[COLLECTIONS] Master branch build failing

classic Classic list List threaded Threaded
10 messages Options
Reply | Threaded
Open this post in threaded view
|

[COLLECTIONS] Master branch build failing

Benedikt Ritter-4
Hi,

it looks like the master branch build for commons-collections is failing
for a while now [1]. Does anybody have the time to look into this?

Benedikt

[1] https://travis-ci.org/apache/commons-collections/jobs/404183011
Reply | Threaded
Open this post in threaded view
|

Re: [COLLECTIONS] Master branch build failing

garydgregory
This has to do with the Java version used to run the build... I've seen it
locally but just switched Java version to avoid it.

Gary

On Thu, Aug 23, 2018, 11:18 Benedikt Ritter <[hidden email]> wrote:

> Hi,
>
> it looks like the master branch build for commons-collections is failing
> for a while now [1]. Does anybody have the time to look into this?
>
> Benedikt
>
> [1] https://travis-ci.org/apache/commons-collections/jobs/404183011
>
Reply | Threaded
Open this post in threaded view
|

Re: [COLLECTIONS] Master branch build failing

sebb-2-2
Huh?

Looks like the failure is due to Clirr is reporting some breaks; not
sure how that can depend on the Java version.


On 23 August 2018 at 18:37, Gary Gregory <[hidden email]> wrote:

> This has to do with the Java version used to run the build... I've seen it
> locally but just switched Java version to avoid it.
>
> Gary
>
> On Thu, Aug 23, 2018, 11:18 Benedikt Ritter <[hidden email]> wrote:
>
>> Hi,
>>
>> it looks like the master branch build for commons-collections is failing
>> for a while now [1]. Does anybody have the time to look into this?
>>
>> Benedikt
>>
>> [1] https://travis-ci.org/apache/commons-collections/jobs/404183011
>>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [COLLECTIONS] Master branch build failing

Benedikt Ritter-4
Am Do., 23. Aug. 2018 um 20:12 Uhr schrieb sebb <[hidden email]>:

> Huh?
>
> Looks like the failure is due to Clirr is reporting some breaks; not
> sure how that can depend on the Java version.
>

Maybe related to Java 8 default methods? I'm just guessing here... Need to
dig into this on the weekend.

Benedikt


>
>
> On 23 August 2018 at 18:37, Gary Gregory <[hidden email]> wrote:
> > This has to do with the Java version used to run the build... I've seen
> it
> > locally but just switched Java version to avoid it.
> >
> > Gary
> >
> > On Thu, Aug 23, 2018, 11:18 Benedikt Ritter <[hidden email]> wrote:
> >
> >> Hi,
> >>
> >> it looks like the master branch build for commons-collections is failing
> >> for a while now [1]. Does anybody have the time to look into this?
> >>
> >> Benedikt
> >>
> >> [1] https://travis-ci.org/apache/commons-collections/jobs/404183011
> >>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
>
Reply | Threaded
Open this post in threaded view
|

Re: [COLLECTIONS] Master branch build failing

Benedikt Ritter-4
Hi,

Am Fr., 24. Aug. 2018 um 09:06 Uhr schrieb Benedikt Ritter <
[hidden email]>:

>
>
> Am Do., 23. Aug. 2018 um 20:12 Uhr schrieb sebb <[hidden email]>:
>
>> Huh?
>>
>> Looks like the failure is due to Clirr is reporting some breaks; not
>> sure how that can depend on the Java version.
>>
>
> Maybe related to Java 8 default methods? I'm just guessing here... Need to
> dig into this on the weekend.
>

It looks like this is related to Clirr handling default methods the wrong
way. So we probably need to fix clirr in order to use it. I found a mirror
of the code on GitHub in Emmanuel Bourg's account [1]. We could probably
try to fix clirr, but I'd like to explore whether it would be okay to drop
clirr entirely and only use jacimp instead (which seems to work with Java
8).

Regards,
Benedikt

[1] https://github.com/ebourg/clirr/issues/1


>
> Benedikt
>
>
>>
>>
>> On 23 August 2018 at 18:37, Gary Gregory <[hidden email]> wrote:
>> > This has to do with the Java version used to run the build... I've seen
>> it
>> > locally but just switched Java version to avoid it.
>> >
>> > Gary
>> >
>> > On Thu, Aug 23, 2018, 11:18 Benedikt Ritter <[hidden email]> wrote:
>> >
>> >> Hi,
>> >>
>> >> it looks like the master branch build for commons-collections is
>> failing
>> >> for a while now [1]. Does anybody have the time to look into this?
>> >>
>> >> Benedikt
>> >>
>> >> [1] https://travis-ci.org/apache/commons-collections/jobs/404183011
>> >>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [hidden email]
>> For additional commands, e-mail: [hidden email]
>>
>>
Reply | Threaded
Open this post in threaded view
|

Re: [COLLECTIONS] Master branch build failing

garydgregory
On Fri, Aug 24, 2018 at 7:41 AM Benedikt Ritter <[hidden email]> wrote:

> Hi,
>
> Am Fr., 24. Aug. 2018 um 09:06 Uhr schrieb Benedikt Ritter <
> [hidden email]>:
>
> >
> >
> > Am Do., 23. Aug. 2018 um 20:12 Uhr schrieb sebb <[hidden email]>:
> >
> >> Huh?
> >>
> >> Looks like the failure is due to Clirr is reporting some breaks; not
> >> sure how that can depend on the Java version.
> >>
> >
> > Maybe related to Java 8 default methods? I'm just guessing here... Need
> to
> > dig into this on the weekend.
> >
>
> It looks like this is related to Clirr handling default methods the wrong
> way. So we probably need to fix clirr in order to use it. I found a mirror
> of the code on GitHub in Emmanuel Bourg's account [1]. We could probably
> try to fix clirr, but I'd like to explore whether it would be okay to drop
> clirr entirely and only use jacimp instead (which seems to work with Java
> 8).
>

It's not clear to me that japicmp is maintained either, it has not had a
release in a long time.

Should we bring in Clirr as a Commons project if the license allows?

Gary


> Regards,
> Benedikt
>
> [1] https://github.com/ebourg/clirr/issues/1
>
>
> >
> > Benedikt
> >
> >
> >>
> >>
> >> On 23 August 2018 at 18:37, Gary Gregory <[hidden email]>
> wrote:
> >> > This has to do with the Java version used to run the build... I've
> seen
> >> it
> >> > locally but just switched Java version to avoid it.
> >> >
> >> > Gary
> >> >
> >> > On Thu, Aug 23, 2018, 11:18 Benedikt Ritter <[hidden email]>
> wrote:
> >> >
> >> >> Hi,
> >> >>
> >> >> it looks like the master branch build for commons-collections is
> >> failing
> >> >> for a while now [1]. Does anybody have the time to look into this?
> >> >>
> >> >> Benedikt
> >> >>
> >> >> [1] https://travis-ci.org/apache/commons-collections/jobs/404183011
> >> >>
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: [hidden email]
> >> For additional commands, e-mail: [hidden email]
> >>
> >>
>
Reply | Threaded
Open this post in threaded view
|

Re: [COLLECTIONS] Master branch build failing

Gilles Sadowski
In reply to this post by Benedikt Ritter-4
On Fri, 24 Aug 2018 15:41:36 +0200, Benedikt Ritter wrote:

> Hi,
>
> Am Fr., 24. Aug. 2018 um 09:06 Uhr schrieb Benedikt Ritter <
> [hidden email]>:
>
>>
>>
>> Am Do., 23. Aug. 2018 um 20:12 Uhr schrieb sebb <[hidden email]>:
>>
>>> Huh?
>>>
>>> Looks like the failure is due to Clirr is reporting some breaks;
>>> not
>>> sure how that can depend on the Java version.
>>>
>>
>> Maybe related to Java 8 default methods? I'm just guessing here...
>> Need to
>> dig into this on the weekend.
>>
>
> It looks like this is related to Clirr handling default methods the
> wrong
> way. So we probably need to fix clirr in order to use it. I found a
> mirror
> of the code on GitHub in Emmanuel Bourg's account [1]. We could
> probably
> try to fix clirr, but I'd like to explore whether it would be okay to
> drop
> clirr entirely and only use jacimp instead (which seems to work with
> Java
> 8).

Japicmp has its own problems; it had to be disabled so that
"Commons RNG" could be released.

Gilles

>
> Regards,
> Benedikt
>
> [1] https://github.com/ebourg/clirr/issues/1
>
>
>>
>> Benedikt
>>
>>
>>>
>>>
>>> On 23 August 2018 at 18:37, Gary Gregory <[hidden email]>
>>> wrote:
>>> > This has to do with the Java version used to run the build...
>>> I've seen
>>> it
>>> > locally but just switched Java version to avoid it.
>>> >
>>> > Gary
>>> >
>>> > On Thu, Aug 23, 2018, 11:18 Benedikt Ritter <[hidden email]>
>>> wrote:
>>> >
>>> >> Hi,
>>> >>
>>> >> it looks like the master branch build for commons-collections is
>>> failing
>>> >> for a while now [1]. Does anybody have the time to look into
>>> this?
>>> >>
>>> >> Benedikt
>>> >>
>>> >> [1]
>>> https://travis-ci.org/apache/commons-collections/jobs/404183011
>>> >>


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [COLLECTIONS] Master branch build failing

sebb-2-2
In reply to this post by garydgregory
On 24 August 2018 at 14:46, Gary Gregory <[hidden email]> wrote:

> On Fri, Aug 24, 2018 at 7:41 AM Benedikt Ritter <[hidden email]> wrote:
>
>> Hi,
>>
>> Am Fr., 24. Aug. 2018 um 09:06 Uhr schrieb Benedikt Ritter <
>> [hidden email]>:
>>
>> >
>> >
>> > Am Do., 23. Aug. 2018 um 20:12 Uhr schrieb sebb <[hidden email]>:
>> >
>> >> Huh?
>> >>
>> >> Looks like the failure is due to Clirr is reporting some breaks; not
>> >> sure how that can depend on the Java version.
>> >>
>> >
>> > Maybe related to Java 8 default methods? I'm just guessing here... Need
>> to
>> > dig into this on the weekend.
>> >
>>
>> It looks like this is related to Clirr handling default methods the wrong
>> way. So we probably need to fix clirr in order to use it. I found a mirror
>> of the code on GitHub in Emmanuel Bourg's account [1]. We could probably
>> try to fix clirr, but I'd like to explore whether it would be okay to drop
>> clirr entirely and only use jacimp instead (which seems to work with Java
>> 8).
>>
>
> It's not clear to me that japicmp is maintained either, it has not had a
> release in a long time.
>
> Should we bring in Clirr as a Commons project if the license allows?

It would be good to adjust the Maven plugin to distinguish source and
binary issues.
At present the HTML report combines these.

> Gary
>
>
>> Regards,
>> Benedikt
>>
>> [1] https://github.com/ebourg/clirr/issues/1
>>
>>
>> >
>> > Benedikt
>> >
>> >
>> >>
>> >>
>> >> On 23 August 2018 at 18:37, Gary Gregory <[hidden email]>
>> wrote:
>> >> > This has to do with the Java version used to run the build... I've
>> seen
>> >> it
>> >> > locally but just switched Java version to avoid it.
>> >> >
>> >> > Gary
>> >> >
>> >> > On Thu, Aug 23, 2018, 11:18 Benedikt Ritter <[hidden email]>
>> wrote:
>> >> >
>> >> >> Hi,
>> >> >>
>> >> >> it looks like the master branch build for commons-collections is
>> >> failing
>> >> >> for a while now [1]. Does anybody have the time to look into this?
>> >> >>
>> >> >> Benedikt
>> >> >>
>> >> >> [1] https://travis-ci.org/apache/commons-collections/jobs/404183011
>> >> >>
>> >>
>> >> ---------------------------------------------------------------------
>> >> To unsubscribe, e-mail: [hidden email]
>> >> For additional commands, e-mail: [hidden email]
>> >>
>> >>
>>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [COLLECTIONS] Master branch build failing

sebb-2-2
On 24 August 2018 at 14:53, sebb <[hidden email]> wrote:

> On 24 August 2018 at 14:46, Gary Gregory <[hidden email]> wrote:
>> On Fri, Aug 24, 2018 at 7:41 AM Benedikt Ritter <[hidden email]> wrote:
>>
>>> Hi,
>>>
>>> Am Fr., 24. Aug. 2018 um 09:06 Uhr schrieb Benedikt Ritter <
>>> [hidden email]>:
>>>
>>> >
>>> >
>>> > Am Do., 23. Aug. 2018 um 20:12 Uhr schrieb sebb <[hidden email]>:
>>> >
>>> >> Huh?
>>> >>
>>> >> Looks like the failure is due to Clirr is reporting some breaks; not
>>> >> sure how that can depend on the Java version.
>>> >>
>>> >
>>> > Maybe related to Java 8 default methods? I'm just guessing here... Need
>>> to
>>> > dig into this on the weekend.
>>> >
>>>
>>> It looks like this is related to Clirr handling default methods the wrong
>>> way. So we probably need to fix clirr in order to use it. I found a mirror
>>> of the code on GitHub in Emmanuel Bourg's account [1]. We could probably
>>> try to fix clirr, but I'd like to explore whether it would be okay to drop
>>> clirr entirely and only use jacimp instead (which seems to work with Java
>>> 8).
>>>
>>
>> It's not clear to me that japicmp is maintained either, it has not had a
>> release in a long time.
>>
>> Should we bring in Clirr as a Commons project if the license allows?

Not sure tha's possible, as it uses LGPL:

http://clirr.sourceforge.net/license.html

> It would be good to adjust the Maven plugin to distinguish source and
> binary issues.
> At present the HTML report combines these.
>
>> Gary
>>
>>
>>> Regards,
>>> Benedikt
>>>
>>> [1] https://github.com/ebourg/clirr/issues/1
>>>
>>>
>>> >
>>> > Benedikt
>>> >
>>> >
>>> >>
>>> >>
>>> >> On 23 August 2018 at 18:37, Gary Gregory <[hidden email]>
>>> wrote:
>>> >> > This has to do with the Java version used to run the build... I've
>>> seen
>>> >> it
>>> >> > locally but just switched Java version to avoid it.
>>> >> >
>>> >> > Gary
>>> >> >
>>> >> > On Thu, Aug 23, 2018, 11:18 Benedikt Ritter <[hidden email]>
>>> wrote:
>>> >> >
>>> >> >> Hi,
>>> >> >>
>>> >> >> it looks like the master branch build for commons-collections is
>>> >> failing
>>> >> >> for a while now [1]. Does anybody have the time to look into this?
>>> >> >>
>>> >> >> Benedikt
>>> >> >>
>>> >> >> [1] https://travis-ci.org/apache/commons-collections/jobs/404183011
>>> >> >>
>>> >>
>>> >> ---------------------------------------------------------------------
>>> >> To unsubscribe, e-mail: [hidden email]
>>> >> For additional commands, e-mail: [hidden email]
>>> >>
>>> >>
>>>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [COLLECTIONS] Master branch build failing

thc202
In reply to this post by garydgregory

On 24/08/18 14:46, Gary Gregory wrote:

> On Fri, Aug 24, 2018 at 7:41 AM Benedikt Ritter <[hidden email]> wrote:
>
>> Hi,
>>
>> Am Fr., 24. Aug. 2018 um 09:06 Uhr schrieb Benedikt Ritter <
>> [hidden email]>:
>>
>>>
>>>
>>> Am Do., 23. Aug. 2018 um 20:12 Uhr schrieb sebb <[hidden email]>:
>>>
>>>> Huh?
>>>>
>>>> Looks like the failure is due to Clirr is reporting some breaks; not
>>>> sure how that can depend on the Java version.
>>>>
>>>
>>> Maybe related to Java 8 default methods? I'm just guessing here... Need
>> to
>>> dig into this on the weekend.
>>>
>>
>> It looks like this is related to Clirr handling default methods the wrong
>> way. So we probably need to fix clirr in order to use it. I found a mirror
>> of the code on GitHub in Emmanuel Bourg's account [1]. We could probably
>> try to fix clirr, but I'd like to explore whether it would be okay to drop
>> clirr entirely and only use jacimp instead (which seems to work with Java
>> 8).
>>
>
> It's not clear to me that japicmp is maintained either, it has not had a
> release in a long time.

It seems to have regular releases, last one 4 days ago:
https://search.maven.org/search?q=g:com.github.siom79.japicmp
https://github.com/siom79/japicmp/tags

Best regards.

>
> Should we bring in Clirr as a Commons project if the license allows?
>
> Gary
>
>
>> Regards,
>> Benedikt
>>
>> [1] https://github.com/ebourg/clirr/issues/1
>>
>>
>>>
>>> Benedikt
>>>
>>>
>>>>
>>>>
>>>> On 23 August 2018 at 18:37, Gary Gregory <[hidden email]>
>> wrote:
>>>>> This has to do with the Java version used to run the build... I've
>> seen
>>>> it
>>>>> locally but just switched Java version to avoid it.
>>>>>
>>>>> Gary
>>>>>
>>>>> On Thu, Aug 23, 2018, 11:18 Benedikt Ritter <[hidden email]>
>> wrote:
>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> it looks like the master branch build for commons-collections is
>>>> failing
>>>>>> for a while now [1]. Does anybody have the time to look into this?
>>>>>>
>>>>>> Benedikt
>>>>>>
>>>>>> [1] https://travis-ci.org/apache/commons-collections/jobs/404183011
>>>>>>
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: [hidden email]
>>>> For additional commands, e-mail: [hidden email]
>>>>
>>>>
>>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]