Unwanted formatting? language info? in plain text messages

[kentk]

I have a message that Outlook 2003 calls a "plain text message", but that
displays in at least two distinct fonts.

I see what appears to be MS PGothic and Courier New in the same plain text
message.

Guessing at relevant details...
The message encoding is UTF8. International Fonts->Default encoding:
Japanese. Proportional font for Japanese: MS PGothic. Composing / reading
plain text: Courier New.

The same substring, "lease" for instance is Courier New in one paragraph,
and MS PGothic in another.

Does this mean that the two substrings "lease" are spelled out in different
character spaces in UTF8?

The above guess doesn't explain things away completely in that the fonts
stop appearing differently when I change the compose / read plain text font
from the default.

Any insight would be appreciated.
-Ken

 

[Pat Willener]

Is this an incoming or outgoing message?

I am also guessing here: could it be that single-byte characters are
displayed with Courier New, and multi-byte chars in MS PGothic?

Could it be that the two instances of the word could be using
single-byte in one (lease), and double-byte in the other (ｌｅï½ï½“ï½…)?

 

[kentk]

Thanks for deciphering an answer out of that. You're right! One lower case
L was character l decimal, whereas the other is ｌ decimal.

Oddly, the characters that appear MS PGothic in the message still identify
as a 108 L, instead of the 65356, which is a Japanese, half-width lower case
L in UTF-8 space.

I suppose Outlook / Exchange keeps more information than it sends, even in a
"plain text" message. Interestingly enough, the L displaying in the wrong
font identifies as a 108 L when copy / pasted out of the reading pane,
forwarded message, etc.

I had to generate a l and a ｌ in another plain text message and compare the
two against the message where this behavior was spotted in order to get a
plausible explanation.

....makes me wonder why I get different information when copy / pasting than
I get visually. If the message contains the characters that I'm copy /
pasting out of them, there's no reason for the fonts to appear different on
characters that are the same character code.

 

[Pat Willener]

Copy/Paste does many mysterious things - I have never been able to
figure out what I will get. The safest way (I know) to get a good result
is to temporarily paste the clipboard contents into Notepad, then copy
it out of there. Copying from the reading pane may also produce
unpredictable results.

 

[kentk]

True, but I've only noticed unexpected behavior with regards to formatting.
I think that in this case, it's copying and pasting as you'd expect. It's
just that the text is rendering in an unexpected font in Outlook.

I should also correct what I wrote previously. It turns out that there are
no half-width roman characters in the Japanese character space in UTF. The
characters that I extracted the codes for are actually in full width latin
space. I confused the full-width for some imaginary half-width because I was
looking only at the lowercase "l" glyph.

Now, I seem to understand how to reproduce this phenomenon on my test rig,
although it's not working on the box that I'm typing on right now.

Set tools->options->mail format-> encoding options to: Auto select + Unicde
UTF-8.

Compose a new rich text message, type a few characters, then enter a glyph
like this ó, and add more text afterwards. Convert that to plain text, and I
end up with MS PGothic looking text before the ó, and Courier New after the ó
in a plain text message.

The phenomenon is profile specific, apparently. Popping my test message
open on the text box with a dummy account yields two fonts, while popping the
test message open on the text box with my profile yields a single font (100%
plain text font, instead of going from International font -> glyph -> plain
text font).

Guess I need to hunt more for whatever setting is contributing to / causing
this difference

Any guesses would be totally welcome. I've exausted my first round of
guesses already.

 

[Pat Willener]

Not much more guesses, except that you mention Rich Text - meaning RTF?
Does the same thing happen if you start in HTML format, then convert to
Plain Text?

Does anything similar happen if you only use Plain Text?

 

[kentk]

Dang. Thanks for thinking with me, though. I appreciate it.

Yes, I do mean RTF when I say Rich Text.

HTML messages don't demonstrate this behavior. It's just Rich Text -> Plain
Text messages that do this.

 

[Pat Willener]

Another reason to stop using RTF. HTML is universally accepted, uses
less space, but much more flexibility and options.

 

[kentk]

Ah, good point. I'm not sure about html being universally accepted, but its
handling in Outlook 2003's message editor does seem more benign.

Also found out yesterday that a conversion from rich text to plain text is
not necessary to reproduce this odd behavior. Retrieving a draft from the
drafts folder / looking through the sent items folder can also yield rich
text / plain text messages where the issue only manifested itself after
saving the draft / sending the message.

I'll see if switching to html somehow circumvents this odd behavior in my
environment.