Bug#511357: Be careful with #511357

2009-01-12 Thread David Weinehall
Sorry for my tone earlier; I realised after I sent my reply that it was
needlessly inflammatory.  Anyway, after posting my reply there has been
further discussions on tp-sv, and it seems that Svenska Språknämnden
still regards W as a variant of V, so sadly it seems that we'll have to
keep Swedish with insane sorting in the future too (unless there's some
way to provide a variant).

Sorry again to Jörgen for my tone, and to Clint for wasting your time
(and thanks for being prepared to do so!).


Regards: David
-- 
 /) David Weinehall  /) Rime on my window   (\
//  ~   //  Diamond-white roses of fire //
\)  http://www.acc.umu.se/~tao/(/   Beautiful hoar-frost   (/



--
To UNSUBSCRIBE, email to debian-glibc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Bug#489946: Slight correction

2009-01-11 Thread David Weinehall
On Sun, Jan 11, 2009 at 05:50:45PM +, Clint Adams wrote:
> On Sun, Jan 11, 2009 at 06:26:55PM +0100, David Weinehall wrote:
> > This patch doesn't seem to apply for me.  Is it reversed?
> > (I dropped it in debian/patches/localedata and added it to series with
> > -p0, I hope that was the right way to do it).
> 
> Ah, yes, you'll need to either disable or tweak
> localedata/first_weekday.diff

After 193 minutes of building glibc (my laptop is certainly no speed
monster...  Are there any plans to split locales from the rest of
libc?), I've tested your patch.  Seems to work just fine!

t...@tal:~$ LC_TIME=sv_FI date +"%x %X"
12.01.2009 00.51.54

(without the patch, the output is: 
t...@tal:~$ LC_TIME=sv_FI date +"%x %X"
2009-01-12 00.55.20
)

Thanks for all the hard work!


Regards: David
-- 
 /) David Weinehall  /) Rime on my window   (\
//  ~   //  Diamond-white roses of fire //
\)  http://www.acc.umu.se/~tao/(/   Beautiful hoar-frost   (/



-- 
To UNSUBSCRIBE, email to debian-glibc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Bug#489946: Slight correction

2009-01-11 Thread David Weinehall
On Sun, Jan 11, 2009 at 06:09:56AM +, Clint Adams wrote:
> On Sun, Jan 11, 2009 at 01:16:00AM +0100, David Weinehall wrote:
> > Disclaimer: I'm certainly not an expert on the Finnish sublocale of
> > Swedish; I just know what the date & time formats should be.
> 
> Please confirm that this triggers the correct behavior, or nudge a
> Swedish-speaking Finn this way.

This patch doesn't seem to apply for me.  Is it reversed?
(I dropped it in debian/patches/localedata and added it to series with
-p0, I hope that was the right way to do it).


Regards: David
-- 
 /) David Weinehall  /) Rime on my window   (\
//  ~   //  Diamond-white roses of fire //
\)  http://www.acc.umu.se/~tao/(/   Beautiful hoar-frost   (/



-- 
To UNSUBSCRIBE, email to debian-glibc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Bug#511357: Be careful with #511357

2009-01-11 Thread David Weinehall
On Sun, Jan 11, 2009 at 04:08:37PM +, Clint Adams wrote:
> On Sun, Jan 11, 2009 at 10:12:25AM +0100, Jörgen Grahn wrote:
> > I was unaware that SAOL was normative for collation order (it's barely
> > normative as a dictionary). I can well imagine them choosing to split
> > out 'w' for the purpose of their dictionary, without claiming that
> > this is the all-purpose sv_SE collation rule.
> 
> David, thoughts?

True, SAOL isn't normative for collation order.  Howver, claiming that
SAOL is anything but normative as *the* dictionary for sv_SE
is, pardon the expression, pure bullshit.  The only other dictionary
that has anything close to the same bearing on the Swedish language is
SAOB, and that one is purely for historical purposes[1].

Anyway, there seem to be agreement on the tp-sv list that the change in
SAOL is indicative of a change in collation order, and I was not the
person who started the thread and suggested the change (I just reported
it to Debian).

The tp-sv list has moved recently though, and I cannot seem to find a
list archive for the new list (though I guess/hope you're already
subscribed to that list Jörgen, and that you'll voice your dissent there
as well?).


Regards: David

[1] There are several *specialised* dictionaries for narrow fields, such
as the word lists published by Tekniska nomenklaturcentralen, etc., but
no other full dictionary.
-- 
 /) David Weinehall  /) Rime on my window   (\
//  ~   //  Diamond-white roses of fire //
\)  http://www.acc.umu.se/~tao/(/   Beautiful hoar-frost   (/



--
To UNSUBSCRIBE, email to debian-glibc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Bug#489946: Slight correction

2009-01-10 Thread David Weinehall
On Sat, Jan 10, 2009 at 11:54:47PM +, Clint Adams wrote:
> On Sat, Jan 10, 2009 at 04:46:44PM +0100, David Weinehall wrote:
> > Slight correction; Finnish time should indeed be written using HH.MM.SS;
> > the part of the bug report concerning dates stand though.
> > 
> > (so, just to be clear: DD.MM., HH.MM.SS).
> > 
> > See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=468849
> > for further reference, but note that this bug is about sv_FI, while that
> > one is about fi_FI.
> 
> Should all the LC_CTIME formats in sv_FI be the same as fi_FI?

TTBOMK, yes.  sv_FI should basically be Finnish formats (including
Finnish collation rules, since the alphabet is the same), Swedish
language.

> If not, what should be different?

Nothing, I hope :)

Disclaimer: I'm certainly not an expert on the Finnish sublocale of
Swedish; I just know what the date & time formats should be.


Regards: David
-- 
 /) David Weinehall  /) Rime on my window   (\
//  ~   //  Diamond-white roses of fire //
\)  http://www.acc.umu.se/~tao/(/   Beautiful hoar-frost   (/



-- 
To UNSUBSCRIBE, email to debian-glibc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Bug#489946: Slight correction

2009-01-10 Thread David Weinehall
Slight correction; Finnish time should indeed be written using HH.MM.SS;
the part of the bug report concerning dates stand though.

(so, just to be clear: DD.MM., HH.MM.SS).

See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=468849
for further reference, but note that this bug is about sv_FI, while that
one is about fi_FI.


Regards: David
-- 
 /) David Weinehall  /) Rime on my window   (\
//  ~   //  Diamond-white roses of fire //
\)  http://www.acc.umu.se/~tao/(/   Beautiful hoar-frost   (/



-- 
To UNSUBSCRIBE, email to debian-glibc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Bug#511357: Sorting wrong for sv_SE

2009-01-10 Thread David Weinehall
On Sat, Jan 10, 2009 at 02:02:26AM +, Clint Adams wrote:
> Does this do the right thing for you?

Yup!  Thanks a million!

BTW, I just noticed
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=506784

That bug is incorrect (though it used to be correct with the old
collation rules for Swedish), and a comment with a reference to this bug
should probably be made to it (closing it).


Regards: David
-- 
 /) David Weinehall  /) Rime on my window   (\
//  ~   //  Diamond-white roses of fire //
\)  http://www.acc.umu.se/~tao/(/   Beautiful hoar-frost   (/



-- 
To UNSUBSCRIBE, email to debian-glibc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Bug#511357: Sorting wrong for sv_SE

2009-01-09 Thread David Weinehall
Package: locales   
Version: 2.7-18 

Until 2006, W was regarded as a variant of V, but when the 13th edition 
of SAOL (Svenska Akademiens OrdLista; the reference dictionary for the  
Swedish language) was introduced, W was a proper letter, after V in 
the alphabet[1].

This means that the reordering rule in sv_SE should be removed. 


Regards: David Weinehall

[1] http://www.saol.se/saol13_pres.html 

Sorry that the reference link is only in Swedish, but news like this are
unlikely to be printed in foreign newspapers...
-- 
 /) David Weinehall  /) Rime on my window   (\
//  ~   //  Diamond-white roses of fire //
\)  http://www.acc.umu.se/~tao/(/   Beautiful hoar-frost   (/



-- 
To UNSUBSCRIBE, email to debian-glibc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Bug#489960: date_fmt incorrect for sv_SE

2008-07-08 Thread David Weinehall
Package: locales
Version: 2.7-12
Severity: minor
Tags: patch

date_fmt uses the wrong time format for the Swedish locale; the included
patch tries to fix this.  The patch also moves the %Y to the same
position it has in %c, and adds a few helpful comments while at it.


Regards: David Weinehall
-- 
 /) David Weinehall <[EMAIL PROTECTED]> /) Rime on my window   (\
//  ~   //  Diamond-white roses of fire //
\)  http://www.acc.umu.se/~tao/(/   Beautiful hoar-frost   (/


--- sv_SE.old   2008-07-09 01:17:05.0 +0300
+++ sv_SE   2008-07-09 01:28:06.0 +0300
@@ -169,14 +169,23 @@
   "";/
   "";/
   ""
+#
+# Appropriate date and time representation (%c)
+#   "%a %e %b %Y %H.%M.%S"
 d_t_fmt   
""
+#
+# Appropriate date representation (%x)   "%Y-%m-%d"
 d_fmt ""
+#
+# Appropriate time representation (%X)   "%H.%M.%S"
 t_fmt ""
 am_pm "";""
 t_fmt_ampm  ""
-date_fmt   "/
-/
-"
+#
+# Appropriate date representation (date(1))   "%a %e %b %Y %H.%M.%S %Z"
+date_fmt   "/
+/
+"
 first_weekday 2
 first_workday 2
 END LC_TIME



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Bug#489946: Incorrect locale information for sv_FI

2008-07-08 Thread David Weinehall
Package: locales
Version: 2.7-12
Severity: minor

The Swedish locale for Finland contains a few errors in the LC_TIME
section; in Finland, dates are written using DD.MM., and times
are written HH:MM:SS; the current sv_FI instead uses the formats used
in Sweden (-MM-DD and HH.MM.SS, respectively).


Regards: David Weinehall
-- 
 /) David Weinehall <[EMAIL PROTECTED]> /) Rime on my window   (\
//  ~   //  Diamond-white roses of fire //
\)  http://www.acc.umu.se/~tao/(/   Beautiful hoar-frost   (/



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Bug#256502: debian/generate-asm.sh contains XSI:isms

2004-06-27 Thread David Weinehall
Package: linux-kernel-headers
Version: 2.5.999-test7-bk-16
Severity: minor
Tags: patch

debian/generate-asm.sh contains use of the XSI:isms `-a' and `-o' in
tests.  This is technically a violation of policy.  The included patch
fixes this.

Rationale:

policy 10.4
http://www.opengroup.org/onlinepubs/009695399/utilities/test.html


Regards: David Weinehall

diff -ur linux-kernel-headers-2.5.999-test7-bk-old/debian/generate-asm.sh 
linux-kernel-headers-2.5.999-test7-bk/debian/generate-asm.sh
--- linux-kernel-headers-2.5.999-test7-bk-old/debian/generate-asm.sh
2004-06-27 17:19:15.0 +0300
+++ linux-kernel-headers-2.5.999-test7-bk/debian/generate-asm.sh
2004-06-27 17:21:20.0 +0300
@@ -12,7 +12,7 @@
cd /usr/include
 fi
 
-if [ ! -d asm-sparc -o ! -d asm-sparc64 ] ; then
+if [ ! -d asm-sparc ] || [ ! -d asm-sparc64 ] ; then
echo E: asm-sparc and asm-sparc64 must exist, or you will have problems
exit 1
 fi
@@ -34,7 +34,7 @@
 EOF
 
# common for sparc and sparc64
-   if [ -f asm-sparc/$h -a -f asm-sparc64/$h ]; then
+   if [ -f asm-sparc/$h ] && [ -f asm-sparc64/$h ]; then
cat >> asm/$h <




Bug#256502: debian/generate-asm.sh contains XSI:isms

2004-06-27 Thread David Weinehall
Package: linux-kernel-headers
Version: 2.5.999-test7-bk-16
Severity: minor
Tags: patch

debian/generate-asm.sh contains use of the XSI:isms `-a' and `-o' in
tests.  This is technically a violation of policy.  The included patch
fixes this.

Rationale:

policy 10.4
http://www.opengroup.org/onlinepubs/009695399/utilities/test.html


Regards: David Weinehall

diff -ur linux-kernel-headers-2.5.999-test7-bk-old/debian/generate-asm.sh 
linux-kernel-headers-2.5.999-test7-bk/debian/generate-asm.sh
--- linux-kernel-headers-2.5.999-test7-bk-old/debian/generate-asm.sh2004-06-27 
17:19:15.0 +0300
+++ linux-kernel-headers-2.5.999-test7-bk/debian/generate-asm.sh2004-06-27 
17:21:20.0 +0300
@@ -12,7 +12,7 @@
cd /usr/include
 fi
 
-if [ ! -d asm-sparc -o ! -d asm-sparc64 ] ; then
+if [ ! -d asm-sparc ] || [ ! -d asm-sparc64 ] ; then
echo E: asm-sparc and asm-sparc64 must exist, or you will have problems
exit 1
 fi
@@ -34,7 +34,7 @@
 EOF
 
# common for sparc and sparc64
-   if [ -f asm-sparc/$h -a -f asm-sparc64/$h ]; then
+   if [ -f asm-sparc/$h ] && [ -f asm-sparc64/$h ]; then
cat >> asm/$h <


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]