Lee Revell wrote: > On Mon, 2005-12-26 at 12:09 -0600, Jason Munro wrote: > >>On 11:54:00 am 26 Dec 2005 Lee Revell <[email protected]> wrote: >> >><snip> >> >>>> Dare I say it, KMail has also been doing the Right Thing for a >>>> long time. It will only line wrap things that you insert by >>>> typing; pastes are left untouched. >>> >>>It seems that of all the popular mail clients only Thunderbird has >>>this problem. AFAICT it's impossible to make it DTRT with inline >>>patches and even if it is the fact that most users get it wrong >>>points to a serious usability/UI issue. >>> >>>Would a patch to add "Don't use Thunderbird/Mozilla Mail" to >>>SubmittingPatches be accepted? Then we can point the Mozilla >>>developers at it (they have shown zero interest in fixing the problem >>>so far) and hopefully this will light a fire under someone. I would second that patch. >>Maybe this is a stupid question but in terms of inline patches what exactly >>would be ideal behavior from a mail client for LKML patch submitters? What >>line lengths are expected to be maintained, preferred encodings, tabs vs. >>spaces, etc? I have noticed that some patch submitters append an EOF after >>the patch, while others do not. Would the ability to pull the patch from >>the message body (assuming there was an agreed upon patch termination >>string) as a separate file/download be useful? Though my client is web >>based it is quite speedy and can handle large folders as well as many >>desktop clients IMHO. I would gladly implement specific features to make >>patch submission for LKML compliant. > > > The specifics do not matter. It does not even have to do what we want > by default when you paste or insert text. There just has to be SOME way > (well, some reasonable way - a global config option is not reasonable) > to insert a text file and paste from the clipboard as-is, no tab->space > conversion, no line wrapping, nothing. And mozilla only does the line-wrapping (with no way that I can find to switch it off). It doesn't do tab->space conversion, that usually (in my experience) results from c&p'ing from an [axe]term which outputs spaces instead of tabs to begin with (well, it does represent a character matrix so I don't really see another way). Ideally (imho) one would like the 'changelog' part to be line-wrapped (to keep it from running into oblivion) but the patch part to be left "as is". There is also the [PATCH] subject prefix and signed-of-by requirements. The only other recommendation (that I recall) is that the changelog and the patch be seperated by '---' - but since this is part of the initial output of the diff command this is done implicitly. I've looked at a few clients and it seems I'm stuck with mozilla for at least a while. Whilst probably the buggiest client there is it does look like it's the best suited for what I want. I might switch to FireFox (which iirc does have an "insert file" feature - which might also solve this problem). For the moment though I'm quickly hacking together a bash script that wraps the sendmail binary that can be used specifically for submitting patches (the intent is to perform certain checks for Signed-of-by lines, correct [PATCH] subject and so forth). If anybody else is interrested I'd be more than happy to share (albeit I suspect the usefullness will be seriously limited). Jaco -- There are only 10 kinds of people in this world, those that understand binary and those that don't. http://www.kroon.co.za/
Attachment:
smime.p7s
Description: S/MIME Cryptographic Signature
- Follow-Ups:
- Re: recommended mail clients
- From: Jeff Garzik <[email protected]>
- Re: recommended mail clients
- From: "Randy.Dunlap" <[email protected]>
- Re: recommended mail clients
- From: Lee Revell <[email protected]>
- Re: recommended mail clients
- From: Lee Revell <[email protected]>
- Re: recommended mail clients
- References:
- [PATCH] ati-agp suspend/resume support (try 2)
- From: Jaco Kroon <[email protected]>
- Re: recommended mail clients [was] [PATCH] ati-agp suspend/resume support (try 2)
- From: Jaco Kroon <[email protected]>
- Re: recommended mail clients [was] [PATCH] ati-agp suspend/resume support (try 2)
- From: Steven Rostedt <[email protected]>
- Re: recommended mail clients [was] [PATCH] ati-agp suspend/resume support (try 2)
- From: Alistair John Strachan <[email protected]>
- Re: recommended mail clients [was] [PATCH] ati-agp suspend/resume support (try 2)
- From: Lee Revell <[email protected]>
- Re: recommended mail clients [was] [PATCH] ati-agp suspend/resume support (try 2)
- From: "Jason Munro" <[email protected]>
- Re: recommended mail clients [was] [PATCH] ati-agp suspend/resume support (try 2)
- From: Lee Revell <[email protected]>
- [PATCH] ati-agp suspend/resume support (try 2)
- Prev by Date: Re: [patch 0/9] mutex subsystem, -V4
- Next by Date: Re: [PATCH] SLAB - have index_of bug at compile time.
- Previous by thread: Re: recommended mail clients [was] [PATCH] ati-agp suspend/resume support (try 2)
- Next by thread: Re: recommended mail clients
- Index(es):