śr., 19 kwi 2023 o 09:37 Greg Huber napisał(a):
>
> > hack to satisfy Roller requirements
>
> Well I don't think its a hack, to have the variable name as part of the
> id does not seem correct to me
>
> eg
>
> entryEdit
>
>
>
> ie the name would be :
>
> entry___tmp__saveDraft
>
> vs
hack to satisfy Roller requirements
Well I don't think its a hack, to have the variable name as part of the
id does not seem correct to me
eg
entryEdit
ie the name would be :
entry___tmp__saveDraft
vs
entry_entryEdit_saveDraft
calculate ID based on action
AND method at the sa
śr., 19 kwi 2023 o 09:19 Greg Huber napisał(a):
> clickById("entry_%{#mainAction}!publish")
>
> This patch fixes the format by calculating the action first:
>
> https://github.com/apache/struts/commit/d7cf72c92eb84437eb9794b56c2525b389cf7900
This is really a hack to satisfy Roller requirements, b
@Greg Huber is the current approach ok?
https://github.com/apache/struts/pull/678
This works but not really related to this formatting issue.
As there is nothing wrong with Struts, it never formatted it correctly in the
first place, which can be seen from the test clickById :
clickById("entry
I'm sorry Yasser if you took this personally - escape logic has
changed and on first thought that was the cause. After investigating
the thing deeper I found it isn't just this but also missing support
for evaluation of ID which bases on action or method.
Previously ID was evaluated on set (in the
from Struts internal behavior.
Best,
Yasser.
From: Greg Huber
Sent: Saturday, April 15, 2023 3:08 PM
To: dev@struts.apache.org
Subject: Re: looking at roller upgrade again
I think it was always wrong as the test is checking for :
entry_%{#mainAction}!pu
I think it was always wrong as the test is checking for :
entry_%{#mainAction}!publish
clickById("entry_%{#mainAction}!publish")
On 14/04/2023 19:17, Yasser Zamani wrote:
Thank you for the explanation Greg. Yes I agree that previously it was
looking better. Currently am wondering how previous
Thank you for the explanation Greg. Yes I agree that previously it was
looking better. Currently am wondering how previous Struts was
generating the id from an evaluated name! Because as you see below,
Struts is and was keeping name property unchanged via introducing a
local var named name:
On 4/3/2023 11:18 AM, Lukasz Lenart wrote:
The change has been introduced here [1] and the problem is that it
replaces any non-alphanumeric character with "_". Also it works on an
unevaluated version of the "name" attribute (in case if the "id"
attribute is not defined). I think this is a bug
There is nothing wrong with struts.
There is a selenium test in roller that checks on the id
clickById("entry_%{#mainAction}!publish");
it now has:
entrymainAction__publish
ie it escapes %{#}! with spaces.
To match other tags, it should evaluate %{#mainAction}
ie using the form below:
Sorry I didn't get what the problem exactly is.
1. Was your app depended to Struts internal behavior of id generation
and so your app is broken now?
2. Or no, Struts itself is broken now by my change?
On 4/11/2023 10:16 AM, Greg Huber wrote:
More housekeeping, the id on the form tag never su
More housekeeping, the id on the form tag never supported %{..} on the
action attribute. ie action="%{#mainAction}!saveDraft"
On 10/04/2023 20:37, Yasser Zamani wrote:
Hi there, please see inline...
On 4/3/2023 11:18 AM, Lukasz Lenart wrote:
The change has been introduced here [1] and the pro
Hi there, please see inline...
On 4/3/2023 11:18 AM, Lukasz Lenart wrote:
The change has been introduced here [1] and the problem is that it
replaces any non-alphanumeric character with "_". Also it works on an
unevaluated version of the "name" attribute (in case if the "id"
attribute is not def
pon., 10 kwi 2023 o 09:57 Greg Huber napisał(a):
>
> Modified version here
>
> https://github.com/gregh3269/struts/tree/WW-5302-unevaluated-id
>
>
>
>myConfig
> action="%{#mainAction}!save" />
> method="save" />
>
> id="myConfig_myConfig_save">
I missed that you do not use the "n
Modified version here
https://github.com/gregh3269/struts/tree/WW-5302-unevaluated-id
myConfig
action="%{#mainAction}!save" />
method="save" />
id="myConfig_myConfig_save">
On 09/04/2023 15:16, Lukasz Lenart wrote:
niedz., 9 kwi 2023 o 12:20 Greg Huber napisał(a):
Testing the
branch WW-5302-unevaluated-id
Seems the same to me. Note the action has a variable: %{#mainAction}!save
All it seems to be doing is escaping action: %{#mainAction}!save :
%{#mainAction}!save
___mainAction}_save
On 09/04/2023 15:16, Lukasz Lenart wrote:
niedz., 9 kwi 2023 o 12:20 Greg Huber
niedz., 9 kwi 2023 o 12:20 Greg Huber napisał(a):
> Testing the branch it is still the same?
Please try now
Regards
--
Łukasz
+ 48 606 323 122 http://www.lenart.org.pl/
-
To unsubscribe, e-mail: dev-unsubscr...@struts.apache.
Testing the branch it is still the same?
myConfig
id="myConfigmainAction__save" >
What we want is:
id="myConfig_myConfig_save" >
##
It needs this from previous email
// determine actual action
ActionMapping mapping = new ActionMapping();
mapping.setName(findString(action)
PR is ready
https://github.com/apache/struts/pull/678
-
To unsubscribe, e-mail: dev-unsubscr...@struts.apache.org
For additional commands, e-mail: dev-h...@struts.apache.org
niedz., 9 kwi 2023 o 10:07 Greg Huber napisał(a):
>
> > > name="entryEdit!saveDraft"/>
>
> I guess what we want.
Great, thanks for reporting this issue :)
https://issues.apache.org/jira/browse/WW-5302
Regards
--
Łukasz
+ 48 606 323 122 http://www.lenart.org.pl/
--
I guess what we want.
###
In the FormButton populateComponentHtmlId(Form form) it evaluates the action based on the text
rather than the "actual" action. If the action is determined from the mapping it renders
correctly. See "// determine actual action" in code below
entryEdit
renders
What about such an approach?
The ID is generated based on the evaluated version of the name attribute.
Regards
--
Łukasz
+ 48 606 323 122 http://www.lenart.org.pl/
-
To unsubscribe, e-mail: dev-unsubscr...@struts.apache.org
pon., 3 kwi 2023 o 10:26 Greg Huber napisał(a):
>
> Thanks.
>
> There is a selenium test in roller that checks on the id
>
> clickById("entry_%{#mainAction}!publish");
>
> it now has:
>
> entrymainAction__publish"
>
> I guess its a matter of modifying the test, as %{ characters etc may not
> b
Thanks.
There is a selenium test in roller that checks on the id
clickById("entry_%{#mainAction}!publish");
it now has:
entrymainAction__publish"
I guess its a matter of modifying the test, as %{ characters etc may not
be desirable the id field?
On 03/04/2023 08:48, Lukasz Lenart wrote
The change has been introduced here [1] and the problem is that it
replaces any non-alphanumeric character with "_". Also it works on an
unevaluated version of the "name" attribute (in case if the "id"
attribute is not defined). I think this is a bug and I'm not sure why
the "escape" method has bee
25 matches
Mail list logo
