ann.py: Further improvements.

[utils.git] / ann.py

diff --git a/ann.py b/ann.py
index ee3d472..4d2c5d1 100644 (file)
--- a/ann.py
+++ b/ann.py
@@ -3,6 +3,9 @@ from urllib.parse import urljoin, urlencode
 import bs4
 soup = lambda cont: bs4.BeautifulSoup(cont, "html.parser")
 
+__all__ = ["anime", "getlist",
+           "error", "incompatible"]
+
 base = "http://www.animenewsnetwork.com/encyclopedia/"
 
 class error(Exception):
@@ -63,6 +66,8 @@ def cstr(soup):
         for el in soup:
             ret += cstr(el)
         return ret
+    elif isinstance(soup, str):
+        return soup
     elif soup is None:
         return None
     else:
@@ -145,6 +150,18 @@ class anime(object):
             return ret
         return int(ret)
 
+    @cproperty
+    def vintage(self):
+        return cstr(self._info("vintage")).strip()
+
+    @cproperty
+    def genres(self):
+        return [cstr(el) for x in (self._info("genres") or []) if isinstance(x, bs4.Tag) for el in x.findAll("a")]
+
+    @cproperty
+    def themes(self):
+        return [cstr(el) for x in (self._info("themes") or []) if isinstance(x, bs4.Tag) for el in x.findAll("a")]
+
     def __repr__(self):
         return "<ann.anime: %r (%i)>" % (self.name, self.id)
 
@@ -157,7 +174,7 @@ class anime(object):
 
 linkpat = re.compile("^/encyclopedia/anime\\.php\\?id=(\d+)$")
 def getlist(name):
-    name = s(name, "^the\s+", "")
+    name = s(name, "^(the|a)\s+", "")
     if len(name) < 1:
         raise error("list() needs a prefix of at least one character")
     fc = name[0]
@@ -169,15 +186,19 @@ def getlist(name):
     ret = []
     ldiv = afind(afind(d, "div", id="maincontent"), "div", "lst")
     for link in ldiv("a", "HOVERLINE"):
-        mn = ""
+        rawname = ""
         for el in link.font:
             if isinstance(el, str):
-                mn += el.strip()
+                rawname += el.strip()
+        mn = rawname.lower()
+        mn = s(mn, "^a\s+", "")
+        mn = mn.replace("\u014d", "ou")
+        mn = mn.replace("\u016b", "uu")
         if mn.lower().startswith(name.lower()):
             m = linkpat.match(link["href"])
             if not m:
                 raise incompatible()
             found = anime.byid(int(m.groups()[0]))
-            found.rawname = mn
+            found.rawname = rawname
             ret.append(found)
     return ret