From: b <rowerynaksiezycu@gmail.com>
Date: Sat, 30 Sep 2023 20:14:15 +0000 (+0000)
Subject: improve regexp handling in some functions
X-Git-Tag: v1.0.22
X-Git-Url: http://bicyclesonthemoon.info/git-projects/?a=commitdiff_plain;h=c88ca49b964a50d81875ab64ccaf2f25c64b89f3;p=botm%2Fcommon-perl

improve regexp handling in some functions
---

diff --git a/botm_common.pm b/botm_common.pm
index 82c62a6..6d76275 100644
--- a/botm_common.pm
+++ b/botm_common.pm
@@ -25,7 +25,7 @@ use Encode ('encode', 'decode');
 
 use Exporter;
 
-our $VERSION     = '1.0.21';
+our $VERSION     = '1.0.22';
 our @ISA         = qw(Exporter);
 our @EXPORT      = ();
 our @EXPORT_OK   = (
@@ -113,6 +113,10 @@ sub make_temp_path {
 	return join_path('/', $dir, $filename);
 }
 
+###################################
+##  ENCODING + SYSTEM FUNCTIONS  ##
+###################################
+
 sub system_encoded {
 	(my $cmd, my @arg) = @_;
 	my @newarg;
@@ -235,8 +239,8 @@ sub read_data_file {
 			next;
 		}
 		
-		$line =~ s/[\n]$//g;
-		$line =~ s/[\r]$//g;
+		$line =~ s/\n$//gs;
+		$line =~ s/\r$//gs;
 		
 		# Empty line - end of header.
 		if ($line eq ''){
@@ -244,15 +248,15 @@ sub read_data_file {
 		}
 		# Line starts with whitespace. It's a continuation of the previous line.
 		# Concatenate the field value, separated by newline.
-		elsif($line =~ /^[ \t](.*)$/){
+		elsif($line =~ /^[ \t]/){
 			if($lastname ne '') {
-				$data{$lastname} .= "\n".$1;
+				$data{$lastname} .= "\n".$';
 			}
 		}
 		# Line starts with a name followed by colon/equal sign. Save the value
-		elsif ($line =~ /^([ -9;-<>-~]+)((:[ \t])|=)(.*)$/) {
+		elsif ($line =~ /^([ -9;-<>-~]+)((:[ \t])|=)/s) {
 			$name = lc($1);
-			$value = $4;
+			$value = $';
 			
 			$data{$name} = $value;
 			
@@ -317,8 +321,8 @@ sub write_data_file {
 			}
 			my $value = $data->{$ind};
 			# convert newlines - add spaces at continuation line
-			$value =~ s/\r//g;
-			$value =~ s/\n/\n /g;
+			$value =~ s/(\r)?\n/\n /g;
+			$value =~ s/\r/\n /g;
 			print $fh "$name: $value\n";
 		}
 	}
@@ -432,8 +436,8 @@ sub read_header_file {
 	my $lastname='';
 	
 	while (defined(my $line = <$fh>)) {
-		$line =~ s/[\n]$//g;
-		$line =~ s/[\r]$//g;
+		$line =~ s/\n$//gs;
+		$line =~ s/\r$//gs;
 		
 		if ($status_line) {
 			$line =~ /^([^ ]+) +([^ ]+)( +([^ ].*))?$/;
@@ -656,9 +660,8 @@ sub split_url {
 	else {
 		$data{'authority'} = $url;
 		if ($data{'authority'} =~ m/[\/\?#]/g) {
-			$ind = pos($data{'authority'})-1;
-			$url = substr($data{'authority'}, $ind); 
-			$data{'authority'} = substr($data{'authority'}, 0, $ind);
+			$data{'authority'} = $`;
+			$url = $&.$';
 		}
 		else {
 			$url = '';
@@ -677,9 +680,8 @@ sub split_url {
 	}
 	$data{'host'} =~ m/\[[^\]]*\]/g;
 	if ($data{'host'} =~ m/:/g) {
-		$ind = pos($data{'host'})-1;
-		$data{'port'} = substr($data{'host'}, $ind+1);
-		$data{'host'} = substr($data{'host'}, 0, $ind);
+		$data{'host'} = $`;
+		$data{'port'} = $';
 	}
 	else {
 		$data{'port'} = '';
@@ -689,9 +691,8 @@ sub split_url {
 	if (($url =~ /^\//) or $relative) {
 		$data{'path'} = $url;
 		if ($data{'path'} =~ m/[\?#]/g) {
-			$ind = pos($data{'path'})-1;
-			$url = substr($data{'path'}, $ind);
-			$data{'path'} = substr($data{'path'}, 0, $ind);
+			$data{'path'} = $`;
+			$url = $&.$';
 		}
 		else {
 			$url = '';
@@ -703,7 +704,7 @@ sub split_url {
 	
 	# query
 	if ($url =~ /^\?/) {
-		$data{'query'} = substr($url, 1);
+		$data{'query'} = $';
 		$ind = index($data{'query'}, '#');
 		if ($ind >= 0) {
 			$url = substr($data{'query'}, $ind);
@@ -719,7 +720,7 @@ sub split_url {
 	
 	# fragment
 	if ($url =~ /^#/) {
-		$data{'fragment'} = substr($url, 1);
+		$data{'fragment'} = $';
 	}
 	else {
 		$data{'fragment'} = '';
@@ -3304,17 +3305,17 @@ sub html_entity_decode {
 			$d .= $1;
 			$t = $2;
 		}
-		elsif ($t =~ /^(\&#?[A-Za-z0-9]+;)(.*)$/s) { # correct encoded character
-			$d .= html_entity_decode_1en($1);
-			$t = $2;
+		elsif ($t =~ /^\&#?[A-Za-z0-9]+;/s) { # correct encoded character
+			$d .= html_entity_decode_1en($&);
+			$t = $';
 		}
-		elsif ($t =~ /^(\&[A-Za-z0-9]+)(.*)$/s) { # encoded character without ";"
-			$d .= html_entity_decode_1en($1);
-			$t = $2;
+		elsif ($t =~ /^\&[A-Za-z0-9]+/s) { # encoded character without ";"
+			$d .= html_entity_decode_1en($&);
+			$t = $';
 		}
-		elsif ($t =~ /^(\&)(.*)$/s) { # invalid "&"
-			$d .= $1;
-			$t = $2;
+		elsif ($t =~ /^\&/s) { # invalid "&"
+			$d .= $&;
+			$t = $';
 		}
 		else {  # nothing left to decode
 			$d .= $t;
@@ -3336,7 +3337,8 @@ sub html_entity_decode_1en {
 	
 	if ($en !~ /;$/) { # name without ";"
 		my $n = substr($en, 1);
-		# we HAVE TO iterate :/
+		# we HAVE TO iterate as we don't know where name terminates :/
+		# why they thought this is a good idea ?
 		foreach my $name (keys %{+HTML_ENTITY_CODE_INF}) {
 			if (index($n, $name) == 0) { # name (beginning of entire string) is valid:
 				# decode