local: Treat dots/periods as digits in destructuring directories.

[automanga.git] / manga / mangafox.py

diff --git a/manga/mangafox.py b/manga/mangafox.py
index 0e8c2e9..c17ef31 100644 (file)
--- a/manga/mangafox.py
+++ b/manga/mangafox.py
@@ -1,8 +1,8 @@
-import urllib.request, re
+import urllib.request, urllib.parse, re
 import bs4, json
 from . import lib, htcache
 soup = bs4.BeautifulSoup
-soupify = lambda cont: soup(cont)
+soupify = lambda cont: soup(cont, "html.parser")
 
 class page(lib.page):
     def __init__(self, chapter, stack, n, url):
@@ -121,21 +121,21 @@ class manga(lib.manga):
                     raise Exception("parse error: weird volume list for %r" % self)
                 for o, ch in enumerate(reversed(cls.findAll("li"))):
                     n = ch.div.h3 or ch.div.h4
-                    name = n.a.string
+                    chid = name = n.a.string
                     for span in ch("span"):
                         try:
                             if "title" in span["class"]:
                                 name += " " + span.string
                         except KeyError:
                             pass
-                    url = n.a["href"]
+                    url = urllib.parse.urljoin(self.url, n.a["href"])
                     if url[-7:] == "/1.html":
                         url = url[:-6]
                     elif self.cure.search(url) is not None:
                         pass
                     else:
                         raise Exception("parse error: unexpected chapter URL for %r: %s" % (self, url))
-                    vol.ch.append(chapter(vol, vol.stack + [(vol, o)], name, name, url))
+                    vol.ch.append(chapter(vol, vol.stack + [(vol, o)], chid, name, url))
                 cvol.append(vol)
             self.cvol = cvol
         return self.cvol
@@ -158,17 +158,18 @@ class library(lib.library):
         self.base = "http://mangafox.me/"
 
     def alphapage(self, pno):
-        page = soupify(htcache.fetch(self.base + ("directory/%i.htm?az" % pno)))
+        abase = self.base + ("directory/%i.htm?az" % pno)
+        page = soupify(htcache.fetch(abase))
         ls = page.find("div", id="mangalist").find("ul", attrs={"class": "list"}).findAll("li")
         ret = []
-        ubase = self.base + "manga/"
         for m in ls:
             t = m.find("div", attrs={"class": "manga_text"}).find("a", attrs={"class": "title"})
             name = t.string
-            url = t["href"]
-            if url[:len(ubase)] != ubase or url.find('/', len(ubase)) != (len(url) - 1):
+            url = urllib.parse.urljoin(abase, t["href"])
+            p = url.find("/manga/")
+            if p < 0 or url.find('/', p + 7) != (len(url) - 1):
                 raise Exception("parse error: unexpected manga URL for %r: %s" % (name, url))
-            ret.append(manga(self, url[len(ubase):-1], name, url))
+            ret.append(manga(self, url[p + 7:-1], name, url))
         return ret
 
     def alphapages(self):