#99243Wrong behaviour when using property and value together
Description
According to the docs, name and value should be ignored, if property is set. I`m using the view helper like this:
<f:form.textfield
id="{id}"
name="customName"
property="{property}"
value="customValue"
placeholder="{placeholder}"
class="input__field form-control{f:if(condition: '{class}', then: ' {class}')}"
errorClass="{errorClass}"
additionalAttributes="{additionalAttributes}"
data="{data}"
aria="{describedby: 'error-label-{id}'}"
/>
This is the result:
<input required="required" aria-describedby="error-label-email" class="input__field form-control" id="email" type="text" name="tx_myextension[customerData][email]" value="customValue">
The name attribute is ignored, but not the value attribute.
I´m not sure, what is wrong. Maybe the code, maybe the docs.
Inside the view helper, I found this part:
* The value is determined as follows:
* * If property mapping errors occurred and the form is re-displayed, the *last submitted* value is returned
* * Else the bound property is returned (only in objectAccessor-mode)
* * As fallback the "value" argument of this ViewHelper is used
if ($this->respectSubmittedDataValue) {
$value = $this->getValueFromSubmittedFormData($value);
} elseif ($this->hasArgument('value')) {
$value = $this->arguments['value'];
} elseif ($this->isObjectAccessorMode()) {
$value = $this->getPropertyValue();
}
I can fix this, if someone tells me, which would be the correct behaviour.
Reviews 2
History 613
-
Gerrit Code Review pushed patch set 1 for main to change 77681 status
NewUnder Review 02/03/23, 13:50 -
Gerrit Code Review pushed patch set 2 for main to change 77681 02/07/23, 07:10
-
Gerrit Code Review pushed patch set 3 for main to change 77681 02/07/23, 08:25
-
Gerrit Code Review pushed patch set 4 for main to change 77681 03/14/23, 10:22
-
Gerrit Code Review pushed patch set 5 for main to change 77681 03/14/23, 12:39
-
Gerrit Code Review pushed patch set 6 for main to change 77681 05/03/23, 09:41
-
Simon Praetorius changed Duplicated by #93527 10/19/23, 09:54
-
Benni Mack changed status
Under ReviewNew 07/29/24, 21:40 -
Gerrit Code Review pushed patch set 1 for main to change 85413 status
NewUnder Review 07/30/24, 09:40 -
Garvin Hicking commented 07/30/24, 09:41
-
Garvin Hicking changed assigned to Garvin Hicking 07/30/24, 09:41
-
Garvin Hicking commented and changed status
Under ReviewResolved done0 %100 % 07/30/24, 12:10Applied in changeset commit:ca8f1f9f9240e80a222e99b7b404226f4caddf90.
-
Benni Mack changed status
ResolvedClosed 10/15/24, 09:30
Previous patch has been abandoned with a discussion about “intentional behavior” of the “value” attribute.
One possible example why both attributes together may make sense is a form to edit a user. For the password field you would want to define property=“password”, but also value=”“ to not expose the current password hash to the form.
Thus, the new patch adjusts the inline code comment to reflect that intention.