Re: ***regular expression***
Thanks all but...
I don't explain well the problem!
So, i'm processing a very big text file ( from A DB ) in this format
number|field 1|field2| field 3| etc. Every line is a record.Every line
contains many url.
I'm interested about specific url and i don't consider the others.
while ( $line =~ s/href=\"([^"]+)\"//) {
$url = $1; ###ALL URL#
if ($url =~ m/\/[^,]+,([^,]+),([^,]+),[^,]+\.html?$/) {
($var1,$var2)=($1,$2); ###INTERESTING URL AND VALUES
}
}
In this way i obtain two values that interested me! It's works!
Now i must substitute these url the_link with
the_link and then i write to a new file with this
substitution on all the field of file!
Can you help me?
perlmagick on panther working
hello again. I've got perlmagick working, at last. it seems that panther has some linking quirk or other than can be avoided by manually specifying libraries in the LIBS line of Makefile.PL. I found this in a completely unrelated bug report about installing DBD::Pg (http://gborg.postgresql.org/project/dbdpg/bugs/bugupdate.php?619), and indeed it works in this case too. If you add this to the LIBS line: -ljpeg -lpng -lTIFF -lBZ2 -lxml2 then nearly all the loader errors in make test go away. I have no idea why. Anyone? The only errors that remain are to do with linking to XDPS, and I believe those will only come up if you are trying to use Apple's X11 instead of the old fink one, which includes the necessary DPS libraries. I suppose it shouldn't come as a surprise that the Apple version does not. But it seems a much better idea to use Apple's X, if possible, and you can get around the DPS problem either by installing the imagemagick-nox package instead (which gives you no X tools, but doesn't seem to mind when it can't find dps), or by editing imagemagick.info to remove the dependency on dps from its configuration. Here's an amended version of Randal's recipe. I'm using the unstable branch, but I expect the same tweaks work for the stable version as it seems to be the same at the moment. 1) edit /sw/fink/10.3/(stable|unstable)/main/finkinfo/graphics/imagemagick.info so that the ConfigureParams line has --without-dps instead of --with-dps 2) "fink install imagemagick" (This will take a while) 3) "fink list imagemagick" 4) Note the version number (n.n.n-n) by the version you have installed. 5) "cd; mkdir Build; cd Build" (getting to a scratchdir) 6) "tar xvfz /sw/src/ImageMagick-n.n.n.tar.gz; cd Image*/PerlMagick/" 7) edit Makefile.PL so that INC begins with -I/sw/include LIBS begins with -L/sw/lib LIBS includes -ljpeg -lpng -lTIFF -lBZ2 -lxml2 after the path definitions (and as well as whatever other specifications are there) 8) "perl Makefile.PL; make all install" 9) "cd; rm -rf Build" it still fails a few tests, as always, but it loads ok and works fine, as far as I can see, with the mainstream formats. I hope this helps someone out. will PS. In the course of these lengthy experiments I also selfupdated to fink 0.6.2, and for all I know that's what made the difference, making these instructions completely redundant :) On 13 Nov 2003, at 15:37, william ross wrote: has anyone had any success installing perlmagick on panther? I'm using the fink-and-make method, which has always worked for me before and is succinctly described by Randal in http://nntp.x.perl.org/group/perl.macosx/3812 fink on 10.3 is in some flux at the moment and they don't recommend installing from source 'for the next few weeks', but that's what you have to do in order to get a version of perlmagick which matches the version of imagemagick you have just installed, so that's what i did, and it seems to have compiled and installed without any bother. the command-line tools work, and the libraries all appear to have ended up where they should. i then go into the perlmagick subdirectory, amend the library and include locations as usual and try to make it. That appears to work too: there are no warnings or complaints from running either the makefile or make. however, on make test (or installing and trying to use the module), I get a flood of errors like this: /sw/temp/ImageMagick-5.5.1/PerlMagick/blib/arch/auto/Image/Magick/ Magick.bundle undefined reference to _xmlSwitchEncoding expected to be defined in a dynamic image which sounds like the old dyld in new clothes? Or maybe a compiler mismatch? curses. I'm using the system 5.8.1rc3 that ships with panther, and fink 0.6.1, both of which expect to use gcc3.3, I believe. fink is selfupdating again at the moment, but that will take a while (all i can get here is 28.8) so I thought I'd enquire. any advice would be most welcome. thanks will
Re: ***regular expression***
On 11/17/03 2:52 AM, "xweb" <[EMAIL PROTECTED]> wrote:
> Thanks all but...
> I don't explain well the problem!
> So, i'm processing a very big text file ( from A DB ) in this format
> number|field 1|field2| field 3| etc. Every line is a record.Every line
> contains many url.
> I'm interested about specific url and i don't consider the others.
>
> while ( $line =~ s/href=\"([^"]+)\"//) {
> $url = $1; ###ALL URL#
> if ($url =~ m/\/[^,]+,([^,]+),([^,]+),[^,]+\.html?$/) {
> ($var1,$var2)=($1,$2); ###INTERESTING URL AND VALUES
> }
> }
> In this way i obtain two values that interested me! It's works!
> Now i must substitute these url the_link with
> the_link and then i write to a new file with this
> substitution on all the field of file!
> Can you help me?
>
The only thing you haven't captured is `the_link`.
open (FH, "> path/to/new/file") || die "$!\n"; # open new file to write to
while ( $line =~ /(.*?)<\/a>/) {
$url = $1; ###ALL URL#
$the_link = $2;
if ($url =~ m/\/[^,]+,([^,]+),([^,]+),[^,]+\.html?$/) {
($var1,$var2)=($1,$2); ###INTERESTING URL AND VALUES
print FH "$the_link"; # print to new file
}
}
close(FH)
Re: Panther problems
On Nov 16, 2003, at 6:45 PM, Vic Norton wrote: I'm sorry about the confusion, Morbus. My ".bash_profile" contains the line MANPATH=/usr/share/man:/man:/usr/X11R6/man:~/man The "set" command, with no parameters, reproduces this as MANPATH=/usr/share/man:/man:/usr/X11R6/man:/Users/vic/man Regardless, I cannot see manuals in either /man or /usr/X11R6/man and "manpath" still produces /usr/share/man:/Users/vic/man My conclusion, in bash the value of the variable MANPATH has no effect whatsoever on "man". Maybe bash doesn't need "setenv", but, so far, it can't tell "man" where to look either. Apparently we're not supposed to set $MANPATH anymore. There's a "manpath" command that "man" calls to automatically figure out the paths (from "man manpath"): Manpath is used by man(1) to determine the search path, so users normally don't need to set the MANPATH environment variable directly. If you *do* set $MANPATH, "manpath" just echoes that, without doing its searching. There is an /etc/manpath.config file that lets you expand the default places that "manpath" searches. On closer looking, and doing several searches across my whole system looking for manpages, I found that I needed to add the following lines: OPTIONAL_MANPATH/opt/local/share/man OPTIONAL_MANPATH/usr/local/share/man The /opt/local reference is because I use Darwinports, which (like fink) puts all its installed files into its own directory. (Note that I have actually changed /usr/bin/perl to point to /opt/bin/perl, which may mean that a lot of what I've written may not be true on your system. ;) Oddly, manpath.config contains the following line: OPTIONAL_MANPATH/usr/local/lib/perl5/5.6.1/man That doesn't seem right for a system running Perl 5.8.x... (The "manpath" command seems, in my opinion, to be a decent solution -- but unfortunately doesn't seem to let a user add in their own paths without modifying the /etc file. It would be nice to have a $MANPATH_MANPATH environment variable that would be appended to whatever magic "manpath" performed.) -- John Labovitz photography www.johnlabovitz.com [EMAIL PROTECTED] +1 503.873.6734
Re: Panther problems
On Nov 16, 2003, at 6:32 PM, John Delacour wrote:
I reinstalled 5.8.1 from source twice and everything went fine until I
upgraded certain modules. In the end I installed 5.8.2 with the Apple
way, ie. to replace 5.8.1, and since then everything has been fine.
All my modules have been updated except for Filter::Simple, which cpan
always reports as being one point behind in spite of updating.
It's because Filter::Simple is installed in two places, and the later
install did not remove or overwrite the first one.
I've been tracking this down on my own system, and finally figured out
what is happening. It's *not* an OS X-specific problem; it happened on
my Redhat system, too.
When there's a module installed with the initial Perl installation (I
think this is called a "core" module), and you try to upgrade that
module, the upgraded module gets installed in a different place than
the original module, yet the @INC include path is such that the
original module gets found first.
Say Filter::Simple is installed under /System/Library/Perl. By either
looking at its version number by hand, or using the CPAN "r" command,
we see that it's 0.79. We update it to 0.80. The update is installed
under /Library/Perl. But our perl's @INC path says to look first in
/System/Library, where it finds the 0.79 version.
I finally "solved" this on my system by tracking down which modules
weren't being updated, and just deleting the older versions -- from
/System/Library/Perl if necessary.
However, the right way to solve this is to use the UNINST=1 flag when
you install Perl modules. From the CPAN FAQ ("man CPAN"):
1) I installed a new version of module X but CPAN keeps saying,
I have
the old version installed
Most probably you do have the old version installed. This
can hap-
pen if a module installs itself into a different directory
in the
@INC path than it was previously installed. This is not
really a
CPAN.pm problem, you would have the same problem when
installing
the module manually. The easiest way to prevent this
behaviour is
to add the argument "UNINST=1" to the "make install" call,
and that
is why many people add this argument permanently by
configuring
o conf make_install_arg UNINST=1
2) So why is UNINST=1 not the default?
Because there are people who have their precise expectations
about
who may install where in the @INC path and who uses which
@INC
array. In fine tuned environments "UNINST=1" can cause
damage.
Personally, this causes me enough hassle that I really wish UNINST=1
*was* the default.
--
John Labovitz photography
www.johnlabovitz.com
[EMAIL PROTECTED]
+1 503.873.6734
Re: Panther problems
At 11:36 am -0800 17/11/03, John Labovitz wrote: > Personally, this causes me enough hassle that I really wish UNINST=1 *was* the > default. Me too! Thanks for the long explanation. I'm now having problems installing Tk. I've force installed but get the same error (coundn/t connect to display ":0" at ...ManinWIndow.pm) when I run scripts as I got when it failed all the tests. All great fun ! JD
Problems installing Tk
When I install Tk I get the same error with all the tests. If I force-install, then I get the same error im MainWindow.pm (though with a different path of course) when I try to run a script that calls Tk. Any hints as to how to fix this? DIED. FAILED tests 1-96 Failed 96/96 tests, 0.00% okay t/JP.couldn't connect to display ":0" at /Users/jd/.cpan/build/Tk-804.025_beta6/blib/lib/Tk/MainWindow.pm line 55. JD
Re: Problems installing Tk
Are you running the X server? It pretty well compiled out the box for me. Failed 5 or 6 tests. There is another problem with Tk on Panther is the the standard file open/close dialogs "bus error". I have not seen a fix to this problem. --Jerry On Nov 17, 2003, at 4:17 PM, John Delacour wrote: When I install Tk I get the same error with all the tests. If I force-install, then I get the same error im MainWindow.pm (though with a different path of course) when I try to run a script that calls Tk. Any hints as to how to fix this? DIED. FAILED tests 1-96 Failed 96/96 tests, 0.00% okay t/JP.couldn't connect to display ":0" at /Users/jd/.cpan/build/Tk-804.025_beta6/blib/lib/Tk/MainWindow.pm line 55. JD
Re: Problems installing Tk
At 9:17 pm + 17/11/03, John Delacour wrote: > When I install Tk I get the same error with all the tests. If I force-install, then > I get the same error im MainWindow.pm (though with a different path of course) when > I try to run a script that calls Tk. > > Any hints as to how to fix this? > > > DIED. FAILED tests 1-96 > Failed 96/96 tests, 0.00% okay > t/JP.couldn't connect to display ":0" at > /Users/jd/.cpan/build/Tk-804.025_beta6/blib/lib/Tk/MainWindow.pm line 55. I now see in the README.darwin ___ For Tk804.025 to build properly on Mac OS X 10.2 (Jaguar) or 10.3 (Panther), Perl must be built dynamic, rather that the default of static. Use a Configure incantation similar to this: sh Configure -des -Duseshrplib [-Dprefix=/path/to/your/private/perl] S. Lidie, 2003/10/27 ___ but that gives me sh-2.05b# Configure -des -Duseshrplib sh: Configure: command not found sh-2.05b# Any ideas? JD
Re: Problems installing Tk
Configure is generally found in the Perl source distribution. The version of Perl that comes with Panther builds Tk ok. Your probably need to do something like ./Configure if the perl source directory is not in your PATH. --Jerry On Nov 17, 2003, at 6:14 PM, John Delacour wrote: At 9:17 pm + 17/11/03, John Delacour wrote: When I install Tk I get the same error with all the tests. If I force-install, then I get the same error im MainWindow.pm (though with a different path of course) when I try to run a script that calls Tk. Any hints as to how to fix this? DIED. FAILED tests 1-96 Failed 96/96 tests, 0.00% okay t/JP.couldn't connect to display ":0" at /Users/jd/.cpan/build/Tk-804.025_beta6/blib/lib/Tk/MainWindow.pm line 55. I now see in the README.darwin ___ For Tk804.025 to build properly on Mac OS X 10.2 (Jaguar) or 10.3 (Panther), Perl must be built dynamic, rather that the default of static. Use a Configure incantation similar to this: sh Configure -des -Duseshrplib [-Dprefix=/path/to/your/private/perl] S. Lidie, 2003/10/27 ___ but that gives me sh-2.05b# Configure -des -Duseshrplib sh: Configure: command not found sh-2.05b# Any ideas? JD
