Hi,
I've a script below which reads the contents of a text file which contains the output of the `ls -lR` in UNIX. This script reads the list of html files contained in the text file, and for every file it checks to see it contains some specified HTML tags or not. If it does not, this script then inserts the specified tag (hardcoded in this script) into the respective HTML file.
## test3b.pl
#!/usr/bin/perl/perl -w
$infile = shift(@ARGV);
open(INFILE, "$infile"); ## reads input file from command line - results of 'ls -lR'
while(<INFILE>){
next if(/totald+/); ## ignore total in the output of ls -lR
next if(/$/);
if(/:$/){
chomp;
s/://;
$path = $_; ## get path name of the file in ls -lR
}
else{
@array = split(/s+/, $_);
$fileName = $array[8];
$timeVal = $array[7];
open(HTMLFILE, "$fileName");
@htmlContents = <HTMLFILE>;
for($i; $i<@htmlContents; $i++){
if($htmlContents[$i] !~ /<head>/){
open(NOHEAD, ">NoHead.txt");
print NOHEAD $path/$fileName; //compile a list of html files which do not have <head>
close(NOHEAD);
}
if($htmlContents[$i] !~ /^</script>history.forward()b/){
if($htmlContents[$i] =~ /<head>/){
s/</head>/</script>history.forward()n</head>/;
open(REPLACEMENT, ">$path/$fileName");
print REPLACEMENT "$htmlContents[$i]"; ## rewrite substituted value back to original file.
close(REPLACEMENT);
}
}
}
close(HTMLFILE);
}
}
close(INFILE);
However, I obtained the error below("Useless use of a variable in void context at ./test3b.pl line 37") below, in which I'm do not how I could resolve it:
/opt/IBM/WebSphere/admin >>./test3b.pl /tmp/lender.txt
Useless use of a variable in void context at ./test3b.pl line 37.
Name "main::timeVal" used only once: possible typo at ./test3b.pl line 19.
Could anyone kindly help me out with this error?
Any suggestions or critique to this script is indeed welcomed.
Thanks
---------------------------------
Celebrate Yahoo!'s 10th Birthday!
Yahoo! Netrospective: 100 Moments of the Web
In article <[Email Removed]>,
[Email Removed] (Danny Fang) writes:
[Email Removed] (Danny Fang) writes:
| QUOTE |
| --0-972884666=:30240 Content-Type: text/plain; charset=us-ascii Hi, I've a script below which reads the contents of a text file which contains the output of the `ls -lR` in UNIX. This script reads the list of html files contained in the text file, and for every file it checks to see it contains some specified HTML tags or not. If it does not, this script then inserts the specified tag (hardcoded in this script) into the respective HTML file. ## test3b.pl #!/usr/bin/perl/perl -w $infile = shift(@ARGV); open(INFILE, "$infile"); ## reads input file from command line - results of 'ls -lR' while(<INFILE>){ next if(/totald+/); ## ignore total in the output of ls -lR next if(/$/); if(/:$/){ chomp; s/://; $path = $_; ## get path name of the file in ls -lR } else{ @array = split(/s+/, $_); $fileName = $array[8]; $timeVal = $array[7]; open(HTMLFILE, "$fileName"); @htmlContents = <HTMLFILE>; for($i; $i<@htmlContents; $i++){ if($htmlContents[$i] !~ /<head>/){ open(NOHEAD, ">NoHead.txt"); print NOHEAD $path/$fileName; //compile a list of html files which do not have <head close(NOHEAD); } if($htmlContents[$i] !~ /^</script>history.forward()b/){ if($htmlContents[$i] =~ /<head>/){ s/</head>/</script>history.forward()n</head>/; open(REPLACEMENT, ">$path/$fileName"); print REPLACEMENT "$htmlContents[$i]"; ## rewrite substituted value back to original file. close(REPLACEMENT); } } } close(HTMLFILE); } } close(INFILE); However, I obtained the error below("Useless use of a variable in void context at ./test3b.pl line 37") below, in which I'm do not how I could resolve it: /opt/IBM/WebSphere/admin >>./test3b.pl /tmp/lender.txt Useless use of a variable in void context at ./test3b.pl line 37. Name "main::timeVal" used only once: possible typo at ./test3b.pl line 19. Could anyone kindly help me out with this error? |
Did you really post the right code? Because that code has
basic syntax errors on line 21 where the programmer has
thought that // introduces a Perl comment, and also left
out quote marks, thereby doing a division by a string.
It should have (probably) generated the error
Bareword found where operator expected
The code does not use strict, which makes me disinclined
to read much more; if someone doesn't ask Perl to help
them get their program correct then they should expect
obscure problems.
I have no idea what line 6 was supposed to be, but it's
wrong.
The open statement is not checked for success and contains
a useless stringification.
The first clause of the for statement looks to me like
the source of the error you cite (it's a warning). The
programmer either didn't understand what a for statement
was or was very careless. Either way, there are so
many serious errors here that there are sufficient
grounds for rejecting the integrity of this program
and it should either be rewritten from requirements or
every line inspected with a magnifying glass.
--
Peter Scott
http://www.perlmedic.com/
http://www.perldebugged.com/
Danny Fang wrote:
| QUOTE |
| Hi, |
Hello,
| QUOTE |
| I've a script below which reads the contents of a text file which contains the output of the `ls -lR` in UNIX. |
Why?? Why not just use File::Find?
| QUOTE |
| This script reads the list of html files contained in the text file, and for every file it checks to see it contains some specified HTML tags or not. If it does not, this script then inserts the specified tag (hardcoded in this script) into the respective HTML file. ## test3b.pl #!/usr/bin/perl/perl -w $infile = shift(@ARGV); open(INFILE, "$infile"); ## reads input file from command line - results of 'ls -lR' |
You should *always* verify that the file opened correctly.
| QUOTE |
| while(<INFILE>){ next if(/totald+/); ## ignore total in the output of ls -lR |
That won't match because there is a space between 'total' and d+.
| QUOTE |
| next if(/$/); |
You are saying skip this line if matches end-of-line but *every* line matches.
| QUOTE |
| if(/:$/){ chomp; s/://; |
Your match isn't anchored so if the dir name contains a colon it will be
removed: './some:dir:' will become './somedir:'.
| QUOTE |
| $path = $_; ## get path name of the file in ls -lR } else{ @array = split(/s+/, $_); $fileName = $array[8]; |
That will not work if the file name contains any whitespace characters.
| QUOTE |
| $timeVal = $array[7]; |
You are not using this variable anywhere else which is why you get the warning
'Name "main::timeVal" used only once'. Also, $array[7] could contain the year
instead of the time depending on how old the file is.
| QUOTE |
| open(HTMLFILE, "$fileName"); |
You should *always* verify that the file opened correctly.
| QUOTE |
| @htmlContents = <HTMLFILE>; for($i; $i<@htmlContents; $i++){ if($htmlContents[$i] !~ /<head>/){ |
You are testing *every* line in the file for the '<head>' tag. This will only
work if every line does contain the tag. It also won't work if the file
contains '<HEAD>' instead of '<head>'.
| QUOTE |
| open(NOHEAD, ">NoHead.txt"); |
You are overwriting the file each time you open it so in the end it will only
contain one line. Also, you should *always* verify that the file opened
correctly.
| QUOTE |
| print NOHEAD $path/$fileName; //compile a list of html files which do not have <head |
You are dividing the numeric value of $fileName by the numeric value of $path
and then printing that value to the file? Comments in Perl begin with # not //.
| QUOTE |
| close(NOHEAD); } if($htmlContents[$i] !~ /^</script>history.forward()b/){ if($htmlContents[$i] =~ /<head>/){ s/</head>/</script>history.forward()n</head>/; open(REPLACEMENT, ">$path/$fileName"); |
Again, you are overwriting the file each time you open it so in the end it
will only contain one line and you should *always* verify that the file opened
correctly.
| QUOTE |
| print REPLACEMENT "$htmlContents[$i]"; ## rewrite substituted value back to original file. close(REPLACEMENT); } } } close(HTMLFILE); } } close(INFILE); However, I obtained the error below("Useless use of a variable in void context at ./test3b.pl line 37") below, in which I'm do not how I could resolve it: |
Which line is number 37?
John
--
use Perl;
program
fulfillment
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.