new lines in the way

[new2perl]

Hi,

I am trying to parse a text file that looks like this:

__
VAR1=15
VAR2=16
VAR=17

FIRSTJOKE=this is my first joke. its isnt that funny but its still
pretty good bla hbls dfd fsdf dsf sdfsd sd fdssd sdfdsf
dsfdsfsdsdfsfsd

SECONDJOKE=fdsf sdfdsf sdfdsfsdfdsfdsfdsfderew
rweeeeeeeeeeeeeeeeeeeeeeeer2 wr ewre rewr er ewewrrewrew rewrewrew
ewrew ewr ewr er ewr ewr er ewr we rew ew
rewrewrewrewrewrererewrewrewrewrewrewrewrwerewrererewreerewrewrerererererer

THIRDJOKE=notfunny
---

My code is like this:

#!/bin/perl

#$/=undef;

$conffile = "/myarea/myconfigfile";

#read in config variables
open ( CONFIGFILE, "< $conffile" ) || die "Can't open email template:
$conffile \n";

while ( <CONFIGFILE> ) {
chop;
if (/.*=.*/) {
($ConfigVar, $Congif_value) = split (/=/);
$ConfigValues{$ConfigVar}=$Congif_value;
}

}
close CONFIGFILE,


while ( ($k,$v) = each %ConfigValues ) {
print "$k => $v\n";
}


# It only prints out the first line of each joke. I played around
with $/ but not working.
# I understand that $/ is set to new line by default. So this explains
why I only get
# the first line of each joke.
# There must be a simple way to fix this? I appreciate any feedback.
Bob


while ( ($k,$v) = each %ConfigValues ) {
print "$k => $v\n";
}

Output:

SECONDJOKE => fdsf sdfdsf sdfdsfsdfdsfdsfdsfderew
rweeeeeeeeeeeeeeeeeeeeeeeer2 wr ewre
THIRDJOKE => notfunny
VAR1 => 15
VAR => 17
VAR2 => 16
FIRSTJOKE => this is my first joke. its isnt that funny but its still
pretty good bla

 

[Mirco Wahab]

I am trying to parse a text file that looks like this:
VAR1=15
VAR2=16
VAR=17

FIRSTJOKE=this is my first joke. its isnt that funny but its still
pretty good bla hbls dfd fsdf dsf sdfsd sd fdssd sdfdsf
dsfdsfsdsdfsfsd
....
....

# It only prints out the first line of each joke. I played around
with $/ but not working.
# I understand that $/ is set to new line by default. So this explains
why I only get
# the first line of each joke.
# There must be a simple way to fix this? I appreciate any feedback.

This is not the real reason. You cant read 'line by line' because
your data items have more than one line. *If* you put an *empty*
new line behind each item, like

VAR1=15

VAR2=16

VAR=17

you could read the file in 'paragraph' mode like:

(pseudocode

$/ = ''; # set para mode
while( <CONFIGFILE> ) {
($ConfigVar, $Congif_value) = split /=/;
$ConfigValues{$ConfigVar} = $Congif_value;
}

and thats it. Otherwise (no empty line between items),
as your example suggests, you have to read in 'slurp'
mode (read the while file into one block), like:


...
my $fname= 'myconfigfile';

open( my $fh, '<', $fname ) or die "Can't read $!";
my $content;
{ local $/; $content = <$fh> }; # read in 'slurp' mode
close $fh;
...

I modified the style of the statements a blittle bit
to match a more comprehensive Perl style. The whole
file would now reside in $content, lets extract
it in one stroke (by regex).

...
my $reg = qr{
^(\w+)\s*=\s* # key follows ^
(.+ # complete text until $ (/m)
(?:\n [^=]* $)* # if available, take more lines w/out '='
) # (note the '$' anchor!)
}mx;
...

The regex has a second part in the second
(values) capture group: (?:\n [^=]* $)*
which reads entire lines if no '=' is
found in there.

Lets apply it:
...
my %Config = $content =~ /$reg/g;
...

You can print it as you used to do:

...
while ( my ($k,$v) = each %Config ) {
print "$k => $v\n\n";
}
...

Regards

M.

 

[anno4000]

Hi,

I am trying to parse a text file that looks like this:

__
VAR1=15
VAR2=16
VAR=17

FIRSTJOKE=this is my first joke. its isnt that funny but its still
pretty good bla hbls dfd fsdf dsf sdfsd sd fdssd sdfdsf
dsfdsfsdsdfsfsd

SECONDJOKE=fdsf sdfdsf sdfdsfsdfdsfdsfdsfderew
rweeeeeeeeeeeeeeeeeeeeeeeer2 wr ewre rewr er ewewrrewrew rewrewrew
ewrew ewr ewr er ewr ewr er ewr we rew ew
rewrewrewrewrewrererewrewrewrewrewrewrewrwerewrererewreerewrewrerererererer

THIRDJOKE=notfunny
---

My code is like this:

#!/bin/perl

No warnings? No strict?

#$/=undef;

$conffile = "/myarea/myconfigfile";

#read in config variables
open ( CONFIGFILE, "< $conffile" ) || die "Can't open email template:
$conffile \n";

while ( <CONFIGFILE> ) {
chop;
if (/.*=.*/) {
($ConfigVar, $Congif_value) = split (/=/);
$ConfigValues{$ConfigVar}=$Congif_value;
}

}
close CONFIGFILE,

^
The comma is an error, it should be a semicolon. This is not the code
you ran. Don't re-type code, copy and paste it.

while ( ($k,$v) = each %ConfigValues ) {
print "$k => $v\n";
}


# It only prints out the first line of each joke.

Sure. You are only printing lines with a "=" in them. What did you
expect.

I played around
with $/ but not working.

"Not working" is no useful error description. What exactly have
you tried?

Your file format is irregular. Some of the assignments are followed
by an empty line (the JOKEs), others aren't (the VARs). So paragraph
mode won't help you much.

# I understand that $/ is set to new line by default. So this explains
why I only get
# the first line of each joke.
# There must be a simple way to fix this? I appreciate any feedback.

The unsuitable format doesn't allow for a really simple way. You'll
have to scan the file, looking for the next line with a "=" in it while
collecting other lines to form the value of the previous assignment.
You can only store away one key/value pair when you have seen the next
one. The last assignment must be treated outside the loop.

my %ConfigValues;
my ( $ConfigVar, $ConfigValue);
while ( <DATA> ) {
chop;
if (/.*=.*/) {
$ConfigValues{$ConfigVar} = $ConfigValue if defined $ConfigVar;
($ConfigVar, $ConfigValue) = split (/=/);
} else {
$ConfigValue .= $_;
}

}
$ConfigValues{$ConfigVar} = $ConfigValue if defined $ConfigVar;
close CONFIGFILE,

while ( my ($k,$v) = each %ConfigValues ) {
print "$k => $v\n";
}

__END__
VAR1=15
VAR2=16
VAR=17

FIRSTJOKE=this is my first joke. its isnt that funny but its still
pretty good bla hbls dfd fsdf dsf sdfsd sd fdssd sdfdsf
dsfdsfsdsdfsfsd

SECONDJOKE=fdsf sdfdsf sdfdsfsdfdsfdsfdsfderew
rweeeeeeeeeeeeeeeeeeeeeeeer2 wr ewre rewr er ewewrrewrew rewrewrew
ewrew ewr ewr er ewr ewr er ewr we rew ew
rewrewrewrewrewrererewrewrewrewrewrewrewrwerewrererewreerewrewrerererererer

THIRDJOKE=notfunny

Anno

 

[anno4000]

Hi,

I am trying to parse a text file that looks like this:

__
VAR1=15
VAR2=16
VAR=17

FIRSTJOKE=this is my first joke. its isnt that funny but its still
pretty good bla hbls dfd fsdf dsf sdfsd sd fdssd sdfdsf
dsfdsfsdsdfsfsd

SECONDJOKE=fdsf sdfdsf sdfdsfsdfdsfdsfdsfderew
rweeeeeeeeeeeeeeeeeeeeeeeer2 wr ewre rewr er ewewrrewrew rewrewrew
ewrew ewr ewr er ewr ewr er ewr we rew ew
rewrewrewrewrewrererewrewrewrewrewrewrewrwerewrererewreerewrewrerererererer

THIRDJOKE=notfunny
---

My code is like this:

#!/bin/perl

No warnings? No strict?

#$/=undef;

$conffile = "/myarea/myconfigfile";

#read in config variables
open ( CONFIGFILE, "< $conffile" ) || die "Can't open email template:
$conffile \n";

while ( <CONFIGFILE> ) {
chop;
if (/.*=.*/) {
($ConfigVar, $Congif_value) = split (/=/);
$ConfigValues{$ConfigVar}=$Congif_value;
}

}
close CONFIGFILE,

^
The comma is an error, it should be a semicolon. This is not the code
you ran. Don't re-type code, copy and paste it.

while ( ($k,$v) = each %ConfigValues ) {
print "$k => $v\n";
}


# It only prints out the first line of each joke.

Sure. You are only printing lines with a "=" in them. What did you
expect?

I played around
with $/ but not working.

"Not working" is no useful error description. What exactly have
you tried?

Your file format is irregular. Some of the assignments are followed
by an empty line (the JOKEs), others aren't (the VARs). So paragraph
mode won't help you much.

# I understand that $/ is set to new line by default. So this explains
why I only get
# the first line of each joke.
# There must be a simple way to fix this? I appreciate any feedback.

The unsuitable format doesn't allow for a really simple way. You'll
have to scan the file, looking for the next line with a "=" in it while
collecting other lines to form the value of the previous assignment.
You can only store away one key/value pair when you have seen the next
one. The last assignment must be treated outside the loop.

my %ConfigValues;
my ( $ConfigVar, $ConfigValue);
while ( <DATA> ) {
chop;
if (/.*=.*/) {
$ConfigValues{$ConfigVar} = $ConfigValue if defined $ConfigVar;
($ConfigVar, $ConfigValue) = split (/=/);
} else {
$ConfigValue .= $_;
}

}
$ConfigValues{$ConfigVar} = $ConfigValue if defined $ConfigVar;
close CONFIGFILE,

while ( my ($k,$v) = each %ConfigValues ) {
print "$k => $v\n";
}

__END__
VAR1=15
VAR2=16
VAR=17

FIRSTJOKE=this is my first joke. its isnt that funny but its still
pretty good bla hbls dfd fsdf dsf sdfsd sd fdssd sdfdsf
dsfdsfsdsdfsfsd

SECONDJOKE=fdsf sdfdsf sdfdsfsdfdsfdsfdsfderew
rweeeeeeeeeeeeeeeeeeeeeeeer2 wr ewre rewr er ewewrrewrew rewrewrew
ewrew ewr ewr er ewr ewr er ewr we rew ew
rewrewrewrewrewrererewrewrewrewrewrewrewrwerewrererewreerewrewrerererererer

THIRDJOKE=notfunny

Anno

 

[Michele Dondi]

# It only prints out the first line of each joke. I played around
with $/ but not working.
# I understand that $/ is set to new line by default. So this explains
why I only get
# the first line of each joke.
# There must be a simple way to fix this? I appreciate any feedback.
Bob

If you know $/, that is you've read

perldoc perlvar

then you should know that what you'ere after is ''.


Michele

 

[new2perl]

# It only prints out the first line of each joke. I played around
with $/ but not working.
# I understand that $/ is set to new line by default. So this explains
why I only get
# the first line of each joke.
# There must be a simple way to fix this? I appreciate any feedback.
Bob

If you know $/, that is you've read

perldoc perlvar

then you should know that what you'ere after is ''.

Michele
--
{$_=pack'B8'x25,unpack'A8'x32,$a^=sub{pop^pop}->(map substr
(($a||=join'',map--$|x$_,(unpack'w',unpack'u','G^<R<Y]*YB='
.'KYU;*EVH[.FHF2W+#"\Z*5TI/ER<Z`S(G.DZZ9OX0Z')=~/./g)x2,$_,
256),7,249);s/[^\w,]/ /g;$ \=/^J/?$/:"\r";print,redo}#JAPH,

Thanks everyone. I got it working. Thanks for your great advice.
Bob