writing get_script as an external routine callable by C

[Franken Sense]

With forum help, I've been able to do pretty good damage with using perl
for input. It's *so much* easier than my alternatives with compiled
languages that I really feel like I'm in touch with the virtue of laziness.

I'm simply too immature with perl to, say, populate a binary tree with the
data or do many of the things that I can do with compiled languages, once I
have the data where I want it. This is, I think, a nice final touch by
Mark Krahn:

#!/usr/bin/perl
# perl m13.pl
use warnings;
use strict;

# open input file
my $filename = 'text43.txt';
open(my $fh, '<', $filename) or
die "cannot open $filename for reading: $!";

# open output file
my $filename2 = 'outfile16.txt';
open(my $gh, '>', $filename2) or
die "cannot open $filename2 for writing: $!";

local $/="";

while ( <$fh> )
{
my ( $verse, @s ) = split;
my $script = join ' ', @s;
print $gh "$verse $script\n";
}


# close input and output files
close($gh) or die("Error closing $filename2: $!");
close($fh) or die("Error closing $filename: $!");

# abridged output:

44:004:002 Being grieved that they taught the people, and preached through
Jesus the resurrection from the dead.
44:004:003 And they laid hands on them, and put them in hold unto the next
day: for it was now eventide.
44:004:004 Howbeit many of them which heard the word believed; and the
number of the men was about five thousand.
44:004:005 And it came to pass on the morrow, that their rulers, and
elders, and scribes,


So now I want main to have to call a routine to get the next $verse and
$script.

@anything = get_script( $verse, \@s)

sub get_script
{

my ( $verse, @s ) = split;
my $script = join ' ', @s;
return something;
}

Not exactly beautiful code, but my first efforts rarely look nice. The
reference for this in the camel book is §6 : Passing References. p 224

Then there's the matter of calling a perl subroutine from C.

#include <EXTERN.h>
#include <perl.h>

static PerlInterpreter *my_perl;

int main(int argc, char **argv, char ** env)
{
char *args[] = {Null};
my_perl = perl_alloc();
perl_construct(my_perl);
perl_parse(my_perl, NULL, argc, argv, NULL);
call_argv("get_script", args);
perl_destruct(my_perl);
perl_free(my_perl);

return 0;
}

The reference here is §21 of the camel book, p. 540.

Does any of this look close?

 

[Franken Sense]

In Dread Ink, the Grave Hand of Ben Morrow Did Inscribe:

Quoth (e-mail address removed):

[snipped and reordered]

Reasonably so. You should be using perldoc perlembed as your reference
rather than the Camel book: the Camel was published shortly before the
release of perl 5.6.0, and a lot has changed since then. Most
importantly, you are missing PERL_SYS_INIT3 and PERL_SYS_TERM which must
be called first and last respectively.

You should also be aware that there are a number of undocumented
initializations that may or may not be necessary depending on your
platform. You can print a C file which ought to reproduce your current
perl binary by running

perl -MExtUtils::Miniperl -ewritemain

Perl embedding is quite subtle, and I would not really recommend it
until you are familiar with writing XS. In order to do anything useful
with the embedded interpreter, you will need to know how to use the perl
API to get at and interpret the Perl values in the program.

You are passing main's argc/v to perl_parse, which means your program
will need to be called with the same command-line arguments as perl
would be. If you want your C program to run a particular Perl program,
you will need to create your own argv array to pass to perl_parse. Note
that you must still pass main's argc/v/env to PERL_SYS_INIT3.

If you really want to try this, you need to start by reading perlembed
and perlcall, and then probably perlapi and perlguts.

With that amount of reading in order to embed perl into C, I'll put that on
the reading list for my next injury. Thank you.

You haven't yet explained why you're trying to do this. Given your
inexperience with perl, it's almost certainly going to be easier to
stick with writing in one language at a time for now.

I've been reading §6. The semantics section does not include an example
with inputs. I think the best example is in Tricks with Parameter Lists
with something like:

sub get_script {
my($verse, $script) =@_;
....

I was thinking that I would try to--thereafter--get these data in a hash,
which is close as I'm going to able to come to getting them into a tree. I
suppose a hash is a tree, why not?

Anyways, below I see
sub configuration {
my %options = @_;
....

If you actually want to pass a ref, your sub needs to look more like

sub get_script {
my ($verse, $sref) = @_;
my $script = join " ", @$sref;

Ok, so a little more of the same terminology with @_ . How does one
pronounce this symbol?

If main were going to open my_file and get the $verse and $script that
we've been using and calls get_script with

getscript($verse, $script);

and instead have:

sub get_script {
my %options = @_;

, what will %options look like? Could you print the hash in alphabetic
order? Can you search in blindingly-fast time for a particular chapter and
verse, indicated by the obvious interpretation of $verse?
--
Frank

When you encounter seemingly good advice that contradicts other seemingly
good advice, ignore them both.
~~ Al Franken,

 

[Jürgen Exner]

If main were going to open my_file and get the $verse and $script that
we've been using and calls get_script with

getscript($verse, $script);

and instead have:

sub get_script {
my %options = @_;

, what will %options look like?

You are passing a list of two scalars to the sub, these are copied into
a hash, which would mean you got a hash with one element, the value of
$verse being the hash key and the value of $script being the hash value
for this element.

Could you print the hash in alphabetic order?

You need to be more specific. Do you want to print the keys or the
values sorted by the alphabetical order of the keys or by the
alpabetical order of the values? Either one is easy enough.

Can you search in blindingly-fast time for a particular chapter and
verse, indicated by the obvious interpretation of $verse?

Can you reconstruct $verse (i.e. the hash key) from that particular
chapter and verse? If yes, then access will be O(1) in the general case.
You can't get any faster than that.

If you cannot reconstruct the hash key based on chapter and verse and
searching for that is _the_ critical operation in your program, then I
would suggest to redesign the datastructure such that a combination of
chapter and verse can be used as the key.

Or use a database system instead.

jue

 

[Jürgen Exner]

Quoth (e-mail address removed):

Umm, I don't think I'm understanding what you're doing here, or you're
not. In general it's best to post complete (but short) programs that you
have actually tested, as it makes things clearer.

If you have

sub get_script {
my %options = @_;

then it expects to be called like

get_script(verse => $verse, script => $script);

and you will then end up with

$options{verse} = $verse;
$options{script} = $script;

If you call it as you have above, you will have a single key which is
the string value of $verse, and its corresponding value will be $script.
I suspect this isn't what you want.

Actually, I think (but I may be wrong) that's exactly what he's looking
for.

If you are trying to build a hash up over successive calls to
get_script, you will need to either declare the hash outside the sub

Good point, missed that one.
And if he makes %options global, then there is little point in calling
that sub, he could just as well do a direct
%options = ($verse -> $script);
instead of the sub call.

In any case, you will want to assign the
provided key and value to different variables, and then insert them into
the hash: assigning to a hash like that replaces everything in it.

Another good catch! So do a
$options{$verse} = $script;
instead of that sub call.

jue

 

[Franken Sense]

In Dread Ink, the Grave Hand of Jürgen Exner Did Inscribe:

Testing now.

Actually, I think (but I may be wrong) that's exactly what he's looking
for.


Good point, missed that one.
And if he makes %options global, then there is little point in calling
that sub, he could just as well do a direct
%options = ($verse -> $script);
instead of the sub call.


Another good catch! So do a
$options{$verse} = $script;
instead of that sub call.

jue

I'll see if I can put that all together; I think I can.

I switched data sets to the second part of the old testament that you might
find at gutenberg.org. I think this is a solid choice, as I didn't want to
get to hung up on the specific form I ran into on my first sortie with
computer exegetics, namely Acts.

I really like the look of these new data:

Canticle of Canticles Chapter 4


Christ sets forth the graces of his spouse: and declares his love for
her.

4:1. How beautiful art thou, my love, how beautiful art thou! thy eyes
are doves' eyes, besides what is hid within. Thy hair is as flocks of
goats, which come up from mount Galaad.

How beautiful art thou. . .Christ again praises the beauties of his
church, which through the whole of this chapter are exemplified by a
variety of metaphors, setting forth her purity, her simplicity, and her
stability.

....

Gosh, I'm gonna have to look at nuns in a different light now. Anyways,
the form here is basically the same: if a line is a scripture, it begins
with a number that will serve as a key for searches.

This is the latest version:

#!/usr/bin/perl
# perl bb2.pl
use warnings;
use strict;

# open input file
my $filename = 'ot4.txt';
open(my $fh, '<', $filename) or
die "cannot open $filename for reading: $!";

# open output file
my $filename2 = 'outfile16.txt';
open(my $gh, '>', $filename2) or
die "cannot open $filename2 for writing: $!";

local $/="";

while ( <$fh> )
{
my ( $verse, @s ) = split;
my $script = join ' ', @s;
print "$verse $script\n";
print $gh "$verse $script\n";
}


# close input and output files
close($gh) or die("Error closing $filename2: $!");
close($fh) or die("Error closing $filename: $!");

# abridged output:

C:\MinGW\source>perl bb1.pl
Canticle of Canticles Chapter 2
Christ caresses his spouse: he invites her to him.
2:1. I am the flower of the field, and the lily of the valleys.
I am the flower of the field. . .Christ professes himself the flower of
mankind,
yea, the Lord of all creatures: and, ver. 2, declares the excellence of
his spo

....
bite and destroy the vines.
2:16. My beloved to me, and I to him who feedeth among the lilies,
2:17. Till the day break, and the shadows retire. Return: be like, my
beloved, t
o a roe, or to a young hart upon the mountains of Bether.

C:\MinGW\source>
--
Frank

The biases the media has are much bigger than conservative or liberal.
They're about getting ratings, about making money, about doing stories that
are easy to cover.
~~ Al Franken,

 

[Franken Sense]

In Dread Ink, the Grave Hand of Franken Sense Did Inscribe:

Gosh, I'm gonna have to look at nuns in a different light now.
http://lomas-assault.net/usenet/(!!hail mary).jpg


Anyways,
the form here is basically the same: if a line is a scripture, it begins
with a number that will serve as a key for searches.

I'm bombing out here on something I think is easy. How do I populate
@books with book1.txt that looks like

Book of Psalms
Book of Proverbs
Ecclesiastes
Solomon's Canticle of Canticles
Book of Wisdom
...
Prophecy of Zacharias
Prophecy of Malachias
First Book of Machabees
Second Book of Machabees

#!/usr/bin/perl
# perl bb3.pl
use warnings;
use strict;

# open input file
my $filename = 'book1.txt';
open(my $fh, '<', $filename) or
die "cannot open $filename for reading: $!";

# open output file
my $filename2 = 'outfile16.txt';
open(my $gh, '>', $filename2) or
die "cannot open $filename2 for writing: $!";

my @books;

while ( <$fh> )
{
chomp;



}


# close input and output files
close($gh) or die("Error closing $filename2: $!");
close($fh) or die("Error closing $filename: $!");
--
Frank

I said that Sean Hannity took residence up Newt Gingrich's butt from 94 to
98. I got that from British intelligence. It turns out he only took up
residence in 95.
~~ Al Franken

 

[Uri Guttman]

BM> chomp( my @books = <$fh> );

BM> If you prefer, you can expand that to three statements:

BM> my @books;
BM> @books = <$fh>;
BM> chomp @books;

BM> You can also use File::Slurp::read_file, which will handle opening the
BM> file, chomping the lines and closing the file for you.

read_file doesn't (yet) chomp lines. that option is in the todo
list. but it does make it easier to read in a file as lines. this should
work fine:

use File::Slurp ;
chomp( my @books = read_file( $file_name ) ) ;

uri

 

[Franken Sense]

In Dread Ink, the Grave Hand of Jürgen Exner Did Inscribe:

I don't quite understand this last sentence.

Another good catch! So do a
$options{$verse} = $script;
instead of that sub call.

jue

Ok, so far so good. Here's what this looks like now:

#!/usr/bin/perl
# perl bb6.pl
use warnings;
use strict;

# open input files
my $filename = 'book1.txt';
open(my $fh, '<', $filename) or
die "cannot open $filename for reading: $!";

my $filename3 = 'ot5.txt';
open(my $hh, '<', $filename3) or
die "cannot open $filename3 for reading: $!";

# open output file
my $filename2 = 'outfile16.txt';
open(my $gh, '>', $filename2) or
die "cannot open $filename2 for writing: $!";

my @books;
my %Scripts;
my %comments;

@books = <$fh>;
chomp @books;
print "@books\n";

local $/="";

while ( <$hh> )
{
my @s = split /\s+/, $_;
my $verse = $s[0];
my $script = join(' ', @s[1..$#s]);
$Scripts{$verse} = $script;
}

print %Scripts;
my @keys = keys %Scripts;
print @keys;

# close input and output files

close($hh) or die("Error closing $filename3: $!");
close($gh) or die("Error closing $filename2: $!");
close($fh) or die("Error closing $filename: $!");

Abridged output:
Book of Psalms Book of Proverbs Ecclesiastes Solomon's Canticle of
Canticles Book of Wisdom Ecclesiasticus Prophecy of Isaias Prophecy
of Jeremias Lamentations of Jeremias Prophecy of Baruch Prophecy of
....
My son, sow not evils in the furrows of injustice, and thou shalt not reap
them sevenfold.Whofirst hath perfect knowledge of her. . .Christ was the
first that had perfect knowledge of heavenly wisdom.50:8.And as the rainbow
giving light in bright clouds, and as the flower of roses in the days of
the spring, and as the lilies that are on the brink of the water, and as
the sweet smelling frankincense in the time of summer.39:21.All the works
of the Lord are exceeding good.4:11.And thou shalt be as the obedient son
of the most High, and he will have mercy on thee more than a mother.29:25.A
sinner that transgresseth the commandment of the Lord, shall fall into an
....
49:7.16:11.It46:14.46:2.30:13.38:30.26:6.25:17.8:20.
4:19.9:19.51:28.38:32.35:17.9:20.23:34.25:25.40:23.25:1
..38:6.25:12.15:1.1:26.23:13.2:5.12:19.10:4.20:19.18:1.19:
8.49:12.14:10.21:26.3:26.Jesus3:31.10:11.34:19.34:18.21:1
4.11:13.7:33.1:9.24:23.31:15.46:9.12:18.3:8.23:1.24:20.4:
4.15:3.25:14.20:22.17:30.41:1.40:6.50:2.1:17.God's2:15.32:
19.26:27.15:20.26:22.38:27.45:17.24:7.6:20.26:23.13:16.47:
4.4:3.30:5.11:25.Wise24:32.16:24.38:21.3:30.45:20.40:4.33:


It appears true that a hash is not ordered like a binary tree.

Where I'm hung up now is in creating a control that separates comments from
scriptures. Scriptures begin with numbers, comments don't. So I was
poking around for a function like isdigit and couldn't find anything.

If comments are to be a hash, then they need something to key on, and all
that comes to mind is the natural numbers. Here's what the pseudosource
looks like:

$my $counter = 0;
while ( <$hh> )
{
my @s = split /\s+/, $_;

if (s[0] is a number)

my $verse = $s[0];
my $script = join(' ', @s[1..$#s]);
$Scripts{$verse} = $script;
else
$counter++;
my $comment = join(' ', @s);
%comments($counter) = $comment;
}

Fishing for tips. I like Ecclesiastes and the sad jews in general like
Janeane Garafalo, Al Franken and Jon Stewart.

 

[Franken Sense]

In Dread Ink, the Grave Hand of Uri Guttman Did Inscribe:

BM> chomp( my @books = <$fh> );

BM> If you prefer, you can expand that to three statements:

BM> my @books;
BM> @books = <$fh>;
BM> chomp @books;

BM> You can also use File::Slurp::read_file, which will handle opening the
BM> file, chomping the lines and closing the file for you.

read_file doesn't (yet) chomp lines. that option is in the todo
list. but it does make it easier to read in a file as lines. this should
work fine:

use File::Slurp ;
chomp( my @books = read_file( $file_name ) ) ;

uri

C:\MinGW\source>perldoc File::Slurp
No documentation found for "File::Slurp".

C:\MinGW\source>
--
Frank

...................... o _______________ _,
` Good Morning! , /\_ _| | .-'_|
`................, _\__`[_______________| _| (_|
] [ \, ][ ][ (_|

 

[Jürgen Exner]

Where I'm hung up now is in creating a control that separates comments from
scriptures. Scriptures begin with numbers, comments don't. So I was
poking around for a function like isdigit and couldn't find anything.

perldoc -q number:
How do I determine whether a scalar is a number/whole/integer/float?

jue

 

[Uri Guttman]

FS> C:\MinGW\source>perldoc File::Slurp
FS> No documentation found for "File::Slurp".

duh! you install it from cpan!

uri

 

[Franken Sense]

In Dread Ink, the Grave Hand of Jürgen Exner Did Inscribe:

perldoc -q number:
How do I determine whether a scalar is a number/whole/integer/float?

jue

How do I determine whether a scalar is a number/whole/integer/float?
Assuming that you don't care about IEEE notations like "NaN" or
"Infinity", you probably just want to use a regular expression.

if (/\D/) { print "has nondigits\n" }
if (/^\d+$/) { print "is a whole number\n" }
if (/^-?\d+$/) { print "is an integer\n" }
if (/^[+-]?\d+$/) { print "is a +/- integer\n" }
if (/^-?\d+\.?\d*$/) { print "is a real number\n" }
if (/^-?(?:\d+(?:\.\d*)?|\.\d+)$/) { print "is a decimal number\n" }
if (/^([+-]?)(?=\d|\.\d)\d*(\.\d*)?([Ee]([+-]?\d+))?$/)
{ print "a C float\n" }

So, I want to test whether the first byte is a digit. I think this is
/\d/. The sad thing is that I can't even put it together at this point.

I have @s populated in paragraph mode and then split. How does the test
condition look with these two: s[0] /\d/ ?

$my $counter = 0;
while ( <$hh> )
{
my @s = split /\s+/, $_;

if (s[0] is a number)

my $verse = $s[0];
my $script = join(' ', @s[1..$#s]);
$Scripts{$verse} = $script;
else
$counter++;
my $comment = join(' ', @s);
%comments($counter) = $comment;
}

 

[Jürgen Exner]

Ok, here comes your last fish.
I _STRONGLY_ suggest to attend a Perl class or a self-study course where
you can learn the fundamentals of Perl in a structured and comprehensive
way instead of digging for random bits and pieces while trying to
implement some program.

In Dread Ink, the Grave Hand of Jürgen Exner Did Inscribe:


How do I determine whether a scalar is a number/whole/integer/float?
Assuming that you don't care about IEEE notations like "NaN" or

[...]

Yeah, I know the answer to that FAQ, there is no need to quote
paragraphs and paragraphs of it.

So, I want to test whether the first byte is a digit.

No, you probably don't. In all likelyhood you want to check if the first
character is a digit. Characters and bytes can be very different things.

I think this is
/\d/. The sad thing is that I can't even put it together at this point.

That's why I strongly recommend a more structured approach to your
learning endeavour.

I have @s populated in paragraph mode and then split. How does the test
condition look with these two: s[0] /\d/ ?

if (s[0] is a number)

If you mean "begins with a digit" as you said above
if ($s[0] =~ m/^\d/)

If you mean "is a number" as you are saying now
if ($s[0] =~ m/^\d+$/) #all characters are digits
or
if (! $s[0] =~ m/\D/) #does not contain any non-digits

Deciphering:
The m-operator is the match operator. It can be omitted if slashes are
used as delimiters, therefore it doesn't show up in the FAQ. I still
prefer to write it sometimes because it makes the program logic more
obvious.

Normally the m (and the s) operator will use $_. If you want them to
operate on a different variable instead, then you need to bind that
variable to the operation, which is achieved by the =~ operator.

And the argument of m is a regular expression, in this case delimited by
slashes:
Case 1 matches if the beginning of the string is followed by a digit
Case 2 matches if the beginning of the string is followed by 1 or more
digits followed by the end of the string.
Case 3 matches if the string contains (at least) one non-digit.

For further details see "perldoc perlretut" and "perldoc perlre".

jue

 

[Tad J McClellan]

In Dread Ink, the Grave Hand of Jürgen Exner Did Inscribe:

So, did you mean "begins with a digit" or not?

if ($s[0] =~ m/^\d/)

If you mean "is a number" as you are saying now
if ($s[0] =~ m/^\d+$/) #all characters are digits
or
if (! $s[0] =~ m/\D/) #does not contain any non-digits

if ($s[0] =~ m/\d/) seems to work.

That will match:

$s[0] = 'this is not the 4th verse';

So, did you mean "begins with a digit" or not?

 

[Franken Sense]

In Dread Ink, the Grave Hand of Franken Sense Did Inscribe:

I have @s populated in paragraph mode and then split. How does the test
condition look with these two: s[0] /\d/ ?

I got that part now:

#!/usr/bin/perl
# perl bb7.pl
use warnings;
use strict;

my $filename3 = 'ot5.txt';
open(my $hh, '<', $filename3) or
die "cannot open $filename3 for reading: $!";

# open output file
my $filename2 = 'outfile16.txt';
open(my $gh, '>', $filename2) or
die "cannot open $filename2 for writing: $!";

my %Scripts;

local $/="";

while ( <$hh> )
{
my @s = split /\s+/, $_;
my $verse = $s[0];

if ($s[0] =~ m/\d/)
{
print $_;
}
my $script = join(' ', @s[1..$#s]);
$Scripts{$verse} = $script;
}


# close input and output files

close($hh) or die("Error closing $filename3: $!");
close($gh) or die("Error closing $filename2: $!");

Abridged output:

1:1. All wisdom is from the Lord God, and hath been always with him,
and is before all time.

1:2. Who hath numbered the sand of the sea, and the drops of rain, and
the days of the world? Who hath measured the height of heaven, and the
breadth of the earth, and the depth of the abyss?

1:3. Who hath searched out the wisdom of God that goeth before all
things?
....
26:28. Two sorts of callings have appeared to me hard and dangerous: a
merchant is hardly free from negligence: and a huckster shall not be
justified from the sins of the lips.

27:1. Through poverty many have sinned: and he that seeketh to be
enriched, turneth away his eye.
....

Thanks for your help.
--
Frank

[Newt Gingrich] is the most unpopular politician in America. His favorable
rating is only four points higher than the Unabomber.
~~ Al Franken, 1996

 

[Franken Sense]

In Dread Ink, the Grave Hand of Jürgen Exner Did Inscribe:

Ok, here comes your last fish.
I _STRONGLY_ suggest to attend a Perl class or a self-study course where
you can learn the fundamentals of Perl in a structured and comprehensive
way instead of digging for random bits and pieces while trying to
implement some program.

In Dread Ink, the Grave Hand of Jürgen Exner Did Inscribe:


How do I determine whether a scalar is a number/whole/integer/float?
Assuming that you don't care about IEEE notations like "NaN" or

[...]

Yeah, I know the answer to that FAQ, there is no need to quote
paragraphs and paragraphs of it.

So, I want to test whether the first byte is a digit.

No, you probably don't. In all likelyhood you want to check if the first
character is a digit. Characters and bytes can be very different things.

I think this is
/\d/. The sad thing is that I can't even put it together at this point.

That's why I strongly recommend a more structured approach to your
learning endeavour.

I have @s populated in paragraph mode and then split. How does the test
condition look with these two: s[0] /\d/ ?

if (s[0] is a number)

If you mean "begins with a digit" as you said above
if ($s[0] =~ m/^\d/)

If you mean "is a number" as you are saying now
if ($s[0] =~ m/^\d+$/) #all characters are digits
or
if (! $s[0] =~ m/\D/) #does not contain any non-digits

if ($s[0] =~ m/\d/) seems to work.

Deciphering:
The m-operator is the match operator. It can be omitted if slashes are
used as delimiters, therefore it doesn't show up in the FAQ. I still
prefer to write it sometimes because it makes the program logic more
obvious.

Normally the m (and the s) operator will use $_. If you want them to
operate on a different variable instead, then you need to bind that
variable to the operation, which is achieved by the =~ operator.

And the argument of m is a regular expression, in this case delimited by
slashes:
Case 1 matches if the beginning of the string is followed by a digit
Case 2 matches if the beginning of the string is followed by 1 or more
digits followed by the end of the string.
Case 3 matches if the string contains (at least) one non-digit.

For further details see "perldoc perlretut" and "perldoc perlre".

jue

Thx. I'll take a look later.
--
Frank

If you put the two Bushs together in their over seven years of their two
presidencies, not one new job has been created. Numbers do not lie. If you
extrapolated from that, if the Bushs had run this country from its very
beginning to the current time, not one American would have ever worked.
We'd be hunter-gatherers.
~~ Al Franken, in response to the 2004 SOTU address

 

[Tad J McClellan]

It appears true that a hash is not ordered

What do you mean "appears"?

The documentation for the data structure you mention says
quite clearly that it is not ordered.

perldoc perldata

... Hashes are unordered...

If comments are to be a hash, then they need something to key on, and all
that comes to mind is the natural numbers.

The data structure that is indexed by numbers is called an "array".

A hash data structure is indexed by strings.

Fishing for tips.

No need to state the painfully obvious.

 

[Uri Guttman]

FS> if ($s[0] =~ m/\d/) seems to work.

seems to work is not a proper logical statement. it works or it
doesn't. does it work for all test cases? for a non-number? for a number
with garbage in front of it? you have to think a bit before you claim
something works and you don't even understand why it may fail.

and i agree. take a perl class. read a whole perl book and do the
exercises. this dribs and drabs approach is not the way to learn a
programming language. it is like learning to drive but pushing all the
pedals, levers and buttons in any order with random pressures and seeing
what may or may not happen.

uri

 

[Jürgen Exner]

In Dread Ink, the Grave Hand of Jürgen Exner Did Inscribe:

So, I want to test whether the first byte is a digit.

No, you probably don't. In all likelyhood you want to check if the first
character is a digit. Characters and bytes can be very different things.
If you mean "begins with a digit" as you said above
if ($s[0] =~ m/^\d/)

If you mean "is a number" as you are saying now
if ($s[0] =~ m/^\d+$/) #all characters are digits
or
if (! $s[0] =~ m/\D/) #does not contain any non-digits

if ($s[0] =~ m/\d/) seems to work.

Between "works" and "seems to work" is a similar difference as between
"well done" and "meant well".

jue

 

[Franken Sense]

In Dread Ink, the Grave Hand of Uri Guttman Did Inscribe:

FS> C:\MinGW\source>perldoc File::Slurp
FS> No documentation found for "File::Slurp".

duh! you install it from cpan!

uri

I'm curious how you treat your clients. I was looking for documentation.

Do you tell them to go back to college? lose weight?

As OP, I certainly didn't ask about the dumb sidebar issue of File::Slurp.
--
Frank

I said that Sean Hannity took residence up Newt Gingrich's butt from 94 to
98. I got that from British intelligence. It turns out he only took up
residence in 95.
~~ Al Franken