Make get_section_url classmethod for performance.

calibre-plugin/dialogs.py

@@ -133,7 +133,7 @@ class RejectUrlEntry:
             if addreasontext:
                 self.note = self.note + ' - ' + addreasontext
 
-        if normalize:
+        if normalize and self.url:
             self.url = getNormalStoryURL(self.url)
 
         self.valid = self.url != None

fanficfare/adapters/__init__.py

@@ -250,13 +250,13 @@ def getNormalStoryURLSite(url):
 
 ## Originally defined for INI [storyUrl] sections where story URL
 ## contains a title that can change, now also used for reject list.
+## waaaay faster with classmethod.
 def get_section_url(url):
-    with lightweight_adapter(url) as adapter:
-        if adapter:
-            return adapter.get_section_url(url)
-        else:
-            logger.debug("WTF? no adapter?")
-            return None
+    cls =  _get_class_for(url)[0]
+    if cls:
+        return cls.get_section_url(url)
+    else:
+        return None
 
 def getAdapter(config,url,anyurl=False):
 

fanficfare/adapters/adapter_fanfictionnet.py

@@ -69,19 +69,26 @@ class FanFictionNetSiteAdapter(BaseSiteAdapter):
         self._setURL("https://"+self.getSiteDomain()\
                          +"/s/"+self.story.getMetadata('storyId')+"/1/"+self.urltitle)
 
-    def get_section_url(self,url):
+    ## here so getSiteURLPattern and get_section_url(class method) can
+    ## both use it.  Note adapter_fictionpresscom has one too.
+    @classmethod
+    def _get_site_url_pattern(cls):
+        return r"https?://(www|m)?\.fanfiction\.net/s/(?P<id>\d+)(/\d+)?(/(?P<title>[^/]+))?/?$"
+
+    @classmethod
+    def get_section_url(cls,url):
         ## minimal URL used for section names in INI and reject list
         ## for comparison
         # logger.debug("pre--url:%s"%url)
-        m = re.match(self.getSiteURLPattern(),url)
+        m = re.match(cls._get_site_url_pattern(),url)
         if m:
-            url = "https://"+self.getSiteDomain()\
+            url = "https://"+cls.getSiteDomain()\
                 +"/s/"+m.group('id')+"/1/"
         # logger.debug("post-url:%s"%url)
         return url
 
     def getSiteURLPattern(self):
-        return r"https?://(www|m)?\.fanfiction\.net/s/(?P<id>\d+)(/\d+)?(/(?P<title>[^/]+))?/?$"
+        return self._get_site_url_pattern()
 
     def _fetchUrl(self,url,parameters=None,extrasleep=1.0,usecache=True):
         ## ffnet(and, I assume, fpcom) tends to fail more if hit too

fanficfare/adapters/adapter_fictionpresscom.py

@@ -44,8 +44,9 @@ class FictionPressComSiteAdapter(FanFictionNetSiteAdapter):
     def getSiteExampleURLs(cls):
         return "https://www.fictionpress.com/s/1234/1/ https://www.fictionpress.com/s/1234/12/ http://www.fictionpress.com/s/1234/1/Story_Title http://m.fictionpress.com/s/1234/1/"
 
-    def getSiteURLPattern(self):
-        return r"https?://(www|m)?\.fictionpress\.com/s/(?P<id>\d+)(/\d+)?(/(?P<title>[^/]+))?/?$"
+    @classmethod
+    def _get_site_url_pattern(cls):
+        return r"https?://(www|m)?\.fanfiction\.net/s/(?P<id>\d+)(/\d+)?(/(?P<title>[^/]+))?/?$"
 
 def getClass():
     return FictionPressComSiteAdapter

fanficfare/adapters/base_adapter.py

@@ -119,7 +119,8 @@ class BaseSiteAdapter(Configurable):
         '''
         return False
 
-    def get_section_url(self,url):
+    @classmethod
+    def get_section_url(cls,url):
         '''
         For adapters that have story URLs that can change.  This is
         applied both to the story URL (saved to metadata as

fanficfare/adapters/base_xenforoforum_adapter.py

@@ -182,12 +182,13 @@ class BaseXenForoForumAdapter(BaseSiteAdapter):
 
         return (is_chapter_url,url)
 
-    def get_section_url(self,url):
+    @classmethod
+    def get_section_url(cls,url):
         ## domain is checked in configuration loop.  Can't check for
         ## storyId, because this is called before story url has been
         ## parsed.
         # logger.debug("pre--url:%s"%url)
-        url = re.sub(re.escape(self.getPathPrefix())+r'threads/.*\.(?P<id>[0-9]+)/',self.getPathPrefix()+r'threads/\g<id>/',url)
+        url = re.sub(re.escape(cls.getPathPrefix())+r'threads/.*\.(?P<id>[0-9]+)/',cls.getPathPrefix()+r'threads/\g<id>/',url)
         # logger.debug("post-url:%s"%url)
         return url