Implement Oglaf's strange navigation (fixes #33)

(also should fix wummel#91)

dosagelib/plugins/o.py

@@ -22,19 +22,32 @@ class OctopusPie(_ParserScraper):
     help = 'Index format: yyyy-mm-dd/nnn-strip-name'
 
 
-class Oglaf(_BasicScraper):
+class Oglaf(_ParserScraper):
     url = 'http://oglaf.com/'
     stripUrl = url + '%s/'
-    imageSearch = compile(tagre("img", "src", r'(http://media\.oglaf\.com/comic/[^"]+)', before="strip"))
+    imageSearch = '//img[@id="strip"]'
-    prevSearch = (
+    # search for "previous story" only
-      # first search for "next page" URLs
+    prevSearch = '//a[div[@id="pvs"]]'
-      compile(tagre("a", "href", r'(/[^"]+/\d+/)') + tagre("div", "id", "nx")),
+    # search for "next page"
-      # then for "prev story"
+    nextSearch = '//a[div[@id="nx"]]'
-      compile(tagre("a", "href", r'(/[^"]+)') + tagre("div", "id", "pvs?")),
+    multipleImagesPerStrip = True
-    )
-    help = 'Index format: stripname'
     adult = True
 
+    def fetchUrls(self, url, data, search):
+        urls = []
+        urls.extend(super(Oglaf, self).fetchUrls(url, data, search))
+        if search == self.imageSearch:
+            try:
+                nexturls = self.fetchUrls(url, data, self.nextSearch)
+            except ValueError:
+                pass
+            else:
+                while nexturls and nexturls[0].startswith(url):
+                    data = self.getPage(nexturls[0])
+                    urls.extend(super(Oglaf, self).fetchUrls(nexturls, data, search))
+                    nexturls = self.fetchUrls(url, data, self.nextSearch)
+        return urls
+
 
 class OhJoySexToy(_WordPressScraper):
     url = 'http://www.ohjoysextoy.com/'