Update CLI download zip.

Also don't retry 404 errors and bump versions.

app.yaml

@@ -1,6 +1,6 @@
 # ffd-retief-hrd fanfictiondownloader
 application: fanfictiondownloader
-version: 4-4-92
+version: 4-4-94
 runtime: python27
 api_version: 1
 threadsafe: true

calibre-plugin/__init__.py

@@ -42,7 +42,7 @@ class FanFictionDownLoaderBase(InterfaceActionBase):
     description         = _('UI plugin to download FanFiction stories from various sites.')
     supported_platforms = ['windows', 'osx', 'linux']
     author              = 'Jim Miller'
-    version             = (1, 8, 11)
+    version             = (1, 8, 13)
     minimum_calibre_version = (1, 13, 0)
 
     #: This field defines the GUI plugin class that contains all the code

calibre-plugin/config.py

@@ -55,7 +55,7 @@ from calibre_plugins.fanfictiondownloader_plugin.dialogs \
             EditTextDialog, RejectUrlEntry)
     
 from calibre_plugins.fanfictiondownloader_plugin.fanficdownloader.adapters \
-    import (getConfigSections, getNormalStoryURL)
+    import getConfigSections
 
 from calibre_plugins.fanfictiondownloader_plugin.common_utils \
     import ( KeyboardConfigDialog, PrefsViewerDialog )

calibre-plugin/ffdl_plugin.py

@@ -660,13 +660,13 @@ class FanFictionDownLoaderPlugin(InterfaceAction):
         #print("update_dialog()")
         
         db = self.gui.current_db
-        book_list = map( self.make_book_id_only, id_list )
+        books = map( self.make_book_id_only, id_list )
 
-        for j, book in enumerate(book_list):
+        for j, book in enumerate(books):
             book['listorder'] = j
-            
+
         LoopProgressDialog(self.gui,
-                           book_list,
+                           books,
                            partial(self.populate_book_from_calibre_id, db=self.gui.current_db),
                            self.update_dialog_finish,
                            init_label=_("Collecting stories for update..."),
@@ -718,6 +718,9 @@ class FanFictionDownLoaderPlugin(InterfaceAction):
             url_list = split_text_to_urls(books)
             books = self.convert_urls_to_books(url_list)
 
+        ## for tweak_fg_sleep
+        options['ffnetcount']=len(filter(lambda x : x['site']=='www.fanfiction.net', books))
+
         options['version'] = self.version
         logger.debug(self.version)
         
@@ -815,6 +818,18 @@ class FanFictionDownLoaderPlugin(InterfaceAction):
         
         options['personal.ini'] = get_ffdl_personalini()
         adapter = get_ffdl_adapter(url,fileform)
+        # reduce foreground sleep time for ffnet when few books.
+        if 'ffnetcount' in options and \
+                adapter.getConfig('tweak_fg_sleep') and \
+                adapter.getSiteDomain() == 'www.fanfiction.net':
+            minslp = float(adapter.getConfig('min_fg_sleep'))
+            maxslp = float(adapter.getConfig('max_fg_sleep'))
+            dwnlds = float(adapter.getConfig('max_fg_sleep_at_downloads'))
+            m = (maxslp-minslp) / (dwnlds-1)
+            b = minslp - m
+            slp = min(maxslp,m*float(options['ffnetcount'])+b)
+            #print("m:%s b:%s = %s"%(m,b,slp))
+            adapter.set_sleep(slp)
 
         ## three tries, that's enough if both user/pass & is_adult needed,
         ## or a couple tries of one or the other
@@ -1576,6 +1591,13 @@ class FanFictionDownLoaderPlugin(InterfaceAction):
                         else:
                             val = val
                         if val != '':
+                            if coldef['datatype'] == 'bool':
+                                if val.lower() in ('t','true','1','yes','y'):
+                                    val = True
+                                elif val.lower() in ('f','false','0','no','n'):
+                                    val = False
+                                else:
+                                    val = None # for tri-state 'booleans'. Yes/No/Null
                             db.set_custom(book_id, val, label=label, commit=False)
 
                     if flag == 'a':
@@ -1622,7 +1644,7 @@ class FanFictionDownLoaderPlugin(InterfaceAction):
 
         if 'Generate Cover' in self.gui.iactions and (book['added'] or not prefs['gcnewonly']):
 
-            logger.debug("Do Generate Cover added:%s gcnewonly:%s"%(book['added'],prefs['gcnewonly']))
+            #logger.debug("Do Generate Cover added:%s gcnewonly:%s"%(book['added'],prefs['gcnewonly']))
             
             # force a refresh if generating cover so complex composite
             # custom columns are current and correct
@@ -1775,6 +1797,7 @@ class FanFictionDownLoaderPlugin(InterfaceAction):
         book['end'] = None
         book['comment'] = '' # note this is a comment on the d/l or update.
         book['url'] = ''
+        book['site'] = ''
         book['added'] = False
         return book
     
@@ -1851,13 +1874,15 @@ class FanFictionDownLoaderPlugin(InterfaceAction):
             book['status'] = _('Not Found')
         else:
             # get normalized url or None.
-            book['url'] = self.is_good_downloader_url(url)
-            if book['url'] == None:
+            urlsitetuple = adapters.getNormalStoryURLSite(url)
+            if urlsitetuple == None:
                 book['url'] = url
                 book['comment'] = _("URL is not a valid story URL.")
                 book['good'] = False
                 book['icon']='dialog_error.png'
                 book['status'] = _('Bad URL')
+            else:
+                (book['url'],book['site'])=urlsitetuple
     
     def get_story_url(self, db, book_id=None, path=None):
         if book_id == None:

defaults.ini

@@ -1215,6 +1215,11 @@ user_agent:
 ## Change this to false to use them anyway.
 never_make_cover: true
 
+## fanfiction.net is blocking people more aggressively.  If you
+## download fewer stories less often you can likely get by with
+## reducing this sleep.
+slow_down_sleep_time:4
+
 ## Extra metadata that this adapter knows about.  See [dramione.org]
 ## for examples of how to use them.
 extra_valid_entries:reviews,favs,follows
@@ -1223,13 +1228,20 @@ extra_valid_entries:reviews,favs,follows
 ## to be romantic pairings.
 ships_label:Pairings
 
-## Date formats used by FFDL.  Published and Update don't have time.
+## Date formats used by FFDL.  Published and Update don't usually have
+## time, but they do now on ffnet.
 ## See http://docs.python.org/library/datetime.html#strftime-strptime-behavior
 ## Note that ini format requires % to be escaped as %%.
 #dateCreated_format:%%Y-%%m-%%d %%H:%%M:%%S
 datePublished_format:%%Y-%%m-%%d %%H:%%M:%%S
 dateUpdated_format:%%Y-%%m-%%d %%H:%%M:%%S
 
+## ffnet used to have a tendency to send out update notices in email
+## before all their servers were showing the update on the first
+## chapter.  It generates another server request and doesn't seem to
+## be needed lately, so now default it to off.
+check_next_chapter:false
+
 [www.fanfiktion.de]
 ## Some sites require login (or login for some rated stories) The
 ## program can prompt you, or you can save it in config.  In

fanficdownloader/adapters/__init__.py

@@ -146,6 +146,13 @@ for x in imports():
             __domain_map[site]=cls
 
 def getNormalStoryURL(url):
+    r = getNormalStoryURLSite(url)
+    if r:
+        return r[0]
+    else:
+        return None
+
+def getNormalStoryURLSite(url):
     if not getNormalStoryURL.__dummyconfig:
         getNormalStoryURL.__dummyconfig = Configuration("test1.com","EPUB")
     # pulling up an adapter is pretty low over-head.  If
@@ -153,10 +160,11 @@ def getNormalStoryURL(url):
     try:
         adapter = getAdapter(getNormalStoryURL.__dummyconfig,url)
         url = adapter.url
+        site = adapter.getSiteDomain()
         del adapter
-        return url
+        return (url,site)
     except:
-        return None;
+        return None
 
 # kludgey function static/singleton
 getNormalStoryURL.__dummyconfig = None
@@ -193,7 +201,7 @@ def getConfigSectionFor(url):
 
 def getClassFor(url):
     ## fix up leading protocol.
-    fixedurl = re.sub(r"(?i)^[htps]+[:/]+","http://",url.strip())
+    fixedurl = re.sub(r"(?i)^[htp]+(s?)[:/]+",r"http\1://",url.strip())
     if not fixedurl.startswith("http"):
         fixedurl = "http://%s"%url
     ## remove any trailing '#' locations.

fanficdownloader/adapters/adapter_fanfictionnet.py

@@ -107,28 +107,31 @@ class FanFictionNetSiteAdapter(BaseSiteAdapter):
         if "not found. Please check to see you are not using an outdated url." in data:
             raise exceptions.FailedToDownload("Error downloading Chapter: %s!  'Chapter not found. Please check to see you are not using an outdated url.'" % url)
 
-        try:
-            # rather nasty way to check for a newer chapter.  ffnet has a
-            # tendency to send out update notices in email before all
-            # their servers are showing the update on the first chapter.
+        if self.getConfig('check_next_chapter'):
             try:
-                chapcount = len(soup.find('select', { 'name' : 'chapter' } ).findAll('option'))
-            # get chapter part of url.
+                ## ffnet used to have a tendency to send out update
+                ## notices in email before all their servers were
+                ## showing the update on the first chapter.  It
+                ## generates another server request and doesn't seem
+                ## to be needed lately, so now default it to off.
+                try:
+                    chapcount = len(soup.find('select', { 'name' : 'chapter' } ).findAll('option'))
+                # get chapter part of url.
+                except:
+                    chapcount = 1
+                chapter = url.split('/',)[5]
+                tryurl = "https://%s/s/%s/%d/"%(self.getSiteDomain(),
+                                                self.story.getMetadata('storyId'),
+                                                chapcount+1)
+                logger.debug('=Trying newer chapter: %s' % tryurl)
+                newdata = self._fetchUrl(tryurl)
+                if "not found. Please check to see you are not using an outdated url." \
+                        not in newdata:
+                    logger.debug('=======Found newer chapter: %s' % tryurl)
+                    soup = bs.BeautifulSoup(newdata)
             except:
-                chapcount = 1
-            chapter = url.split('/',)[5]
-            tryurl = "https://%s/s/%s/%d/"%(self.getSiteDomain(),
-                                            self.story.getMetadata('storyId'),
-                                            chapcount+1)
-            logger.debug('=Trying newer chapter: %s' % tryurl)
-            newdata = self._fetchUrl(tryurl)
-            if "not found. Please check to see you are not using an outdated url." \
-                    not in newdata:
-                logger.debug('=======Found newer chapter: %s' % tryurl)
-                soup = bs.BeautifulSoup(newdata)
-        except:
-            pass
-        
+                pass
+            
         # Find authorid and URL from... author url.
         a = soup.find('a', href=re.compile(r"^/u/\d+"))
         self.story.setMetadata('authorId',a['href'].split('/')[2])

fanficdownloader/adapters/adapter_fimfictionnet.py

@@ -101,7 +101,7 @@ class FimFictionNetSiteAdapter(BaseSiteAdapter):
         # if "/images/missing_story.png" in data:
         #     raise exceptions.StoryDoesNotExist(self.url)
         
-        if "This story has been marked as having adult content." in data:
+        if "This story has been marked as having adult content. Please click below to confirm you are of legal age to view adult material in your country." in data:
             raise exceptions.AdultCheckRequired(self.url)
         
         if self.password:
@@ -249,18 +249,20 @@ class FimFictionNetSiteAdapter(BaseSiteAdapter):
                     value = unicode(value)
                 self.story.setMetadata(metakey, value)
 
-        #Sequel links and group links are each bundled into story_group_list containers.
-        #Rather than mess around examining the header text, which is outside the containers,
-        #one can tell the two link types apart by examining them directly.
+        ## Groups and sequels code from FaceDeer
         allGroupLists = soup.findAll('ul', {'id':'story_group_list'})
         for groupList in allGroupLists:
             for groupName in groupList.findAll('a', {'href':re.compile('^/group/')}):
                 self.story.addToList("groupsUrl", 'http://'+self.host+groupName["href"]) 
                 self.story.addToList("groups",stripHTML(groupName).replace(',', ';'))
-            for sequel in groupList.findAll('a', {'class':'story_link'}):
+
+        sequelStoryHeader = soup.find('h1', {'class':'header-stories'}, text="Sequels")
+        if not sequelStoryHeader == None:
+            sequelContainer = sequelStoryHeader.parent.parent
+            for sequel in sequelContainer.findAll('a', {'class':'story_link'}):
                 self.story.addToList("sequelsUrl", 'http://'+self.host+sequel["href"]) 
                 self.story.addToList("sequels", stripHTML(sequel).replace(',', ';'))
-
+                
         #The link to the prequel is embedded in the description text, so erring
         #on the side of caution and wrapping this whole thing in a try block.
         #If anything goes wrong this probably wasn't a valid prequel link.

fanficdownloader/adapters/adapter_test1.py

@@ -50,6 +50,7 @@ class TestSiteAdapter(BaseSiteAdapter):
     def extractChapterUrlsAndMetadata(self):
         idstr = self.story.getMetadata('storyId')
         idnum = int(idstr)
+        self.do_sleep()
 
         if idnum >= 1000:
             logger.warn("storyId:%s - Custom INI data will be used."%idstr)
@@ -296,6 +297,7 @@ Some more longer description.  "I suck at summaries!"  "Better than it sounds!"
 
     def getChapterText(self, url):
         logger.debug('Getting chapter text from: %s' % url)
+        self.do_sleep()
         if self.story.getMetadata('storyId').startswith('670') or \
                 self.story.getMetadata('storyId').startswith('672'):
             time.sleep(1.0)

fanficdownloader/adapters/base_adapter.py

@@ -76,6 +76,8 @@ class BaseSiteAdapter(Configurable):
         self.username = "NoneGiven" # if left empty, site doesn't return any message at all.
         self.password = ""
         self.is_adult=False
+
+        self.override_sleep = None
         
         self.opener = u2.build_opener(u2.HTTPCookieProcessor(),GZipProcessor())
         ## Specific UA because too many sites are blocking the default python UA.
@@ -147,8 +149,7 @@ class BaseSiteAdapter(Configurable):
 
     # Assumes application/x-www-form-urlencoded.  parameters, headers are dict()s
     def _postUrl(self, url, parameters={}, headers={}):
-        if self.getConfig('slow_down_sleep_time'):
-            time.sleep(float(self.getConfig('slow_down_sleep_time')))
+        self.do_sleep()
 
         ## u2.Request assumes POST when data!=None.  Also assumes data
         ## is application/x-www-form-urlencoded.
@@ -166,17 +167,31 @@ class BaseSiteAdapter(Configurable):
             return self.opener.open(url.replace(' ','%20'),urllib.urlencode(parameters)).read()
         else:
             return self.opener.open(url.replace(' ','%20')).read()
+
+    def set_sleep(self,val):
+        print("\n===========\n set sleep time %s\n==========="%val)
+        self.override_sleep = val
     
+    def do_sleep(self):
+        if self.override_sleep:
+            time.sleep(float(self.override_sleep))
+        elif self.getConfig('slow_down_sleep_time'):
+            time.sleep(float(self.getConfig('slow_down_sleep_time')))
+        
     # parameters is a dict()
     def _fetchUrl(self, url, parameters=None):
-        if self.getConfig('slow_down_sleep_time'):
-            time.sleep(float(self.getConfig('slow_down_sleep_time')))
+        self.do_sleep()
 
         excpt=None
         for sleeptime in [0, 0.5, 4, 9]:
             time.sleep(sleeptime)	
             try:
                 return self._decode(self._fetchUrlRaw(url,parameters))
+            except u2.HTTPError, he:
+                if he.code == 404:
+                    logger.warn("Caught an exception reading URL: %s  Exception %s."%(unicode(url),unicode(he)))
+                    excpt=he
+                    break # break out on 404
             except Exception, e:
                 excpt=e
                 logger.warn("Caught an exception reading URL: %s  Exception %s."%(unicode(url),unicode(e)))
@@ -367,7 +382,7 @@ class BaseSiteAdapter(Configurable):
             
         # Don't want body tags in chapter html--writers add them.
         # This is primarily for epub updates.
-        retval = re.sub(r"</?body>\r?\n?","",retval)
+        retval = re.sub(r"</?body[^>]*>\r?\n?","",retval)
         
         if self.getConfig("replace_br_with_p"):
             # Apply heuristic processing to replace <br> paragraph

index.html

@@ -76,7 +76,7 @@
 	    If you have any problems with this application, please
 	    report them in
 	    the <a href="http://groups.google.com/group/fanfic-downloader">FanFictionDownLoader Google Group</a>.  The
-	    <a href="http://4-4-91.fanfictiondownloader.appspot.com">Previous Version</a> is also available for you to use if necessary.
+	    <a href="http://4-4-92.fanfictiondownloader.appspot.com">Previous Version</a> is also available for you to use if necessary.
 	  </p>
 	  <div id='error'>
 	    {{ error_message }}

plugin-defaults.ini

@@ -1207,6 +1207,15 @@ never_make_cover: true
 ## reducing this sleep.
 slow_down_sleep_time:4
 
+## ffnet is sensitive to too many hits.  Users are sensitive to long
+## waits during the initial metadata collection in the foreground.
+## When used, these settings will speed up metadata downloads in the
+## foreground linearly.
+tweak_fg_sleep:true
+min_fg_sleep:1.0
+max_fg_sleep:4.0
+max_fg_sleep_at_downloads: 10
+
 ## Extra metadata that this adapter knows about.  See [dramione.org]
 ## for examples of how to use them.
 extra_valid_entries:reviews,favs,follows
@@ -1215,13 +1224,20 @@ extra_valid_entries:reviews,favs,follows
 ## to be romantic pairings.
 ships_label:Pairings
 
-## Date formats used by FFDL.  Published and Update don't have time.
+## Date formats used by FFDL.  Published and Update don't usually have
+## time, but they do now on ffnet.
 ## See http://docs.python.org/library/datetime.html#strftime-strptime-behavior
 ## Note that ini format requires % to be escaped as %%.
 #dateCreated_format:%%Y-%%m-%%d %%H:%%M:%%S
 datePublished_format:%%Y-%%m-%%d %%H:%%M:%%S
 dateUpdated_format:%%Y-%%m-%%d %%H:%%M:%%S
 
+## ffnet used to have a tendency to send out update notices in email
+## before all their servers were showing the update on the first
+## chapter.  It generates another server request and doesn't seem to
+## be needed lately, so now default it to off.
+check_next_chapter:false
+
 [www.fanfiktion.de]
 ## Some sites require login (or login for some rated stories) The
 ## program can prompt you, or you can save it in config.  In