Collection of i18n patches

[tkikuchi]

This patch works for

1. _at_ substitution for archive.

2. MIME decoded subject and message body in admindb
pending post list.

3. i18n personalization.

4. Header/Footer charset adjusting for multiple
charset. (works for Latin-15/1 problem, I believe.)

5. i18n message held notice.

6. fix for scrubber bug (atatched message).

7. i18n checkdbs.

This collection of patch is for most recent CVS
(2.1.4b) as of 12/25/2003 and I hope this is included
in 2.1.4-release.

[tkikuchi (tkikuchi-users)]

Logged In: YES
user_id=67709

Update of patch. espacially:

1. improve in 4 above

2. Use of CJKCodecs instead of Japanese and Korean. Now we
can use Chinese.

[tkikuchi (tkikuchi-users)]

The file i18n-patch.20031226.txt was added: i18n-patch collection

[tkikuchi (tkikuchi-users)]

Logged In: YES
user_id=67709

u Uploading CJKCodecs 1.0 because their site has only CVS
and no package to download.

[tkikuchi (tkikuchi-users)]

Logged In: YES
user_id=67709

You may need patch email package to properly encode plain
text with base64. (And others for Japanese messages)

[tkikuchi (tkikuchi-users)]

The file email.patch.20031226.txt was added: email-patch (2003/12/26)

[tkikuchi (tkikuchi-users)]

Logged In: YES
user_id=67709

Sorry but CJKCodecs was too big to upload in this patch
area. You can download it from
http://mm.tkikuchi.net/CJKCodecs-1.0.tar.gz

Cheers,

[bwarsaw (bwarsaw)]

Logged In: YES
user_id=12800

Can you explain more what you mean by:

"You may need patch email package to properly encode plain
text with base64. (And others for Japanese messages)"

[bwarsaw (bwarsaw)]

Logged In: YES
user_id=12800

Okay, I see the patches to email/Charset.py and
email/Encoders.py. I think I understand the fix to
Encoders.py, but not really the fix to Charset.py. Could
you explain in more detail? Also, please provide test cases
for these changes.

[bwarsaw (bwarsaw)]

Logged In: YES
user_id=12800

See attached Encoders.py-for-py234.txt for a patch to
Encoders.py that doesn't break the Python 2.3.x test suite.

[bwarsaw (bwarsaw)]

The file Encoders.py-for-py234.txt was added: None

[bwarsaw (bwarsaw)]

Logged In: YES
user_id=12800

In Scrubber.py, can you explain more why this part is there:

+ # TK: We (Japanese) need to stringify and re-generate
the message
+ # instance because multiple charsets are used.
+ try:
+ msg = message_from_string(str(msg))
+ except UnicodeError:
+ pass

IOW, why do you need to stringify and then re-parse the
message? Can you provide an example message that this fixes?

[bwarsaw (bwarsaw)]

Logged In: YES
user_id=12800

The Utils.py patch to include oneline() had some problems:

- it didn't import email.Header or define UEMPTYSTRING
- it shouldn't use oneline as a local var if the function is
called oneline

Both are fixable.

[bwarsaw (bwarsaw)]

Logged In: YES
user_id=12800

I have applied all but the Decorate.py chunk and the
Scrubber.py chunk. For Scrubber, please answer the
questions below. For Decorate, I'm a little uncomfortable
with apply this big of a patch so late in the game.

My suggestion would be to split the remaining two patches up
into separate (new) SF tracker items. I'm closing this one.

[tkikuchi (tkikuchi-users)]

Logged In: YES
user_id=67709

Thank you and OK, Barry.
I have submitted a new patch for Scrubber.py which is more
descriptive about the problem, I believe.