Bug#555978: debian-policy: Forbid duplicate fields in control files
Russ Allbery
Version: 3.8.3.0
Severity: wishlist
Currently, so far as I can see, nothing in the general discussion of the
syntax of control files prohibits repetition of the same field name in
one paragraph. For example:
Package: foo
Package: bar
However, I believe this should always be a syntax error. Lintian has a
check for this:
Tag: debian-control-with-duplicate-fields
Severity: serious
Certainty: certain
Info: One of the paragraphs of your debian/control contains the same
field more than once. This can lead to unexpected behaviour in
<tt>dpkg</tt> and <tt>apt</tt>.
(I haven't checked the assertion about unexpected behavior) and this
check is among those that ftpmaster uses to reject packages. I think
we should prohibit such duplicate fields in Policy as well.
-- System Information:
Debian Release: squeeze/sid
APT prefers testing
APT policy: (990, 'testing'), (500, 'unstable'), (1, 'experimental')
Architecture: i386 (i686)
Kernel: Linux 2.6.30-2-686-bigmem (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
debian-policy depends on no packages.
debian-policy recommends no packages.
Versions of packages debian-policy suggests:
ii doc-base 0.9.5 utilities to manage online documen
-- no debconf information
--
To UNSUBSCRIBE, email to debian-bugs-...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org
Raphael Hertzog
> Tag: debian-control-with-duplicate-fields
> Severity: serious
> Certainty: certain
> Info: One of the paragraphs of your debian/control contains the same
> field more than once. This can lead to unexpected behaviour in
> <tt>dpkg</tt> and <tt>apt</tt>.
>
> (I haven't checked the assertion about unexpected behavior) and this
$ LANG=C dpkg-gencontrol -plibvte9
dpkg-gencontrol: error: syntax error in debian/control at line 29: duplicate field Architecture found
It's checked at build time at least.
> check is among those that ftpmaster uses to reject packages. I think
> we should prohibit such duplicate fields in Policy as well.
OK for me.
Cheers,
--
Raphaël Hertzog
Russ Allbery
> Currently, so far as I can see, nothing in the general discussion of the
> syntax of control files prohibits repetition of the same field name in
> one paragraph. For example:
> Package: foo
> Package: bar
> However, I believe this should always be a syntax error. Lintian has a
> check for this:
> Tag: debian-control-with-duplicate-fields
> Severity: serious
> Certainty: certain
> Info: One of the paragraphs of your debian/control contains the same
> field more than once. This can lead to unexpected behaviour in
> <tt>dpkg</tt> and <tt>apt</tt>.
> (I haven't checked the assertion about unexpected behavior) and this
> check is among those that ftpmaster uses to reject packages. I think we
> should prohibit such duplicate fields in Policy as well.
dpkg-dev checks this at build time, so this definitely seems to be the
right move. Here is a patch.
Objections or seconds?
diff --git a/policy.sgml b/policy.sgml
index 87b9795..99ab0ff 100644
--- a/policy.sgml
+++ b/policy.sgml
@@ -2398,6 +2398,11 @@ Package: libc6
</p>
<p>
+ Each paragraph may contain at most one instance of a particular
+ field name.
+ </p>
+
+ <p>
Many fields' values may span several lines; in this case
each continuation line must start with a space or a tab.
Any trailing spaces or tabs at the end of individual
--
Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/>
Emilio Pozuelo Monfort
On 11/06/10 18:58, Russ Allbery wrote:
> dpkg-dev checks this at build time, so this definitely seems to be the
> right move. Here is a patch.
>
> Objections or seconds?
>
> diff --git a/policy.sgml b/policy.sgml
> index 87b9795..99ab0ff 100644
> --- a/policy.sgml
> +++ b/policy.sgml
> @@ -2398,6 +2398,11 @@ Package: libc6
> </p>
>
> <p>
> + Each paragraph may contain at most one instance of a particular
> + field name.
> + </p>
> +
> + <p>
> Many fields' values may span several lines; in this case
> each continuation line must start with a space or a tab.
> Any trailing spaces or tabs at the end of individual
>
Seconded,
Emilio
Julien Cristau
> Objections or seconds?
>
> diff --git a/policy.sgml b/policy.sgml
> index 87b9795..99ab0ff 100644
> --- a/policy.sgml
> +++ b/policy.sgml
> @@ -2398,6 +2398,11 @@ Package: libc6
> </p>
>
> <p>
> + Each paragraph may contain at most one instance of a particular
> + field name.
> + </p>
> +
> + <p>
> Many fields' values may span several lines; in this case
> each continuation line must start with a space or a tab.
> Any trailing spaces or tabs at the end of individual
>
Seconded.
Cheers,
Julien
Jakub Wilk
>diff --git a/policy.sgml b/policy.sgml
>index 87b9795..99ab0ff 100644
>--- a/policy.sgml
>+++ b/policy.sgml
>@@ -2398,6 +2398,11 @@ Package: libc6
> </p>
>
> <p>
>+ Each paragraph may contain at most one instance of a particular
>+ field name.
>+ </p>
>+
>+ <p>
> Many fields' values may span several lines; in this case
> each continuation line must start with a space or a tab.
> Any trailing spaces or tabs at the end of individual
Seconded.
--
Jakub Wilk
Charles Plessy
>
> diff --git a/policy.sgml b/policy.sgml
> index 87b9795..99ab0ff 100644
> --- a/policy.sgml
> +++ b/policy.sgml
> @@ -2398,6 +2398,11 @@ Package: libc6
> </p>
>
> <p>
> + Each paragraph may contain at most one instance of a particular
> + field name.
> + </p>
> +
> + <p>
> Many fields' values may span several lines; in this case
> each continuation line must start with a space or a tab.
> Any trailing spaces or tabs at the end of individual
Hi Russ,
as a non-native speaker, I have difficulties with the use of 'may' in your
patch: if fields may be unique, they also may be not unique, so what is the
message in this sentence? It does not give me the impression that the goal
is to discourage the use of the same field name twice in the same paragraph.
How about “A paragraph should not contain data fields having the same name.”
Have a nice day,
--
Charles Plessy
Illkirch, France
Emilio Pozuelo Monfort
> as a non-native speaker, I have difficulties with the use of 'may' in your
> patch: if fields may be unique, they also may be not unique, so what is the
> message in this sentence? It does not give me the impression that the goal
> is to discourage the use of the same field name twice in the same paragraph.
See http://legal-dictionary.thefreedictionary.com/May
"2) in statutes, and sometimes in contracts, the word "may" must be read in
context to determine if it means an act is optional or mandatory, for it may be
an imperative."
You ought to read it as "must" in this context.
Not a native speaker either, so I may be mistaken :)
Cheers,
Emilio
Andrew McMillan
> Le Fri, Jun 11, 2010 at 09:58:28AM -0700, Russ Allbery a écrit :
> >
> > diff --git a/policy.sgml b/policy.sgml
> > index 87b9795..99ab0ff 100644
> > --- a/policy.sgml
> > +++ b/policy.sgml
> > @@ -2398,6 +2398,11 @@ Package: libc6
> > </p>
> >
> > <p>
> > + Each paragraph may contain at most one instance of a particular
> > + field name.
> > + </p>
> > +
> > + <p>
> > Many fields' values may span several lines; in this case
> > each continuation line must start with a space or a tab.
> > Any trailing spaces or tabs at the end of individual
>
> Hi Russ,
>
> as a non-native speaker, I have difficulties with the use of 'may' in your
> patch: if fields may be unique, they also may be not unique, so what is the
> message in this sentence? It does not give me the impression that the goal
> is to discourage the use of the same field name twice in the same paragraph.
>
> How about “A paragraph should not contain data fields having the same name.”
Hi Charles,
In this sense 'may' should be read as 'must', however I think that if it
causes readability issues for non-native english speakers then the word
'must' should actually be used...
"Each paragraph must contain at most one instance of a particular
field name."
Is that clearer?
Cheers,
Andrew.
--
------------------------------------------------------------------------
andrew (AT) morphoss (DOT) com +64(272)DEBIAN
Building more free and open source software for New Zealanders
------------------------------------------------------------------------
Russ Allbery
> On Sat, 2010-06-12 at 18:18 +0900, Charles Plessy wrote:
>> as a non-native speaker, I have difficulties with the use of 'may' in your
>> patch: if fields may be unique, they also may be not unique, so what is the
>> message in this sentence? It does not give me the impression that the goal
>> is to discourage the use of the same field name twice in the same paragraph.
>> How about “A paragraph should not contain data fields having the same name.”
> In this sense 'may' should be read as 'must', however I think that if it
> causes readability issues for non-native english speakers then the word
> 'must' should actually be used...
> "Each paragraph must contain at most one instance of a particular
> field name."
> Is that clearer?
Yes, indeed, that's what I meant; I'll go with that (assuming that doesn't
have a different clarity problem that I'm missing).
--
Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/>
--
Russ Allbery
> Andrew McMillan <and...@morphoss.com> writes:
>> On Sat, 2010-06-12 at 18:18 +0900, Charles Plessy wrote:
>>> as a non-native speaker, I have difficulties with the use of 'may' in
>>> your patch: if fields may be unique, they also may be not unique, so
>>> what is the message in this sentence? It does not give me the
>>> impression that the goal is to discourage the use of the same field
>>> name twice in the same paragraph.
>>> How about “A paragraph should not contain data fields having the same
>>> name.”
>> In this sense 'may' should be read as 'must', however I think that if
>> it causes readability issues for non-native english speakers then the
>> word 'must' should actually be used...
>> "Each paragraph must contain at most one instance of a particular
>> field name."
>> Is that clearer?
> Yes, indeed, that's what I meant; I'll go with that (assuming that
> doesn't have a different clarity problem that I'm missing).
Hm, actually, better (slightly less awkward, I think):
A paragraph must not contain more than one instance of a particular
field name.
--
Steve Langasek
> Russ Allbery <r...@debian.org> writes:
> > Andrew McMillan <and...@morphoss.com> writes:
> >> On Sat, 2010-06-12 at 18:18 +0900, Charles Plessy wrote:
> >>> as a non-native speaker, I have difficulties with the use of 'may' in
> >>> your patch: if fields may be unique, they also may be not unique, so
> >>> what is the message in this sentence? It does not give me the
> >>> impression that the goal is to discourage the use of the same field
> >>> name twice in the same paragraph.
> >>> How about “A paragraph should not contain data fields having the same
> >>> name.”
> >> In this sense 'may' should be read as 'must', however I think that if
> >> it causes readability issues for non-native english speakers then the
> >> word 'must' should actually be used...
> >> "Each paragraph must contain at most one instance of a particular
> >> field name."
> >> Is that clearer?
> > Yes, indeed, that's what I meant; I'll go with that (assuming that
> > doesn't have a different clarity problem that I'm missing).
> Hm, actually, better (slightly less awkward, I think):
> A paragraph must not contain more than one instance of a particular
> field name.
Seconded.
Cheers,
--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer http://www.debian.org/
slan...@ubuntu.com vor...@debian.org
Charles Plessy
>
> A paragraph must not contain more than one instance of a particular
> field name.
Hi Russ,
This one is much clearer, thank you for the changes. I second them.
Thanks also to for the other answers, who made me progress in English.
Have a nice day,
--
Charles
--
Cyril Brulebois
> This one is much clearer, thank you for the changes. I second them.
Need to sign your mail.
Mraw,
KiBi.
Cyril Brulebois
> Not for Policy work. Some people do, but I never check them. I
> figure the consequences here just aren't enough for there to be much
> risk of someone impersonating someone else, and there's always a
> Policy maintainer looking over the change and making sure it makes
> sense anyway.
Alright, sorry for the noise…
Russ Allbery
Not for Policy work. Some people do, but I never check them. I figure
the consequences here just aren't enough for there to be much risk of
someone impersonating someone else, and there's always a Policy maintainer
looking over the change and making sure it makes sense anyway.
--
Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/>
--
Russ Allbery
> Hm, actually, better (slightly less awkward, I think):
> A paragraph must not contain more than one instance of a particular
> field name.
This has now been merged for the next release.