Issue with Big5 mail spooling

[DaveSmix]

Hi all,

I'm hoping someone can help me out here, as I'm hitting a brick wall with internet searches...

In our application, broadcast emails are generated for anywhere between 1 and 6,000 subscribers - the emails themselves are either HTML or Text, with charset set to Big5 in the cfmail tag.

For the most part, this works fine however when a fairly high volume of messages are sent, we have noticed that a number emails received have corrupted characters (lots of ???s), even though the correct character encoding is used.

The same recipients have no problems with emails that are sent in smaller groups, using the exact same code.

I'm starting to think that, as part of the spooling process with a high volume of messages, CF is somehow corrupting the message characters.

Can anybody provide any thoughts/ideas what might be happening or causing this to occur? Does anyone have any documentation of the process that CFMAIL follows to spool mail messages, or any gotchas about multi-lingual systems?

EDIT: sorry, forgot to give details 🙂 - CFMX 6.1 on Windows 2k3, using Exchange 6.5 as the mail server

Cheers
Dave

[coderWil]

Can you supply some of the code?

What parts of the email are being corrupted, pulled from the database or a variable set previous to the mail loop, or static text?

I have always used the HTML setting and put the meta tag for Big-5 directly into the email so that the recipients machine can translate it locally.

[DaveSmix]

Sample code attached... (also, it only appears to affect text emails - HTML emails, so far, have not had a problem)

To, Subject and Content are all pulled from the database - when corruption occures, it affects all Big5 characters in the subject and body (western characters are unaffected)

Whats driving me crazy is that only some messages in a large broadcast are affected, but if I send the same broadcast to the same recipient as a single email, they receive it fine....

[BKBK]

And when you use charset="utf-8" ?

[Newsgroup_User]

DaveSmix wrote:
> messages are sent, we have noticed that a number emails received have corrupted
> characters (lots of ???s), even though the correct character encoding is used.

no, if you're seeing ?? then the encoding somewhere along the line is
wrong (or at least not what you think it is). it may be the data for one
set of emails is different from another.