DresdenCodak: Fix and improve scraper

The scraper was broken due to the site layout changing. The structure is
lacking so there are many unique cases to deal with.

As the comic is separated into two storylines, one finished and one
ongoing, and a series of one-offs that don't fit in either storyline
I've put each of the series into its own subdirectory.

dosagelib/plugins/d.py

@@ -329,17 +329,78 @@ class DreamKeepersPrelude(_ParserScraper):
 
 
 class DresdenCodak(_ParserScraper):
-    url = 'http://dresdencodak.com/'
-    startUrl = url + 'cat/comic/'
-    firstStripUrl = url + '2007/02/08/pom/'
-    imageSearch = '//section[d:class("entry-content")]//img[d:class("aligncenter")]'
+    from datetime import datetime
+
+    url = "https://dresdencodak.com/"
+    firstStripUrl = url + "2005/06/08/the-tomorrow-man/"
+    imageSearch = '(//section[d:class("entry-content")]//img[d:class("size-full") and not (contains(@alt, "revious") or contains(@alt,"irst") or contains(@alt,"ext"))])[1]'
+    textSearch = '//section[d:class("entry-content")]//p[(4 < position()) and (position() < (last() - 1))]'
+    textOptional = True
     prevSearch = '//a[img[contains(@src, "prev")]]'
     latestSearch = '//a[d:class("tc-grid-bg-link")]'
     starter = indirectStarter
 
-    # Blog and comic are mixed...
-    def shouldSkipUrl(self, url, data):
-        return not data.xpath(self.imageSearch)
+    # Haven't found a better way to distinguish whether or not a page is part
+    # of Hob than by the date prefix.
+    date_format = "%Y-%m-%d"
+    hob_start = datetime.strptime("2007-02-08", date_format)
+    hob_end = datetime.strptime("2008-10-22", date_format)
+
+    pagenumber_re = compile("(?:[0-9]+-)*[^0-9]+_([0-9]+)(?:a|-1|_001)?\.jpg$")
+
+    def getPrevUrl(self, url, data):
+        # Fix skipping newest One-Off
+        if url == self.url + "2010/06/03/dark-science-01/":
+            newurl = self.url + "category/oneoffs/"
+            return self.fetchUrl(
+                newurl, self.getPage(newurl), self.latestSearch
+            )
+        return super(DresdenCodak, self).getPrevUrl(url, data)
+
+    def namer(self, image_url, page_url):
+        import os.path
+
+        filename = image_url.rsplit("/", 1)[-1]
+        # The archives are divided into three parts:
+        # Dark Science, Hob and One-Offs
+        if filename.startswith("ds"):
+            filename = filename[:2] + "_" + filename[2:]
+        elif filename == "84_new.jpg":
+            # Single anomalous page
+            filename = "ds_84.jpg"
+        elif filename == "cyborg_time.jpg":
+            filename = os.path.join("Dark Science", "84b.jpg")
+        elif "act_4" in filename:
+            filename = os.path.join("Dark Science", "80b.jpg")
+        elif "act_3" in filename:
+            filename = os.path.join("Dark Science", "38b.jpg")
+        elif "act_2" in filename:
+            filename = os.path.join("Dark Science", "18b.jpg")
+
+        if filename.startswith("ds_") or "-dark_science_" in filename:
+            # Dark Science
+            import re
+
+            pagenumber = re.match(self.pagenumber_re, filename).group(1)
+            filename = os.path.join(
+                "Dark Science", "{0:0>3}".format(pagenumber)
+            )
+        elif "/" not in filename:
+            # Hob
+            from datetime import datetime
+
+            date_prefix = page_url.rsplit("/", 5)[-5:-2]
+            date = datetime(*(int(i) for i in date_prefix))
+            if self.hob_start <= date <= self.hob_end:
+                filename = os.path.join("Hob", filename)
+            else:
+                # One-Offs
+                year_day_prefix = date.strftime("%Y-%m-%d")
+                filename = os.path.join(
+                    "One-Offs", "{0}-{1}".format(year_day_prefix, filename)
+                )
+
+        return filename
 
 
 class DrFun(_ParserScraper):