Shoudn't search in Jenkins be case-insensitive by default?

[Marcin Zajączkowski]

Hi,

The search mechanism in Jenkins is case-sensitive by default. You may not noticed that as currently that property isn't taken into account very often. However, working recently on JENKINS-40718 [1] (Search by build param values in Build history widget) I've been asked to implement considering that property there and in general it seem to be a movement to consider it also in other places [2].

For me personally case-sensitive search by default is not a very intuitive option. As an user I would prefer to see all possible result and if needed (too many results) I would consider switching it to be case-sensitive. Currently there will be just no results in specific cases (e.g. searching for failed jobs for in a given job with "fail" keyword will not result anything (you will need to use "FAIL" or "FAILED" - upper case) - new people could think that searching by build/run status is not supported - I for instance don't know about a possibility to change case (in)sensitivity in search up until recently).

I don't know if there were any strong arguments to have searching case-sensitive by default a few years ago or it was just set to that mode. Therefore, I would ask what do you think about that and about a possibility to change that default behavior (to be case-insensitive)?


[1] - https://issues.jenkins-ci.org/browse/JENKINS-40718
[2] - https://github.com/jenkinsci/jenkins/pull/2683#discussion_r100706395


Marcin

--
http://blog.solidsoft.info/ - Solid Soft - Working code is not enough

[Marcin Zajączkowski]

And a link to current code:
https://github.com/jenkinsci/jenkins/blob/76b5ba7f4641102917d551764e275f3d6044186e/core/src/main/java/hudson/search/UserSearchProperty.java#L28-L29

[Daniel Beck]

> On 08.03.2017, at 11:37, Marcin Zajączkowski <msz...@wp.pl> wrote:
>
> I don't know if there were any strong arguments to have searching case-sensitive by default a few years ago or it was just set to that mode. Therefore, I would ask what do you think about that and about a possibility to change that default behavior (to be case-insensitive)?

I'd be in favor. That default has never made sense to me.

[Baptiste Mathus]

+1, definitely. Case insensitive seems wayyyy more sensible to me.

--
You received this message because you are subscribed to the Google Groups "Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-dev+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-dev/C747F65B-7B16-4D90-86A0-4332B3C596DA%40beckweb.net.

For more options, visit https://groups.google.com/d/optout.

[Baptiste Mathus]

I've started that very unscientific poll FWIW :) https://twitter.com/bmathus/status/839432020916854784

[Marcin Zajączkowski]

On Wednesday, March 8, 2017 at 1:39:23 PM UTC+1, Baptiste Mathus wrote:

I've started that very unscientific poll FWIW :) https://twitter.com/bmathus/status/839432020916854784

Participants of your poll seem to agree as well:
4% - case sensitive (as now)
81% - case insensitive
15% - WAT?

I created https://issues.jenkins-ci.org/browse/JENKINS-42645 and I plan to work on the code to change the default behavior unless there will be a good argument against it.

However, in addition it would be good to consider (probably as a separate issue to do not postpone the main action) a possibility to change that behavior for users that haven't intentionally opt-in to have case sensitive search. Unfortunately it can be impossible to determine that situation. Therefore maybe it would be beneficial (to prevent a "regression" related to JENKINS-40718 and similar changes) to reset that behavior and require from an user to opt-in to it once more - however, it could be somehow clumsy in the implementation (e.g. two values in XML for the same feature - old legacy and the new one). Suggestions for that are especially welcome.

Marcin

 

2017-03-08 12:00 GMT+01:00 Baptiste Mathus <m...@batmat.net>:

+1, definitely. Case insensitive seems wayyyy more sensible to me.

2017-03-08 11:56 GMT+01:00 Daniel Beck <m...@beckweb.net>:

> On 08.03.2017, at 11:37, Marcin Zajączkowski <msz...@wp.pl> wrote:
>
> I don't know if there were any strong arguments to have searching case-sensitive by default a few years ago or it was just set to that mode. Therefore, I would ask what do you think about that and about a possibility to change that default behavior (to be case-insensitive)?

I'd be in favor. That default has never made sense to me.

--
You received this message because you are subscribed to the Google Groups "Jenkins Developers" group.

To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-de...@googlegroups.com.

[Marcin Zajączkowski]

I proposed changes with implementing for existing users: https://github.com/jenkinsci/jenkins/pull/2801

In addition I created a new issue for existing users: https://issues.jenkins-ci.org/browse/JENKINS-42701 . Implementation suggestions are welcome.

Marcin

[Marcin Zajączkowski]

On 2017-03-13 00:29, Marcin Zajączkowski wrote:
> I proposed changes with implementing for existing users:
> https://github.com/jenkinsci/jenkins/pull/2801

Should be for new and anonymous users here.

Marcin

>
> In addition I created a new issue for existing users:
> https://issues.jenkins-ci.org/browse/JENKINS-42701 . Implementation
> suggestions are welcome.
>
> Marcin
>
>
> On Thursday, March 9, 2017 at 11:24:02 PM UTC+1, Marcin Zajączkowski wrote:
>>
>> On Wednesday, March 8, 2017 at 1:39:23 PM UTC+1, Baptiste Mathus wrote:
>>>
>>> I've started that very unscientific poll FWIW :)
>>> https://twitter.com/bmathus/status/839432020916854784
>>>
>>
>> Participants of your poll seem to agree as well:
>> 4% - case sensitive (as now)
>> 81% - case insensitive
>> 15% - WAT?
>>
>> I created https://issues.jenkins-ci.org/browse/JENKINS-42645 and I plan
>> to work on the code to change the default behavior unless there will be a
>> good argument against it.
>>
>> However, in addition it would be good to consider (probably as a separate
>> issue to do not postpone the main action) a possibility to change that
>> behavior for users that haven't intentionally opt-in to have case sensitive
>> search. Unfortunately it can be impossible to determine that situation.
>> Therefore maybe it would be beneficial (to prevent a "regression" related

>> to JENKINS-40718 <https://issues.jenkins-ci.org/browse/JENKINS-40718> and

>> similar changes) to reset that behavior and require from an user to opt-in
>> to it once more - however, it could be somehow clumsy in the implementation
>> (e.g. two values in XML for the same feature - old legacy and the new one).
>> Suggestions for that are especially welcome.
>>
>> Marcin
>>
>>
>>
>>>
>>> 2017-03-08 12:00 GMT+01:00 Baptiste Mathus <m...@batmat.net>:
>>>
>>>> +1, definitely. Case insensitive seems wayyyy more sensible to me.
>>>>
>>>> 2017-03-08 11:56 GMT+01:00 Daniel Beck <m...@beckweb.net>:
>>>>
>>>>>
>>>>>> On 08.03.2017, at 11:37, Marcin Zajączkowski <msz...@wp.pl> wrote:
>>>>>>
>>>>>> I don't know if there were any strong arguments to have searching
>>>>> case-sensitive by default a few years ago or it was just set to that mode.
>>>>> Therefore, I would ask what do you think about that and about a possibility
>>>>> to change that default behavior (to be case-insensitive)?
>>>>>
>>>>> I'd be in favor. That default has never made sense to me.
>>>>>
>>>>> --
>>>>> You received this message because you are subscribed to the Google
>>>>> Groups "Jenkins Developers" group.
>>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>>> an email to jenkinsci-de...@googlegroups.com.
>>>>> To view this discussion on the web visit
>>>>> https://groups.google.com/d/msgid/jenkinsci-dev/C747F65B-7B16-4D90-86A0-4332B3C596DA%40beckweb.net
>>>>> .
>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>
>>>>
>>>>
>>>
>

--
http://blog.solidsoft.info/ - Working code is not enough