+            raise exceptions.RejectImage("Animated gifs come out purely--not going to use it.")
+
         if scaled:
             img.size = (nwidth, nheight)
             export = True
@@ -224,7 +230,7 @@ langs = {
     "Devanagari":"hi",
 
     ## These are from/for AO3:
-    
+
     u'العربية':'ar',
     u'беларуская':'be',
     u'Български език':'bg',
@@ -325,7 +331,7 @@ class InExMatch:
         else:
             retval = self.match == value
             #print(">>>>>>>>>>>>>%s==%s r: %s,%s=%s"%(self.match,value,self.negate,retval, self.negate != retval))
-            
+
         return self.negate != retval
 
     def __str__(self):
@@ -338,7 +344,7 @@ class InExMatch:
         else:
             s='='
         return u'InExMatch(%s %s%s %s)'%(self.keys,f,s,self.match)
-    
+
 ## metakey[,metakey]=~pattern
 ## metakey[,metakey]==string
 ## *for* part lines.  Effect only when trailing conditional key=~regexp matches
@@ -358,7 +364,7 @@ def set_in_ex_clude(setting):
             match = InExMatch(line)
             dest.append([match,condmatch])
     return dest
-              
+
 ## Two or three part lines.  Two part effect everything.
 ## Three part effect only those key(s) lists.
 ## pattern=>replacement
@@ -433,7 +439,7 @@ class Story(Configurable):
 
     def join_list(self, key, vallist):
         return self.getConfig("join_string_"+key,u", ").replace(SPACE_REPLACE,' ').join(map(unicode, vallist))
-                
+
     def setMetadata(self, key, value, condremoveentities=True):
 
         # keep as list type, but set as only value.
@@ -445,20 +451,20 @@ class Story(Configurable):
                 self.metadata[key]=conditionalRemoveEntities(value)
             else:
                 self.metadata[key]=value
-                
+
         if key == "language":
             try:
                 # getMetadata not just self.metadata[] to do replace_metadata.
                 self.setMetadata('langcode',langs[self.getMetadata(key)])
             except:
                 self.setMetadata('langcode','en')
-                
+
         if key == 'dateUpdated' and value:
             # Last Update tags for Bill.
             self.addToList('lastupdate',value.strftime("Last Update Year/Month: %Y/%m"))
             self.addToList('lastupdate',value.strftime("Last Update: %Y/%m/%d"))
 
-        
+
     def do_in_ex_clude(self,which,value,key):
         if value and which in self.in_ex_cludes:
             include = 'include' in which
@@ -487,7 +493,7 @@ class Story(Configurable):
             if include and keyfound and not found:
                 value = None
         return value
-        
+
 
     def doReplacements(self,value,key,return_list=False,seen_list=[]):
         value = self.do_in_ex_clude('include_metadata_pre',value,key)
@@ -526,7 +532,7 @@ class Story(Configurable):
                         # print("replacement,value:%s,%s->%s"%(replacement,value,regexp.sub(replacement,value)))
                         value = regexp.sub(replacement,value)
                         retlist = [value]
-                    
+
         for val in retlist:
             retlist = map(partial(self.do_in_ex_clude,'include_metadata_post',key=key),retlist)
             retlist = map(partial(self.do_in_ex_clude,'exclude_metadata_post',key=key),retlist)
@@ -610,7 +616,7 @@ class Story(Configurable):
                                                     self.getMetadata('author', removeallentities, doreplacements)))
 
         self.extendList("extratags",self.getConfigList("extratags"))
-            
+
         if self.getMetadataRaw('seriesUrl'):
             self.setMetadata('seriesHTML',linkhtml%('series',self.getMetadata('seriesUrl', removeallentities, doreplacements),
                                                     self.getMetadata('series', removeallentities, doreplacements)))
@@ -703,10 +709,10 @@ class Story(Configurable):
                 for val in retlist:
                     newretlist.extend(self.doReplacements(val,listname,return_list=True))
                 retlist = newretlist
-                
+
             if removeallentities:
                 retlist = map(removeAllEntities,retlist)
-                
+
             retlist = filter( lambda x : x!=None and x!='' ,retlist)
 
         # reorder ships so b/a and c/b/a become a/b and a/b/c.  Only on '/',

included_dependencies/gif.py

@@ -0,0 +1,283 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+"""
+A pure Python GIF metadata extractor.
+Supports adjustable detail to fine-tune performance.
+
+Example code and full epydoc docstrings included.
+
+Uses:
+ - Identifying whether a GIF is static or animated.
+ - Extracting the dimensions, pixel aspect ratio, number of frames, loop count,
+    global palette or palette size, and background color.
+ - Extracting comments and other plaintext.
+ - Testing for various structural errors.
+
+TODO:
+ - Provide basic support for XMP Metadata extraction
+   - http://en.wikipedia.org/wiki/Extensible_Metadata_Platform#Location_in_file_types
+   - http://www.matthewflickinger.com/lab/whatsinagif/bits_and_bytes.asp
+   - Generate test GIF with http://code.google.com/p/python-xmp-toolkit/
+
+Changelog:
+ - 0.2.2: Audited the code and made some corrections.
+ - 0.2.1: 40% speed improvement (went from 15 to 9 seconds for 1000 images)
+ - 0.2.0: Feature-complete
+ - 0.1.0: Initial release
+"""
+
+__appname__ = "gif.py"
+__author__  = "Stephan Sokolow (deitarion/SSokolow)"
+__version__ = "0.2.2"
+__license__ = "PSF License 2.4 or higher (The Python License)"
+
+#{ Check Types (enum, numerical ordering is significant)
+CHECK_IS_GIF_FILE   = 0  #: Just check for a valid GIF header.
+CHECK_IS_ANIMATED   = 1  #: Check whether the file has more than one frame.
+CHECK_COUNT_FRAMES  = 2  #: Count the number of frames in the file.
+CHECK_PARSE_PALETTE = 3  #: Parse the palette and resolve the background color.
+CHECK_READ_COMMENTS = 4  #: Load comments (can sometimes be large) into the L{GifInfo} object.
+CHECK_READ_ALL_TEXT = 5  #: Also load the contents of Plain Text extension blocks.
+
+CHECK_ALL = CHECK_READ_ALL_TEXT #: alias to allow for future modifications
+
+#{  Warning Codes (bitfield)
+WARN_NONE = 0        #: No warnings
+WARN_BAD_IMG = 1     #: Corruption (of the [sub]block size field(s)) or truncation detected in an image block
+WARN_BAD_EXT = 2     #: Corruption (of the [sub]block size field(s)) or truncation detected in an extension block
+WARN_BAD_SIZE = 4    #: An image block specifies dimensions exceeding the global canvas size
+WARN_BAD_BGCOLOR = 8 #: The background color index specified is greater than the palette size
+WARN_EOF  = 16       #: File is missing it's trailer. (Corrupt elsewhere, truncated, or breaking spec by using EOF as the terminator.)
+WARN_TRUNC  = 32     #: File is definitively either truncated or corrupt. (An EOF was encountered part-way through a structure.)
+WARN_LOOP_POS = 64   #: Netscape Application Extension block (animation-control) was present but not first in the file.
+#}
+
+import struct
+
+#{ Structures used by GifInfo
+gifHeaderStruct = struct.Struct('<xxxxxxHHBBB')  #: File header
+gifImageStruct = struct.Struct('<HHHHB')         #: Image block header
+gifExtenStruct = struct.Struct('<BB')            #: Top-level extension block header
+gifNetscapeStruct = struct.Struct('<BBH')        #: NETSCAPE Loop-control sub-block
+gifColorTripleStruct = struct.Struct('<BBB')     #: RGB palette element
+gifPlaintextStruct = struct.Struct('<BHHHHBBBB') #: Plain Text block header
+#}
+
+class BadHeaderException(Exception):
+    """Raised when no valid GIF header is found"""
+
+class GifInfo(object):
+    """A class for loading and storing metadata from GIF files.
+
+    Accepts paths and file-like objects.
+
+    When using L{CHECK_ALL}, this can also be used to walk past a valid
+    GIF file in an un-delimited byte stream in order to identify the point at
+    which the following file starts. (It doesn't C{fh.seek(0)} or C{fh.close()} after
+    use)
+    """
+    warnFlags = WARN_NONE  #: A bit field of C{WARN_*} flags set by L{__init__}
+    checkLevel = CHECK_ALL #: Default C{CHECK_*} level used by L{__init__}
+
+    #{ Pre-defined "unset" values for GIF Metadata
+    path = None        #: The path to the file, if one was passed to L{__init__}
+    version = None     #: C{87a} or C{89a}
+    width = None
+    height = None
+    loopCount = None
+    pixelAspect = None
+    paletteSize = None #: Always calculated if a global palette is present
+    palette = None     #: The palette as a list of integer RGB tuples. Requires L{CHECK_PARSE_PALETTE}.
+    bgColor = None     #: Global background color as an RGB tuple. Requires L{CHECK_PARSE_PALETTE}.
+    comments = None    #: Text in Comment (0xFE) extension blocks as a list of strings. Requires L{CHECK_READ_COMMENTS}
+    otherText = None   #: Text in "Plain Text" (0x01) extension blocks as a list of strings. Requires L{CHECK_READ_ALL_TEXT}
+    frameCount = 0
+    #}
+
+    def __init__(self, fh, checkLevel=checkLevel):
+        """
+        @param fh: A path or file-like object for a GIF file.
+        @param checkLevel: A C{CHECK_*} constant.
+
+        @raises BadHeaderException: The given file lacks a valid GIF header.
+        @raises IOError: The underlying C{open()} system call failed.
+        """
+        self.checkLevel = checkLevel
+        if isinstance(fh, basestring):
+            self.path = fh
+            fh = open(fh, 'rb')
+
+        header = fh.read(gifHeaderStruct.size)
+        if len(header) < gifHeaderStruct.size:
+            raise BadHeaderException("File is too small to be a GIF")
+
+        self.version = header[3:6]
+        if header[0:3] != 'GIF' or self.version not in ['87a', '89a']:
+            raise BadHeaderException("File does not have a recognizable GIF header")
+        elif self.checkLevel <= CHECK_IS_GIF_FILE:
+            return
+
+        self.width, self.height, GCTF_Byte, bgColor, self.pixelAspect = gifHeaderStruct.unpack(header)
+
+        if self.pixelAspect:
+            self.pixelAspect = (self.pixelAspect + 15) / 64.0
+
+        rawPalette = self._getPalette(fh, GCTF_Byte)
+        self.paletteSize = int(len(rawPalette) / 3)
+        if self.checkLevel >= CHECK_PARSE_PALETTE:
+            self.palette = []
+            for pos in range(0, self.paletteSize):
+                self.palette.append(gifColorTripleStruct.unpack_from(rawPalette, pos * 3))
+
+        if self.paletteSize and bgColor > self.paletteSize:
+            self.warnFlags = self.warnFlags | WARN_BAD_BGCOLOR
+        elif self.palette:
+            self.bgColor = self.palette[bgColor]
+
+        # Iterate blocks
+        self.firstBlock = True
+        blocktype = self._read(fh, 1)
+        while not blocktype == chr(0x3B) and not self.warnFlags & WARN_EOF:
+            self._blockHandlers.get(blocktype, lambda x, y:'')(self, fh)
+            if self.checkLevel <= CHECK_IS_ANIMATED and self.frameCount > 1:
+                return
+
+            self.firstBlock = False
+            blocktype = self._read(fh, 1)
+
+        del self.firstBlock
+
+    def _handleImageBlock(self, fh):
+        """"""
+        self.frameCount += 1
+        try:
+            x, y, w, h, LCTF_Byte = gifImageStruct.unpack(self._read(fh, gifImageStruct.size))
+        except:
+            self.warnFlags = self.warnFlags | WARN_EOF | WARN_TRUNC
+            return
+
+        if x + w > self.width or y + h > self.height:
+            self.warnFlags = self.warnFlags | WARN_BAD_SIZE
+
+        self._getPalette(fh, LCTF_Byte) # Skip the local color table if present
+        fh.read(1)                      # Skip the LZW minimum code size.
+
+        # Skip content and test for the block terminator
+        if not self._skipSubBlocks(fh): # For example, if it's a zero-length string like EOF would return.
+            self.warnFlags = self.warnFlags | WARN_BAD_IMG
+
+    def _handleGenericExtensionBlock(self, fh):
+        """@todo: Rewrite this so extension block types have method handlers."""
+        try:
+            extType, blkSize = gifExtenStruct.unpack(self._read(fh, gifExtenStruct.size))
+        except:
+            self.warnFlags = self.warnFlags | WARN_EOF | WARN_TRUNC
+            return
+        startOffset = fh.tell()
+
+        if extType == 0x01 and self.checkLevel >= CHECK_READ_ALL_TEXT: # Plain Text Block
+            self._read(fh, gifPlaintextStruct.size)
+            self.otherText = self.otherText or []
+            blkSize = self._read(fh, 1)
+            while blkSize and blkSize != '\x00':
+                self.otherText.append(self._read(fh, ord(blkSize)))
+                blkSize = self._read(fh, 1)
+        elif extType == 0xFE and self.checkLevel >= CHECK_READ_COMMENTS: # Comment Block
+            self.comments = self.comments or []
+            blkSize = self._read(fh, 1)
+            while blkSize and blkSize != '\x00':
+                self.comments.append(self._read(fh, ord(blkSize)))
+                blkSize = self._read(fh, 1)
+        elif extType == 0xFF: # Application Block
+            if blkSize == 0x0B and self._read(fh, blkSize) == "NETSCAPE2.0":
+                try:
+                    a, b, self.loopCount = gifNetscapeStruct.unpack(self._read(fh, gifNetscapeStruct.size))
+                except:
+                    self.warnFlags = self.warnFlags | WARN_EOF | WARN_TRUNC
+                    return
+
+                if a != 3 and b != 1:
+                    self.warnFlags = self.warnFlags | WARN_BAD_EXT
+
+                if not self.firstBlock:
+                    self.warnFlags = self.warnFlags | WARN_LOOP_POS
+        else:
+            fh.seek( startOffset + blkSize ) # Skip the contents
+
+        # Test for the block terminator
+        if not self._skipSubBlocks(fh):
+            self.warnFlags = self.warnFlags | WARN_BAD_EXT
+
+    def _getPalette(self, handle, bitfield):
+        """Using the size value from C{bitfield},
+        load the palette at C{handle}'s current file pointer position."""
+        if bitfield & int("10000000", 2):
+            nBits = bitfield & int("00000111", 2)
+            tableSize = 3 * 2**( nBits + 1 )
+            return handle.read(tableSize)
+        else:
+            return ''
+
+    def _read(self, handle, size):
+        """Attempt to read the specified number of bytes. Set L{WARN_EOF} if
+        fewer are received."""
+        content = handle.read(size)
+        if len(content) < size:
+            self.warnFlags = self.warnFlags | WARN_EOF
+        return content
+
+    def _skipSubBlocks(self, handle):
+        """Skip sub-blocks beginning at the current file pointer position
+        using fseek."""
+        offset = handle.tell()
+        blkSize = handle.read(1)
+        while blkSize and blkSize != '\x00':
+            offset += ord(blkSize) + 1
+            handle.seek(offset)
+            blkSize = handle.read(1)
+        return blkSize
+
+    _blockHandlers = {
+            chr(0x2C) : _handleImageBlock,
+            chr(0x21) : _handleGenericExtensionBlock,
+        }
+
+def gif_is_animated(path):
+    """A simple convenience function for testing whether a GIF is animated.
+    @rtype: C{bool}
+    """
+    return GifInfo(file(path,'rb'), CHECK_IS_ANIMATED).frameCount > 1
+
+if __name__ == '__main__':
+    from optparse import OptionParser
+    parser = OptionParser(description=__doc__.split('\n\n')[0],
+            version="%%prog v%s" % __version__, usage="%prog <path> ...")
+
+    opts, args = parser.parse_args()
+
+    if args:
+        for fpath in args:
+            try:
+                info = GifInfo(fpath, CHECK_COUNT_FRAMES)
+                warnFlags = (
+                        (info.warnFlags & WARN_BAD_IMG     and 'I' or ' ') +
+                        (info.warnFlags & WARN_BAD_EXT     and 'X' or ' ') +
+                        (info.warnFlags & WARN_BAD_SIZE    and 'C' or ' ') +
+                        (info.warnFlags & WARN_BAD_BGCOLOR and 'B' or ' ') +
+                        (info.warnFlags & WARN_EOF         and 'E' or ' ') +
+                        (info.warnFlags & WARN_TRUNC       and 'T' or ' ') +
+                        (info.warnFlags & WARN_LOOP_POS    and 'L' or ' ')
+                    )
+                print "[%s](%3s Frames): %s" % (warnFlags, info.frameCount, info.path)
+            except BadHeaderException, err:
+                print "%s: %s" % (str(err), fpath)
+        print "\nWarning Flags:"
+        print " I = Image Chunk Corruption/Truncation"
+        print " X = Extension Chunk Corruption/Truncation"
+        print " C = Image Chunk Dimensions Exceed Global Canvas"
+        print " B = Bad Background Color (Index Exceeds Palette Size)"
+        print " E = Unexpected EOF Encountered (Missing Image Terminator)"
+        print " T = EOF Encountered Within A Block Header (Corrupt or Truncated File)"
+        print " L = Loop-control block misplaced within the file"
+        print
+        print "Note: A nearly-threefold speed-up can be had by using CHECK_IS_ANIMATED rather than CHECK_COUNT_FRAMES"

makeplugin.py

@@ -36,7 +36,7 @@ if __name__=="__main__":
 
     os.chdir('../included_dependencies')
     # 'a' for append
-    files=['six.py','bs4','html5lib','chardet']
+    files=['gif.py','six.py','bs4','html5lib','chardet']
     createZipFile("../"+filename,"a",
                   files,
                   exclude=exclude)