[gentoo-dev] `paludis --info' is not like `emerge --info'

[gentoo-dev] `paludis --info' is not like `emerge --info'

[Jeroen Roovers]

     Hi folks,


both in #gentoo and more importantly in bug reports, people are often
asked to provide their `emerge --info', which gives a quick and
very useful overview of the most important bits found on the system that
wants support or (allegedly) exhibits a bug. Based on that information
it is usually very easy to determine whether an issue arose because of,
say, a problematic combination of packages, a configuration error or
some unsupported compiler features.

Paludis users often seem to think that `paludis --info' and `emerge
--info' are interchangeable, whereas a quick inspection reveals to
even the most casual user that the former only intends to inform about
paludis' own compile time and run time configuration, whereas the
latter includes information about a number of critical system
packages, the global USE and USE_EXPAND variables and GENTOO_MIRRORS
variable, to name a few.

In short, `paludis --info' is not a replacement, and when `emerge
--info' is asked for in a bug report, post *that*. There are many
examples of what should be included out here, so if you can't or won't
use emerge, then you'll have to copy the information manually.
Remarking that you "don't use emerge" is not a valid reason not to post
it - post both `(emerge|paludis) --info' if you must - having both
there will help greatly and is necessary for the time being[1].


Kind regards,
     jer


[1] Future development could see the release of a unified, fully
supported `$PM--info' feature or perhaps a separate tool, and
this side issue is most likely off topic on gentoo-dev@.

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Dawid Węgliński]

On Wednesday 18 of February 2009 23:22:12 Jeroen Roovers wrote:
> In short, `paludis --info' is not a replacement, and when `emerge
> --info' is asked for in a bug report, post *that*.

Hi Jeroen.
If you ask me to post a emerge --info you will get very, but very outdated 
info. Not much useful. Keep that in mind :).

Cheers,
Dawid Węgliński

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Jeroen Roovers]

On Wed, 18 Feb 2009 23:35:00 +0100
Dawid Węgliński <cla@gentoo.org> wrote:

> On Wednesday 18 of February 2009 23:22:12 Jeroen Roovers wrote:
> > In short, `paludis --info' is not a replacement, and when `emerge
> > --info' is asked for in a bug report, post *that*.
> 
> Hi Jeroen.
> If you ask me to post a emerge --info you will get very, but very
> outdated info. Not much useful. Keep that in mind :).

Then assemble the info by hand or write a replacement tool that
provides the same info. I have outlined what is lacking in `paludis
--info' and you're merely underlining that.


Kind regards,
     jer

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Ciaran McCreesh]

[-- Attachment #1: Type: text/plain, Size: 1152 bytes --]

On Wed, 18 Feb 2009 23:22:12 +0100
Jeroen Roovers <jer@gentoo.org> wrote:
> Paludis users often seem to think that `paludis --info' and `emerge
> --info' are interchangeable, whereas a quick inspection reveals to
> even the most casual user that the former only intends to inform about
> paludis' own compile time and run time configuration, whereas the
> latter includes information about a number of critical system
> packages, the global USE and USE_EXPAND variables and GENTOO_MIRRORS
> variable, to name a few.

...which is why you ask for 'paludis --info pkg', not 'paludis --info'.
That way you get accurate information about pkg, even when users have
per-package use flags configured. Much more useful than Portage's
misleading output, which just gives global configuration and
completely ignores everything else.

Heck, Paludis even tells you this if you run --info without a spec:

> No packages were specified on the command line, so detailed
> information is not available (Paludis can display detailed information
> for both installed and installable packages).

Please try harder next time.

-- 
Ciaran McCreesh

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Jeroen Roovers]

On Wed, 18 Feb 2009 22:55:06 +0000
Ciaran McCreesh <ciaran.mccreesh@googlemail.com> wrote:

> ...which is why you ask for 'paludis --info pkg', not 'paludis
> --info'.

Spread the word!


     jer

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Ciaran McCreesh]

[-- Attachment #1: Type: text/plain, Size: 369 bytes --]

On Thu, 19 Feb 2009 00:03:24 +0100
Jeroen Roovers <jer@gentoo.org> wrote:
> On Wed, 18 Feb 2009 22:55:06 +0000
> Ciaran McCreesh <ciaran.mccreesh@googlemail.com> wrote:
> > ...which is why you ask for 'paludis --info pkg', not 'paludis
> > --info'.
> 
> Spread the word!

http://git.pioto.org/gitweb/paludis.git?a=commitdiff;h=86dc61e

-- 
Ciaran McCreesh

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Stelian Ionescu]

[-- Attachment #1: Type: text/plain, Size: 770 bytes --]

On Wed, 2009-02-18 at 23:24 +0000, Ciaran McCreesh wrote:
> On Thu, 19 Feb 2009 00:03:24 +0100
> Jeroen Roovers <jer@gentoo.org> wrote:
> > On Wed, 18 Feb 2009 22:55:06 +0000
> > Ciaran McCreesh <ciaran.mccreesh@googlemail.com> wrote:
> > > ...which is why you ask for 'paludis --info pkg', not 'paludis
> > > --info'.
> > 
> > Spread the word!
> 
> http://git.pioto.org/gitweb/paludis.git?a=commitdiff;h=86dc61e

could the output of paludis --info be made a little less verbose by
eliminating repository information - perhaps except then one containing
the package's ebuild - and info about ebuild phases being executed(like
">>> Starting builtin_initmisc" etc...) ?

-- 
Stelian Ionescu a.k.a. fe[nl]ix
Quidquid latine dictum sit, altum videtur.


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Ciaran McCreesh]

[-- Attachment #1: Type: text/plain, Size: 596 bytes --]

On Thu, 19 Feb 2009 00:39:28 +0100
Stelian Ionescu <stelian.ionescu-zeus@poste.it> wrote:
> could the output of paludis --info be made a little less verbose by
> eliminating repository information - perhaps except then one
> containing the package's ebuild

That's useful information, so no.

> and info about ebuild phases being executed(like ">>> Starting
> builtin_initmisc" etc...) ?

That's there in case something goes wrong. If you unexpectedly get a
"rm blah: permission denied" or whatever, it's helpful if you know
*why* something's being removed.

-- 
Ciaran McCreesh

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Jeroen Roovers]

On Wed, 18 Feb 2009 23:24:25 +0000
Ciaran McCreesh <ciaran.mccreesh@googlemail.com> wrote:

> http://git.pioto.org/gitweb/paludis.git?a=commitdiff;h=86dc61e

Thanks. I hope it helps.


     jer

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Jeroen Roovers]

On Wed, 18 Feb 2009 22:55:06 +0000
Ciaran McCreesh <ciaran.mccreesh@googlemail.com> wrote:

> Heck, Paludis even tells you this if you run --info without a spec:
> 
> > No packages were specified on the command line, so detailed
> > information is not available (Paludis can display detailed
> > information for both installed and installable packages).

Apparently users aren't picking that up, which is why I started this
thread.

> Please try harder next time.

Who? What?


     jer

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Peter Alfredsen]

Paludis --info does not work for me. Plz2fix.
In particular, have a look at
http://bugs.gentoo.org/show_bug.cgi?id=262277#c4
Where both emerge --info and paludis --info is posted. Sometimes, more
is less. 

While it may be useful to post all the information about the
package-manager for *you*, it's not helping us (bug-wranglers) any. We
need one screenfull of text that concisely shows us the state of your
system. We really can guess a lot from the emerge --info output. We
don't really want to care which package manager you're using, apart
from a discrete text.

Also, we still get people who just post their paludis --info. Could you
please put in *bold letters* at the top of that output that it's not
the info they'll need when reporting bugs.

Thanks,
loki_val

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Ciaran McCreesh]

[-- Attachment #1: Type: text/plain, Size: 930 bytes --]

On Sat, 4 Apr 2009 15:17:30 +0200
Peter Alfredsen <loki_val@gentoo.org> wrote:
> While it may be useful to post all the information about the
> package-manager for *you*, it's not helping us (bug-wranglers) any. We
> need one screenfull of text that concisely shows us the state of your
> system.

There is no magic way of determining which bits of information are or
are not useful. 

> We really can guess a lot from the emerge --info output.

'guess'. Like how you have to guess what use flags are really being
used for the package in question, because it doesn't tell you?

> Also, we still get people who just post their paludis --info. Could
> you please put in *bold letters* at the top of that output that it's
> not the info they'll need when reporting bugs.

It's in bright pink at the bottom. If people aren't reading that, tell
them to start paying attention or buy a mac.

-- 
Ciaran McCreesh

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Thilo Bangert]

[-- Attachment #1: Type: text/plain, Size: 416 bytes --]

> 'guess'. Like how you have to guess what use flags are really being
> used for the package in question, because it doesn't tell you?

i'd like to ask the developers of package managers to standardize this. 
having <packagmanager> --info be the same no matter who you like best is 
incredibly usefull.

while we are at it, emerge --info output may or may not be made even more 
usefull...

thanks
Thilo

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Matti Bickel]

[-- Attachment #1: Type: text/plain, Size: 610 bytes --]

Thilo Bangert <bangert@gentoo.org> wrote:
> > 'guess'. Like how you have to guess what use flags are really being
> > used for the package in question, because it doesn't tell you?
> 
> i'd like to ask the developers of package managers to standardize this. 
> having <packagmanager> --info be the same no matter who you like best is 
> incredibly usefull.

++
Please, do it.
Even an educated guess is better than nothing, raising the probability
bug-wranglers can handle the bug even before it hits other devs' inboxes.
-- 
Regards, Matti Bickel
Signed/Encrypted email preferred (key 4849EC6C)

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Timothy Redaelli]

[-- Attachment #1: Type: text/plain, Size: 889 bytes --]

On Saturday 04 April 2009 18:12:09 Thilo Bangert wrote:
> > 'guess'. Like how you have to guess what use flags are really being
> > used for the package in question, because it doesn't tell you?
>
> i'd like to ask the developers of package managers to standardize this.
> having <packagmanager> --info be the same no matter who you like best is
> incredibly usefull.
>
> while we are at it, emerge --info output may or may not be made even more
> usefull...

i think it's better to develop an "emerge --info package" like "paludis --info 
package".
Since it's better to have an ebuild-scope environment than a global one (for 
package.use/bashrc)

-- 
Timothy `Drizzt` Redaelli
FreeSBIE Developer, Gentoo Developer, GUFI Staff
There are two major products that come out of Berkeley: LSD and UNIX.
We don't believe this to be a coincidence.      -- Jeremy S. Anderson

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Andrew D Kirch]

Timothy Redaelli wrote:
> On Saturday 04 April 2009 18:12:09 Thilo Bangert wrote:
>   
>>> 'guess'. Like how you have to guess what use flags are really being
>>> used for the package in question, because it doesn't tell you?
>>>       
>> i'd like to ask the developers of package managers to standardize this.
>> having <packagmanager> --info be the same no matter who you like best is
>> incredibly usefull.
>>
>> while we are at it, emerge --info output may or may not be made even more
>> usefull...
>>     
>
> i think it's better to develop an "emerge --info package" like "paludis --info 
> package".
> Since it's better to have an ebuild-scope environment than a global one (for 
> package.use/bashrc)
>
>   
I think it's best as a general rule to NEVER _EVER_ under any
circumstances emulate paludis.

Andrew

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Andrew Gaffney]

Andrew D Kirch wrote:
> I think it's best as a general rule to NEVER _EVER_ under any
> circumstances emulate paludis.

While I'm not personally a fan of paludis, it doesn't help anyone to post crap
like that to any mailing list. Please take it elsewhere. Thanks.

-- 
Andrew Gaffney                                 http://dev.gentoo.org/~agaffney/
Gentoo Linux Developer            Catalyst/Genkernel + Release Engineering Lead

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Andrew D Kirch]

Andrew Gaffney wrote:
> Andrew D Kirch wrote:
>   
>> I think it's best as a general rule to NEVER _EVER_ under any
>> circumstances emulate paludis.
>>     
>
> While I'm not personally a fan of paludis, it doesn't help anyone to post crap
> like that to any mailing list. Please take it elsewhere. Thanks.
>
>   
Why is it inappropriate to discuss the poor UI, and implementation of
software we use especially in open source?  Maybe if we're closed to
valid argument against poor methodology we can fail like everyone else
who develops closed minded (Debian) and closed source (long list here)
software.  Larry the Cow is deeply disappointed in you sir.

Why is it not appropriate to note the prolonged damage that paludis and
its associated personalities have done to the Gentoo community?  This
damage and resulting tree situation caused many to stop using Gentoo,
myself included for a time.  The diminished quality of the portage tree,
and the open hostility of those involved with paludis caused injury to
Gentoo, and it's reputation which is both significant and lasting.

Why is it not appropriate to note that commandline arguments for paludis
read like war and peace, and are a leading cause of repetitive stress
injury in the open source community?  And if a development mailing list
where the merits of paludis and portage are being debated is not the
correct forum to note the manifest shortcomings of one, then where is it? 

Andrew

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Nirbheek Chauhan]

On Mon, Apr 6, 2009 at 9:19 AM, Andrew D Kirch <trelane@trelane.net> wrote:
> Why is it inappropriate to discuss the poor UI, and implementation of
> software we use especially in open source?  Maybe if we're closed to
> valid argument against poor methodology we can fail like everyone else
> who develops closed minded (Debian) and closed source (long list here)
> software.  Larry the Cow is deeply disappointed in you sir.
>

Discuss, sure. Flamebait, no. If you flame, the other side gets a
license to flame. Let's keep things civil, and restrict ourselves to
technical critique. If you want things to prevent war, measure
yourself with the same yardstick as you expect others to be.

That will be all,

Thank you.

-- 
~Nirbheek Chauhan

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Andrew D Kirch]

Nirbheek Chauhan wrote:
> On Mon, Apr 6, 2009 at 9:19 AM, Andrew D Kirch <trelane@trelane.net> wrote:
>   
>> Why is it inappropriate to discuss the poor UI, and implementation of
>> software we use especially in open source?  Maybe if we're closed to
>> valid argument against poor methodology we can fail like everyone else
>> who develops closed minded (Debian) and closed source (long list here)
>> software.  Larry the Cow is deeply disappointed in you sir.
>>
>>     
>
> Discuss, sure. Flamebait, no. If you flame, the other side gets a
> license to flame. Let's keep things civil, and restrict ourselves to
> technical critique. If you want things to prevent war, measure
> yourself with the same yardstick as you expect others to be.
>
> That will be all,
>
> Thank you.
>
>   
I reject the premise that war should always be prevented.  I am however
concerned that criticism where criticism is due is flame baiting.

Andrew

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Richard Freeman]

Andrew D Kirch wrote:
> I reject the premise that war should always be prevented.  I am however
> concerned that criticism where criticism is due is flame baiting.
> 

The original post did not contain any constructive criticism of paludis 
at all.  It simply stated that under no circumstances should anybody 
ever adopt anything that was done first in paludis.  That is simply 
flamebait.

If you think there is a good reason not to have an emerge --info 
<package> feature by all means state it.  However, let's discuss issues 
based on their merits -- preferably their technical merits.  I fully 
acknowledge that other non-technical aspects of decisions also matter, 
but they need to be talked about constructively.

Ironically the relationship between the paludis and portage camps is 
probably the best I've seen it in the last year or two.  Maybe I just 
don't get on IRC often enough.  :)

I think that taking these steps is a good move.  Package maintainers 
should be focusing on the ebuild's environment and not the particular 
program being used to invoke the ebuild. If it becomes clear that the 
problem is that a package manager isn't complying with the PMS then by 
all means point fingers at the guilty parties.

Gentoo is about choice...

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Timothy Redaelli]

[-- Attachment #1: Type: text/plain, Size: 654 bytes --]

On Monday 06 April 2009 05:29:01 Andrew D Kirch wrote:
<cut>
> I think it's best as a general rule to NEVER _EVER_ under any
> circumstances emulate paludis.
I think that everyone may have a good idea, also Microsoft (TM).
I only take the idea. I don't care about who has it ;)
I think that it's better to have an ebuild-scope information instead of the 
global-one.
It's only my opinion, I don't want to make flame :)

-- 
Timothy `Drizzt` Redaelli
FreeSBIE Developer, Gentoo Developer, GUFI Staff
There are two major products that come out of Berkeley: LSD and UNIX.
We don't believe this to be a coincidence.      -- Jeremy S. Anderson

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[gentoo-dev] Re: `paludis --info' is not like `emerge --info'

[Duncan]

Timothy Redaelli <drizzt@gentoo.org> posted
200904061132.26894.drizzt@gentoo.org, excerpted below, on  Mon, 06 Apr
2009 11:32:21 +0200:

> I only take the idea. I don't care about who has it ;)

Who has it isn't a particular problem, but in the context you mentioned, 
whether it has been patented unfortunately is. =:^(

But thankfully I'm not aware of any Gentoo PM patents, and yes, some way 
of getting exactly what's applied to the package would be useful.  Of 
course, a simple --pretend yields the package-specific USE flags, but 
there's no way to see the effect of what's in /etc/portage/env, for 
instance (and /etc/portage/patches for those still running Ed Catmur's 
patching scripts stuff, which sure helps when one is grabbing patches off 
bugs, google, etc, to allow building a particular package with a new gcc, 
for instance!), without specifically asking.

Well, most of it can be seen by studying the emerge output logs, but 
that's package maintainer level, not something bug wranglers should have 
to do.  If there was a nice neat post-env post-patch-script report that 
summarized everything nicely for the wranglers, it would certainly help 
-- and could go a long way toward encouraging integration of something 
similar to those patch scripts directly in portage, as well.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Peter Volkov]

[-- Attachment #1: Type: text/plain, Size: 310 bytes --]

В Вск, 05/04/2009 в 22:41 +0200, Timothy Redaelli пишет:
> i think it's better to develop an "emerge --info package" 

It already exists. But regretfully it does not provide useful output
even about the package's USE flags so I suspect it was implemented for
different purposes.

-- 
Peter.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Jeroen Roovers]

On Mon, 06 Apr 2009 12:11:02 +0400
Peter Volkov <pva@gentoo.org> wrote:

> В Вск, 05/04/2009 в 22:41 +0200, Timothy Redaelli пишет:
> > i think it's better to develop an "emerge --info package" 
> 
> It already exists. But regretfully it does not provide useful output
> even about the package's USE flags so I suspect it was implemented for
> different purposes.

What would we need other than the output of, say, `emerge -vp
<package>' attached to the bottom or top?


Kind regards,
     jer

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Peter Volkov]

В Срд, 08/04/2009 в 16:36 +0200, Jeroen Roovers пишет:
> On Mon, 06 Apr 2009 12:11:02 +0400
> Peter Volkov <pva@gentoo.org> wrote:
> > В Вск, 05/04/2009 в 22:41 +0200, Timothy Redaelli пишет:
> > > i think it's better to develop an "emerge --info package" 
> > 
> > It already exists. But regretfully it does not provide useful output
> > even about the package's USE flags so I suspect it was implemented for
> > different purposes.
> 
> What would we need other than the output of, say, `emerge -vp
> <package>' attached to the bottom or top?

CFLAGS, compiler/libc version and other information that can change
since package was built but still helps to reproduce problem.


And it's really great that now we have simple tool to ask users to show
us CFLAGS. Thank you Zac.

-- 
Peter.

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Zac Medico]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Peter Volkov wrote:
> В Вск, 05/04/2009 в 22:41 +0200, Timothy Redaelli пишет:
>> i think it's better to develop an "emerge --info package" 
> 
> It already exists. But regretfully it does not provide useful output
> even about the package's USE flags so I suspect it was implemented for
> different purposes.
> 

In svn r13297 I've added support for displaying the USE similarly to
how they are displayed in the merge list:

http://sources.gentoo.org/viewcvs.py/portage?view=rev&rev=13297
- --
Thanks,
Zac
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.11 (GNU/Linux)

iEYEARECAAYFAknc91QACgkQ/ejvha5XGaOgrwCgsauJVvIqpDacLt2h4nVs1aL1
sqsAn0UctKfWz1WyyW5cmI0QUN7zrowy
=U33D
-----END PGP SIGNATURE-----

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Sebastian Pipping]

Just an idea without knowing all the details:
Could

 - "--emerge-info" and "--emerge-info-verbose"
   options be added to paludis or

 - a converter script be written to convert
   paludis output to feel like emerge output

to solves this issue?  Would that technically
be possible?  Is paludis returning a superset
of the information returned by portage?



Sebastian

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Ciaran McCreesh]

[-- Attachment #1: Type: text/plain, Size: 638 bytes --]

On Mon, 13 Apr 2009 14:21:37 +0200
Sebastian Pipping <webmaster@hartwork.org> wrote:
> Just an idea without knowing all the details:
> Could
> 
>  - "--emerge-info" and "--emerge-info-verbose"
>    options be added to paludis or
> 
>  - a converter script be written to convert
>    paludis output to feel like emerge output
> 
> to solves this issue?  Would that technically
> be possible?

Sure, but we're not going to do it. All the information provided by
paludis --info is necessary and useful. Taking any of it out would be
removing things that are required to determine the cause of a bug.

-- 
Ciaran McCreesh

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Sebastian Pipping]

Ciaran McCreesh wrote:
> Taking any of it out would be
> removing things that are required to determine the cause of a bug.

I doubt that's true in most cases:  The information needed to
finding the bug is a (possibly small) subset of the whole.

I don't see how additionally providing a stripped down version of
paludis info hurts anybody, as the subset either holds what the
wrangler needs or it doesn't, in which case they can still request
more information for the reporter.

I suggest that you take this as a usability request from the users
of your software.  Isn't that a great chance to make several people
happier every day in the future by just a few lines of code on your end?
Would they raise their voice if it did not matter?



Sebastian

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Ciaran McCreesh]

[-- Attachment #1: Type: text/plain, Size: 1394 bytes --]

On Mon, 13 Apr 2009 23:50:07 +0200
Sebastian Pipping <webmaster@hartwork.org> wrote:
> Ciaran McCreesh wrote:
> > Taking any of it out would be
> > removing things that are required to determine the cause of a bug.
> 
> I doubt that's true in most cases:  The information needed to
> finding the bug is a (possibly small) subset of the whole.

Yes, but the computer doesn't know which subset it is, just like it
doesn't know which subset of emerge --info is useful. There's nothing
in there that's never relevant.

> I don't see how additionally providing a stripped down version of
> paludis info hurts anybody, as the subset either holds what the
> wrangler needs or it doesn't, in which case they can still request
> more information for the reporter.

Or we could just provide the information the wrangler will need
straight away. Much easier.

> I suggest that you take this as a usability request from the users
> of your software.  Isn't that a great chance to make several people
> happier every day in the future by just a few lines of code on your
> end? Would they raise their voice if it did not matter?

This isn't a usability request. Making the "use paludis --info cat/pkg"
text stand out more was a usability request, and I was happy to make
that change. This is a few noxious trolls whining in an attempt to cause
trouble.

-- 
Ciaran McCreesh

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Richard Freeman]

Ciaran McCreesh wrote:
> This isn't a usability request. Making the "use paludis --info cat/pkg"
> text stand out more was a usability request, and I was happy to make
> that change. This is a few noxious trolls whining in an attempt to cause
> trouble.
> 

For those who have concerns with the paludis output - is there anything 
present in emerge --info that the paludis output omits?  I'd see that as 
a completely legitimate concern.

Also - if a better way of organizing the paludis output would make it 
easier to find what you're looking for I'd think that would be a 
constructive piece of criticism.

It does seem a bit odd to ask somebody to imitate the emerge formatting 
of the output just to say that it matches.  If somebody wants to create 
a script to do it they can post it online so that everybody who cares 
can run it and not have to look at evil paludis output...  :)

Now, if were were talking about data that was an input to some kind of 
automated routine then that would be a different story.  However, in 
that case we might start by xmlifying it or something...

[gentoo-dev] Re: `paludis --info' is not like `emerge --info'

[Duncan]

Richard Freeman <rich0@gentoo.org> posted 49E3D8DC.7090502@gentoo.org,
excerpted below, on  Mon, 13 Apr 2009 20:29:16 -0400:

> It does seem a bit odd to ask somebody to imitate the emerge formatting
> of the output just to say that it matches.  If somebody wants to create
> a script to do it they can post it online so that everybody who cares
> can run it and not have to look at evil paludis output...  :)

Actually, such a script was my first thought as well.  I could envision 
particularly cranky bug wranglers closing bugs as NEEDINFO, linking the 
script and saying reopen when the requested output is provided.

That's certainly not being particularly cooperative, but a flat "we won't 
implement it ourselves" isn't being particularly cooperative either, and 
this would be one way I could see particularly cranky bug wranglers 
handling it.

But having the script for bug wranglers to use is certainly more user 
friendly, and changing the formatting to put the familiar stuff all 
together in one spot, preferably top of the output, would be wrangler 
friendly as well. =:^)

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Thomas Anderson]

[-- Attachment #1: Type: text/plain, Size: 999 bytes --]

On Wed, Feb 18, 2009 at 11:22:12PM +0100, Jeroen Roovers wrote:
>      Hi folks,
> --info' is asked for in a bug report, post *that*. There are many
> examples of what should be included out here, so if you can't or won't
> use emerge, then you'll have to copy the information manually.
> Remarking that you "don't use emerge" is not a valid reason not to post
> it - post both `(emerge|paludis) --info' if you must - having both
> there will help greatly and is necessary for the time being[1].
> 

Folks, I'd like to direct your attention to bug #269067. In this case,
had the bugwrangler(yngwin) asked for paludis --info the bug could have
been RESO INVALID without it getting to me. I wouldn't have wasted ~30
minutes deciding that the user was being crazy.

So yes, paludis --info can come in handy, and I found it useful in this
case.

-- 
---------
Thomas Anderson
Gentoo Developer
/////////
Areas of responsibility:
AMD64, Secretary to the Gentoo Council
---------

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[Nirbheek Chauhan]

On Sun, May 10, 2009 at 2:44 AM, Thomas Anderson <gentoofan23@gentoo.org> wrote:
> Folks, I'd like to direct your attention to bug #269067. In this case,
> had the bugwrangler(yngwin) asked for paludis --info the bug could have
> been RESO INVALID without it getting to me. I wouldn't have wasted ~30
> minutes deciding that the user was being crazy.
>
> So yes, paludis --info can come in handy, and I found it useful in this
> case.
>

1. It was a paludis bug, of course paludis --info came in handy (are
you trying to jest? ;p)
2. You found it useful because you knew the syntax, and where to look
for what -- it is relevant to you. Not to everyone else
3. Also, last comment on the bug:

"The emerge --info and paludis --info I reported above are from the wrong
machine.  I'll update the results on Monday when I get back to the correct
machine.  It is a Xeon W5580 cpu which should be compiling as x86_64.  I
believe the bug is real.  Sorry for the confusion."

Oops? =p

-- 
~Nirbheek Chauhan

Re: [gentoo-dev] `paludis --info' is not like `emerge --info'

[David Leverton]

On Sunday 10 May 2009 04:23:25 Nirbheek Chauhan wrote:
> 1. It was a paludis bug, of course paludis --info came in handy (are
> you trying to jest? ;p)

It's most likely not a Paludis bug; do you really think that no-one's ever 
tried to compile Qt4 on amd64 with Paludis until now?  I'm guessing a 
misconfiguration, but we'll have to wait for the real paludis --info output 
(NOT emerge --info, because that doesn't say anything about the Paludis 
configuration) to be sure.

> 2. You found it useful because you knew the syntax, and where to look
> for what -- it is relevant to you. Not to everyone else

It's useful and relevant if and only if it contains information that helps 
diagnose the problem.  In the likely event that it's a misconfiguration, that 
applies to paludis --info, but probably not emerge --info, unless he made the 
same mistake with both.

> 3. Also, last comment on the bug:
>
> "The emerge --info and paludis --info I reported above are from the wrong
> machine.  I'll update the results on Monday when I get back to the correct
> machine.  It is a Xeon W5580 cpu which should be compiling as x86_64.  I
> believe the bug is real.  Sorry for the confusion."
>
> Oops? =p

If anything, that's a point in favour of tanderson's argument.  If the --info 
the user had posted had been right, then both the paludis and emerge output 
are equally useful, because they both indicate that it isn't an amd64 system 
at all.  On the other hand, if it turns out that on the correct system, 
Paludis is misconfigured and Portage isn't, then only paludis --info will 
help.