Topics

"First" 4.19.8 files available


Bob Jacobsen
 

The proposed JMRI 4.19.7 files are available in the usual way at:

https://builds.jmri.org/jenkins/job/testreleases/job/4.19.8/15/

(Please use build 15; the task has been making tentative builds from the branch over the last two weeks)

Feedback appreciated. I would like to release this later today or tomorrow morning if the files are OK.

Bob

--
Bob Jacobsen
@BobJacobsen


Dan Boudreau
 

Downloaded, installed, and tested all good on Win10.

 

Dan

 

From: Bob Jacobsen
Sent: Friday, June 19, 2020 3:48 PM
To: jmri@jmri-developers.groups.io
Subject: [jmri-developers] "First" 4.19.8 files available

 

The proposed JMRI 4.19.7 files are available in the usual way at:

https://builds.jmri.org/jenkins/job/testreleases/job/4.19.8/15/

(Please use build 15; the task has been making tentative builds from the branch over the last two weeks)

Feedback appreciated. I would like to release this later today or tomorrow morning if the files are OK.

Bob

--
Bob Jacobsen
rgj1927@...




 


Dave Sand
 

Works fine with macOS 10.15.5.

Dave Sand

----- Original message -----
From: Bob Jacobsen <@BobJacobsen>
To: jmri@jmri-developers.groups.io
Subject: [jmri-developers] "First" 4.19.8 files available
Date: Friday, June 19, 2020 2:48 PM

The proposed JMRI 4.19.7 files are available in the usual way at:

https://builds.jmri.org/jenkins/job/testreleases/job/4.19.8/15/

(Please use build 15; the task has been making tentative builds from the branch over the last two weeks)

Feedback appreciated. I would like to release this later today or tomorrow morning if the files are OK.

Bob

--
Bob Jacobsen
@BobJacobsen


Dave Sand
 

The scripts that use AbstractShutDownTask fail with a method not found.

The execute method has been deprecated and it appears that Jython tries to use the run method.

AC_PowerControl.py:class ShutDown(jmri.implementation.AbstractShutDownTask):

AutoDispatcher2.py:from jmri.implementation import AbstractShutDownTask

IoT/JMRI_TcpPeripheral.py:class TcpPeripheral_ShutDown(jmri.implementation.AbstractShutDownTask):

ShutDownExample.py:class MyShutDownTask(jmri.implementation.AbstractShutDownTask):

TurnoutStatePersistence.py:class PersistTurnoutStateTask(jmri.implementation.AbstractShutDownTask):


Changing the method and removing the return True resolves the problem.

Dave Sand


----- Original message -----
From: Dave Sand <ds@...>
Subject: Re: [jmri-developers] "First" 4.19.8 files available
Date: Friday, June 19, 2020 3:52 PM

Works fine with macOS 10.15.5.

Dave Sand



----- Original message -----
From: Bob Jacobsen <rgj1927@...>
Subject: [jmri-developers] "First" 4.19.8 files available
Date: Friday, June 19, 2020 2:48 PM

The proposed JMRI 4.19.7 files are available in the usual way at:


(Please use build 15; the task has been making tentative builds from the branch over the last two weeks)

Feedback appreciated. I would like to release this later today or tomorrow morning if the files are OK.

Bob

--
Bob Jacobsen








Pete Cressman
 

Installs and runs on win 10 with NEC and LocoBuffer.
The WhereUsed tool throws an NPE at line 942 of OBlock.java. This is a log.debug statement having an argument that requires the presence of a sensor in the OBlock.
PR8715 rewrites the debug statement.
Sorry I didn't know about this tool until yesterday or I would have done something about it sooner.


On Friday, June 19, 2020, 12:48:20 PM PDT, Bob Jacobsen <rgj1927@...> wrote:


The proposed JMRI 4.19.7 files are available in the usual way at:


(Please use build 15; the task has been making tentative builds from the branch over the last two weeks)

Feedback appreciated. I would like to release this later today or tomorrow morning if the files are OK.

Bob

--
Bob Jacobsen






Bob Jacobsen
 

There’s now a new 4.19.8 build with installers at:

https://builds.jmri.org/jenkins/job/testreleases/job/4.19.8/16/

This includes Pete’s fix below, plus fixing the script that use AbstractShutDownTask

Would appreciate people checking their basic functionality.

Bob

On Jun 20, 2020, at 10:46 AM, Pete Cressman <pete_cressman@...> wrote:

Installs and runs on win 10 with NEC and LocoBuffer.
The WhereUsed tool throws an NPE at line 942 of OBlock.java. This is a log.debug statement having an argument that requires the presence of a sensor in the OBlock.
PR8715 rewrites the debug statement.
Sorry I didn't know about this tool until yesterday or I would have done something about it sooner.


Dan Boudreau
 

Downloaded, installed, and tested okay on Win10.

 

Dan

 

From: Bob Jacobsen
Sent: Monday, June 22, 2020 6:41 PM
To: jmri@jmri-developers.groups.io
Subject: Re: [jmri-developers] "First" 4.19.8 files available

 

There’s now a new 4.19.8 build with installers at:

https://builds.jmri.org/jenkins/job/testreleases/job/4.19.8/16/

This includes Pete’s fix below, plus fixing the script that use AbstractShutDownTask

Would appreciate people checking their basic functionality. 

Bob

> On Jun 20, 2020, at 10:46 AM, Pete Cressman <pete_cressman@...> wrote:
>
> Installs and runs on win 10 with NEC and LocoBuffer.
> The WhereUsed tool throws an NPE at line 942 of OBlock.java. This is a log.debug statement having an argument that requires the presence of a sensor in the OBlock.
> PR8715 rewrites the debug statement.
> Sorry I didn't know about this tool until yesterday or I would have done something about it sooner.
>


 


Dave Sand
 

Works fine on macOS 10.15.5. The TurnoutStatePersistence.py script works properly.

Dave Sand

----- Original message -----
From: Bob Jacobsen <@BobJacobsen>
To: jmri@jmri-developers.groups.io
Subject: Re: [jmri-developers] "First" 4.19.8 files available
Date: Monday, June 22, 2020 5:41 PM

There’s now a new 4.19.8 build with installers at:

https://builds.jmri.org/jenkins/job/testreleases/job/4.19.8/16/

This includes Pete’s fix below, plus fixing the script that use AbstractShutDownTask

Would appreciate people checking their basic functionality.

Bob

On Jun 20, 2020, at 10:46 AM, Pete Cressman <pete_cressman@...> wrote:

Installs and runs on win 10 with NEC and LocoBuffer.
The WhereUsed tool throws an NPE at line 942 of OBlock.java. This is a log.debug statement having an argument that requires the presence of a sensor in the OBlock.
PR8715 rewrites the debug statement.
Sorry I didn't know about this tool until yesterday or I would have done something about it sooner.


Andrew Crosland
 

Tested with SPROG on Win10 Pro.

Andrew

------ Original Message ------
From: "Bob Jacobsen" <@BobJacobsen>
To: jmri@jmri-developers.groups.io
Sent: 22/06/2020 23:41:18
Subject: Re: [jmri-developers] "First" 4.19.8 files available

There’s now a new 4.19.8 build with installers at:

https://builds.jmri.org/jenkins/job/testreleases/job/4.19.8/16/

This includes Pete’s fix below, plus fixing the script that use AbstractShutDownTask

Would appreciate people checking their basic functionality.

Bob

On Jun 20, 2020, at 10:46 AM, Pete Cressman <pete_cressman@...> wrote:

Installs and runs on win 10 with NEC and LocoBuffer.
The WhereUsed tool throws an NPE at line 942 of OBlock.java. This is a log.debug statement having an argument that requires the presence of a sensor in the OBlock.
PR8715 rewrites the debug statement.
Sorry I didn't know about this tool until yesterday or I would have done something about it sooner.




--
Andrew Crosland


Gert (Speed) Muller
 

Downloaded, installed, and MQTT and NCE Simulator tested fine on Ubuntu 20.04, using openjdk version "11.0.7" 2020-04-14
Speed


Pete Cressman
 

JMRI.4.19.8+Rbab3d0f33.exe fails at the same test, same line.
Is the link below correct?

Although the release note can announce this, I'm afraid it will become generally noticeable for most people who use warrants.  It's convenient to configure "Dark" blocks at yards and stub storage sidings so they can be used as origins and destinations of warrants.

On Monday, June 22, 2020, 03:41:32 PM PDT, Bob Jacobsen <rgj1927@...> wrote:


There’s now a new 4.19.8 build with installers at:

https://builds.jmri.org/jenkins/job/testreleases/job/4.19.8/16/

This includes Pete’s fix below, plus fixing the script that use AbstractShutDownTask

Would appreciate people checking their basic functionality.

Bob

> On Jun 20, 2020, at 10:46 AM, Pete Cressman <pete_cressman@...> wrote:
>
> Installs and runs on win 10 with NEC and LocoBuffer.
> The WhereUsed tool throws an NPE at line 942 of OBlock.java. This is a log.debug statement having an argument that requires the presence of a sensor in the OBlock.
> PR8715 rewrites the debug statement.
> Sorry I didn't know about this tool until yesterday or I would have done something about it sooner.
>




Bob Jacobsen
 

The released files for downloading are:

• OS X / macOS: https://github.com/JMRI/JMRI/releases/download/v4.19.8/JMRI.4.19.8+Rbab3d0f33.dmg
sha256: 18a2e69d2aadce903a6eceab5edcb2cb95da0870434e7e7977beface3bc26347
• Windows: https://github.com/JMRI/JMRI/releases/download/v4.19.8/JMRI.4.19.8+Rbab3d0f33.exe
sha256: 14e784bd5eab03f6c91fbba75a441d6a124b2e39705d9a81b7284a862862c4c5
• Linux: https://github.com/JMRI/JMRI/releases/download/v4.19.8/JMRI.4.19.8+Rbab3d0f33.tgz
sha256: 10fc17166c2789694d114c1221efa36321242e043f570a37f96502b379257678

Could you confirm whether the fault is still in those? If we need to make a 4.19.9 with a fix, then we can do that, so if the problem is present in 4.19.8, _PLEASE_ develop the fix starting far enough back that _your_ fix can be merged. For example, start with

git fetch
git checkout origin/v4.19.8
git checkout -b my-new-branch

If I made a mistake while disassembling and reassembling the previous patch for this problem to put it in the right branch for merging, I’m sorry that happened. I would have been better if that hadn’t been necessary.

Bob

On Jun 23, 2020, at 12:18 PM, Pete Cressman <pete_cressman@...> wrote:

JMRI.4.19.8+Rbab3d0f33.exe fails at the same test, same line.
Is the link below correct?

Although the release note can announce this, I'm afraid it will become generally noticeable for most people who use warrants. It's convenient to configure "Dark" blocks at yards and stub storage sidings so they can be used as origins and destinations of warrants.

On Monday, June 22, 2020, 03:41:32 PM PDT, Bob Jacobsen <@BobJacobsen> wrote:


There’s now a new 4.19.8 build with installers at:

https://builds.jmri.org/jenkins/job/testreleases/job/4.19.8/16/

This includes Pete’s fix below, plus fixing the script that use AbstractShutDownTask

Would appreciate people checking their basic functionality.

Bob

On Jun 20, 2020, at 10:46 AM, Pete Cressman <pete_cressman@...> wrote:

Installs and runs on win 10 with NEC and LocoBuffer.
The WhereUsed tool throws an NPE at line 942 of OBlock.java. This is a log.debug statement having an argument that requires the presence of a sensor in the OBlock.
PR8715 rewrites the debug statement.
Sorry I didn't know about this tool until yesterday or I would have done something about it sooner.



Andrew Crosland
 

Tested on Raspbian Buster.

Andrew

------ Original Message ------
From: "Bob Jacobsen" <@BobJacobsen>
To: jmri@jmri-developers.groups.io
Sent: 22/06/2020 23:41:18
Subject: Re: [jmri-developers] "First" 4.19.8 files available

There’s now a new 4.19.8 build with installers at:

https://builds.jmri.org/jenkins/job/testreleases/job/4.19.8/16/

This includes Pete’s fix below, plus fixing the script that use AbstractShutDownTask

Would appreciate people checking their basic functionality.

Bob

On Jun 20, 2020, at 10:46 AM, Pete Cressman <pete_cressman@...> wrote:

Installs and runs on win 10 with NEC and LocoBuffer.
The WhereUsed tool throws an NPE at line 942 of OBlock.java. This is a log.debug statement having an argument that requires the presence of a sensor in the OBlock.
PR8715 rewrites the debug statement.
Sorry I didn't know about this tool until yesterday or I would have done something about it sooner.




--
Andrew Crosland


steve young
 

Hi Folks,

I found that the Web Throttle is broken, see https://github.com/JMRI/JMRI/issues/8734
I've uploaded a fix for head of master, however unsure if it's worth me uploading the fix for this on the 4.19.8 branch?

I'm tempted to hold off to see if any other 4.19.8 issues emerge, or should I upload one to be used only IF there's a 4.19.9 and could just be deleted if there's not?

Steve.


Randall Wood <rhwood@...>
 

I would go ahead and create a PR to fix this in the 4.19.8 branch. It’s easier for us to close it without merging than it is for us to remember to create a PR later.

Randall

On Jun 26, 2020, at 06:58, steve young via groups.io <icklesteve@...> wrote:

Hi Folks,

I found that the Web Throttle is broken, see https://github.com/JMRI/JMRI/issues/8734
I've uploaded a fix for head of master, however unsure if it's worth me uploading the fix for this on the 4.19.8 branch?

I'm tempted to hold off to see if any other 4.19.8 issues emerge, or should I upload one to be used only IF there's a 4.19.9 and could just be deleted if there's not?

Steve.