|
Home > Archive > Debian Developers > January 2005 > Depends: and commands used in maintainer scripts
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Depends: and commands used in maintainer scripts
|
|
| Frank Küster 2005-01-26, 7:51 am |
| Hi,
what is the reason why in the following sentence in Policy:
,----
| The Depends field should also be used if the postinst, prerm or postrm
| scripts require the package to be present in order to run.
`----
the word "should" is used, not "must"? I'm asking here (not on -policy)
because I assume there must be a technical reason for it, but I really
can't think of any.
If a package is missing a Depends, and therefore will routinely fail in
prerm or postrm --remove, isn't that a release-critical bug?
TIA, Frank
--=20
Frank K=FCster
Inst. f. Biochemie der Univ. Z=FCrich
Debian Developer
| |
| Joel Aelwyn 2005-01-26, 6:03 pm |
| On Wed, Jan 26, 2005 at 11:32:19AM +0100, Frank Küster wrote:
> Hi,
>
> what is the reason why in the following sentence in Policy:
>
> ,----
> | The Depends field should also be used if the postinst, prerm or postrm
> | scripts require the package to be present in order to run.
> `----
>
> the word "should" is used, not "must"? I'm asking here (not on -policy)
> because I assume there must be a technical reason for it, but I really
> can't think of any.
>
> If a package is missing a Depends, and therefore will routinely fail in
> prerm or postrm --remove, isn't that a release-critical bug?
Because policy, unlike RFCs, does not use normative declarations such as
SHOULD and MUST (note the reason for uppercasing them in RFCs - to indicate
that they are, in fact, normative).
--
Joel Aelwyn <fenton@debian.org> ,''`.
: :' :
`. `'
`-
| |
| John Hasler 2005-01-26, 6:03 pm |
| Joel Aelwyn writes:
> Because policy, unlike RFCs, does not use normative declarations such as
> SHOULD and MUST...
>From debian-policy:
In the normative part of this manual, the words must, should and may, and
the adjectives required, recommended and optional, are used to
distinguish the significance of the various guidelines in this policy
document. Packages that do not conform to the guidelines denoted by must
(or required) will generally not be considered acceptable for the Debian
distribution. Non-conformance with guidelines denoted by should (or
recommended) will generally be considered a bug, but will not
necessarily render a package unsuitable for distribution. Guidelines
denoted by may (or optional) are truly optional and adherence is left to
the maintainer's discretion.
--
John Hasler
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Jeroen van Wolffelaar 2005-01-26, 6:03 pm |
| On Wed, Jan 26, 2005 at 01:21:38PM -0600, John Hasler wrote:
> Joel Aelwyn writes:
>
> From debian-policy:
> In the normative part of this manual, the words must, should and may, and
> the adjectives required, recommended and optional, are used to
> distinguish the significance of the various guidelines in this policy
> document. Packages that do not conform to the guidelines denoted by must
> (or required) will generally not be considered acceptable for the Debian
> distribution. Non-conformance with guidelines denoted by should (or
> recommended) will generally be considered a bug, but will not
> necessarily render a package unsuitable for distribution. Guidelines
> denoted by may (or optional) are truly optional and adherence is left to
> the maintainer's discretion.
As an additional note to this, the 'generally not be considered
acceptable for the Debian distribution' is for as far Sarge is
concerned, canonically defined at [1].
--Jeroen
[1] http://release.debian.org/sarge_rc_policy.txt
--
Jeroen van Wolffelaar
Jeroen@wolffelaar.nl (also for Jabber & MSN; ICQ: 33944357)
http://Jeroen.A-Eskwadraat.nl
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Goswin von Brederlow 2005-01-26, 6:03 pm |
| Joel Aelwyn <fenton@debian.org> writes:
> On Wed, Jan 26, 2005 at 11:32:19AM +0100, Frank Küster wrote:
>
> Because policy, unlike RFCs, does not use normative declarations such as
> SHOULD and MUST (note the reason for uppercasing them in RFCs - to indicate
> that they are, in fact, normative).
> --
> Joel Aelwyn <fenton@debian.org> ,''`.
It should still be must so failure to do so is a serious policy
violation (violation of a 'must' or 'required' directive).
Just my 2c,
Goswin
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Joel Aelwyn 2005-01-26, 6:03 pm |
| On Wed, Jan 26, 2005 at 01:21:38PM -0600, John Hasler wrote:
> Joel Aelwyn writes:
>
> In the normative part of this manual, the words must, should and may, and
> the adjectives required, recommended and optional, are used to
> distinguish the significance of the various guidelines in this policy
> document. Packages that do not conform to the guidelines denoted by must
> (or required) will generally not be considered acceptable for the Debian
> distribution. Non-conformance with guidelines denoted by should (or
> recommended) will generally be considered a bug, but will not
> necessarily render a package unsuitable for distribution. Guidelines
> denoted by may (or optional) are truly optional and adherence is left to
> the maintainer's discretion.
Hmmm. That contradicts what I got told the last time I filed a wishlist bug
asking for the policy to clearly indicate the normative usages. Oh well,
either times change or someone was mistaken.
I *like* normative usages. I just also like them to be obvious.
--
Joel Aelwyn <fenton@debian.org> ,''`.
: :' :
`. `'
`-
| |
| Frank Küster 2005-01-27, 7:54 am |
| X-Posting to -policy, because this might be a bug in Policy. I'm not
subscribed to -policy, please Cc me unless you keep -devel in.
Goswin von Brederlow <brederlo@informatik.uni-tuebingen.de> wrote:
> Joel Aelwyn <fenton@debian.org> writes:
>
[...][vbcol=seagreen]
> It should still be must so failure to do so is a serious policy
> violation (violation of a 'must' or 'required' directive).
Do others also think this is an error in policy? Nobody would object if
I raise severity of such a bug and address it in an NMU (which I'm going
to do for a different RC bug, anyway)?
Regards, Frank
--=20
Frank K=FCster
Inst. f. Biochemie der Univ. Z=FCrich
Debian Developer
| |
| Frank Küster 2005-01-27, 7:54 am |
| Frank K=FCster <frank@debian.org> wrote:
> Do others also think this is an error in policy? Nobody would object if
> I raise severity of such a bug and address it in an NMU (which I'm going
> to do for a different RC bug, anyway)?
I can answer the second question myself: According to
http://release.debian.org/sarge_rc_policy.txt, this is a *must* for
sarge, so the severity is RC anyway.
Regards, Frank
--=20
Frank K=FCster
Inst. f. Biochemie der Univ. Z=FCrich
Debian Developer
| |
| Andrew Suffield 2005-01-27, 7:54 am |
| On Wed, Jan 26, 2005 at 11:32:19AM +0100, Frank K?ster wrote:
> what is the reason why in the following sentence in Policy:
>
> ,----
> | The Depends field should also be used if the postinst, prerm or postrm
> | scripts require the package to be present in order to run.
> `----
>
> the word "should" is used, not "must"? I'm asking here (not on -policy)
> because I assume there must be a technical reason for it, but I really
> can't think of any.
>
> If a package is missing a Depends, and therefore will routinely fail in
> prerm or postrm --remove, isn't that a release-critical bug?
Failing to remove is a grave bug anyway. Policy doesn't really
matter. Not every possible bug is written into policy.
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -><- |
|
|
|
|
|