Decode HTML entities correctly.

[automanga.git] / manga / batoto.py

diff --git a/manga/batoto.py b/manga/batoto.py
index cab29b5..baa602f 100644 (file)
--- a/manga/batoto.py
+++ b/manga/batoto.py
@@ -1,6 +1,7 @@
 import urllib, re, BeautifulSoup
 import lib, htcache
 soup = BeautifulSoup.BeautifulSoup
+soupify = lambda cont: soup(cont, convertEntities=soup.HTML_ENTITIES)
 
 def byclass(el, name, cl):
     for ch in el.findAll(name):
@@ -28,7 +29,7 @@ class page(lib.page):
 
     def iurl(self):
         if self.ciurl is None:
-            page = soup(htcache.fetch(self.url))
+            page = soupify(htcache.fetch(self.url))
             img = nextel(page.find("div", id="full_image")).img
             self.ciurl = img["src"].encode("us-ascii")
         return self.ciurl
@@ -60,7 +61,7 @@ class chapter(lib.pagelist):
     pnre = re.compile(r"page (\d+)")
     def pages(self):
         if self.cpag is None:
-            pg = soup(htcache.fetch(self.url))
+            pg = soupify(htcache.fetch(self.url))
             cpag = []
             for opt in pg.find("select", id="page_select").findAll("option"):
                 url = opt["value"].encode("us-ascii")
@@ -83,6 +84,7 @@ class manga(lib.manga):
         self.url = url
         self.cch = None
         self.stack = []
+        self.cnames = None
 
     def __getitem__(self, i):
         return self.ch()[i]
@@ -93,7 +95,7 @@ class manga(lib.manga):
     cure = re.compile(r"/read/_/(\d+)/[^/]*")
     def ch(self):
         if self.cch is None:
-            page = soup(htcache.fetch(self.url))
+            page = soupify(htcache.fetch(self.url))
             cls = byclass(page, u"table", u"chapters_list")
             if cls.tbody is not None:
                 cls = cls.tbody
@@ -117,6 +119,26 @@ class manga(lib.manga):
             self.cch = rch
         return self.cch
 
+    def altnames(self):
+        if self.cnames is None:
+            page = soupify(htcache.fetch(self.url))
+            cnames = None
+            for tbl in page.findAll("table", attrs={"class": "ipb_table"}):
+                if tbl.tbody is not None: tbl = tbl.tbody
+                for tr in tbl.findAll("tr"):
+                    if u"Alt Names:" in tr.td.text:
+                        nls = nextel(tr.td)
+                        if nls.name != u"td" or nls.span is None:
+                            raise Exception("Weird altnames table in " + self.id)
+                        cnames = [nm.text.strip() for nm in nls.findAll("span")]
+                        break
+                if cnames is not None:
+                    break
+            if cnames is None:
+                raise Exception("Could not find altnames for " + self.id)
+            self.cnames = cnames
+        return self.cnames
+
     def __str__(self):
         return self.name
 
@@ -130,31 +152,61 @@ class library(lib.library):
 
     def byid(self, id):
         url = self.base + "comic/_/comics/" + id
-        page = soup(htcache.fetch(url))
+        page = soupify(htcache.fetch(url))
         title = page.find("h1", attrs={"class": "ipsType_pagetitle"})
         if title is None:
             raise KeyError(id)
         return manga(self, id, title.string.strip(), url)
 
-    mure = re.compile(r"/comic/_/comics/([^/]*)$")
-    def search(self, expr):
-        resp = urllib.urlopen(self.base + "forums/index.php?app=core&module=search&do=search&fromMainBar=1",
-                              urllib.urlencode({"search_term": expr, "search_app": "ccs:database:3"}))
-        try:
-            page = soup(resp.read())
-        finally:
-            resp.close()
-        none = page.find("p", attrs={"class": "no_messages"})
-        if none is not None and u"No results" in none.text:
-            return []
-        ret = []
-        for child in page.find("div", id="search_results").ol.childGenerator():
-            if isinstance(child, BeautifulSoup.Tag) and child.name == u"li":
-                info = child.find("div", attrs={"class": "result_info"})
-                url = info.h3.a["href"].encode("us-ascii")
-                m = self.mure.search(url)
+    def _search(self, pars):
+        p = 1
+        while True:
+            _pars = dict(pars)
+            _pars["p"] = str(p)
+            resp = urllib.urlopen(self.base + "search?" + urllib.urlencode(_pars))
+            try:
+                page = soupify(resp.read())
+            finally:
+                resp.close()
+            rls = page.find("div", id="comic_search_results").table
+            if rls.tbody is not None:
+                rls = rls.tbody
+            hasmore = False
+            for child in rls.findAll("tr"):
+                if child.th is not None: continue
+                if child.get("id", u"")[:11] == u"comic_rowo_": continue
+                if child.get("id") == u"show_more_row":
+                    hasmore = True
+                    continue
+                link = child.td.strong.a
+                url = link["href"].encode("us-ascii")
+                m = self.rure.search(url)
                 if m is None: raise Exception("Got weird manga URL: %r" % url)
                 id = m.group(1)
-                name = info.h3.a.string.strip()
-                ret.append(manga(self, id, name, url))
-        return ret
+                name = link.text.strip()
+                yield manga(self, id, name, url)
+            p += 1
+            if not hasmore:
+                break
+
+    rure = re.compile(r"/comic/_/([^/]*)$")
+    def search(self, expr):
+        if not isinstance(expr, unicode):
+            expr = expr.decode("utf8")
+        return self._search({"name": expr.encode("utf8"), "name_cond": "c"})
+
+    def byname(self, prefix):
+        if not isinstance(prefix, unicode):
+            prefix = prefix.decode("utf8")
+        for res in self._search({"name": prefix.encode("utf8"), "name_cond": "s"}):
+            if res.name[:len(prefix)].lower() == prefix.lower():
+                yield res
+            else:
+                for aname in res.altnames():
+                    if aname[:len(prefix)].lower() == prefix.lower():
+                        yield manga(self, res.id, aname, res.url)
+                        break
+                else:
+                    if False:
+                        print "eliding " + res.name
+                        print res.altnames()