Re: syntax differences and performance
And the clouds parted, and Adamiec, Larry said...
Given the following lines:
$MESSAGE .= PApplication Fee: B . $price . Dollars U.S./B;
$MESSAGE .= PApplication Fee: B$price Dollars U.S./B;
Is there a performance issue (or any other issue) for using either of
the above lines? Does it make any difference which is used?
Suppose it changed to the following:
$MESSAGE .= PApplication Fee: B . $query-param('price') .
Dollars U.S./B;
or
$MESSAGE .= PApplication Fee: B$query-param('price') Dollars
U.S./B;
I have used the first two lines without any noticeable difference, but I
didn't use any sort of debugging/timer tools either. I am debating with
myself about which line from the second set I should use. In all
instances the correct value of price was inserted into the variable
$MESSAGE.
Larry
Hi Larry-
The first thing I would suggest is to check out the Benchmark::Timer
module off of CPAN. You can time just about any section of your code
and print a report at the end to help you see where your optimization
time would best be spent. Let me know if you would like some example
code to see how to use it.
Using this and a simple (contrived) script, I ran some tests and it
looks like the inline versions (foo $var bar) might be slightly slower
than their append counterparts (foo . $var . bar), although the
timings sometimes drifted in the other direction based on how many trials
I ran (more trials == inline gets faster), a result I suspect can be
attributed to caching at some level. Basically, there wasn't enough
deviation to be able to point to either one as significantly slower than
the other. One of the heavier technical perldocs may have the answer, but
I did a quick scan through 'perldoc perltoc' and nothing jumped off the
page at me. YMMV.
So I would suggest running *your*own* tests on some test data of your own
to get a better idea what will work best in *your* environment. Also keep
in mind that unless your script does several thousand of these assignments
or your benchmarking shows that the majority of your script's runtime is
spent in these assignments, you probably won't see much of a difference
in performance by figuring out which one saves you 3 ms over the other. :)
I'm all for optimizing code just for the sake of good programming, but
if you're trying to ameliorate a performance problem, you want to look for
the part of your script where it's spending the most time.
HTH-
Brian
PS - $query-param('price') should be $query-param{'price'} - curly
brackets for hashes, parens for subroutines.
/~~\
| Brian Gerard Dawn is nature's way of telling you to|
| First initial + 'lists' go to bed. And to just stay there until |
| at technobrat dot com the evil yellow disk is gone again. |
\__/
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
http://learn.perl.org/ http://learn.perl.org/first-response
Re: File Size Calculator
And the clouds parted, and SilverFox said...
Hi all, I'm trying to writing a script that will allow a user to enter a
number and that number will be converted into KB,MB or GB depending on the
size of the number. Can someone point me in the right direction?
Example:
user enter: 59443
Script will output: 58M
SilverFox
Here's a little chunk that should give you about what you're looking
for, up to Tebibytes (2**40 bytes). Note that I used the binary
prefixes[1] (Kibi, Mebi, Gibi, Tebi) as opposed to the base-10 versions
(Kilo, Mega, Giga, Tera). Feel free to change them if you're so
inclined. :)
--- Begin Chunk ---
our %ByteCount = (
B = 1,
KiB = 2**10,
MiB = 2**20,
GiB = 2**30,
TiB = 2**40
);
sub prettybyte {
my $bytes = shift;
foreach my $unit ( qw{ TiB GiB MiB KiB B } ) {
if ($bytes = $ByteCount{$unit}) {
return sprintf(%4.3f $unit, $bytes/$ByteCount{$unit});
}
}
}
--- End Chunk ---
HTH[2]-
Brian
[1] http://www.alcyone.com/max/reference/physics/binary.html
-anyone remember offhand the URL to the /. story on these, btw?
[2] I'm a little rushed at the moment, so I don't have time to fill
in any details of how it works. Let me know if you want/need an
explanation and I'll be happy to provide one. :)
/~~\
| Brian Gerard Some drink at the fountain of|
| First initial + 'lists' knowledge...others just gargle. |
| at technobrat dot com |
\__/
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
http://learn.perl.org/ http://learn.perl.org/first-response
Re: File Size Calculator
And the clouds parted, and Brian Gerard said... [1] http://www.alcyone.com/max/reference/physics/binary.html -anyone remember offhand the URL to the /. story on these, btw? ...never mind. Found it. (uncaught typo on my first google query... DOH!) http://slashdot.org/articles/99/08/10/0259245.shtml /~~\ | Brian Gerard Give me liberty or give me something | | First initial + 'lists' of equal or lesser value | | at technobrat dot com from your glossy 32-page catalog. | \__/ -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] http://learn.perl.org/ http://learn.perl.org/first-response
Favorite email parser?
Hi all- I'm about to start a project that goes a little something like this... Right now, alerts get automatically emailed to a central address which due to the volume is largely ignored. You know the drill: Ooo! It would be great if we got notified when resource X is running low! Then we could anticipate the need for it and get ahead of the game. Repeat for resources A-Z, AA-ZZ, etc, etc and you get a mailbox that quickly becomes filled with alerts that no one reads, largely due to the fact that the reading alone would be a full time job. We can receive anywhere from a couple of hundred to a couple of thousand of these per day. So I'm planning to set up the address to forward to a backline server, where my Brilliantly Written(tm) perl script will crunch said alerts. I'd like it to chop them up into little bits and insert the bits into an intelligently-structured database of some sort. I'm planning on Postgres at this point. Then there will be another script, available to the proper people via an internal website, where queries can be made against the database according to a yet-to-be-designed web interface. The main thing I want is a simplification of the alert viewing. So you might see 3245 instances of alert Y, which could be drilled down into based on various criteria, rather than the first page (out of 65) of 50 alerts all of the same type. Statistical trending and other such pie-chart-able things may show up in the future, but not for now. So my questions are these: 1) Does this sound like a reasonable approach to the problem, given the relatively low amount of detail I've provided? Anyone dealt with this type of thing before and have any gotchas for me? 2) From what I've read, Postgres beats MySQL performance-wise, which is why I chose it, but IANADBA. Any votes on a free DB to use for this? Obviously, it needs to have a perl interface or CPAN module available. Regardless of which DB you like, what's your favorite module for interacting with it? ...and, umm, if you could point me to any Getting Started docs, that would be just lovely. :) 3) Which module would you suggest for parsing email? There are a few of them out there, and I really don't have the time to try each one out until I find the one I like the best. :) Answers to any or all of the above gratefully received! aTdHvAaNnKcSe- Brian /~~\ | Brian Gerard O'Reilly is to a system administrator| | First initial + 'lists' as a shoulder length | | at technobrat dot com latex glove is to a veterinarian. | \__/ -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] http://learn.perl.org/ http://learn.perl.org/first-response
Re: Favorite email parser?
Forgot to mention the platform. This will be on a FreeBSD system, and the emails will be in the standard flat file format (mbox). Didn't want anyone going too deep on the M$/Exchange side of the house. :) /~~\ | Brian GerardRed meat isn't bad for you. Fuzzy | | First initial + 'lists' blue-green meat is bad for you. | | at technobrat dot com | \__/ -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] http://learn.perl.org/ http://learn.perl.org/first-response
Re: matching hypenated strings
And the clouds parted, and deb said...
## begin ##
while ($name = DATA) {
$name =~ /(\w*)\.*/;
$name{$1}++;
$name =~ /(\w+)/;
print $ \n;
}
__DATA__
tibor.test.net
mars.test.net
moon-bx-r.test.net
moon-bs-d.test.net
moon-bt-321.test.net
## end ##
This works for hostnames without hyphens, but when there is a hyphen in the
name, everything after the hyphen is ignored. I've been trying things like
$name =~ /[a-z]*\-*\-*/ with no luck. The data coming into the expression
may or may not be fully qualified, so I can't just take everything to the left
of .test.net, and the domain name may be different at times, anyway.
So what I'm left with finding an expression that will match any alphanumeric,
with 0 or more embedded dashes. It sounds simple, but I can't seem to find
it.
What am I missing?
Two things:
1) The regex you're looking for is likely /[-\w]+/, which says match one
or more dashes or word characters. This will slurp up everything up to
the first non-word, non-dash character.
2) You can probably simplify your script to
## begin ##
while (DATA) {
(print $ \n and $name{$}++) if /[-\w+]+/;
}
__DATA__
tibor.test.net
mars.test.net
moon-bx-r.test.net
moon-bs-d.test.net
moon-bt-321.test.net
## end ##
Its output is :
ksh$ ./mchname
tibor
mars
moon-bx-r
moon-bs-d
moon-bt-321
ksh$
HTH-
Brian
/~~\
| Brian Gerard A child of five could understand |
| First initial + 'lists' this! Fetch me a child of five. |
| at technobrat dot com |
\__/
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: matching hypenated strings
And the clouds parted, and Rob Dixon said... That's a misuse of 'and' Brian. It says that the hash element should be incremented only if the print succeeds, which isn't what was intended. If you mean a code block then use a code block. Mea Culpa. You're absolutely correct. Thanks for the correction. :) /~~\ | Brian GerardCrime does not pay... as well as politics. | | First initial + 'lists'| | at technobrat dot com | \__/ -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: regexp matching- real simple!
And the clouds parted, and Nandita Mullapudi said...
This has to be really easy- but I cant get it to work:
I want my script to carry out a loop only if the number stored in
$diff is positive,
so i say
if ($diff =~m/^[-\d+]/) {blah blah}
but it won't work. why?
TIA
-Nandita
Hi Nandita,
The problem is that you're trying to match a character class instead of the
pattern you think you're matching.
Your pattern: ^[-\d+]
Which means : At the beginning of the line, match a single dash, digit, or
plus sign.
In other words, it will match '-23 fnords', '42', '--Socrates', '++$foo', and
a slew of other things you never indended. :) A character class matches a
single character, no more.
What you probably want is the ($diff == abs($diff)) solution already proposed,
since you've stated that $diff should be a number.
That being said, the pattern you were trying for would probably be: \A-\d+\Z
Which says: At the beginning of the string, match a single dash, followed by
one or more digits, followed by the end of the string.
HTH-
Brian
/~~\
| Brian Gerard I don't suffer from insanity.|
| First initial + 'lists'I enjoy every minute of it. |
| at technobrat dot com |
\__/
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: regexp matching- real simple!
And the clouds parted, and Jeff 'japhy' Pinyan said...
Well, your regex is matching either a digit, a -, or a + as the first
character in $diff. If you wanted to match a POSITIVE number, you'd be
better with /^+?\d/, but WHY are you using a regex for a MATH problem?
if ($diff 0) { ... }
Bah. That's the trouble with starting a reply, getting otherwise occupied
before completing it, and then sending it in late.
What he said. :)
/~~\
| Brian Gerard We now return to our regularly |
| First initial + 'lists' scheduled flame-throwing. |
| at technobrat dot com |
\__/
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Why is parsing your own form data a bad idea?
And the clouds parted, and Dan Anderson said... This seems kind of silly. Can anyone explain to me why this is? Beats me. I've been rolling my own cgi-handlers since perl4 with no discernable ill effects. :) Let me know if you want some sample code. Brian /~~\ | Brian GerardI'm sorry, are the voices | | First initial + 'lists' in my head bothering you? | | at technobrat dot com | \__/ -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Training in the Middle and Far East
And the clouds parted, and Rajeev Prasad said... computer science is no science at all. It is only layers of information. It is physics and mathematics which makes this number crunching plastic do its job. that is science. rest of the stuff as Jerry said it is a BIGENNER's list. so it will be like the way it is. Hi Rajeev- I'm afraid I must disagree. Science can be defined (in this sense) as knowledge or a system of knowledge covering general truths or the operation of general laws especially as obtained and tested through scientific method[1], and lest I be accused of circular reasoning, the scientific method referred to above would be principles and procedures for the systematic pursuit of knowledge involving the recognition and formulation of a problem, the collection of data through observation and experiment, and the formulation and testing of hypotheses.[2] It is true that mathematics and physics (as well as chemistry) are integral parts (excuse the pun) of computer science, just as mathematics can be said to be an integral part of quantum mechanics. They are prerequisite bases of knowledge without which the composite science would be well nigh impossible to discuss, much less work in. This does not mean that the composite science is not a science in its own right. Computer scientists propose theorems and design experiments to test said theorems. Their theories predict behaviors resulting from given conditions and can be demonstrated to be valid or not. Big O notation, analysis of algorithms, cryptology, information theory, pattern recognition, graph theory, computational biology; all of these are examples of tools, branches, and applications of a real, definable, rigorous science. Try reading through one or two volumes of Knuth's[3] The Art of Computer Programming and try to tell me that computer science isn't a real science. ;) Sorry, I know far too many brilliant CS researchers to let this comment go by unanswered. :) Respectfully, Brian [1] http://www.m-w.com/cgi-bin/dictionary?book=Dictionaryva=science Definition 3a. [2] http://www.m-w.com/cgi-bin/dictionary?book=Dictionaryva=scientific+method [3] http://www-cs-staff.stanford.edu/~knuth/index.html /~~\ | Brian Gerard Any sufficiently advanced technology is | | First initial + 'lists' indistinguishable from a perl script.| | at technobrat dot com | \__/ -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Training in the Middle and Far East
And the clouds parted, and Rajeev Prasad said... SNIP! And the clouds parted, and Rajeev Prasad said... computer science is no science at all. It is only layers of information. It is physics and mathematics which makes this number crunching plastic do its job. that is science. rest of the stuff as Jerry said it is a BIGENNER's list. so it will be like the way it is. I'm afraid I must disagree. Science can be defined (in this sense) as knowledge or a system of knowledge covering general truths or the operation of general laws especially as obtained and tested through scientific method[1], and lest I be accused of circular reasoning, the scientific method referred to above would be principles and procedures for the systematic pursuit of knowledge involving the recognition and formulation of a problem, the collection of data through observation and experiment, and the formulation and testing of hypotheses.[2] SNIP! [1] http://www.m-w.com/cgi-bin/dictionary?book=Dictionaryva=science Definition 3a. [2] http://www.m-w.com/cgi-bin/dictionary?book=Dictionaryva=scientific+method Hi again Rajeev- I think perhaps we have a basic disagreement on the definition of science. I have quoted what I accept as a rigorous definition (with cites from a well known authority on the English language, Merriam-Webster). Regardless of whether you have corresponding citations, I would be interested to know what you consider science to be. Clearly, you rate mathematics, physics, and chemistry as sciences (I would agree). Examples aside, what is a good definition of science as far as you are concerned? Brian /~~\ | Brian Gerard The moon is covered with the| | First initial + 'lists' results of astronomical odds.| | at technobrat dot com | \__/ -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: What is my name
And the clouds parted, and Kevin Pfeiffer said...
Here's a line that will give you exactly that.
($PROGNAME = $0) =~ s|(\.?\.?(/[^/]+)*/)?([^/]+)|$3|;
I have two questions...
I came up with this earlier today:
my ($program) = $0 =~ /([^\/]+)$/;
Doesn't it do the same? (Question 1)
Sure looks that way, yes. :) I realized after you posted your (much,
_much_) simpler regex that the reason mine was so convoluted was that
the script I was using it in originally needed to capture the path the
script was invoked with as well as the script name. For just the script
name your solution should perform admirably.
If I don't escape the slash in the char class -- i.e. /([^\/]+)$/ -- I get
this error:
Unmatched [ in regex; marked by -- HERE in m/([ -- HERE ^/ at ./test-0
line 7.
This makes no sense to me (since this is Perl and not sed or something)...
(implied Question 2)
The reason is that you're using '/' as the delimiter for the m// operator,
and '/' isn't automatically escaped in a character class (it's not a
metacharacter). I got away with it because I was using '|' as the delimiter
in my substitution.
What I mean is this. Talking about the m// and s/// operators, you can
use just about anything in place of the '/'s. (We won't discuss using the
'x' option to m// and s///, for the sake of brevity and simplicity.)
m/pattern/ #is the same as
m;pattern; #which is the same as
m|pattern|
and
s/pattern/replacement/ #is equivalent to
s#pattern#replacement#
...or even...
s{pattern}{replacement}
...but I digress. :)
So to avoid what the Camel Book refers to as 'leaning toothpick syndrome'
when matching unix pathnames, I use a delimiter other than '/' so that I
don't have to always type '\/' when I want to match a slash in the path.
As for why it wouldn't lose its specialness inside of a character class,
that only applies to regex metacharacters ('.', '*', '+', etc). When perl
looks for the pattern to match it looks for everything up to the first
matching delimiter before trying to compile it. So when you said
m/([^/]+)$/
perl tried to use
([^
as its regex, which isn't valid. That's why it was complaining about an
unmatched [. It only saw one. :)
HTH-
Brian
/~~\
| Brian Gerard I am Pooh of Borg. Bother is|
| First initial + 'lists'futile; hunny will be assimilated. |
| at technobrat dot com |
\__/
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: initialising a list of variables
And the clouds parted, and Rob Dixon said...
Gary Stainburn wrote:
SNIP!
Hi Gary.
Just one more offering:
$_ = '' foreach my ($fred, $ginger);
HTH,
Rob
...and in the spirit of TMTOWTDI...
map {$_=} my ($fred, $ginger);
:)
/~~\
| Brian Gerard Do not taunt Happy Fun Ball.|
| First initial + 'lists'|
| at technobrat dot com |
\__/
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: What is my name
And the clouds parted, and [EMAIL PROTECTED] said...
Hi,
I have a little problem. I have script test.pl and inside this script I want to
know what is my name (I mean this test.pl.
test.pl script
---
my $this_scripts_name_is = who_am_I();
sub who_am_I
{
my $my_name = ; # what I have to put to this to get test.pl scripts
name
return $my_name;
}
I ran into exactly this issue, but I wanted to just get the actual script
name. $0 returns the entire path the script was invoked as. ie - $0 could
be ../../foo/bar/blarch/scriptname.pl, but I just wanted scriptname.pl.
Here's a line that will give you exactly that.
($PROGNAME = $0) =~ s|(\.?\.?(/[^/]+)*/)?([^/]+)|$3|;
...or, to put it in the above terms:
my $this_scripts_name_is = who_am_I();
sub who_am_I
{
my $PROGNAME;
($PROGNAME = $0) =~ s|(\.?\.?(/[^/]+)*/)?([^/]+)|$3|;
return $PROGNAME;
}
HTH-
Brian
/~~\
| Brian Gerard Overflow on /dev/null; |
| First initial + 'lists'please empty the bit bucket.|
| at technobrat dot com |
\__/
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: script to test a file.
And the clouds parted, and Wiggins d Anconia said...
Rick Bragg wrote:
Hi,
I want to write a script that will test the contents of a file.
The file being tested will pass only if it contains nothing more
than an
ip address on one line. Does anyone have a sample of a simple regex to
accomplish this?
/\d+\.\d+\.\d+\.\d+\n?/s
Beware of insufficient regexes in regex clothing. The above will
certainly find one or more digits followed by a dot, followed by one or
more digits followed by a dot, etc. But it does *NOT* match an IP
address...
/^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}$/
Is at least closer, but it is still not sufficient, as IP addresses are
bounded, I believe, at 255 so 336.47.894.0, will match the above, but is
not a valid IP address.
How strict do you need to be, how sure are you about the data in the files?
Good point.
At it's most basic, to match any valid IPv4 address alone on a line, I would
use
/\A((1?\d{1,2}|2([0-4]\d|5[0-5]))\.){3}(1?\d{1,2}|2([0-4]\d|5[0-5]))\z/
Which will match a string containing only a dotted quad from 0.0.0.0 to
255.255.255.255, inclusive. Note that I'm assuming the string in
question contains the _entire_ contents of the file.
A couple of things to note, though:
* A valid IP is actually any 32 bit integer from 0 to 4294967295 (try
putting 3639555427 in Mozilla (or supposedly any other well-behaved
browser... in other words, _not_ IE)... it will take you to Google
because that number is actually 216.239.53.99 - the address for
www.google.com - in base-10).
* This pattern doesn't exclude RFC1918 reserved addresses (10/8,
172.16/12, and 192.168/16), or multicasts.
* It doesn't take netmasks into account. 1.2.3.255/16 is a valid
host address, while 1.2.3.255/24 is not (it's a broadcast).
Taking these sorts of things into account could make for a bit hairier
solution. ;) Caveat user.
/~~\
| Brian GerardI'm writing an unauthorized autobiography. |
| First initial + 'lists'|
| at technobrat dot com |
\__/
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
