Removing Special characters out of DCR

I need to remove some characters that get inserted into a DCR when users copy and paste from Word or other editors.

In one example I have seen the the characters â€“. I looked at the hex codes and they are E2 80 AND 93. I tried to use the line '$path =- s [\xE2]g;' but it does not replace the character.

Any ideas?

Thanks

Find more posts tagged with

Comments

ISCBorisB

g;' but it does not replace the character. Any ideas?Thanks
I need to remove some characters that get inserted into a DCR when users copy and paste from Word or other editors.

In one example I have seen the the characters â€“. I looked at the hex codes and they are E2 80 AND 93. I tried to use the line '$path =- s [\xE2]

Please tell us that your line of Perl code is a typo. Funny thing is, it's syntactically correct.
What it does however has notning to do to pattern search/replace.

$path =~ s/\xE2/*/g;

pkrug

Please tell us that your line of Perl code is a typo. Funny thing is, it's syntactically correct.
What it does however has notning to do to pattern search/replace.

$path =~ s/\xE2/*/g;

Typo? No I do not think so. I use the same type of line to replace linefeeds with
. The line I use is "$path =~ s [\n][
]g;" It works just fine.

I assumed that I could use the same type of line to replace the hex code that I saw but it is not working...

Hyperlink To Onther Report.rptdesign

Update In Database with Hyberlink.rptdesign

ISCBorisB

Typo? No I do not think so

Please reread your original post carefully. You will see that you have minus(-) instead of tilda(~).

Please note that although you can use almost anything as a pattern delimiter square brackets are bad choice IMO.
They denote regex class so you may not be able to use this syntax element if needed.

my $path = 'âx';
print "$path\n";
$path =~ s/\xE2/*/g;
print "$path\n";

...

C:\temp>perl foo.pl
Γx
*x
C:\temp>

pkrug

Please reread your original post carefully. You will see that you have minus(-) instead of tilda(~).

Please note that although you can use almost anything as a pattern delimiter square brackets are bad choice IMO.
They denote regex class so you may not be able to use this syntax element if needed.
my $path = 'âx';
print "$path\n";
$path =~ s/\xE2/*/g;
print "$path\n";

...

C:\temp>perl foo.pl
Γx
*x
C:\temp>

Thanks for your reply and your suggestion! I checked and I am using a tilde.

One big thing I did not mention I am using the perl code in a template.

I created a test perl file and ran the test and the special characters were removed. If I added it to my template they were not. I can remove other characters like linefeeds (in the template), substitute 'a' for 'b' etc. But I can not remove 'â' or other characters like it.

Here is a snippet of the perl code in the template.

sub NTfilter
{
my ($path) = @_;

$path =~ s [\x93][anything]g;
$path =~ s [\x80]g;

$path =~ s/\xE2/*/g;
$path =~ s/\x93/*/g;

return $path;
}

Thanks once again.

ISCBorisB

I created a test perl file and ran the test and the special characters were removed. If I added it to my template they were not.

Ah... You are not the first one who's getting burned by Perl 5.8 native UTF-8 support. Do the following once before translation:

$path = Encode::encode_utf8($path);
$path =~ s...

pkrug

Ah... You are not the first one who's getting burned by Perl 5.8 native UTF-8 support. Do the following once before translation:

$path = Encode::encode_utf8($path);
$path =~ s...

I tried that but it did not translate xe2 to something else.
Any more ideas? I really appreciate the help.

$path =~ Encode::encode_utf8($path);
$path =~ s [\n][
]g; # translate all '\n' into '
'
$path =~ s/\xe2/*/g;
$path =~ s [\x93][anything]g;

Adam Stoller

Keep in mind that PT processing of values does some encoding and that the values you see in the raw DCR may not be the values that are being processed by the perl code in your PT...

ISCBorisB

I tried that but it did not translate xe2 to something else.
Any more ideas? I really appreciate the help.

$path =~ Encode::encode_utf8($path);
$path =~ s [\n][
]g; # translate all '\n' into '
'
$path =~ s/\xe2/*/g;
$path =~ s [\x93][anything]g;

$path =~ Encode::encode_utf8($path); Is highlighted symbol a typo? It should not be there.
Also, do yourself a favor and use capital hexadeciaml digits in this notation: $path =~ s/\xE2/*/gi;
(That may be a bit paranoid though - it used to be an issue looong time ago. May be it's not anymore, I never bothered to check)

Do you know Perl? Please note that if the answer is 'No', it's not a problem. I'll just shape my posts differently

Adam Stoller

Sigh - the first substitution in this post had me really confused until I decided to respond to it - below is the original post with the code formatted so that it reads correctly in the forums (ISCBorisB already pointed out the error in the first line):

I tried that but it did not translate xe2 to something else.
Any more ideas? I really appreciate the help.

$path =~ Encode::encode_utf8($path);
$path =~ s [\n][<br />]g; # translate all '\n' into '<br />'
$path =~ s/\xe2/*/g;
$path =~ s [\x93][anything]g;

Estimate.xml

watermark.rptdesign

draft_portrait.gif