Decode HTML entities correctly.

[automanga.git] / manga / mangafox.py

diff --git a/manga/mangafox.py b/manga/mangafox.py
index c95a77b..ef84eb0 100644 (file)
--- a/manga/mangafox.py
+++ b/manga/mangafox.py
@@ -1,33 +1,8 @@
 import urllib, re
-import BeautifulSoup
+import BeautifulSoup, json
 import lib, htcache
 soup = BeautifulSoup.BeautifulSoup
-
-class imgstream(lib.imgstream):
-    def __init__(self, url):
-        self.bk = urllib.urlopen(url)
-        ok = False
-        try:
-            if self.bk.getcode() != 200:
-                raise IOError("Server error: " + str(self.bk.getcode()))
-            self.ctype = self.bk.info()["Content-Type"]
-            self.clen = int(self.bk.info()["Content-Length"])
-            ok = True
-        finally:
-            if not ok:
-                self.bk.close()
-
-    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):
@@ -43,12 +18,12 @@ class page(lib.page):
 
     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
@@ -74,7 +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)
@@ -117,7 +92,7 @@ def nextel(el):
             return el
 
 class manga(lib.manga):
-    cure = re.compile(r"/v\d+/c[\d.]+/$")
+    cure = re.compile(r"/c[\d.]+/$")
     
     def __init__(self, lib, id, name, url):
         self.lib = lib
@@ -135,7 +110,7 @@ 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"})
             cvol = []
             for i, vn in enumerate(reversed(vls)):
@@ -181,7 +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/"
@@ -195,7 +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)
 
@@ -233,9 +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)