Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
| From | Eli the Bearded <*@eli.users.panix.com> |
|---|---|
| Newsgroups | comp.misc |
| Subject | Re: email threading |
| Date | 2019-04-08 01:30 +0000 |
| Organization | Some absurd concept |
| Message-ID | <eli$1904072103@qaz.wtf> (permalink) |
| References | <20190328113109.20e2a9ce.rsw@therandymon.com> <87o95nuli0.fsf_-_@violet.siamics.net> <eli$1904031259@qaz.wtf> <87v9zqtmyn.fsf@violet.siamics.net> |
In comp.misc, Ivan Shmakov <ivan@siamics.net> wrote:
> Back when I've considered Trn some 15 years ago, my chief issue
> was its lack of MIME support. Which is to say, I was expecting
> the NUA to convert whatever encoding used in the article to the
> one I use on my terminal for incoming messages; and at least to
> properly declare the one I use for outgoing ones.
Mime support exists but is slightly sucky in trn 4. Specifically it does
most parts of mime but makes zero effort to convert specified charset
into desired charset. And there are a few bugs that bite any multibyte
character encoding (namely UTF-8).
Declaring any headers you want is a simple matter of configuration.
Enforcing that the headers are accurate is not in the code, I've
modified my local version to warn on inaccurate use of "charset=UTF-8"
or on posting without that for UTF-8 content.
I live with the other bugs, but if you *need* that, trn4 won't be fun.
> Of course, as I have since embraced monoculturalism and largely
> ceased to participate in non-English Internet fora, I may as
> well give it another try.
This patch of mine makes UTF-8 display work in article bodies, headers
still have issues:
--- trn-4.0-test77/art.c.orig 2010-09-02 02:12:26.000000000 -0400
+++ trn-4.0-test77/art.c 2018-03-01 20:15:59.000000000 -0500
@ -468,7 +468,7 @@
outpos += 2;
}
else { /* other control char */
- if (dont_filter_control) {
+ if (dont_filter_control || (*bufptr & 0x80)) {
if (outputok)
putchar(*bufptr);
outpos++;
> A minor point is that with Emacs, I can alter the Gnus source and
> effect the changes immediately -- without even restarting Gnus
> proper (much less Emacs as a whole), which tends to make hacking
> a lot more fun (than going through the build-restart loop.)
I find dental work more pleasant than Emacs. Hacking C, even if a
recompile is needed, is way more my style.
The big issue, once you start diving into trn code, is how much of it is
rn code that is highly optimized for fast reading on slow network
connections. So, say, the code reads line at a time off of the NNTP
wire, decodes base64 in three (source) character blocks and displays
that applying line wrapping as needed. A UTF-8 state machine could be
shoe-horned in, for the simple case of wrapping UTF-8 lines correctly,
but the code described is already in six functions across three source
files already. And then if input is not UTF-8 but output output is, or
vice-versa, that's more code and more complicated.
The article selection stuff is worse vis-a-vis UTF-8 because of all of
the English-centric tricks it applies to provide nice shortening of
names and subjects that don't fit the column count available.
> > The code to do this is in wumpus.c, a nod to early computer gaming.
> A version of the game is still available in modern distributions,
It might be there, but that doesn't mean people use (play) it. I think
for most people only fortune is a game they use, for me I find factor to
be most handy.
> FSF associate member #7257 http://am-1.org/~ivan/
:r! /usr/games/factor 7257
7257: 3 41 59
Elijah
------
but for all it's faults, trn is really easy to port to a new unix-like system
Back to comp.misc | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
email and time to think RS Wood <rsw@therandymon.com> - 2019-03-28 11:31 -0400
Re: email and time to think Huge <Huge@nowhere.much.invalid> - 2019-03-28 15:33 +0000
Re: email and time to think Jerry Peters <jerry@example.invalid> - 2019-03-28 20:26 +0000
Re: email and time to think RS Wood <rswood@therandymon.com> - 2019-03-28 21:27 -0400
Re: email and time to think Mike Spencer <mds@bogus.nodomain.nowhere> - 2019-03-29 03:17 -0300
Re: email and time to think Dan Purgert <dan@djph.net> - 2019-03-29 10:05 +0000
Re: email and time to think Mike Spencer <mds@bogus.nodomain.nowhere> - 2019-03-29 16:46 -0300
Re: email and time to think RS Wood <rsw@therandymon.com> - 2019-03-29 21:49 -0400
Re: email and time to think Mike Spencer <mds@bogus.nodomain.nowhere> - 2019-03-31 17:27 -0300
Re: email and time to think Jerry Peters <jerry@example.invalid> - 2019-03-29 20:20 +0000
Re: email and time to think Rich <rich@example.invalid> - 2019-03-29 21:17 +0000
Re: email and time to think scott@alfter.diespammersdie.us (Scott Alfter) - 2019-04-01 16:28 +0000
Re: email and time to think Rich <rich@example.invalid> - 2019-04-02 01:29 +0000
Re: email and time to think RS Wood <rsw@therandymon.com> - 2019-04-02 21:00 -0400
Re: email and time to think Rich <rich@example.invalid> - 2019-04-03 14:33 +0000
email threading Ivan Shmakov <ivan@siamics.net> - 2019-04-03 16:00 +0000
Re: email threading Eli the Bearded <*@eli.users.panix.com> - 2019-04-03 16:59 +0000
Re: email threading scott@alfter.diespammersdie.us (Scott Alfter) - 2019-04-05 16:26 +0000
Re: email threading RS Wood <rsw@therandymon.com> - 2019-04-05 21:55 -0400
Re: email threading Ivan Shmakov <ivan@siamics.net> - 2019-04-07 05:15 +0000
Re: email threading Eli the Bearded <*@eli.users.panix.com> - 2019-04-08 01:30 +0000
Re: email threading Pabst Blue Ribbon <pabst@blue.ribbon> - 2019-04-10 14:18 +0000
Re: email threading Dmitry Alexandrov <321942@gmail.com> - 2019-04-11 04:12 +0300
Re: email and time to think Rich <rich@example.invalid> - 2019-03-29 21:11 +0000
Re: email and time to think Richard Kettlewell <invalid@invalid.invalid> - 2019-03-31 15:32 +0100
Re: email and time to think Oregonian Haruspex <no_email@invalid.invalid> - 2019-03-29 06:36 +0000
Re: email and time to think visiblink <visiblink@mail.invalid> - 2019-03-29 19:56 -0700
Re: email and time to think Dostoevsky <dosdoevsky@protonmail.com> - 2019-03-29 23:35 -0500
csiph-web