Decode HTML entities correctly.

[automanga.git] / manga / mangafox.py

diff --git a/manga/mangafox.py b/manga/mangafox.py
index 572853c..ef84eb0 100644 (file)
--- a/manga/mangafox.py
+++ b/manga/mangafox.py
@@ -1,25 +1,8 @@
-import urllib
-import BeautifulSoup
+import urllib, re
+import BeautifulSoup, json
 import lib, htcache
 soup = BeautifulSoup.BeautifulSoup
-
-class imgstream(lib.imgstream):
-    def __init__(self, url):
-        self.bk = urllib.urlopen(url)
-        self.ctype = self.bk.info()["Content-Type"]
-        self.clen = int(self.bk.info()["Content-Length"])
-
-    def fileno(self):
-        return self.bk.fileno()
-
-    def close(self):
-        self.bk.close()
-
-    def read(self, sz = None):
-        if sz is None:
-            return self.bk.read()
-        else:
-            return self.bk.read(sz)
+soupify = lambda cont: soup(cont, convertEntities=soup.HTML_ENTITIES)
 
 class page(lib.page):
     def __init__(self, chapter, stack, n, url):
@@ -29,17 +12,24 @@ class page(lib.page):
         self.manga = self.volume.manga
         self.n = n
         self.id = str(n)
+        self.name = u"Page %s" % n
         self.url = url
         self.ciurl = None
 
     def iurl(self):
         if self.ciurl is None:
-            page = soup(htcache.fetch(self.url))
+            page = soupify(htcache.fetch(self.url))
             self.ciurl = page.find("div", id="viewer").find("img", id="image")["src"]
         return self.ciurl
 
     def open(self):
-        return imgstream(self.iurl())
+        return lib.stdimgstream(self.iurl())
+
+    def __str__(self):
+        return self.name
+
+    def __repr__(self):
+        return "<mangafox.page %r.%r.%r.%r>" % (self.manga.name, self.volume.name, self.chapter.name, self.name)
 
 class chapter(lib.pagelist):
     def __init__(self, volume, stack, id, name, url):
@@ -59,7 +49,7 @@ class chapter(lib.pagelist):
 
     def pages(self):
         if self.cpag is None:
-            pg = soup(htcache.fetch(self.url + "1.html"))
+            pg = soupify(htcache.fetch(self.url + "1.html"))
             l = pg.find("form", id="top_bar").find("div", attrs={"class": "l"})
             if len(l.contents) != 3:
                 raise Exception("parse error: weird page list for %r" % self)
@@ -102,6 +92,8 @@ def nextel(el):
             return el
 
 class manga(lib.manga):
+    cure = re.compile(r"/c[\d.]+/$")
+    
     def __init__(self, lib, id, name, url):
         self.lib = lib
         self.id = id
@@ -118,9 +110,9 @@ class manga(lib.manga):
 
     def vols(self):
         if self.cvol is None:
-            page = soup(htcache.fetch(self.url))
+            page = soupify(htcache.fetch(self.url))
             vls = page.find("div", id="chapters").findAll("div", attrs={"class": "slide"})
-            self.cvol = []
+            cvol = []
             for i, vn in enumerate(reversed(vls)):
                 name = vn.find("h3", attrs={"class": "volume"}).contents[0].strip()
                 vid = name.encode("utf8")
@@ -139,10 +131,15 @@ class manga(lib.manga):
                         except KeyError:
                             pass
                     url = n.a["href"].encode("us-ascii")
-                    if url[-7:] != "/1.html":
+                    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)], chid, name, url[:-6]))
-                self.cvol.append(vol)
+                    vol.ch.append(chapter(vol, vol.stack + [(vol, o)], chid, name, url))
+                cvol.append(vol)
+            self.cvol = cvol
         return self.cvol
 
     def __str__(self):
@@ -159,7 +156,7 @@ class library(lib.library):
         self.base = "http://mangafox.me/"
 
     def alphapage(self, pno):
-        page = soup(htcache.fetch(self.base + ("directory/%i.htm?az" % pno)))
+        page = soupify(htcache.fetch(self.base + ("directory/%i.htm?az" % pno)))
         ls = page.find("div", id="mangalist").find("ul", attrs={"class": "list"}).findAll("li")
         ret = []
         ubase = self.base + "manga/"
@@ -173,7 +170,7 @@ class library(lib.library):
         return ret
 
     def alphapages(self):
-        page = soup(htcache.fetch(self.base + "directory/?az"))
+        page = soupify(htcache.fetch(self.base + "directory/?az"))
         ls = page.find("div", id="mangalist").find("div", id="nav").find("ul").findAll("li")
         return int(ls[-2].find("a").string)
 
@@ -211,9 +208,17 @@ class library(lib.library):
             ls = self.alphapage(pno)
             i = 0
 
+    def search(self, expr):
+        resp = urllib.urlopen(self.base + ("ajax/search.php?term=%s" % urllib.quote(expr)))
+        try:
+            rc = json.load(resp)
+        finally:
+            resp.close()
+        return [manga(self, id.encode("utf8"), name, self.base + ("manga/%s/" % id.encode("utf8"))) for num, name, id, genres, author in rc]
+
     def byid(self, id):
         url = self.base + ("manga/%s/" % id)
-        page = soup(htcache.fetch(url))
+        page = soupify(htcache.fetch(url))
         if page.find("div", id="title") is None:
             # Assume we got the search page
             raise KeyError(id)