# HG changeset patch
# User Igor Chubin <igor@chub.in>
# Date 1332850165 -7200
# Node ID 5a003076eb11745b9826e00c536f5969a70d6bb3
# Parent  c3adf6452eda6173fad805255218bfd59d594e6d
-w for web support (alpha)

diff -r c3adf6452eda -r 5a003076eb11 new-words.py
--- a/new-words.py	Sat Nov 12 14:03:54 2011 +0100
+++ b/new-words.py	Tue Mar 27 14:09:25 2012 +0200
@@ -36,7 +36,10 @@
             'it' : 'italian',
             'uk' : 'ukrainian',
         }
-        self.stemmer = Stemmer.Stemmer(stemmer_algorithm[language])
+        try:
+            self.stemmer = Stemmer.Stemmer(stemmer_algorithm[language])
+        except:
+            self.stemmer = None
         self.linked_words = linked_words
 
     def normalize(self, word):
@@ -44,7 +47,10 @@
         while word in self.linked_words and not word in word_chain:
             word_chain.append(word)
             word = self.linked_words[word]
-        return self.stemmer.stemWord(word.lower())
+        if self.stemmer:
+            return self.stemmer.stemWord(word.lower())
+        else:
+            return word.lower()
 
     def best_word_from_group(self, wordpairs_group):
         """Returns the word that is the most relevant to the wordpairs_group.
@@ -239,6 +245,12 @@
     dest="vocabulary_filename")
 
 parser.add_option(
+    "-w", "--web",
+    help="Web browser version",
+    action="store_true",
+    dest="web")
+
+parser.add_option(
     "-2", "--two-words",
     help="find 2 words' sequences",
     action="store_true",
@@ -571,17 +583,60 @@
             result += [lines[i]]
     return result
 
+def web_editor(output):
+    from twisted.internet import reactor
+    from twisted.web.server import Site
+    from twisted.web.static import File
+    from twisted.web.resource import Resource
+    import json
+
+    word_list = []
+
+    for o in output:
+        a = re.split('\s+', o.strip(), 2)
+        a = a + ['']*(3-len(a))
+        word_list.append({'number':a[0], 'word':a[1], 'comment':a[2]})
+    
+    print "Loaded ", len(word_list)
+
+    new_words_html = "/home/igor/hg/new-words/web"
+
+    class JSONPage(Resource):
+        isLeaf = True
+        def render_GET(self, request):
+            return json.dumps({"word_list": word_list})
+
+    class SaveJSON(Resource):
+        isLeaf = True
+        def render_POST(self, request):
+            print json.loads(request.args["selected_words"][0])
+            return json.dumps({"status": "ok"})
+
+    json_page = JSONPage()
+    save_json = SaveJSON()
+
+    resource = File(new_words_html)
+    resource.putChild("json", json_page)
+    resource.putChild("save", save_json)
+
+    factory = Site(resource)
+    reactor.listenTCP(8880, factory)
+    reactor.run()
+
+
 def filter_get_words_group_words_add_stat(args):
     vocabulary = load_vocabulary()
     notes = load_notes(notes_filenames())
 
+    input_lines = []
     if len(args) > 0:
-        if 'http://' in args[0]:
-            input_lines = readlines_from_url(args[0])
-        else:
-            input_lines = readlines_from_file(args[0])
+        for arg in args:
+            if 'http://' in arg:
+                input_lines += readlines_from_url(arg)
+            else:
+                input_lines += readlines_from_file(arg)
     else:
-        input_lines = readlines_from_stdin()
+        input_lines += readlines_from_stdin()
 
     if len(input_lines) == 0:
         print >> sys.stderr, "Nothing to do, standard input is empty, exiting."
@@ -676,6 +731,8 @@
 
     if ('non_interactive' in config or 'text_stats' in config):
         codecs.getwriter("utf-8")(sys.stdout).write("".join(output))
+    elif config.get('web', False):
+        web_editor(output)
     else:
         (_, temp1) = tempfile.mkstemp(prefix='new-word')
         (_, temp2) = tempfile.mkstemp(prefix='new-word')
@@ -759,6 +816,9 @@
 if options.no_words_grouping:
     config['no_words_grouping'] = True
 
+if options.web:
+    config['web'] = True
+
 filter_get_words_group_words_add_stat(args)
 
 #if options.function: