Sam Varshavchik wrote:
Hatuka*nezumi - IKEDA Soji writes:
sqwebmail-5.1.1-noflowed.patch:
o "wbnoflowed" user authentication-time option.
Old "flowed" format (RFC2646) is not suitable for some (e.g.
East-Asian) languages. So until most of MUAs implement current
"flowed" format (RFC3676) correctly, any option to disable "flowed"
format will be desired by users of such languages.
"wbnoflowed" user option controls behaviour of message composition
as:
0 - composing "flowed" format (default).
1 - composing "fixed" format.
2 - composing "fixed" format, and textarea of composition form
hard-wraps the text (recommended for non-flowed format).
What I would really like to see is a different approach.
I'd like to a new flag to unicode.h, and use that to set the textarea
wrapping automatically, based on the composition character set.
I have given up this approach. Because some (e.g. all of multibyte)
character sets contain multiple subsets of vary scripts (e.g. Latin, Greek,
Cyrilic, and Ideographs).
But: modern browsers that default to UTF-8 will send composed text as
UTF-8, and UTF-8 should use flowed format.
So, after thinking about it, I believe that when typing in Asian
characters you must explicitly hit enter at the end of every line, which
inserts a hard linebreak.
This is just what users of Asian characters do --- Someones prefer inserting
hard linebreak manually. But most users use auto-wrap feature of MUA which
insert hard linebreak automatically so that lines are made even on columns.
"wbnoflowed=2" option emulates the latter behaviour (*).
Now, RFC 3767 is a separate issue. I wasn't aware of this -- I'll need to
read this, when I have some free time, and see what needs to be done in
order to generate messages using RFC 3767 coding, while still being able
to display all existing RFC 2646 or RFC 3767-formatted messages.
In some (most of Asian etc.) scripts, word separator (such as SPACE) won't be
used. Linebreak may occur all character boundaries, except some limitations.
So RFC 2646 didn't fit for text of those scripts. RFC 3767 extends flowed
format to handle such situations.
By the way, note that linebreaking behaviour also shouldn't be determined by
(character set or) script. Texts in some scripts composed as either
space-separated or by-character-breakable occasionally. Examples are Hangul
and some Indic scripts.
(*) I'm planning to implement Unicode line breaking algorithm defined by
UAX#14 on SqWebMail. But it'll need at very least some weeks.
--- nezumi
8 messages in net.sourceforge.lists.courier-sqwebmailRe: [sqwebmail] patches: No flowed fo...
.patch, .patch
