And I think it's annoying that this section of code is annoying:
if (ValidatorTrim(value).length == 0)
return true;
Meaning that you need to put a required field validator as well as a
regex validator, when the regex validator could easily do both.
-----Original Message-----
From: Michael Weinhardt [mailto:[EMAIL PROTECTED]]
Sent: Thursday, 16 May 2002 11:46 a.m.
To: [EMAIL PROTECTED]
Subject: Re: [DOTNET] Winform Regex Validator and Multiline TextBox
comments inline
On Wed, 15 May 2002 19:05:21 -0400, Marsh, Drew <[EMAIL PROTECTED]>
wrote:
>I said:
>
>> <codeSnippet language="ECMAScript">
>> function RegularExpressionValidatorEvaluateIsValid(val) {
>> var value = ValidatorGetValue(val.controltovalidate);
>> if (ValidatorTrim(value).length == 0)
>> return true;
>> var rx = new RegExp(val.validationexpression);
>> var matches = rx.exec(value);
>> return (matches != null && value == matches[0]);
>> }
>> </codeSnippet>
>>
>> Your problem is that you need to set *an option*.
>
>Actually, I take that back. You don't need to set an option for the
>test to succeed... thought you might have needed global or multiline,
>but you
don't.
>I'm actually perplexed at this point because I can't figure out a piece
>of their code... specifically the value == matches[0] portion.
>
>Let's say you have a value of in the control you're validating
>"33\r\n33" and your regex is "33". Stepping through their code, the
>exec actually succeeds so matches != null. However, value == matches[0]
>only ever going
to
>be true if the whole string is a match! Ahhhhhhhhhaaaaaaaaaaa! So it
>looks like this code can't do a portioned match which is absolutely
>insane.
What you're saying makes sense. I originally thought that the regex val
control should match the whole text, but now I can't see why you
shouldn't have the option to match it all, or a portion of it? Either
approach is
valid: it's up to the developer to work out which technique is good
enough to make data is valid.
>Here's
>a quick ECMAScript windows script to prove the case:
>
><codeSnippet language="ECMAScript">
>var value = "anything 33 anything";
>
>var regex = new RegExp("33");
>var matches = regex.exec(value);
>
>WScript.Echo(matches[0]);
>WScript.Echo(value == matches[0]);
></codeSnippet>
>
>So basically what this means is you need to make your regex ".*33.*"
>and everything will be good to go! Why the hell they added the value ==
>matches[0] makes no sense to me as matches[0] != null would have
>covered
all
>the bases as far as I can see. Also, they could have just used
>RegExp::test instead of RegExp:: exec and avoided these problems
>altogether.
>
I tried the .*33.* in a multiline textarea, and it worked ok if "33" was
on the first row but, not if 33 was on the 2nd ie aa\r\n33.
I've also had trouble applying options. eg
/IN/i
I could not make it work in either a single or multiline scenario,
either lower or upper case. This is my intepretation of the help, so it
may be off.
Thanks for the other reply too, Drew!
>HTH,
>Drew
>.NET MVP
>
>You can read messages from the DOTNET archive, unsubscribe from DOTNET,
>or subscribe to other DevelopMentor lists at
>http://discuss.develop.com.
You can read messages from the DOTNET archive, unsubscribe from DOTNET,
or subscribe to other DevelopMentor lists at http://discuss.develop.com.
You can read messages from the DOTNET archive, unsubscribe from DOTNET, or
subscribe to other DevelopMentor lists at http://discuss.develop.com.
