Changeset 909 for metadata

Timestamp:

11/17/10 13:25:55 (14 years ago)

Author:

dietuyt

Message:

Fix for UTF-8 issues.

File:

• metadata/trunk/toolkit/scripts/lrt2cmdi.py (modified) (5 diffs)

metadata/trunk/toolkit/scripts/lrt2cmdi.py

@@ -5 +5 @@
 
 import urllib, csv, datetime, xml.etree.ElementTree as ElementTree
+from curses.ascii import ascii
 
 class CmdiFile:
@@ -96 +97 @@
                 newLanguageNode = ElementTree.Element("ISO639")
                 newCodeNode = ElementTree.Element("iso-639-%s-code" % iso639Type)
-                newCodeNode.text = isoList[language]
+                keyLang = language.encode("utf-8")
+                newCodeNode.text = isoList[keyLang]
                 newLanguageNode.append(newCodeNode)
                 languagesNode.append(newLanguageNode)
@@ -214 +216 @@
 def loadInfo():
     csvFile = csv.reader(urllib.urlopen("http://www.clarin.eu/export_resources").readlines())
+    #csvFile =[l.decode('utf-8') for l in rawCsvFile]
+    
     #csvFile = csv.reader(urllib.urlopen("resources.csv").readlines())  
     linenr = 0
@@ -224 +228 @@
             colnr = 0
             for field in fieldList:
-                newDict[linenr][fieldList[colnr].replace(" ", "_")] = l[colnr]
+                newDict[linenr][fieldList[colnr].replace(" ", "_").decode('utf-8')] = l[colnr].decode('utf-8')
                 colnr += 1 
         linenr += 1
@@ -233 +237 @@
     dictionary = dict()
     for l in csvFile:
-        dictionary[l[1]] = l[0]
+        dictionary[l[1]] = unicode(l[0])
+
     return dictionary