From a99101d95ce4057d81830c347809c6ccfd1ca24a Mon Sep 17 00:00:00 2001
From: Damien Elmes <git@ichi2.net>
Date: Wed, 29 Feb 2012 13:33:55 +0900
Subject: [PATCH] initMapping(); csv tweaks

---
 anki/importing/csvfile.py | 21 ++++++---------------
 anki/importing/noteimp.py | 15 +++++++++++----
 tests/test_importing.py   |  2 +-
 3 files changed, 18 insertions(+), 20 deletions(-)

diff --git a/anki/importing/csvfile.py b/anki/importing/csvfile.py
index 2c0018621..0ddc3cbab 100644
--- a/anki/importing/csvfile.py
+++ b/anki/importing/csvfile.py
@@ -20,7 +20,7 @@ class TextImporter(NoteImporter):
         self.tagsToAdd = []
 
     def foreignNotes(self):
-        self.sniff()
+        self.open()
         # process all lines
         log = []
         notes = []
@@ -31,12 +31,7 @@ class TextImporter(NoteImporter):
         else:
             reader = csv.reader(self.data, self.dialect, doublequote=True)
         for row in reader:
-            try:
-                row = [unicode(x, "utf-8") for x in row]
-            except UnicodeDecodeError, e:
-                raise ImportFormatError(
-                    type="encodingError",
-                    info=_("Please save in UTF-8 format. Click help for info."))
+            row = [unicode(x, "utf-8") for x in row]
             if len(row) != self.numFields:
                 log.append(_(
                     "'%(row)s' had %(num1)d fields, "
@@ -54,7 +49,7 @@ class TextImporter(NoteImporter):
         self.fileobj.close()
         return notes
 
-    def sniff(self):
+    def open(self):
         "Parse the top line and determine the pattern and number of fields."
         # load & look for the right pattern
         self.cacheFile()
@@ -79,15 +74,11 @@ class TextImporter(NoteImporter):
                 del self.data[0]
             self.updateDelimiter()
         if not self.dialect and not self.delimiter:
-            raise ImportFormatError(
-                type="encodingError",
-                info=_("Couldn't determine format of file."))
+            raise Exception("unknownFormat")
 
     def updateDelimiter(self):
         def err():
-            raise ImportFormatError(
-                type="encodingError",
-                info=_("File is not encoded in UTF-8."))
+            raise Exception("unknownFormat")
         self.dialect = None
         sniffer = csv.Sniffer()
         delims = [',', '\t', ';', ':']
@@ -123,7 +114,7 @@ class TextImporter(NoteImporter):
 
     def fields(self):
         "Number of fields."
-        self.sniff()
+        self.open()
         return self.numFields
 
     def noteFromFields(self, fields):
diff --git a/anki/importing/noteimp.py b/anki/importing/noteimp.py
index 235227ec0..603af1373 100644
--- a/anki/importing/noteimp.py
+++ b/anki/importing/noteimp.py
@@ -51,15 +51,22 @@ class NoteImporter(Importer):
         "The number of fields."
         return 0
 
-    def maybeChecksum(self, data, unique):
-        if not unique:
-            return ""
-        return fieldChecksum(data)
+    def initMapping(self):
+        flds = [f['name'] for f in self.model['flds']]
+        # truncate to provided count
+        flds = flds[0:self.fields()]
+        # if provided count is greater, pad
+        flds = flds + [None] * (self.fields() - len(flds))
+        self.mapping = flds
 
     def foreignNotes(self):
         "Return a list of foreign notes for importing."
         assert 0
 
+    def open(self):
+        "Open file and ensure it's in the right format."
+        assert 0
+
     def importNotes(self, notes):
         "Convert each card into a note, apply attributes and add to col."
         # gather checks for duplicate comparison
diff --git a/tests/test_importing.py b/tests/test_importing.py
index ccf23e5ce..d4834fb31 100644
--- a/tests/test_importing.py
+++ b/tests/test_importing.py
@@ -80,7 +80,7 @@ def test_csv():
     deck = getEmptyDeck()
     file = unicode(os.path.join(testDir, "support/text-2fields.txt"))
     i = TextImporter(deck, file)
-    i.mapping = ['Front', 'Back']
+    i.initMapping()
     i.run()
     # four problems - too many & too few fields, a missing front, and a
     # duplicate entry