GNU bug report logs - #17528
[mktemp] Not a bug.... a feature proposal

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 17528 in the body.
You can then email your comments to 17528 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Matthieu Fronton <m <at> tthieu.fr>
To: bug-coreutils <at> gnu.org
Subject: [mktemp] Not a bug.... a feature proposal
Date: Mon, 19 May 2014 15:27:50 +0200

[Message part 1 (text/plain, inline)]

Hello,

First i'd like to apologize because this message is about for a new feature
proposal, not a bug.
I send it to the only mail I have found in the man page. Sorry for that.

I just wanted to propose to add the last generate temporary file's full path
in a predefined environment variable.
I can create a patch or send a pull request but I'll first need to know
where your source code is and how can we contribute, even for these kind of
tiny little feature.

Regards,
Matthieu Fronton.

[Message part 2 (text/html, inline)]

Message #12 received at 17528-done <at> debbugs.gnu.org (full text, mbox):

From: Eric Blake <eblake <at> redhat.com>
To: Matthieu Fronton <m <at> tthieu.fr>, 17528-done <at> debbugs.gnu.org
Subject: Re: bug#17528: [mktemp] Not a bug.... a feature proposal
Date: Mon, 19 May 2014 10:08:24 -0600

[Message part 1 (text/plain, inline)]

tag 17528 notabug
thanks

On 05/19/2014 07:27 AM, Matthieu Fronton wrote:
> Hello,
> 
> First i'd like to apologize because this message is about for a new feature
> proposal, not a bug.
> I send it to the only mail I have found in the man page. Sorry for that.

One person's feature request is another person's bug for missing
functionality, so you mailed the correct place.

> 
> I just wanted to propose to add the last generate temporary file's full path
> in a predefined environment variable.

Umm, how do you propose to do that?  Child processes cannot modify the
environment of the parent.  The only way to invoke mktemp is by spawning
a child process - but that means that there is no way for mktemp to
modify a predefined environment variable.  That is, unless you are
proposing to add a new shell builtin (so that a child process is not
spawned), but then your patch would need to be against the shell (we
don't maintain the shell here in coreutils).

> I can create a patch or send a pull request but I'll first need to know
> where your source code is and how can we contribute, even for these kind of
> tiny little feature.

You are welcome to try and write a patch, but it sounds fundamentally
futile based on how environment variables work.  As such, I'm going to
close this request as not a bug, but feel free to continue replying on
this thread if you have more questions or comments.

Read this for more ideas about contributing other patches:
http://git.savannah.gnu.org/cgit/coreutils.git/tree/HACKING

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

[signature.asc (application/pgp-signature, attachment)]

Message #15 received at 17528 <at> debbugs.gnu.org (full text, mbox):

From: Pádraig Brady <P <at> draigBrady.com>
To: Matthieu Fronton <m <at> tthieu.fr>
Cc: 17528 <at> debbugs.gnu.org
Subject: Re: bug#17528: [mktemp] Not a bug.... a feature proposal
Date: Mon, 19 May 2014 17:16:34 +0100

On 05/19/2014 02:27 PM, Matthieu Fronton wrote:
> Hello,
> 
> First i'd like to apologize because this message is about for a new feature
> proposal, not a bug.
> I send it to the only mail I have found in the man page. Sorry for that.
> 
> I just wanted to propose to add the last generate temporary file's full path
> in a predefined environment variable.
> I can create a patch or send a pull request but I'll first need to know
> where your source code is and how can we contribute, even for these kind of
> tiny little feature.

That should be supported already:

  my_temp_file=$(mktemp)

thanks,
Pádraig

Message #18 received at 17528 <at> debbugs.gnu.org (full text, mbox):

From: Matthieu Fronton <m <at> tthieu.fr>
To: Pádraig Brady <P <at> draigbrady.com>
Cc: 17528 <at> debbugs.gnu.org
Subject: Re: bug#17528: [mktemp] Not a bug.... a feature proposal
Date: Tue, 20 May 2014 08:41:38 +0200

Aouch !
I now realize that's what I have done on some previous scripts.
Cannot explain what happened to me yesterday. Looks like my brain was on strike.

Thanks.
Matthieu

On Mon, May 19, 2014 at 6:16 PM, Pádraig Brady <P <at> draigbrady.com> wrote:
> On 05/19/2014 02:27 PM, Matthieu Fronton wrote:
>> Hello,
>>
>> First i'd like to apologize because this message is about for a new feature
>> proposal, not a bug.
>> I send it to the only mail I have found in the man page. Sorry for that.
>>
>> I just wanted to propose to add the last generate temporary file's full path
>> in a predefined environment variable.
>> I can create a patch or send a pull request but I'll first need to know
>> where your source code is and how can we contribute, even for these kind of
>> tiny little feature.
>
> That should be supported already:
>
>   my_temp_file=$(mktemp)
>
> thanks,
> Pádraig

Message #21 received at 17528-done <at> debbugs.gnu.org (full text, mbox):

From: Matthieu Fronton <m <at> tthieu.fr>
To: Eric Blake <eblake <at> redhat.com>
Cc: 17528-done <at> debbugs.gnu.org
Subject: Re: bug#17528: [mktemp] Not a bug.... a feature proposal
Date: Tue, 20 May 2014 08:50:29 +0200

On Mon, May 19, 2014 at 6:08 PM, Eric Blake <eblake <at> redhat.com> wrote:
> tag 17528 notabug
> thanks
>
> On 05/19/2014 07:27 AM, Matthieu Fronton wrote:
>> Hello,
>>
>> First i'd like to apologize because this message is about for a new feature
>> proposal, not a bug.
>> I send it to the only mail I have found in the man page. Sorry for that.
>
> One person's feature request is another person's bug for missing
> functionality, so you mailed the correct place.
>
>>
>> I just wanted to propose to add the last generate temporary file's full path
>> in a predefined environment variable.
>
> Umm, how do you propose to do that?  Child processes cannot modify the
> environment of the parent.  The only way to invoke mktemp is by spawning
> a child process - but that means that there is no way for mktemp to
> modify a predefined environment variable.  That is, unless you are
> proposing to add a new shell builtin (so that a child process is not
> spawned), but then your patch would need to be against the shell (we
> don't maintain the shell here in coreutils).
>

I didn't realise this at first but sounds obvious now I have switch my
brain back on.

Thanks
Matthieu

> --
> Eric Blake   eblake redhat com    +1-919-301-3266
> Libvirt virtualization library http://libvirt.org
>

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.