Updating a text file

[Mercedissa1@netscape.net]

I have a text file with a records like:

smith, James, Dr., 115 Fourth Street, Chicago, IL, 32012, $20.00: [email]james_naismith@aol.com[/email]

and another text file with placeholders like:

Hello $last_name, $first_name:

I want to read the first textfile and update the second one.

Any ideas!

__________________________________________________ ________________
New! Unlimited Netscape Internet Service.
Only $9.95 a month -- Sign up today at [url]http://isp.netscape.com/register[/url]
Act now to get a personalized email address!

Netscape. Just the Net You Need.

[Jan Eden]

Hi,

[email]Mercedissa1@netscape.net[/email] wrote:

>I have a text file with a records like:
>
>smith, James, Dr., 115 Fourth Street, Chicago, IL, 32012, $20.00:
>james_naismith@aol.com
>
>and another text file with placeholders like:
>
>Hello $last_name, $first_name:
>
>I want to read the first textfile and update the second one.
>
>Any ideas!

This is a very general question. Have you already tried something? Have your read an Introduction to Perl (O'Reilly's "Learning Perl" is highly recommended)?

Basically, what you have is a data file and a template file. You should open the data file and read the data into a hash (using pattern matching), then merge the hash and the template file (replacing your placeholders), putting out a scalar variable for each set of data. A very simple solution (onceyou established the hash) could be a subroutine like this:

my $build_page = sub {
# takes a parameter hash as its argument
my %param = ( @_, );
open TEMPLATEFILE, $templatepath or die "could not open file '$templatepath': $!";
my $template = join '', <TEMPLATEFILE>;
# now replace %%placeholders%% in $template with values from %param hash
$template =~ s{ %% ( .*? ) %% }{ exists( $param{$1} ) ? $param{$1} : ''}gex;
return $template;
}

It is not a good idea to choose strings like $last_name as your placeholders, since such strings are considered variable names in Perl.

This might not sound too helpful, but if you deliver a piece of code, one of the Perl wizards on this list will certainly help you.

- Jan
--
If all else fails read the instructions. - Donald Knuth

[R.]

Jan Eden wrote:
 
>
> This is a very general question. Have you already tried something? Have your read an Introduction to Perl (O'Reilly's "Learning Perl" is highly recommended)?
>
> Basically, what you have is a data file and a template file. You should open the data file and read the data into a hash (using pattern matching), then merge the hash and the template file (replacing your placeholders), putting out a scalar variable for each set of data.[/ref]

I would do it the other way around. I am guessing here that the OP wants to use the second file as a tameplate, as you suggest also, but consider the use of a template. He probably does not want to actually update the original, but to spawn personalized copies for each customer/recipient record in the data file.

Since the text of the template is the material that would be repeated each time through the loop, the text is what he should slurp. Then he can go through the data-file record by record, making a personalized version from the template for each one.

Of course I could be wrong--the OP might really want to modify the template itself. In that case, he would first have to decide which record he would want to take the values from

For the time being, though, I am assuming that he simply misstated the procdure. If he has the template in memory as a concatenated string, he can substitute in the proper variables for the place holders with each record

my $template_string = get_file_contents();
my $contact = {};
my $count = 0;
while (<DATA_FILE>) {
chomp;
($contact->{'last name'}, $contact->{'first name'} ...$conatact->{'current balance'}) = split /,\s+/, $_;
(my $inerpolated_string = $template_string) =~ s/%(.*?)%/$contact->{$1}/g;
open(OUT_FILE, 'boiler_plate' . ++$count . '.txt'
print OUT_FILE $inerpolated_string;
}

This should work alright. Took a few rounds to get the typos out, but the following worked from the command line:
my $template_string = 'Hello, %first name% %last name%, how are you';

my $contact = {};
my $count = 0;
while (<DATA>) {
chomp;
($contact->{'last name'}, $contact->{'first name'}) = split /,\s+/, $_;
(my $interpolated_string = $template_string) =~ s/%(.*?)%/$contact->{$1}/g;
open(OUT_FILE, '>boiler_plate' . ++$count . '.txt');
print OUT_FILE $interpolated_string;
}
__DATA__
Smith, John
Doe, Jack

The file rendered were:
boiler_plate1.txt
Hello, John Smith, how are you
boiler_plate2.txt
Hello, Jack Doe, how are you

Joseph

[Mame]

Thanks, I will give it a try!

"R. Joseph Newton" wrote: 
> >
> > This is a very general question. Have you already tried something? Have your read an Introduction to Perl (O'Reilly's "Learning Perl" is highly recommended)?
> >
> > Basically, what you have is a data file and a template file. You should open the data file and read the data into a hash (using pattern matching), then merge the hash and the template file (replacing your placeholders), putting out a scalar variable for each set of data.[/ref]
>
> I would do it the other way around. I am guessing here that the OP wants to use the second file as a tameplate, as you suggest also, but consider the use of a template. He probably does not want to actually update the original, but to spawn personalized copies for each customer/recipient record in the data file.
>
> Since the text of the template is the material that would be repeated each time through the loop, the text is what he should slurp. Then he can go through the data-file record by record, making a personalized version from the template for each one.
>
> Of course I could be wrong--the OP might really want to modify the template itself. In that case, he would first have to decide which record he would want to take the values from
>
> For the time being, though, I am assuming that he simply misstated the procdure. If he has the template in memory as a concatenated string, he can substitute in the proper variables for the place holders with each record
>
> my $template_string = get_file_contents();
> my $contact = {};
> my $count = 0;
> while (<DATA_FILE>) {
> chomp;
> ($contact->{'last name'}, $contact->{'first name'} ...$conatact->{'current balance'}) = split /,\s+/, $_;
> (my $inerpolated_string = $template_string) =~ s/%(.*?)%/$contact->{$1}/g;
> open(OUT_FILE, 'boiler_plate' . ++$count . '.txt'
> print OUT_FILE $inerpolated_string;
> }
>
> This should work alright. Took a few rounds to get the typos out, but the following worked from the command line:
> my $template_string = 'Hello, %first name% %last name%, how are you';
>
> my $contact = {};
> my $count = 0;
> while (<DATA>) {
> chomp;
> ($contact->{'last name'}, $contact->{'first name'}) = split /,\s+/, $_;
> (my $interpolated_string = $template_string) =~ s/%(.*?)%/$contact->{$1}/g;
> open(OUT_FILE, '>boiler_plate' . ++$count . '.txt');
> print OUT_FILE $interpolated_string;
> }
> __DATA__
> Smith, John
> Doe, Jack
>
> The file rendered were:
> boiler_plate1.txt
> Hello, John Smith, how are you
> boiler_plate2.txt
> Hello, Jack Doe, how are you
>
> Joseph
>
> --
> To unsubscribe, e-mail: org
> For additional commands, e-mail: org
> <http://learn.perl.org/> <http://learn.perl.org/first-response>[/ref]

[Jan]

R. Joseph Newton wrote:
 
>>
>>This is a very general question. Have you already tried something?
>>Have your read an Introduction to Perl (O'Reilly's "Learning Perl"
>>is highly recommended)?
>>
>>Basically, what you have is a data file and a template file. You
>>should open the data file and read the data into a hash (using
>>pattern matching), then merge the hash and the template file
>>(replacing your placeholders), putting out a scalar variable for
>>each set of data.[/ref]
>
>I would do it the other way around. I am guessing here that the OP
>wants to use the second file as a tameplate, as you suggest also, but
>consider the use of a template. He probably does not want to
>actually update the original, but to spawn personalized copies for
>each customer/recipient record in the data file.
>
>Since the text of the template is the material that would be repeated
>each time through the loop, the text is what he should slurp. Then
>he can go through the data-file record by record, making a
>personalized version from the template for each one.
>[/ref]
You are probably right and that's what I meant. My description was probablya bit misleading. The way I do it is create the hash, slurp the template file in a scalar and produce an output string for each data set in the hash (which can be written to a new file, named after the respective hash key.

- Jan
--
These are my principles and if you don't like them... well, I have others. - Groucho Marx