[18100] in Perl-Users-Digest
Perl-Users Digest, Issue: 260 Volume: 10
daemon@ATHENA.MIT.EDU (Perl-Users Digest)
Fri Feb 9 18:15:59 2001
Date: Fri, 9 Feb 2001 15:15:27 -0800 (PST)
From: Perl-Users Digest <Perl-Users-Request@ruby.OCE.ORST.EDU>
To: Perl-Users@ruby.OCE.ORST.EDU (Perl-Users Digest)
Message-Id: <981760526-v10-i260@ruby.oce.orst.edu>
Content-Type: text
Perl-Users Digest Fri, 9 Feb 2001 Volume: 10 Number: 260
Today's topics:
Re: Problem with regex and if/else expression <mothra@nowhereatall.com>
Re: Problem with regex and if/else expression (LK)
Re: Problem with regex and if/else expression gnari@my-deja.com
Re: Problem with regex and if/else expression <faerloche@arkansas.net>
Re: Problem with regex and if/else expression <godzilla@stomp.stomp.tokyo>
Re: Radical readdir suggestion <ldo@geek-central.gen.new_zealand>
Re: Regular Expression Question bonjaa@my-deja.com
Re: Regular Expression Question <elijah@workspot.net>
Re: Regular Expression Question <elijah@workspot.net>
Re: Specifying the length of regular expression <elijah@workspot.net>
Re: splitting a string on the / character (Mahesh A)
Re: This is driving me nuts and I need a guru <jdf@pobox.com>
UNIX core dumps - Segmentation Fault hilljroberts@my-deja.com
Re: Untangling NT Logon Scripts <jdf@pobox.com>
Wanted: Perl Programmer for Chatologica Metasearch Mods <horace700@my-deja.com>
XML: "Flattening" XML to dot notation - best module? <mkruse@netexpress.net>
Digest Administrivia (Last modified: 16 Sep 99) (Perl-Users-Digest Admin)
----------------------------------------------------------------------
Date: Fri, 09 Feb 2001 11:24:32 -0800
From: mothra <mothra@nowhereatall.com>
Subject: Re: Problem with regex and if/else expression
Message-Id: <3A8443F0.ACA87E4@nowhereatall.com>
LK wrote:
>
> On Fri, 09 Feb 2001 16:37:00 GMT, lkenny@fisheries.org (LK) wrote:
>
> >On Fri, 09 Feb 2001 08:09:48 -0800, mothra <mothra@nowhereatall.com>
> >wrote:
> >
> >>LK wrote:
> >>>
> >>> I am trying to get the program below to search a file for a match with
> >>> a regular expression. When it makes that match I want that line to be
> >>> stored in the variable $ms. I though I had programed this properly,
> >>> but I am wrong. When I run it this way, no matter what $num is, the
> >>> "else" is executed and not the "if". If I take out the "else"
> >>> portion, no matter what $num is the last line of the file is stored in
> >>> $ms.
> >>[snipped]
> >>> $num = $FORM{'msnumber'};
> >>> open(FILE, $file) || &dienice;
> >>> while(<FILE>){
> >>> if($_ !=~ /$num/){
> >>> $ms = $_;
> >>[snipped]
> >>change the if to this
> >> if(/$num/) {
> >> $ms = $_;
> >>}
> >>
> >>Hope this helps
> >>
> >>Mothra!
> >
> >Thanks for the help, mothra. This gets me out of that problem of only
> >getting the final line in the file.
> >But there are other intereting issues that arise. else seems to take
> >over the program when it is kept in. No matter what $num is, the else
> >will execute. When I take else out, it runs properly.
> >Any ideas?
> >
> >LK
>
> Actually, I kno why it is doing this, just don' know the work around
> for it yet
>
> LK
Hi LK,
one question for you is: once a match is made ($ms) do you need to
continue searching the file?
if not you can add a last command in the if statement to kick out of the
loop.
Mothra!
------------------------------
Date: Fri, 09 Feb 2001 19:39:27 GMT
From: lkenny@fisheries.org (LK)
Subject: Re: Problem with regex and if/else expression
Message-Id: <3a8446d7.15323532@wingate>
On Fri, 09 Feb 2001 11:24:32 -0800, mothra <mothra@nowhereatall.com>
wrote:
>LK wrote:
>>
>> On Fri, 09 Feb 2001 16:37:00 GMT, lkenny@fisheries.org (LK) wrote:
>>
>> >On Fri, 09 Feb 2001 08:09:48 -0800, mothra <mothra@nowhereatall.com>
>> >wrote:
>> >
>> >>LK wrote:
>> >>>
>> >>> I am trying to get the program below to search a file for a match with
>> >>> a regular expression. When it makes that match I want that line to be
>> >>> stored in the variable $ms. I though I had programed this properly,
>> >>> but I am wrong. When I run it this way, no matter what $num is, the
>> >>> "else" is executed and not the "if". If I take out the "else"
>> >>> portion, no matter what $num is the last line of the file is stored in
>> >>> $ms.
>> >>[snipped]
>> >>> $num = $FORM{'msnumber'};
>> >>> open(FILE, $file) || &dienice;
>> >>> while(<FILE>){
>> >>> if($_ !=~ /$num/){
>> >>> $ms = $_;
>> >>[snipped]
>> >>change the if to this
>> >> if(/$num/) {
>> >> $ms = $_;
>> >>}
>> >>
>> >>Hope this helps
>> >>
>> >>Mothra!
>> >
>> >Thanks for the help, mothra. This gets me out of that problem of only
>> >getting the final line in the file.
>> >But there are other intereting issues that arise. else seems to take
>> >over the program when it is kept in. No matter what $num is, the else
>> >will execute. When I take else out, it runs properly.
>> >Any ideas?
>> >
>> >LK
>>
>> Actually, I kno why it is doing this, just don' know the work around
>> for it yet
>>
>> LK
>
>Hi LK,
>
>one question for you is: once a match is made ($ms) do you need to
>continue searching the file?
>if not you can add a last command in the if statement to kick out of the
>loop.
>
>Mothra!
No. Once I get the match I want to capture the line that $num is in
and store it in $ms. But the problem seems to be that when the file
is being searched, if the first line isn't matched it kicks right over
to the else. I need a way to keep the program from going to else
until the entire file is searched for the match. I haven't come up
with anything practical yet. But if you have any ideas I'd appreciate
the help.
Thanks,
LK
------------------------------
Date: Fri, 09 Feb 2001 20:50:49 GMT
From: gnari@my-deja.com
Subject: Re: Problem with regex and if/else expression
Message-Id: <961l75$kgh$1@nnrp1.deja.com>
In article <3a8446d7.15323532@wingate>,
lkenny@fisheries.org (LK) wrote:
>
> No. Once I get the match I want to capture the line that $num is in
> and store it in $ms. But the problem seems to be that when the file
> is being searched, if the first line isn't matched it kicks right over
> to the else. I need a way to keep the program from going to else
> until the entire file is searched for the match. I haven't come up
> with anything practical yet. But if you have any ideas I'd appreciate
> the help.
>
you have the exit inside the while, so you abort the search at the
first fail. you want something like:
$num = $FORM{'msnumber'};
open(FILE, $file) || &dienice;
$ms='';
while(<FILE>){
if(/$num/){
$ms = $_;
last;
}
}
close(FILE);
if ($ms eq '') {
print "Content-type: text/html\n\n";
print "<html><body>STUFF";
print "</body></html>\n";
exit;
}
(note: I did not try this , so there might be bugs here,
but this is the general idea)
also remember to use strict and -w
gnari
Sent via Deja.com
http://www.deja.com/
------------------------------
Date: Fri, 09 Feb 2001 15:07:55 -0600
From: Faerloche <faerloche@arkansas.net>
Subject: Re: Problem with regex and if/else expression
Message-Id: <8tm88tc7vndujopb3rq3glm34v1s7nid49@4ax.com>
On Fri, 09 Feb 2001 19:39:27 GMT, lkenny@fisheries.org (LK) wrote:
[snipped]
>>> >>> $num = $FORM{'msnumber'};
>>> >>> open(FILE, $file) || &dienice;
>>> >>> while(<FILE>){
>>> >>> if($_ !=~ /$num/){
>>> >>> $ms = $_;
>>> >>[snipped]
>>> >>change the if to this
>>> >> if(/$num/) {
>>> >> $ms = $_;
>>> >>}
>>> >>
>>Hi LK,
>>
>>one question for you is: once a match is made ($ms) do you need to
>>continue searching the file?
>>if not you can add a last command in the if statement to kick out of the
>>loop.
>>
>>Mothra!
>
>No. Once I get the match I want to capture the line that $num is in
>and store it in $ms. But the problem seems to be that when the file
>is being searched, if the first line isn't matched it kicks right over
>to the else. I need a way to keep the program from going to else
>until the entire file is searched for the match. I haven't come up
>with anything practical yet. But if you have any ideas I'd appreciate
>the help.
>
>Thanks,
>
>LK
$num = $FORM{'msnumber'};
open(FILE, $file) || &dienice;
while(<FILE>) {
next if !~ m/$num/) {
$ms = $_;
last;
}
If you want the loop to continue after the first match, remove the
last.
-Faerloche
------------------------------
Date: Fri, 09 Feb 2001 13:15:51 -0800
From: "Godzilla!" <godzilla@stomp.stomp.tokyo>
Subject: Re: Problem with regex and if/else expression
Message-Id: <3A845E07.42DA5E87@stomp.stomp.tokyo>
LK wrote:
> mothra wrote:
> >LK wrote:
> >> LK wrote:
> >> > mothra wrote:
> >> > >>LK wrote:
(lots snipped)
> > one question for you is: once a match is made ($ms) do
> > you need to continue searching the file?
> > if not you can add a last command in the if statement
> > to kick out of the loop.
> >Mothra!
> No. Once I get the match I want to capture the line that $num is in
> and store it in $ms. But the problem seems to be that when the file
> is being searched, if the first line isn't matched it kicks right over
> to the else. I need a way to keep the program from going to else
> until the entire file is searched for the match. I haven't come up
> with anything practical yet. But if you have any ideas I'd appreciate
> the help.
Mothra is right. You need to exit your loop if a match
is found. Your first match should 'turn off' your else
routine following. You cannot do this while your else
statement is 'inside' your while loop. Each line which
does not match, kicks in your else, regardless if a
match could be found later.
Follow what Mothra has suggested and some. Add his last;
statement to exit your loop. My 'and some' is to add a
control to activate or skip over your 'else' routine.
This is your original code, my style:
$num = $FORM{'msnumber'};
open(FILE, $file) || &dienice;
while(<FILE>)
{
if($_ !=~ /$num/)
{ $ms = $_; }
else
{
print "(your html code)";
exit;
}
}
close(FILE);
Change your code using Mothra's last; and adding a control:
$num = $FORM{'msnumber'};
$control = "false";
open(FILE, $file) || &dienice;
while(<FILE>)
{
if($_ !=~ /$num/)
{
$ms = $_;
$control = "true";
last;
}
close(FILE);
if ($control eq "false")
{
print "(your html code)";
exit;
}
Others have commented on your !=~ syntax and
if this is what you really want to do. Be sure
to note, using last; in your loop will cause
$ms to be set to $_ for your first non-match
or match, depending on how you handle this.
Without a last; included, your $ms will be
equal to the last $_ found which meets whatever
criteria you set; $ms will be reassigned values
as your while loop proceeds with the last value
being the final value for $ms.
Godzilla!
------------------------------
Date: Sat, 10 Feb 2001 12:03:09 +1300
From: Lawrence DčOliveiro <ldo@geek-central.gen.new_zealand>
Subject: Re: Radical readdir suggestion
Message-Id: <ldo-A36364.12030910022001@news.wave.co.nz>
In article <x766impzen.fsf@home.sysarch.com>, Uri Guttman
<uri@sysarch.com> wrote:
>>>>>> "LD" == Lawrence DčOliveiro <ldo@geek-central.gen.new_zealand>
>>>>>> writes:
>
> LD> There is a subtle distinction between how the file system is
> LD> _implemented_ and how it _works_. On UNIX, there is no OS call to
> return
> LD> the pathname of the current working directory (strange, but true).
> LD> That's why you need those "." and ".." entries in _every_
> LD> directory--they are an essential part of the mechanism for
> navigating
> LD> your way around the filesystem tree. They are not part of the
> LD> directory's _contents_ per se.
>
>they are normal directory entries and are hard links to the current and
>parent dirs.
No, they are not normal directory entries, because you _cannot_ create
your own hard links with the same meanings as either "." or ".." (try
it). Neither can you remove the "." or ".." entries, like you can with
any other directory entry. Thus, they are very clearly NOT part of the
normal directory hierarchy.
> LD> (Strange that you should choose to burden every directory on every
> UNIX
> LD> filesystem with the extra space needed to hold these special
> entries,
> LD> rather than just fix the oversight in the original design of the
> kernel,
> LD> but there you go...)
>
>no, you are mistaken. the whole concept of a unix filesystem was to
>seperate the file name from its location.
And why was this a good idea?
so when you are in some random
>dir how could you find its parent dir? all you have in the process is an
>inode of the dir you are in. you use the .. and . dir entries to
>navigate.
Why is this a feature?
------------------------------
Date: Fri, 09 Feb 2001 19:09:27 GMT
From: bonjaa@my-deja.com
Subject: Re: Regular Expression Question
Message-Id: <961f8u$ehg$1@nnrp1.deja.com>
Thanks for all these replies. Here is why I asked this question:
I have a big text file (several MBs, which I read into one string by
undefining $/) containing only genomic sequences. Each line of the file
has exactly 50 letters (A, C, T, G, or N), now these N's are used to
separate one fragment from another, so you would have something like
ACTTGGCATTGCNNNNNN
NNAGCCTAGG
on two consecutive lines, with ACTTGGCATTGC belonging to the first
fragment and AGCCTAGG belonging to the second. You could also have all
of them on one line:
ACTTGGCATTGCNNNNNNNNAGCCTAGG
One more requirement: the total length of the consecutive N's should be
greater than 10, less than that, the two sequences on either side of
the N's still belong to the same fragment, and should not be separated.
So ten or more consecutive N's are like delimiters, basically I want a
regular expression that says one or more N's followed by an optional
newline symbol (if the consecutive N's are on two lines) followed by
one or more N's, with the total number of N's greater than 10.
If a regular expression could be easily defined, than I can just use
the split function to get the individual fragments.
Bon.
Sent via Deja.com
http://www.deja.com/
------------------------------
Date: 9 Feb 2001 21:10:26 GMT
From: Eli the Bearded <elijah@workspot.net>
Subject: Re: Regular Expression Question
Message-Id: <eli$0102091607@qz.little-neck.ny.us>
In comp.lang.perl.misc, David H. Adler <dha@panix6.panix.com> wrote:
> On 8 Feb 2001 23:16:34 GMT, Damian James <damian@qimr.edu.au> wrote:
> >Note that the OP wanted to test the length of the _match_, not the line.
> >DHA's solution in this thread did this.
> Nevertheless, I yield to ETB for the solution in the painful
> category. :-)
s/ETB/EtB/, please. And if you look at the OP's new comment in the
thread: wanting to use the RE in split(), mine is the only solution
that actually works.
Elijah
------
except that I solved for a match of minimum ten rather than more than ten
------------------------------
Date: 9 Feb 2001 21:26:14 GMT
From: Eli the Bearded <elijah@workspot.net>
Subject: Re: Regular Expression Question
Message-Id: <eli$0102091611@qz.little-neck.ny.us>
In comp.lang.perl.misc, <bonjaa@my-deja.com> wrote:
> I have a big text file (several MBs, which I read into one string by
> undefining $/) containing only genomic sequences. Each line of the file
> has exactly 50 letters (A, C, T, G, or N), now these N's are used to
> separate one fragment from another, so you would have something like
Well, solving for just ten or more Ns should be easy. The ABC
problem was a bit more complicated. You'll find it much easier
if you kill the line breaks, first though.
$geneseq =~ tr:A-Z::cd; # kill non-capital letters
@frags = split(/N{10,}/, $geneseq);
> So ten or more consecutive N's are like delimiters, basically I want a
> regular expression that says one or more N's followed by an optional
> newline symbol (if the consecutive N's are on two lines) followed by
> one or more N's, with the total number of N's greater than 10.
This is a bit more complicated, but if we can assume that there
will be at most one newline in the sequence (possibly false, if
Ns go one for multiple very short lines), then this can do it:
(?=N{10,}|[N\n]{11,})[N\n]+
> If a regular expression could be easily defined, than I can just use
> the split function to get the individual fragments.
By not mentioning split() originally, people here wanted to solve
a different problem. But see my post <eli$0102081913@qz.little-neck.ny.us>
for a split() compatible RE for your original, harder, question.
Elijah
------
thinks nothing of writing programs to write REs
------------------------------
Date: 9 Feb 2001 21:00:45 GMT
From: Eli the Bearded <elijah@workspot.net>
Subject: Re: Specifying the length of regular expression
Message-Id: <eli$0102091554@qz.little-neck.ny.us>
In comp.lang.perl.misc, Greg Bacon <gbacon@hiwaay.net> wrote:
> Ian Boreham <iboreham@my-deja.com> wrote:
> : > If you really need to do it with a single regex, try:
> : > ^(?=[ABC]{10,})A+B*C+$
> : > or some variation to taste.
> When matching strings from non-regular languages, I find that
> verifying the match with additional checks, e.g.,
> if (/^(A+B*C+)$/ && length($1) > 10) { ... }
> as Dominus wrote, makes for much clearer code than doing it all in the
> regular expression.
No kidding. But where's the fun in that? See my post
<eli$0102081913@qz.little-neck.ny.us> for my solution, which
does not require that ugly anchoring. :^)
Elijah
------
really this is a trivial RE
------------------------------
Date: Fri, 09 Feb 2001 22:56:18 GMT
From: maheshasolkar@yahoo.com (Mahesh A)
Subject: Re: splitting a string on the / character
Message-Id: <3a847493.278525178@news>
Now, I am not refering to any specific contributer, but I don't get
all this. People can spend hours discussing a map, a fish or may be an
elephant or a sea gherkin which has nothing to do with perl (Ooops..
map does !!), but not a minute or a half to give a simple answer for
present and reference to a manpage for later.
If this is what knowledgeable people are.. I'd rather stay away...
-
On 08 Feb 2001 04:29:22 -0800, merlyn@stonehenge.com (Randal L.
Schwartz) wrote:
>>>>>> "Studio" == Studio 51 <leekembel@hotmail.com> writes:
>
>Studio> My point is that a lot of "helpful" replies seem to just be
>Studio> "read the man pages". That is not helpful,
>
>Yes it is. If a clueful person can determine that the answer is
>derivable directly from the manpages, the original poster should be
>directed there first. But if someone says "I've read manpage XYZ,
and
>still don't understand it", I'm sure you'll find that we don't then
>say "well go back and read it again". That *would* be rude.
Instead,
>we try to help explain if we can.
>
>But why should you waste the resources of Usenet as a simple Manpage
>Copy-N-Paste service? That's silly.
>
>From the "worth repeating 27 more times department":
>
> THIS IS NOT A HELP DESK.
> THIS IS A COMMUNITY POTLUCK.
> DON'T BOGART THE POTATO SALAD.
> BRING A DISH IF YOU CAN.
> DON'T WHINE IF NOBODY BROUGHT ICE CREAM.
>
>There.
>
>--
>Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503
777 0095
><merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
>Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
>See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl
training!
------------------------------
Date: 09 Feb 2001 17:16:00 -0500
From: Jonathan Feinberg <jdf@pobox.com>
Subject: Re: This is driving me nuts and I need a guru
Message-Id: <y9vfh59b.fsf@pobox.com>
"John W" <jwgws@hotZEROSPAMmail.com> writes:
> Yes, but if I understood correctly (perhaps I didn't), Jonathon's reference
> was that he was also killfiling *me*.
I didn't mean to imply such a thing; I was mistaken, as the OP
suggests. *knolp*!
--
Jonathan Feinberg jdf@pobox.com Sunny Brooklyn, NY
http://pobox.com/~jdf
------------------------------
Date: Fri, 09 Feb 2001 22:30:58 GMT
From: hilljroberts@my-deja.com
Subject: UNIX core dumps - Segmentation Fault
Message-Id: <961r32$ppu$1@nnrp1.deja.com>
I have a basic question about core dumps. Everytime my program runs it
core dumps and doesn't do any editing to any files that it is supposed
to. However if I run it in debug mode, the entire program runs fine,
outputs and saves the necessary files, and does not core dump. Does
anyone know if there is something that I just forgot to do? I know it
has got to be some stupid error. Also, the program runs fine on
Windows based machines, it does not even display any errors from the
interpreter. What the heck!?!?!?!
Sent via Deja.com
http://www.deja.com/
------------------------------
Date: 09 Feb 2001 17:11:38 -0500
From: Jonathan Feinberg <jdf@pobox.com>
Subject: Re: Untangling NT Logon Scripts
Message-Id: <4ry3ik11.fsf@pobox.com>
"rude_al" <i.know.nothing@mindless.com> writes:
> I need some way of scanning the text of each file looking for CALL
> and NET USE commands. I've been told Perl is the way forward,
> anyone got any suggestions?
Perl is certainly the language of choice for a job like this. If you
buy the book "Learning Perl" and do all of the exercises (about a
week's work, or less if you really dig in) you will have most of the
skills you need for a job like the one you describe. What you do not
then know, you will know how to learn (by reading the docs that come
with perl).
You might also try to convince your employers to foot the bill for
some classroom training.
You will find this newsgroup very helpful for specific Perl-related
questions, especially if you explain exactly what you don't
understand, and tell folks what you've already read in an effort to
figure it out.
HTH.
--
Jonathan Feinberg jdf@pobox.com Sunny Brooklyn, NY
http://pobox.com/~jdf
------------------------------
Date: Fri, 09 Feb 2001 20:46:24 GMT
From: Horace <horace700@my-deja.com>
Subject: Wanted: Perl Programmer for Chatologica Metasearch Mods
Message-Id: <961kuu$k4i$1@nnrp1.deja.com>
Wanted: Perl programmer for paid services to incorporate specific
cookie management functions in Chatologica metasearch. Chatologica is
a socket based metasearcher similar to Anaconda. If interested, please
respond to horace700@my-deja.com
Sent via Deja.com
http://www.deja.com/
------------------------------
Date: Fri, 9 Feb 2001 16:21:14 -0600
From: "Matt Kruse" <mkruse@netexpress.net>
Subject: XML: "Flattening" XML to dot notation - best module?
Message-Id: <3a846cd4$0$1545@wodc7nh0.news.uu.net>
There seem to be quite a few XML modules available, and I've started looking
through some of them (specifically XML::Simple) to see if there are any that
provide an easy method to do the following:
I want to "flatten" XML into dot notation. For example:
<foo>
<bar>X</bar>
<bar2>Y</bar2>
</foo>
Becomes:
foo.bar = X
foo.bar2 = Y
Assuming the XML will be very simple (no attributes, all text values, etc)
then what would be the easiest way to accomplish this? Are there any modules
that will let me simply step through each element in the XML document and
give me it's fully qualified name?
Thanks.
--
Matt Kruse
http://www.mattkruse.com/
------------------------------
Date: 16 Sep 99 21:33:47 GMT (Last modified)
From: Perl-Users-Request@ruby.oce.orst.edu (Perl-Users-Digest Admin)
Subject: Digest Administrivia (Last modified: 16 Sep 99)
Message-Id: <null>
Administrivia:
The Perl-Users Digest is a retransmission of the USENET newsgroup
comp.lang.perl.misc. For subscription or unsubscription requests, send
the single line:
subscribe perl-users
or:
unsubscribe perl-users
to almanac@ruby.oce.orst.edu.
| NOTE: The mail to news gateway, and thus the ability to submit articles
| through this service to the newsgroup, has been removed. I do not have
| time to individually vet each article to make sure that someone isn't
| abusing the service, and I no longer have any desire to waste my time
| dealing with the campus admins when some fool complains to them about an
| article that has come through the gateway instead of complaining
| to the source.
To submit articles to comp.lang.perl.announce, send your article to
clpa@perl.com.
To request back copies (available for a week or so), send your request
to almanac@ruby.oce.orst.edu with the command "send perl-users x.y",
where x is the volume number and y is the issue number.
For other requests pertaining to the digest, send mail to
perl-users-request@ruby.oce.orst.edu. Do not waste your time or mine
sending perl questions to the -request address, I don't have time to
answer them even if I did know the answer.
------------------------------
End of Perl-Users Digest V10 Issue 260
**************************************
