import/lists/autocompleteFolder

oml/meta/__init__.py

@@ -2,8 +2,7 @@
 # vi:si:et:sw=4:sts=4:ts=4
 from __future__ import division
 
-import logging
-logger = logging.getLogger('meta')
+import stdnum.isbn
 
 import abebooks
 import loc
@@ -13,6 +12,10 @@ import worldcat
 import google
 import duckduckgo
 
+import logging
+logger = logging.getLogger('meta')
+
+
 providers = [
     ('openlibrary', 'olid'),
     ('loc', 'lccn'),
@@ -32,6 +35,8 @@ def find(title, author=None, publisher=None, date=None):
     return results
 
 def lookup(key, value):
+    if not isvalid_id(key, value):
+        return {}
     data = {key: value}
     ids = [(key, value)]
     provider_data = {}
@@ -59,4 +64,13 @@ def lookup(key, value):
                 data[k_] = v_
     return data
 
+def isvalid_id(key, value):
+    if key in ('isbn10', 'isbn13'):
+        if 'isbn%d'%len(value) != key or not stdnum.isbn.is_valid(value):
+            return False
+    if key == 'asin' and len(value) != 10:
+        return False
+    if key == 'olid' and not (value.startswith('OL') and value.endswith('M')):
+        return False
+    return True
 

oml/meta/abebooks.py

@@ -9,10 +9,11 @@ import lxml.html
 import logging
 logger = logging.getLogger('meta.abebooks')
 
+base = 'http://www.abebooks.com'
+
 def get_ids(key, value):
     ids = []
     if key in ('isbn10', 'isbn13'):
-        base = 'http://www.abebooks.com'
         url = '%s/servlet/SearchResults?isbn=%s&sts=t' % (base, id)
         data = read_url(url)
         urls = re.compile('href="(/servlet/BookDetailsPL[^"]+)"').findall(data)
@@ -24,21 +25,20 @@ def get_ids(key, value):
 
 def lookup(id):
     logger.debug('lookup %s', id)
-    return {}
-
-def get_data(id):
-    info = {}
-    base = 'http://www.abebooks.com'
+    data = {}
     url = '%s/servlet/SearchResults?isbn=%s&sts=t' % (base, id)
-    data = read_url(url)
-    urls = re.compile('href="(/servlet/BookDetailsPL[^"]+)"').findall(data)
+    html = read_url(url)
+    urls = re.compile('href="(/servlet/BookDetailsPL[^"]+)"').findall(html)
+    keys = {
+        'pubdate': 'date'
+    }
     if urls:
         details = '%s%s' % (base, urls[0])
-        data = read_url(details)
-        doc = lxml.html.document_fromstring(data)
+        html = read_url(details)
+        doc = lxml.html.document_fromstring(html)
         for e in doc.xpath("//*[contains(@id, 'biblio')]"):
             key = e.attrib['id'].replace('biblio-', '')
             value = e.text_content()
-            if value and key not in ('bookcondition', 'binding'):
-                info[key] = value
-    return info
+            if value and key not in ('bookcondition', 'binding', 'edition-amz'):
+                data[keys.get(key, key)] = value
+    return data

oml/meta/duckduckgo.py

@@ -37,6 +37,6 @@ def find(title, author=None, publisher=None, date=None):
             done.add(isbn)
             if len(isbn) == 10:
                 done.add(stdnum.isbn.to_isbn13(isbn))
-            if len(isbn) == 13:
+            if len(isbn) == 13 and isbn.startswith('978'):
                 done.add(stdnum.isbn.to_isbn10(isbn))
     return results

oml/meta/lookupbyisbn.py

@@ -45,9 +45,9 @@ def lookup(id):
     }
     for key in keys:
         r[key] = find_re(data, '<span class="title">%s:</span>(.*?)</li>'% re.escape(keys[key]))
-        if r[key] == '--':
-            r[key] = ''
-        if key == 'pages' and r[key]:
+        if r[key] == '--' or not r[key]:
+            del r[key]
+        if key == 'pages' and key in r:
             r[key] = int(r[key])
     desc = find_re(data, '<h2>Description:<\/h2>(.*?)<div ')
     desc = desc.replace('<br /><br />', ' ').replace('<br /> ', ' ').replace('<br />', ' ')