This line needs to be followed with these two lines:#!/usr/bin/perl
Google says that "Ansvarig vardenhet" is Swedish for "Responsible care unit". If that is true, then this print statement (and its twin further down) is lying. They are not reading 'ansvarig_vardenhet_utf8.txt'; they are reading 'icd-10-codes.txt' and 'female1.txt'. If you just need reassurance about what point your program has reached, then use something like print "P1\n"; and print "P2\n";. Otherwise, include the actual file names. Do not lie; it will just confuse the next person who works on the program.print "reading 'ansvarig_vardenhet_utf8.txt'\n";
As written, if the ICD file is missing, then your program will seem to run correctly, but will silently fail to do any replacements. Silent failure is not acceptable. die() is used to halt a program on a critical error; it is often coupled with or so that the die happens when the first part of a statement fails. The $! variable contains the reason for the failure.open(WIN, "icd-10-codes.txt");
Since we added "use strict", we now have to declare all variables, such as $line. Also, $line needs to have its newline removed, otherwise the ICD description will have the newline embedded in it, which will mis-format your replacements. The chomp() function removes a trailing newline if one is present.while($line = <WIN>) {
Since each line has exactly two fields, you should name them. You can split() any extra fields into a generic array, to help find any future problems with the file formatting.@columns = split(/\t/, $line);
That line is clearly for debugging, but if you use it, the output will be indistinct from its neighbors. You need a terminating newline.#print "$columns[1]";
When you need to inspect more than a single simple variable, use Data::Dumper. In fact, use it to inspect even single simple variables, too.foreach my $value (@columns) { #print "$value\n" }
This would need "or die" if it was anything more complex than *reading* a file on a *local* filesystem. It is fine as-is.close(WIN);
Note that you have not preserved *any* data from the ICD file, so no ICD lookups can be done below this point! We will discuss how to fix this later, by copying the needed info to a hash or array.
(OK, before I added the my before @columns, you did technically preserve *some* ICD data, but since it was only the data from the last line in the file, it hardly counts.)
Fix as described above, including chomp().print "reading 'ansvarig_vardenhet_utf8.txt'\n"; open(IN, "female1.txt"); while($line = <IN>) {
No. This is nonsense. It would only make sense if you had a file with a single line, which contained all the keys and values separated by spaces. Instead of a hash, we need an array (declared, of course).%col = split(/ /, $line);
Use Data::Dumper.#print "$columns[1]\n";
"for" is easier to type than "foreach", and either of them work. We have already changed from hash to array.foreach my $val (%col) {
The == operator compares two vars *numerically*. The eq operator compares two vars as strings, so it is the one you want. In other words, "3" == "3.0", but "3" ne "3.0". Also, if you have just determined that $val is an exact match, then s/// is overkill; just replace $var by assigning to it.if($val == $columns[0]){ $val =~ s/$val/$columns[1]/g;
This looks confused. print IN $anything_at_all will write to the IN file. Are you trying to overwrite the data in 'female1.txt'? Luckily, Perl will not let you, because you opened the file as read-only. Your original description says "...should be replaced with ICD-code files corresponding...", but you are not outputting the entire line. Is the '>>' a debugging marker, or does it indicate that you are trying to append to a file? Going by your original description, I think you need:print IN ">>$val\n";
Again, fine when reading locally.close(IN);
You are a beginner in Perl. (See PERL as shibboleth and the Perl community.)I am a beginner in PERL.
The problem with your *approach* is just the failure to preserve the ICD data from the first half of the program so that the second half can use it. This can be done by declaring %lookup_icd above the first while() loop, and then within the while() loop, use:
$lookup_icd{$icd_code} = $icd_code_and_text;
With a few more additions, it would look like this:
#!/usr/bin/perl use strict; use warnings; use Data::Dumper; $Data::Dumper::Useqq = 1; my $icd_path = 'icd-10-codes.txt'; my $in_path = 'female1.txt'; my $out_path = 'female1_corrected.txt'; open my $icd_fh, '<', $icd_path or die "Cannot open '$icd_path': $!"; #print "Reading '$icd_path'\n"; my %lookup_icd; while ( my $line = <$icd_fh> ) { chomp $line; my ( $lookup_code, $icd_code_and_text ) = split /\t/, $line; # print Dumper $lookup_code, $icd_code_and_text; if ( exists $lookup_icd{$lookup_code} ) { warn "Replacing $lookup_code;" . " was '$lookup_icd{$lookup_code}'," . " now '$icd_code_and_text'\n"; } $lookup_icd{$lookup_code} = $icd_code_and_text; } close $icd_fh; #print Dumper \%lookup_icd; #print "Reading '$in_path'\n"; #print "Writing '$out_path'\n"; open my $in_fh, '<', $in_path or die "Cannot open '$in_path': $!"; open my $out_fh, '>', $out_path or die "Cannot open '$out_path': $!"; while ( my $line = <$in_fh> ) { chomp $line; my @cols = split / /, $line; for my $possible_icd (@cols) { my $replacement_icd = $lookup_icd{$possible_icd}; if ($replacement_icd) { $possible_icd = $replacement_icd; } } print {$out_fh} join( ' ', @cols ), "\n"; } close $in_fh; close $out_fh or warn "Cannot close '$out_path': $!\nSome data may not have been + written.";
I strongly recommend spending some time with either of the books I mentioned above. You have ventured a little past "baby Perl", and so now will be better served by tutorial than by blind exploration.
In reply to Re: comparing file contents
by Util
in thread comparing file contents
by kumaar1986
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |