phplib-users — A list for users of phplib to talk / discuss phplib

R: [Phplib-users] RE: javascript links (was: Auth working with a twist)

[Gaetano G. <giu...@se...>]

IMHO you might supply different (or the same) addresses to the href and =
onclick tags, in order to satisfy browsers both with and without =
javascript enabled...=20
Or am I completely wrong in doing this?

Gaetano

>=20
> I had not thought of that.  Do you know if that is=20
> "officially" supported as
> part of the HTML spec?  Time to do some homework...

> >
> >
> > IMHO, the correct way to do JavaScript links is
> >
> > 	<a href=3D"javascript:myFunction();">Blah</a>
> >
> > rather than using onClick. Creating a link to an undefined=20
> anchor is a
> > kludge. I can't say that this will fix your problem, but=20
> the extra karma
> > from cleaner code can't hurt.
> >

FW: [Phplib-users] OOHForms and RegEx

[Andres B. Z. <aba...@ei...>]

Thanks Andrew... You are right. It works now... But I still have a couple of
questions.

why do the original  regexs ( example: "^[a-Z]*$" ) work as I expected in
the intranet server but fail in the internet host? it seems like regex has a
different behavior in the intranet server setup and in the internet host
setup...
is it any special recommendation or paper on regex use in OOHforms?

Thanks again
Andres

-----Original Message-----
From: Andrew Hull [mailto:ah...@ni...]
Sent: Viernes, 11 de Octubre de 2002 10:45 a.m.
To: Andres Barreto Zicare
Subject: Re: [Phplib-users] OOHForms and RegEx


Andres Barreto Zicare wrote:

- snip -

>But if I change
>	"valid_regex"=>"^[a-z]*$"
>for
>	"valid_regex"=>"^[a-Z]*$"
>
>
this _is_ invalid since lowercase comes after uppercase. in any event, i
think u mean [a-zA-Z]

- snip -

a.g.h

[Phplib-users] RE: javascript links (was: Auth working with a twist)

[Rob H. <rob...@ws...>]

I had not thought of that.  Do you know if that is "officially" supported as
part of the HTML spec?  Time to do some homework...

Rob Hutton
Web Safe
www.wsafe.com

> -----Original Message-----
> From: Layne Weathers [mailto:la...@if...]
> Sent: Monday, October 14, 2002 9:30 AM
> To: 'phplib-users'
> Cc: rob...@ws...
> Subject: javascript links (was: Auth working with a twist)
>
>
> > <a href="#" onClick="document.all.innerFrameName.src =
> > 'pagename.php?again=yes'; return true">
> >
> > now it does not do the immediate refresh.  I would expect this on one
> > browser, but I was seeing it cross platform, cross browser.
> >
> > Now, the problem is if I change it to a # and click on it a
> > second time, it
> > does not do anything, but if I leave off the #, then it
> > doesn't work at all
> > on Mozilla or Netscape.  AHHH!
>
>
> IMHO, the correct way to do JavaScript links is
>
> 	<a href="javascript:myFunction();">Blah</a>
>
> rather than using onClick. Creating a link to an undefined anchor is a
> kludge. I can't say that this will fix your problem, but the extra karma
> from cleaner code can't hurt.
>
>
> Layne Weathers
> Ifworld Inc.
>
>
>

RE: [Phplib-users] Next auth issue, php variable scope issue

[Rob H. <rob...@ws...>]

I will work from here and see if I can figure something out.

Rob Hutton
Web Safe
www.wsafe.com 

> -----Original Message-----
> From: php...@li...
> [mailto:php...@li...]On Behalf Of Giancarlo
> Sent: Monday, October 14, 2002 4:34 AM
> To: php...@li...
> Subject: Re: [Phplib-users] Next auth issue, php variable scope issue
> 
> 
> Il 19:48, domenica 13 ottobre 2002, Rob Hutton ha scritto:
> > This does not seem to be an auth issue per say, but maybe Gian 
> or someone
> > can help.  Here is my app structure
> >
> > index.php (wrapper page) -> instantiates sitePage (Object that does the
> > work)
> >
> > sitePage has several functions that are called to process a 
> template.  If I
> > put the page_open inside one of those functions, then the sess, 
> auth, etc
> > variables never show up in $GLOBALS.  If I move the page_open 
> to index.php,
> > then I get them in $GLOBALS correctly.
> >
> 
> That's a new usage. Interesting. The idea of a  page class is 
> coming to my mind.
> I've tried it and it seems to work to me, with reg_glob off. 
> This is a modified defauth.php3
> 
> Gian
> 
> 
> 
> 
> 
> 

RE: [Phplib-users] Next auth issue, php variable scope issue

[Rob H. <rob...@ws...>]

  If you read page.inc you will see that there is a global variable
definition for all of the features that should result in them going into the
$GLOBALS array as that is what the key work global does.  So it looks to me
as if this should be working.

Rob Hutton
Web Safe
www.wsafe.com

> -----Original Message-----
> From: php...@li...
> [mailto:php...@li...]On Behalf Of Michael
> Chaney
> Sent: Monday, October 14, 2002 12:06 AM
> To: Phplib-Users
> Subject: Re: [Phplib-users] Next auth issue, php variable scope issue
>
>
> On Sun, Oct 13, 2002 at 01:48:23PM -0400, Rob Hutton wrote:
> > This does not seem to be an auth issue per say, but maybe Gian
> or someone
> > can help.  Here is my app structure
> >
> > index.php (wrapper page) -> instantiates sitePage (Object that does the
> > work)
> >
> > sitePage has several functions that are called to process a
> template.  If I
> > put the page_open inside one of those functions, then the sess,
> auth, etc
> > variables never show up in $GLOBALS.  If I move the page_open
> to index.php,
> > then I get them in $GLOBALS correctly.
>
> This is one of the reasons that I removed the "includes" when I wrote
> phpauth and simply made the header and footer into functions.  No
> question about scope, it's not global so you're forced to specify if
> that's what you need.

[Phplib-users] javascript links (was: Auth working with a twist)

[Layne W. <la...@if...>]

> <a href="#" onClick="document.all.innerFrameName.src =
> 'pagename.php?again=yes'; return true">
>
> now it does not do the immediate refresh.  I would expect this on one
> browser, but I was seeing it cross platform, cross browser.
>
> Now, the problem is if I change it to a # and click on it a
> second time, it
> does not do anything, but if I leave off the #, then it
> doesn't work at all
> on Mozilla or Netscape.  AHHH!


IMHO, the correct way to do JavaScript links is

	<a href="javascript:myFunction();">Blah</a>

rather than using onClick. Creating a link to an undefined anchor is a
kludge. I can't say that this will fix your problem, but the extra karma
from cleaner code can't hurt.


Layne Weathers
Ifworld Inc.

Re: [Phplib-users] Next auth issue, php variable scope issue

[Giancarlo <gia...@na...>]

Il 19:48, domenica 13 ottobre 2002, Rob Hutton ha scritto:
> This does not seem to be an auth issue per say, but maybe Gian or someone
> can help.  Here is my app structure
>
> index.php (wrapper page) -> instantiates sitePage (Object that does the
> work)
>
> sitePage has several functions that are called to process a template.  If I
> put the page_open inside one of those functions, then the sess, auth, etc
> variables never show up in $GLOBALS.  If I move the page_open to index.php,
> then I get them in $GLOBALS correctly.
>

That's a new usage. Interesting. The idea of a  page class is 
coming to my mind.
I've tried it and it seems to work to me, with reg_glob off. 
This is a modified defauth.php3

Gian

Re: [Phplib-users] Next auth issue, php variable scope issue

[Michael C. <mdc...@mi...>]

On Sun, Oct 13, 2002 at 01:48:23PM -0400, Rob Hutton wrote:
> This does not seem to be an auth issue per say, but maybe Gian or someone
> can help.  Here is my app structure
> 
> index.php (wrapper page) -> instantiates sitePage (Object that does the
> work)
> 
> sitePage has several functions that are called to process a template.  If I
> put the page_open inside one of those functions, then the sess, auth, etc
> variables never show up in $GLOBALS.  If I move the page_open to index.php,
> then I get them in $GLOBALS correctly.

This is one of the reasons that I removed the "includes" when I wrote
phpauth and simply made the header and footer into functions.  No
question about scope, it's not global so you're forced to specify if
that's what you need.

It's as easy when using phplib, particularly with php4, simply make sure
that everything is explicitly $GLOBALS['varname'].

Michael
-- 
Michael Darrin Chaney
mdc...@mi...
http://www.michaelchaney.com/

[Phplib-users] Next auth issue, php variable scope issue

[Rob H. <rob...@ws...>]

This does not seem to be an auth issue per say, but maybe Gian or someone
can help.  Here is my app structure

index.php (wrapper page) -> instantiates sitePage (Object that does the
work)

sitePage has several functions that are called to process a template.  If I
put the page_open inside one of those functions, then the sess, auth, etc
variables never show up in $GLOBALS.  If I move the page_open to index.php,
then I get them in $GLOBALS correctly.

BTW.  THANKS MUCH TO GIAN AND SEVERAL OTHERS THAT HAVE HELPED TREMENDOUSLY
IN FIGURING THIS OUT!!!!!!!

Rob Hutton
Web Safe
www.wsafe.com

**********************************************************************

Introducing Symantec Client Security - Integrated Anti-Virus,
Firewall, and Intrusion Detection for the Client.

Learn more:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=2&PID=11624271

View our Symantec Client Security Demo:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=3&PID=11624271

Download the Symantec Client Security Fact Sheet:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=4&PID=11624271

Download the Symantec Client Security Brochure:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=5&PID=11624271

RE: [Phplib-users] Auth working with a twist

[Rob H. <rob...@ws...>]

OK, I figured out what it was.  You guys are gonna love this.

I have JavaScript doing the redirect (as I mentioned earlier).  But in the
href of the anchor I had "", but not target or anything, so the anchor looks
something like this (it does other things, this is just an example).

<a href="" onClick="document.all.innerFrameName.src =
'pagename.php?again=yes'; return true">

This way, it displayed the page in innerFrameName, then immediately
refreshed it.  But adding a # to the href so the anchor became

<a href="#" onClick="document.all.innerFrameName.src =
'pagename.php?again=yes'; return true">

now it does not do the immediate refresh.  I would expect this on one
browser, but I was seeing it cross platform, cross browser.

Now, the problem is if I change it to a # and click on it a second time, it
does not do anything, but if I leave off the #, then it doesn't work at all
on Mozilla or Netscape.  AHHH!

Rob Hutton
Web Safe
www.wsafe.com


> -----Original Message-----
> From: Giancarlo [mailto:gia...@na...]
> Sent: Sunday, October 13, 2002 4:02 AM
> To: rob...@ws...
> Cc: phplib-users
> Subject: Re: [Phplib-users] Auth working with a twist
>
>
> Il 02:44, domenica 13 ottobre 2002, hai scritto:
>
> >   Just to run down again, I have a page with an IFRAME imbedded
> in it.  The
> > problem that I am seeing is that the content in the IFRAME is actually
> > being loaded, and then immediately reloaded by the browser.  So
> since I am
> > using default auth, I see the login page flash, then the protected page
> > appears, because the first time through the again is stripped
> off, then the
> > page is reloaded, and it falls through the login_if.
>
> But if you saw the loginform somehow, that means your auth there does not
> exist anymore.
> Once $auth->unauth() or auth->logout has bee called (and it is
> called before
> splashing the form), how can then you see again a protected page?
> From that
> moment, the $auth in your session says you're not authed...
> So, on any reload, it would be not login_if that decides, but the
> auth->start
> in page.inc, that happens before login_if.
> So the problem I think is that the session is not freezed for
> some reason.
>
> Inside login_if, basically you are unauthed, the session is
> freezed, and the
> script exits. So if you could place some echoes just after each of these
> necessary calls, and see what the $auth->auth[uid] is like, we
> could have a
> clue. Or you could try the test defauth.php3 and showoff.php3
> pages with your
> 'features' in page_open, and see what happens (if the counters increase
> mainly)..
>
> That login_if can easily become a mini app, that logins/registers users
> etc... It is already a bit out_of_place there, inside the auth class.
>
> Gian
>
> >   So, first of all, has anyone seen a browser refresh a frame
> immediately
> > when it is loaded before.  Is this common behavior?  I've never seen it.
> >   Secondly, the regex that removes the "again" looks like it removes
>
> > everything after the ?.
>
> Till the next '&' or end_of_string I think it is...
>
>
>   This isn't a good idea IMHO because you can't pass
> > variables to the page.  Am I reading it wrong? and if not, should it be
> > made more specific to just remove the "again".  I'm not a regex person.
> >
> >
> > Rob Hutton
> > Web Safe
> > www.wsafe.com
> >
> > **********************************************************************
> >
> > Introducing Symantec Client Security - Integrated Anti-Virus,
> > Firewall, and Intrusion Detection for the Client.
> >
> > Learn more:
> > http://enterprisesecurity.symantec.com/symes238.cfm?JID=2&PID=11624271
> >
> > View our Symantec Client Security Demo:
> > http://enterprisesecurity.symantec.com/symes238.cfm?JID=3&PID=11624271
> >
> > Download the Symantec Client Security Fact Sheet:
> > http://enterprisesecurity.symantec.com/symes238.cfm?JID=4&PID=11624271
> >
> > Download the Symantec Client Security Brochure:
> > http://enterprisesecurity.symantec.com/symes238.cfm?JID=5&PID=11624271
> >
> >
> >
> >
> > -------------------------------------------------------
> > This sf.net email is sponsored by:ThinkGeek
> > Welcome to geek heaven.
> > http://thinkgeek.com/sf
> > _______________________________________________
> > Phplib-users mailing list
> > Php...@li...
> > https://lists.sourceforge.net/lists/listinfo/phplib-users
>
>

Re: [Phplib-users] Auth working with a twist

[Giancarlo <gia...@na...>]

>
> Inside login_if, basically you are unauthed, the session is freezed, and
> the script exits. So if you could place some echoes just after each of
> these necessary calls, and see what the $auth->auth[uid] is like, we could
> have a clue. Or you could try the test defauth.php3 and showoff.php3 pages
> with your 'features' in page_open, and see what happens (if the counters
> increase mainly)..


Or try this in login_if (snapshot)

           $this->unauth();  
           $sess->freeze();   ##### moved here
###### you could even try $this->logout() instead of the two previos
    
     if (! $this->start()) {
           # See what form
           if ( ($trig == "reg") || ($trig == "log") ) 
              $m=$trig;
           else
              $m=$this->mode;  ## or default

           if ($m == "log")
              $this->auth_loginform();
           else
              $this->auth_registerform();
## moved   $sess->freeze();  ####moved from here
           exit;



There a subtle difference between unauth() and logout(), I think it mimics 
yahoo's mail 'logout' or 'exit completely' (dunno relly, I am forced to see 
the italian version of yahoo...).

In the first case something is kept from your previous login, eg the name you 
typed in the form (uname), some $auth->auth[log_*] info etc.
There is also a kind of bug I found in phplib, that when you are done 
logout() your evetual $user object was still there global and set. That can 
cause some unadverted programmin erroers, so I modified the logout() to unset 
and unregister all eventual $user obj. I still didn't place this in the 
lighter unauth() function, but I think it should be there too.

So you can try logout() instead of unauth, in login_if. You'll loose the 
uname to place ready in tle 'name' field of the loginform but...
Please, as I cannot reproduce, some feedback. 

Gian

Re: [Phplib-users] Auth working with a twist

[Giancarlo <gia...@na...>]

Il 02:44, domenica 13 ottobre 2002, hai scritto:

>   Just to run down again, I have a page with an IFRAME imbedded in it.  The
> problem that I am seeing is that the content in the IFRAME is actually
> being loaded, and then immediately reloaded by the browser.  So since I am
> using default auth, I see the login page flash, then the protected page
> appears, because the first time through the again is stripped off, then the
> page is reloaded, and it falls through the login_if.

But if you saw the loginform somehow, that means your auth there does not 
exist anymore. 
Once $auth->unauth() or auth->logout has bee called (and it is called before 
splashing the form), how can then you see again a protected page? From that 
moment, the $auth in your session says you're not authed... 
So, on any reload, it would be not login_if that decides, but the auth->start 
in page.inc, that happens before login_if.
So the problem I think is that the session is not freezed for some reason. 

Inside login_if, basically you are unauthed, the session is freezed, and the 
script exits. So if you could place some echoes just after each of these 
necessary calls, and see what the $auth->auth[uid] is like, we could have a 
clue. Or you could try the test defauth.php3 and showoff.php3 pages with your 
'features' in page_open, and see what happens (if the counters increase 
mainly)..

That login_if can easily become a mini app, that logins/registers users 
etc... It is already a bit out_of_place there, inside the auth class.

Gian

>   So, first of all, has anyone seen a browser refresh a frame immediately
> when it is loaded before.  Is this common behavior?  I've never seen it.
>   Secondly, the regex that removes the "again" looks like it removes

> everything after the ?.

Till the next '&' or end_of_string I think it is...


  This isn't a good idea IMHO because you can't pass
> variables to the page.  Am I reading it wrong? and if not, should it be
> made more specific to just remove the "again".  I'm not a regex person.
>
>
> Rob Hutton
> Web Safe
> www.wsafe.com
>
> **********************************************************************
>
> Introducing Symantec Client Security - Integrated Anti-Virus,
> Firewall, and Intrusion Detection for the Client.
>
> Learn more:
> http://enterprisesecurity.symantec.com/symes238.cfm?JID=2&PID=11624271
>
> View our Symantec Client Security Demo:
> http://enterprisesecurity.symantec.com/symes238.cfm?JID=3&PID=11624271
>
> Download the Symantec Client Security Fact Sheet:
> http://enterprisesecurity.symantec.com/symes238.cfm?JID=4&PID=11624271
>
> Download the Symantec Client Security Brochure:
> http://enterprisesecurity.symantec.com/symes238.cfm?JID=5&PID=11624271
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Phplib-users mailing list
> Php...@li...
> https://lists.sourceforge.net/lists/listinfo/phplib-users

RE: [Phplib-users] Auth working with a twist

[Rob H. <rob...@ws...>]

So it is Gian's fault ;-).  Seriously, I wonder if it is a php thing.

It only seems to happen inside a frame though...

Rob Hutton
Web Safe
www.wsafe.com 

**********************************************************************

Introducing Symantec Client Security - Integrated Anti-Virus,
Firewall, and Intrusion Detection for the Client.

Learn more:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=2&PID=11624271

View our Symantec Client Security Demo:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=3&PID=11624271

Download the Symantec Client Security Fact Sheet:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=4&PID=11624271

Download the Symantec Client Security Brochure:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=5&PID=11624271


> -----Original Message-----
> From: php...@li...
> [mailto:php...@li...]On Behalf Of Mike Green
> Sent: Saturday, October 12, 2002 9:05 PM
> To: rob...@ws...
> Cc: Phplib-Users
> Subject: Re: [Phplib-users] Auth working with a twist
> 
> 
> Rob Hutton wrote:
> 
> >    So, first of all, has anyone seen a browser refresh a frame 
> immediately
> > when it is loaded before.
> 
> Yes.
> 
> I had this happen to me for the first time yesterday when I put a 
> statement on a
> page which had the following form:
>     session_start();
>     $reset = false;
>     if ($reset) {
>         unset($_SESSION);
>     }
> If I made $reset = true I observed the same phenomenon that you 
> report: the
> browser immediately refreshed the page. This was/is not a problem 
> for what I am
> doing, since I only set $reset = true at certain points during 
> the development
> when I have made changes to a class which cause the objects to be 
> no longer
> valid if they are being saved by the session. And I suspect that 
> what I am doing
> is a bad way to do even what I want to do. But for the sake of 
> just getting on
> with the job and since I don't expect to keep the above statements in the
> production code, I haven't pursued the issue further.
> 
> I'm reporting this here because there is a possibility that it 
> could give a clue
> to what is happening in your case.
> 
> I am not using PHPLib at this point in this project. So the 
> browser immediately
> refreshing the page is not specially a PHPLib problem. It looks, 
> however, as if
> it can be caused by starting a session and then un-setting the $_SESSION
> variable. Thus it might be that PHPLib has something embedded in 
> it which has
> the same effect as that bit of kludgy code of mine above.
> 
> For what it is worth...
> 
> Cheers!
> 
> Mike Green
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Phplib-users mailing list
> Php...@li...
> https://lists.sourceforge.net/lists/listinfo/phplib-users
> 

Re: [Phplib-users] Auth working with a twist

[Mike G. <Mik...@sa...>]

Rob Hutton wrote:

>    So, first of all, has anyone seen a browser refresh a frame immediately
> when it is loaded before.

Yes.

I had this happen to me for the first time yesterday when I put a statement on a
page which had the following form:
    session_start();
    $reset = false;
    if ($reset) {
        unset($_SESSION);
    }
If I made $reset = true I observed the same phenomenon that you report: the
browser immediately refreshed the page. This was/is not a problem for what I am
doing, since I only set $reset = true at certain points during the development
when I have made changes to a class which cause the objects to be no longer
valid if they are being saved by the session. And I suspect that what I am doing
is a bad way to do even what I want to do. But for the sake of just getting on
with the job and since I don't expect to keep the above statements in the
production code, I haven't pursued the issue further.

I'm reporting this here because there is a possibility that it could give a clue
to what is happening in your case.

I am not using PHPLib at this point in this project. So the browser immediately
refreshing the page is not specially a PHPLib problem. It looks, however, as if
it can be caused by starting a session and then un-setting the $_SESSION
variable. Thus it might be that PHPLib has something embedded in it which has
the same effect as that bit of kludgy code of mine above.

For what it is worth...

Cheers!

Mike Green

[Phplib-users] Auth working with a twist

[Rob H. <rob...@ws...>]

  Thanks to Gian and others working on auth with the snapshot.  It works
almost.  What I am seeing is weird, but I am seeing it cross browser.
  Just to run down again, I have a page with an IFRAME imbedded in it.  The
problem that I am seeing is that the content in the IFRAME is actually being
loaded, and then immediately reloaded by the browser.  So since I am using
default auth, I see the login page flash, then the protected page appears,
because the first time through the again is stripped off, then the page is
reloaded, and it falls through the login_if.
  So, first of all, has anyone seen a browser refresh a frame immediately
when it is loaded before.  Is this common behavior?  I've never seen it.
  Secondly, the regex that removes the "again" looks like it removes
everything after the ?.  This isn't a good idea IMHO because you can't pass
variables to the page.  Am I reading it wrong? and if not, should it be made
more specific to just remove the "again".  I'm not a regex person.


Rob Hutton
Web Safe
www.wsafe.com

**********************************************************************

Introducing Symantec Client Security - Integrated Anti-Virus,
Firewall, and Intrusion Detection for the Client.

Learn more:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=2&PID=11624271

View our Symantec Client Security Demo:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=3&PID=11624271

Download the Symantec Client Security Fact Sheet:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=4&PID=11624271

Download the Symantec Client Security Brochure:
http://enterprisesecurity.symantec.com/symes238.cfm?JID=5&PID=11624271

Re: [Phplib-users] OOHForms and RegEx

[Dr T. S. <ta...@sa...>]

On Fri, 11 Oct 2002, Andres Barreto Zicare wrote:

> But if I change
> 	"valid_regex"=>"^[a-z]*$"
> for
> 	"valid_regex"=>"^[a-Z]*$"


try using "valid_regex"=>"^[a-zA-Z]*$"

HTH

Tarique

-- 
=============================================================
PHP Applications for E-Biz: http://www.sanisoft.com

Indian PHP User Group: http://groups.yahoo.com/group/in-phpug
=============================================================

[Phplib-users] OOHForms and RegEx

[Andres B. Z. <aba...@ei...>]

Hello all!
I have been using the PHPLib's OOHForm class with success in a linux redhat
7.2 (apache, mysql & php4) intranet web server. But the same scripts fail
when hosted in the internet. The problem is with the regex support... Most
of the "valid_regex" values do not work. For example: If I try the following
piece of code it works.

$f->add_element(array("type"=>"text",
			    "name"=>"username",
			    "minlength"=>4,
			    "maxlength"=>12,
			    "length_e"=>"Invalid Username. [ 4 to 12 chars long ]",
			    "valid_regex"=>"^[a-z]*$",
			    "valid_e"=>"Invalid Username.",
			    "value"=>"$username"));

But if I change
	"valid_regex"=>"^[a-z]*$"
for
	"valid_regex"=>"^[a-Z]*$"

I have the answer:

Warning: REG_ERANGE in ..../class/oohforms.inc on line 476
Invalid Username.

Next is the piece of code at line 476 in oohforms.inc

  function self_validate($val) {
    if (!is_array($val)) $val = array($val);
    reset($val);
    while (list($k,$v) = each($val)) {
      if ($this->length_e && (strlen($v) < $this->minlength))
        return $this->length_e;
      if ($this->valid_e && (($this->icase &&
            !eregi($this->valid_regex,$v)) ||
           (!$this->icase &&
            !ereg($this->valid_regex,$v))))      // line 476
		return $this->valid_e;
    }
    return false;
  }

Any help will be very welcome

I am not a native english speaker so excuse my poor english language
proficiency.

Thanks

Andres

Re: [Phplib-users] OOHforms Textarea with minlength

[Marko K. <M.K...@os...>]

sorry, forgot that!

Re: [Phplib-users] OOHforms Textarea with minlength

[Marko K. <M.K...@os...>]

The solution is so easy:

you need to copy the function self_validate($val) from of_text.inc
into of_textarea.inc, as well as some var definitions in the textarea
class.

I appended my updated version of the latter originating from Joe's
snapshot which is able to test minlength and maxlength at the same time
as well as enables you to use regex validation, like

$f->add_element(array('type'=>'textarea',
                             'name'=>'comment',
                             'rows'=>6,
                             'cols'=>40,
                             'wrap'=>20,
                             'minlength'=>1,
                             'maxlength'=>3,
                             'length_e'=>'at least 1 up to 3 chars',
                             'valid_regex'=>'^[a-z]*$',
                             'valid_e'=>'only characters allowed!',
                             'value'=>'This is a test text area!'));


Marko

Re: [Phplib-users] OOHforms Textarea with minlength

[Dr T. S. <ta...@sa...>]

On Fri, 11 Oct 2002, Marko Kaening wrote:

Isn't there a patch which Vinay had submitted on the site?

We had fixed this one, mail me offlist if you cant find it

Tarique


> I could reproduce your problem!
> Syntax is okay, analogous to the text input...
> Looks like this is a phplib quirck!
> But, sorry, no solution, just wanted to underline that there is a problem
> indeed.
> 
> Marko
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Phplib-users mailing list
> Php...@li...
> https://lists.sourceforge.net/lists/listinfo/phplib-users
> 

-- 
=============================================================
PHP Applications for E-Biz: http://www.sanisoft.com

Indian PHP User Group: http://groups.yahoo.com/group/in-phpug
=============================================================

Re: [Phplib-users] OOHforms Textarea with minlength

[Marko K. <M.K...@os...>]

I could reproduce your problem!
Syntax is okay, analogous to the text input...
Looks like this is a phplib quirck!
But, sorry, no solution, just wanted to underline that there is a problem
indeed.

Marko

[Phplib-users] OOHforms Textarea with minlength

[Uli L. <U....@Li...>]

Hello List!

I want to require an input in an textarea-field. I tried a minlength, as I
do with "normal" text-fields. But there is no error-message, if the field is
left blank:

$f->add_element(array(	"type"=>"textarea",
						"name"=>"gast_text",
						"cols"=>"69",
						"rows"=>"10",
						"minlength"=>"10",
						"length_e"=>"Irgendeinen Kommentar müssen Sie schon abgeben..."));

Is there any other syntax, to require an input? I can't find the soloution
in the manual...

Thank you for your help...

Uli Linn

------------------------------------------------------
Uli Linn                    LINN Internet Marketing
Fon:  +49-6831-976789       Waldstraße 19
Fax:  +49-6831-976788       D-66763 Dillingen
Mail: U....@Li...     URL: http://www.LinnWeb.de

RE: [Phplib-users] handle perm in auth

[Rob H. <rob...@ws...>]

Not a dead horse, maybe hairless soon, but not dead.  I agree on one hand,
on the other hand I will tell you the same thing I told Gian, and that is
that this is somewhat of a square peg in a round whole.  What I am looking
for it permx + 1 layer of abstraction.  I have to have it, so I will write
permxx when I get there and post it.

Rob Hutton
Web Safe
www.wsafe.com

> -----Original Message-----
> From: php...@li...
> [mailto:php...@li...]On Behalf Of Joe
> Stewart
> Sent: Thursday, October 10, 2002 1:13 PM
> To: phplib-users
> Subject: Re: [Phplib-users] handle perm in auth
>
>
> Hope I'm not beating a dead horse...
>
> On Wed, Oct 09, 2002 at 08:44:21PM -0400, Rob Hutton wrote:
> > It's tied to the user, not the page.  So I can't be an author
> of one page
> > and a reviewer of another page.
> >
>
> The argument of have_perm can be seen as the page's required permisssion.
>
> Permissions per page or parts of a page can be accomplished easily.  The
> perm->have_perm checks whether the authenticated user has the permission.
>
> Posessing author privileges doesn't stop them from having reviewer
> privileges also.
>
> Joe
>
>
> > Rob Hutton
> > Web Safe
> > www.wsafe.com
> >
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Phplib-users mailing list
> Php...@li...
> https://lists.sourceforge.net/lists/listinfo/phplib-users
>
>

Re: [Phplib-users] handle perm in auth

[Joe S. <jo...@be...>]

To keep the conversation going...

On Wed, Oct 09, 2002 at 09:57:15PM -0400, Rob Hutton wrote:
> I think you are trying to fit a square peg in a round hole.  That kind of
> bitwise math is too complex to reasonably maintain for the average db admin.
> IMHO it would be better to have a user_perms table like:
> 
> uid
> permname
> permlevel
> 

You should be able to generate the permissions array from a db if needed.  
The example in local.inc is just that, and example.

> Then it would be easy to see in the db what perms are set and still fits in
> with the current scheme.
> 

I don't want to be tied to a db for each perm check.

I agree that using binary arithmetic to generate individual permissions 
causes problems.

Give the permx.class a try.  Applied to the atomic permissions example in 
local.inc the permissions array would look like:

var $permissions = array(
   "user"       => true,
   "author"     => true,
   "editor"     => true,
   "supervisor" => true,
   "admin"      => true
);

The permx.class is in the patches snapshot and here:

http://sourceforge.net/tracker/index.php?func=detail&aid=617072&group_id=31885&atid=403613

It's been posted for everyone's feedback and ridicule.

Joe

> Rob Hutton
> Web Safe
> www.wsafe.com
> 

Re: [Phplib-users] handle perm in auth

[Joe S. <jo...@be...>]

Hope I'm not beating a dead horse...

On Wed, Oct 09, 2002 at 08:44:21PM -0400, Rob Hutton wrote:
> It's tied to the user, not the page.  So I can't be an author of one page
> and a reviewer of another page.
> 

The argument of have_perm can be seen as the page's required permisssion.

Permissions per page or parts of a page can be accomplished easily.  The
perm->have_perm checks whether the authenticated user has the permission.

Posessing author privileges doesn't stop them from having reviewer 
privileges also.

Joe 


> Rob Hutton
> Web Safe
> www.wsafe.com
>