translation messages questions

[Mariano Absatz]

Hi Julian,

I wanted to do a small revamp to the Spanish translations and I have a couple
of questions...

Some of the reports are complete text messages, but without headers:
- deleted.content.message.txt
- deleted.filename.message.txt
- deleted.virus.message.txt
- disinfected.report.txt
how do you create the headers for this messages? The point is, in all of them
I'm using accented characters (in Spanish) in plain ISO-8859-1 8bit (no
encoding).

Will these messages eventually get the following headers?:
MIME-Version: 1.0
Content-type: text/plain; charset=ISO-8859-1
Content-transfer-encoding: 8BIT

or will they get Q-P encoded and get the following headers?:
MIME-Version: 1.0
Content-type: text/plain; charset=ISO-8859-1
Content-transfer-encoding: Quoted-Printable

or should I manually Q-P encode them and will they get the latter headers?




Other reports are complete messages with headers:
- sender.content.report.txt
- sender.error.report.txt
- sender.filename.report.txt
- sender.spam.rbl.report.txt
- sender.spam.report.txt
- sender.spam.sa.report.txt
- sender.tcp.report.txt (this one I have to translate yet)
- sender.virus.report.txt
- stored.content.message.txt
- stored.filename.message.txt
- stored.virus.message.txt

So here I can put the headers I want... alas, checking them I see that I
said:
Content-transfer-encoding: Quoted-Printable
but put the accented characters in 8BIT :-(

The easiest (and clearest) fix would be to change the header to show the 8-
bit nature of the message. I know ZMailer will process this OK and if the
remote server doesn't support the 8BITMIME extension, will encode it and
change the header appropiately, but will Sendmail, Exim & Postfix do it also?
If not, I will leave the header as is and will manually Q-P encode the
messages :-(




Yet, other reports are meant to be inserted in-line in an already existing
message:
- inline.sig.html
- inline.sig.txt
- inline.spam.warning.txt
- inline.warning.html
- inline.warning.txt

HTML inline is no problem, since I'm using portable character entities for
accented characters. However, the text messages are indeed a problem... do we
know (or can we force) the headers of these messages?

The same considerations as the first group apply to these...




Last, but not least, the file:
- languages.conf
has a bunch of short phrases of standalone words...

Are all these used in generated headers? Or some go in the body of a message?

The point is, as I saw it, although headers that contain accented characters
are properly translated into Q-P as per RFC2047, this encoding is usually not
decoded by mail clients when they don't know the header (as is the case of
the X-MailScanner-xxx: headers)...

If this is so, I would probably remove all the accented characters since, at
least in Spanish, it is usually easy to understand a word although it doesn't
have the correct accented characters.

If you get to reply today, I'll try to do this right away... otherwise, I'll
try to do it on Monday morning (GMT-03:00)...

Regards.

--
Mariano Absatz
El Baby
----------------------------------------------------------
I work 40 hours a week to be this poor.