use of useless void in context

Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: use of useless void in context by moritz (Cardinal) on Feb 07, 2009 at 17:35 UTC
I think that $1 is not set outside of the s/// marks That's correct. `$content =~ s/(\$[a-zA-Z0-9\{\'\}_]+)/ if $1(eval($1)) /ge;` That won't work. At the very least it's spelled `if (...) { ... }` (but I guess it won't work when an expression is expected). That said, why do you want that `if` at all? evaling an empty string isn't an error, so I don't see why you want to avoid it.	[reply] [d/l] [select]
Re: use of useless void in context by JavaFan (Canon) on Feb 07, 2009 at 17:36 UTC
`$content =~ s/(\$[a-zA-Z0-9\{\'\}_]+)/ eval($1) /ge if $1;` The '$1' after the if would refer to the $1 set by the last successful match. Also, if you want to eval the content of $1, no need to do both `eval($1)` and /e. Just /e should do. `$content =~ s/(\$[a-zA-Z0-9\{\'\}_]+)/ if $1(eval($1)) /ge;` That doesn't make any sense. Since you don't tell us what you try to do, I won't bother offering suggestions.	[reply] [d/l] [select]
Re^2: use of useless void in context by romandas (Pilgrim) on Feb 09, 2009 at 10:58 UTC
I agree. From what I can tell, the OP could use a backreference instead ( \1, \2, etc ), but I can't quite puzzle out what the OP is trying to accomplish.	[reply]
Re^3: use of useless void in context by JavaFan (Canon) on Feb 09, 2009 at 11:14 UTC
Considering the OP is using all his backreferences outside of the pattern itself, I don't see where the OP should use a \1 instead of a $1.	[reply]
Re: use of useless void in context by Anonymous Monk on Feb 07, 2009 at 17:49 UTC
I created a table in our mysql database for page variables, so the pages can load dynamically. I then call certain variables in text that is in a database record. Since I cannot use variables directly in mysql, I put them as merge fields, such as: ##variable## or something. There are several different types of variables, which were created over time, so I have accounted for each of them. Here is the subroutine I wrote to put these into a hash for storage while the system is running. We then later replace each of those merge fields: sub Get_Page_Vars { my ($temp_vars,$type) = @_; my $sth = $dbh->prepare (qq{ SELECT `name`,`value`,`add_name` FROM + `page_vars` WHERE `type` = ? OR `type2` = ? ORDER BY add_name,id,d}) +; $sth->execute($type,$type); while(my ($db_name,$content,$_add_name) = $sth->fetchrow_array()) +{ if($content =~ /(\$[a-zA-Z0-9\{\'\}_]+)/) { $content =~ s/(\$[a-zA-Z0-9\{\'\}_]+)/eval($temp_vars{$1}) +/ge; $content =~ s/(\$[a-zA-Z0-9\{\'\}_]+)/$temp_vars{eval($1)} +/ge; $content =~ s/(\$[a-zA-Z0-9\{\'\}_]+)/eval($vars{$1})/ge; $content =~ s/(\$[a-zA-Z0-9\{\'\}_]+)/$vars{eval($1)}/ge; $content =~ s/(\$[a-zA-Z0-9\{\'\}_]+)/eval($1)/ge; } if($content =~ /\<templ_var ([a-zA-Z0-9\{\'\}_]+)\>/) { $content =~ s/\<templ_var ([a-zA-Z0-9\{\'\}_]+)\>/eval($va +rs{$1})/ge; $content =~ s/\<templ_var ([a-zA-Z0-9\{\'\}_]+)\>/$vars{ev +al($1)}/ge; $content =~ s/\<templ_var ([a-zA-Z0-9\{\'\}_]+)\>/eval($te +mp_vars{$1})/ge; $content =~ s/\<templ_var ([a-zA-Z0-9\{\'\}_]+)\>/$temp_va +rs{eval($1)}/ge; $content =~ s/\<templ_var ([a-zA-Z0-9\{\'\}_]+)\>/eval($1) +/ge; } if($content =~ /{{([a-zA-Z0-9\{\'\}_]+)}}/) { $content =~ s/{{([a-zA-Z0-9\{\'\}_]+)}}/eval($vars{$1})/ge +; $content =~ s/{{([a-zA-Z0-9\{\'\}_]+)}}/$vars{eval($1)}/ge +; } $content =~ s\|<br>\|<br />\|g; if ($content && ($content =~ /<code>/i && $content =~ /<\/code +>/i)) { while($content =~ /<code>/i && $content =~ /<\/code>/i) { my ($code1,$code2,$_do_code); ($content,$code1) = split /<code>/, $content, 2; ($code2,$ocontent) = split /<\/code>/, $code1, 2; $_do_code = eval($code2); $content = $content . $_do_code . $ocontent; } } if ($content && ($content =~ /<syscode>/i && $content =~ /<\/s +yscode>/i)) { while($content =~ /<syscode>/i && $content =~ /<\/syscode> +/i) { my ($syscode1,$syscode2); ($content,$syscode1) = split /<syscode>/i, $content, 2 +; ($syscode2,$ocontent) = split /<\/syscode>/i, $syscode +1, 2; eval{$syscode2}; $content = $content . $ocontent; } } if ($_add_name) { $content = qq~<!--Start $db_name-->$content<!--End $db_nam +e-->~; } $temp_vars{$db_name} = $content; } $sth->finish(); return(%temp_vars); } [download] I had to go add a bunch of stuff because there were soooo many useless void errors. I know there are modules already that will do this more effieiently, however, I wrote this years ago, have not had time to upgrade the system to a better one. that is on my to-do list for the future though. I certainly would love any feedback, negative or positive. thank you.	[reply] [d/l]
Re^2: use of useless void in context by toolic (Bishop) on Feb 07, 2009 at 18:33 UTC
`\w` is a lot less typing than `a-zA-Z0-9_` You can cut down on back-whackin' by choosing a different delimiter: `split m{</syscode>}i, $syscode, 2;` Update: johngg++. Added m. I often botch untested code.	[reply] [d/l] [select]
Re^3: use of useless void in context by johngg (Canon) on Feb 07, 2009 at 19:03 UTC
`split {</syscode>}i, $syscode, 2;` According to the documentation, "If "/" is the delimiter then the initial m is optional" so once you choose a different delimiter I think you need the m. Therefore the code should be `split m{</syscode>}i, $syscode, 2;` [download] I hope this is of interest. Cheers, JohnGG	[reply] [d/l] [select]
Re^4: use of useless void in context by jwkrahn (Monsignor) on Feb 07, 2009 at 20:30 UTC
Re^2: use of useless void in context by moritz (Cardinal) on Feb 07, 2009 at 17:50 UTC
I think you would be better off with a template system like HTML::Template, instead of rolling a crippled one on your own.	[reply]
Re^2: use of useless void in context by missingthepoint (Friar) on Feb 08, 2009 at 06:01 UTC
I certainly would love any feedback, negative or positive. That sub's not very readable. Perhaps: sub Get_Page_Vars { my ($temp_vars,$type) = @_; my $sth = $dbh->prepare (qq{ SELECT `name`,`value`,`add_name` FROM + `page_vars` WHERE `type` = ? OR `type2` = ? ORDER BY add_name,id,d}) +; $sth->execute($type, $type); while( my ($db_name,$content,$_add_name) = $sth->fetchrow_array()) + { { local $_ = $content; my $ident_key = qr/[\w{}']+/; my $phash_pat = qr/[$]$ident_key/; if (/$phash_pat/) { s/($phash_pat)/ eval($temp_vars{$1}) /ge; s/($phash_pat)/ $temp_vars{eval($1)} /ge; s/($phash_pat)/ eval($vars{$1}) /ge; s/($phash_pat)/ $vars{eval($1)} /ge; s/($phash_pat)/ eval($1) /ge; } if (/\<templ_var ($ident_key)\>/) { s/\<templ_var ($ident_key)\>/ eval($vars{$1}) /g +e; s/\<templ_var ($ident_key)\>/ $vars{eval($1)} /g +e; s/\<templ_var ($ident_key)\>/ eval($temp_vars{$1}) /g +e; s/\<templ_var ($ident_key)\>/ $temp_vars{eval($1)} /g +e; s/\<templ_var ($ident_key)\>/ eval($1) /g +e; } if (/{{$ident_key}}/) { s/{{($ident_key)}}/ eval($vars{$1}) /ge; s/{{($ident_key)}}/ $vars{eval($1)} /ge; } s\|<br>\|<br />\|gi; $content = $_; } if ($content && ($content =~ /<code>/i && $content =~ /<\/code +>/i)) { while($content =~ /<code>/i && $content =~ /<\/code>/i) { my ($code1,$code2,$_do_code); ($content,$code1) = split /<code>/, $content, 2; ($code2,$ocontent) = split /<\/code>/, $code1, 2; $_do_code = eval($code2); $content = $content . $_do_code . $ocontent; } } if ($content && ($content =~ /<syscode>/i && $content =~ /<\/s +yscode>/i)) { while($content =~ /<syscode>/i && $content =~ /<\/syscode> +/i) { my ($syscode1,$syscode2); ($content,$syscode1) = split /<syscode>/i, $content, 2 +; ($syscode2,$ocontent) = split /<\/syscode>/i, $syscode +1, 2; eval{$syscode2}; $content = $content . $ocontent; } } if ($_add_name) { $content = qq~<!--Start $db_name-->$content<!--End $db_nam +e-->~; } $temp_vars{$db_name} = $content; } $sth->finish(); return(%temp_vars); } [download] Less visual noise. Life is denied by lack of attention, whether it be to cleaning windows or trying to write a masterpiece... -- Nadia Boulanger	[reply] [d/l]


We don't bite newbies here... much
	PerlMonks