| Index
» Projects
» phpMyEdit
» Bug list
» Bug #284 |
 |
|
Viewing Bug Advanced Details |
[ Jump to Bugnotes ]
[ Bug History ]
[ Print ]
|
|
ID |
Category |
Severity |
Reproducibility |
Date Submitted |
Last Update |
|
0000284 |
documentation |
feature |
always |
2004-11-27 18:57 |
2004-11-27 22:02 |
|
|
|
Reporter |
marianov |
View Status |
public |
|
|
Assigned To |
|
|
Priority |
normal |
Resolution |
open |
Platform |
Firefox |
|
Status |
acknowledged |
Duplicate ID |
|
OS |
Linux |
|
Projection |
none |
|
Version |
|
|
ETA |
none |
|
Product Version |
5.4 (stable) |
|
|
Product Build |
|
|
|
Votes |
1 |
|
|
|
Summary |
Browser refresh causes duplicate inserts. |
|
Description |
When the user clicks refresh on the browser after an insert PME inserts a duplicated record in the table.
I havent found a way to avoid it (aside from using a unique index, which is not always posible).
In other php apps i have used something like:
if (!isset($_SESSION['pcount']))
$_SESSION['pcount'] = 0; //$pcount is a "form post counter"
function formrefresh() { return $_REQUEST['pcount'] !=
$_SESSION['pcount']++; }
and then when building each form I add:
<input type=hidden name=pcount value=<?=$_SESSION['pcount'] ; ?> >
So before taking action I can check that the user is not submitting a form like this:
if (!formrefresh() ) {
//insert the record
} else {
// show a new empty form or redirect to the list of records
}
Do you think it would be easy to add this kind of feature to PME ?
Thanks,
Mariano |
|
Steps To Reproduce |
|
|
Additional Information / Solution |
|
|
Attached Files |
|
|
|
|
Bug Notes |
michal
2004-11-27 20:31
|
[this is not website/homepage related, more docuemntation issue.]
yes,
have a look at trigger insert before (to stop the insert proccess), and $opts['cgi']['persist'] option (to keep the hidden variable). |
|
|
nepto
2004-11-27 21:55
|
Well, concerning to addition of this into core class, I do not thing it is a good idea. I do not want to mess/bloat core class with things like $_SESSION and so on.
upravené: 2004-11-27 21:57 |
|
|
hbernard
2004-11-27 22:02
|
Agreed.
This is (as michal said) a documentation issue. |
|
|
|
|