Alberto Simões (a...@alfarrabio.di.uminho.pt)

Interesting is that if I change that print hidden() by a print of the
real html code, the value printed is not the correct either.

So, it seems the browser is confused when rendering the page. CGI posts
al the defined variables back again?

Alberto

Suppose you have a CGI, and param("foo") is "bar".

if (param("foo") eq "bar") {
    print start_form;
    print hidden("foo","ugh");   ## prints "bar" as the value, not "ugh"
    print end_form;
}

To correct the behaviour...

if (param("foo") eq "bar") {
    print start_form;
    param("foo", "ugh");
    print hidden("foo","ugh");   ## prints "bar" as the value, not "ugh"
    print end_form;
}

Alberto

So, this is OSX, 10.4, and I'm not very used to it yet, so, it might be
my fault to have these errors:

...
../lib/locale...............................NOK 115# The following
locales
#
#       C C POSIX POSIX af_ZA af_ZA.ISO8859-1 af_ZA.ISO8859-15
#       ...
#       zh_HK.Big5HKSCS zh_HK.UTF-8 zh_TW zh_TW.Big5 zh_TW.UTF-8
#
# tested okay.
#
# The following locales
#
#       eu_ES eu_ES.ISO8859-1 eu_ES.ISO8859-15 eu_ES.UTF-8
#
# had problems.
#
../lib/locale...............................FAILED tests 99, 105-106,
108-109, 113, 115
         Failed 7/117 tests, 94.02% okay
../lib/Math/BigInt/t/bare_mbf...............ok
...

He! Nice colection. Now, anybody up to DRAW the table? :)

Hi!

I have a module with lot of Encoding problems (just don't know why, and
maybe its fault of the modules I use).

Meanwhile, Encode could simplify my life, but I think it is just
available for 5.8.x. Is this correct?

Thanks.
Alberto

Ok, this might be useless, but maybe you like to know:

All tests successful, 1 test and 64 subtests skipped.
Files=135, Tests=2252, 798 wallclock secs (299.99 cusr + 96.80 csys =
396.79 CPU)

On a PIII 1Ghz/256MB RAM, running Slackware Linux

Cheers,
Alberto

Cheers,
Alb

I would say EU::MM is much better than M::B at the moment.
Just my thoughts..

Alb

Also, it would be a nice idea to log the searches, so that searched
modules documentation can be built if there is interest from the users.

All tests successful.
u=1.92  s=0.68  cu=147.91  cs=14.59  scripts=848  tests=87641
make[2]: Leaving directory `/tmp/perl-5.8.6-RC1'
make[1]: Leaving directory `/tmp/perl-5.8.6-RC1'

User configure, make and make test. Not installing it.
Cheers,
ambs

Yes, to incorporate some kind of cpan+ or cpan with hability to install
M::B modules is __VERY__ important.

  Dear Perl 5 Porters and brian,

Today my cpan does not install a module using Module::Builder.
--->

Package seems to come without Makefile.PL.
   (The test -f "/root/.cpan/build/Lingua-PT-Abbrev-0.02/Makefile.PL" 
returned fa lse.)
   Writing one on our own (setting NAME to LinguaPTAbbrev)

--->

I think I did already install modules based on Built.PL before. Now, I
can't notice what is the problem. I have the latest Bundle::CPAN.
What am I doing wrong (or cpan is doing wrong?))

THanks.

Alberto

Hi all

I remember some time ago we found a "bug" on using utf-8 on DB_File.
Now, I am using DB_File again with utf-8 and I need some help.

I'm using DB_File to store utf-8 data. I can try to encode to iso-8859-1
but I would lost a set of characters :-(

At the moment, when accessing that data I get the bytes, and not the
utf-8 string.

Thanks for any help
Cheers.