February 2018 - Pharo-dev - lists.pharo.org

[Pharo 7.0-dev] Build #502: 19702-glmSubscriptions-not-needed-in-SubscriptionRegistry
by ci-pharo-ci-jenkins2＠inria.fr Feb. 6, 2018

Feb. 6, 2018

There is a new Pharo build available! The status of the build #502 was: SUCCESS. The Pull Request #777 was integrated: "19702-glmSubscriptions-not-needed-in-SubscriptionRegistry" Pull request url: https://github.com/pharo-project/pharo/pull/777 Issue Url: https://pharo.fogbugz.com/f/cases/19702 Build Url: https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending%20pull%20request%2…

1 0

[Pharo 7.0-dev] Build #501: 21240-update-iceberg-to-067
by ci-pharo-ci-jenkins2＠inria.fr Feb. 6, 2018

Feb. 6, 2018

There is a new Pharo build available! The status of the build #501 was: SUCCESS. The Pull Request #800 was integrated: "21240-update-iceberg-to-067" Pull request url: https://github.com/pharo-project/pharo/pull/800 Issue Url: https://pharo.fogbugz.com/f/cases/21240 Build Url: https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending%20pull%20request%2…

1 0

[Pharo 7.0-dev] Build #500: 21236-Sometimes-the-debugger-opens-with-the-context-one-off
by ci-pharo-ci-jenkins2＠inria.fr Feb. 6, 2018

Feb. 6, 2018

There is a new Pharo build available! The status of the build #500 was: SUCCESS. The Pull Request #797 was integrated: "21236-Sometimes-the-debugger-opens-with-the-context-one-off" Pull request url: https://github.com/pharo-project/pharo/pull/797 Issue Url: https://pharo.fogbugz.com/f/cases/21236 Build Url: https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending%20pull%20request%2…

1 0

Weird ZnClient benchmarking result
by Sven Van Caekenberghe Feb. 6, 2018

Feb. 6, 2018

Hi, While benchmarking ZnClient I encountered something weird and counter intuitive that I cannot explain. This was on macOS running Pharo 7 and 6. First you need to setup a fast local HTTP server, can be anything, for example: $ brew install nginx $ brew services start nginx $ curl http://localhost:8080 In Pharo, the benchmark code looks as follows: ZnClient new in: [ :client | [ client get: 'http://localhost:8080' ] bench ]. On my machine this gives about 2500 req/s. ZnClient uses object logging: it emits special event objects that contain information regarding what happens (full details and timing information) using an Announcer. Subscribers can then decide what to do with this detailed information. This mechanism is always active, but there is an option to turn it totally off for maximum performance: ZnClient new in: [ :client | client loggingOff. [ client get: 'http://localhost:8080' ] bench ]. This gives about 3000 req/s, a non-trivial + 500 req/s improvement. Now comes the weird thing. GT contains a custom view for Announcers, inspect: ZnLogEvent announcer. And select the Announcements tab. Leave this window open. Now, the first expression (with full logging enabled and an active subscriber with a UI) is way FASTER than first, it yields about 7000 to 8000 req/s. Since the benchmark runs in the main UI thread, no actual drawing happens until the expression finishes. Running both expression inside a Time profiler show how in the first case significant time (actually *most* time ;-) is spent in the logging: In the second case, with logging disabled, that changes to: One has to take into account that the request has to be processed by the server, so from the client's perspective, most time will always be spent waiting for a reply, #readResponse. With the logging enabled, probably less time is spent waiting, since the client does a bit more work. Still what should be slower is more than 2 times as fast ! I have a feeling that this has to do with scheduling (probably interacting with network IO), as going deep into the last #readResponse above, I see: But I still cannot understand why doing apparently more work results in faster throughput. Can anyone explain this behaviour ? Thx, Sven

6 8

ESUG 2018 Call for Participation
by Stephane Ducasse Feb. 6, 2018

Feb. 6, 2018

Call-for-Presentations 26rd International Smalltalk Joint Conference The ESUG board is pleased to announce that the 26th ESUG conference/summer-school will be held in Cagliari, Italy 10-14 September 2018; with Camp Smalltalk 8-9 September 2018. The conference is co-organized by University of Cagliari and the Agile Group of the faculty of computer science. http://www.esug.org/wiki/pier/Conferences/2018/ This call includes: - Developer Forum - Smalltalk Technology Award - International Workshop http://www.esug.org/wiki/pier/Conferences/2018/International-Workshop-IWST_… - Student Volunteer http://www.esug.org/wiki/pier/Conferences/2018/Student-volunteers You can support the ESUG conference in many different ways: =============================================== - Sponsor the conference. New sponsoring packages are described at http://www.esug.org/wiki/pier/About/BecomeSponsor - Submit a talk, a software or a paper to one of the events. See below. - Attend the conference. We'd like to beat the previous record of attendance. Students can get free registration and hosting if they enrol into the the Student Volunteers program. See below. Developers Forum: International Smalltalk Developers Conference Call For Participation ================ We are looking for YOUR experience on using Smalltalk. You will have 30 min for presentations and 45-60 min for hand-ons tutorial. The list of topics for the normal talks and tutorials includes, but is not limited to the following: -Micro Services, Container, Cloud, Big Data, -XP practices, Development tools, Experience reports -Model driven development, Web development, Team management -Meta-Modeling, Security, New libraries and frameworks -Educational material, Embedded systems and robotics -SOA and Web services, Interaction with other programming languages -Teaching Pearls and Show us Your Business -Show your business 10 min session (Get prepared!!) How to submit? ============ Submissions deadline is 15 of June 2018 Notification of acceptance will be on done on the fly. Pay attention: the places are limited so do not wait till the last minute to apply. Prospective presenters should submit a request to Stephane.Ducasse[at]inria.fr AND USE THE following header [ESUG 2018 Developers] Title: Please follow the template below the email will be automatically processed! Name: Email: Abstract: Bio: Any presentation not respecting this form will be discarded automatically. International Workshop on Smalltalk Technologies ====================================== Read the page: *http://www.esug.org/wiki/pier/Conferences/2018/International-Workshop-IWST_18* Technology Award Competition ======================== The top 3 teams with the most innovative software will receive, respectively, 500 Euros, 300 Euros and 200 Euros during an awards ceremony at the conference. Developers of any Smalltalk-based software are welcome to compete. More information at http://www.esug.org/wiki/pier/Conferences/2018/Innovation-Technology-Awards Student Volunteer Program ===================== If you are a student wanting to attend ESUG, have you considered being a student volunteer? Student volunteers help keep the conference running smoothly; in return, they have free accommodations, while still having most of the time to enjoy the conference. More information at http://www.esug.org/wiki/pier/Conferences/2018/Student-volunteers We hope to see you there and have fun together. -------------------------------------------- StÃ©phane Ducasse http://stephane.ducasse.free.fr http://www.synectique.eu / http://www.pharo.org 03 59 35 87 52 Assistant: Julie Jonas FAX 03 59 57 78 50 TEL 03 59 35 86 16 S. Ducasse - Inria 40, avenue Halley, Parc Scientifique de la Haute Borne, BÃ¢t.A, Park Plaza Villeneuve d'Ascq 59650 France

1 0

Feedback about "Contribute a Fix to Pharo" (For windows users)
by Nicolai Hess Feb. 6, 2018

Feb. 6, 2018

The setup finally somehow works: I followed the guide at https://github.com/pharo-project/pharo/wiki/Contribute-a-fix-to-Pharo * wget -O- get.pharo.org/70+vm | bash Does not work (windows msys-console) It starts donwloading something but not pharo I tried to do at seperate steps: Only the image * wget -O- get.pharo.org/70 | bash Works * wget -O- get.pharo.org/vm | bash works, but is the wrong VM I guess, I can not open the image * wget -O- get.pharo.org/vm70 | bash works, seems to load the right vm * ./pharo-ui Pharo.image Starts the vm but complains about : "Pharo cannot write to the changes file named u:\tmp\pharo\Pharo.changes." I am sure the file is writable. The vm reports to be built on aug 27 2017, is this right? Setting up Iceberg: Yes, need option 2, that is, I use the commandline to clone a repository and add it to iceberg. Cloning from iceberg directly does not work (long file names). Ok, now I see some uncommitted changes, the guide says: "you may need to invoke synchronize" But how? There is a "synchronize repository... " and "Synchronize full repository ..." menu entry. Which one? The first synchronize does not show me how to actually do anything and the second one shows a intimidating warning :) FInally I removed my fork (it was pretty old), did a new fork and sstarted with a fresh image. Finally, the initial setup up is done, I now try to do some work with it and report back nicolai

4 9

[Pharo 7.0-dev] Build #499: 21237-Remove-UndefinedObjectsuspend
by ci-pharo-ci-jenkins2＠inria.fr Feb. 6, 2018

Feb. 6, 2018

There is a new Pharo build available! The status of the build #499 was: SUCCESS. The Pull Request #798 was integrated: "21237-Remove-UndefinedObjectsuspend" Pull request url: https://github.com/pharo-project/pharo/pull/798 Issue Url: https://pharo.fogbugz.com/f/cases/21237 Build Url: https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending%20pull%20request%2…

1 0

[Bug] Sometimes the debugger opens with the context one off
by Marcus Denker Feb. 5, 2018

Feb. 5, 2018

Hi, This one was annoying me for a while: sometimes, when you put a halt the debugger opens but with exactly the method you put the halt in not selected and hidden (you need to scroll up!). I analysed a bit the situation and the case were this happens is when you put a halt and then run a test to trigger the halt. Normal halts are ok. https://pharo.manuscript.com/f/cases/21236/Sometimes-the-debugger-opens-wit… fix to review: https://github.com/pharo-project/pharo/pull/797

1 0

nil suspend
by Aliaksei Syrel Feb. 5, 2018

Feb. 5, 2018

Hi, I was looking at nil (already funny) and found a very interesting method: suspend "Kills off processes that didn't terminate properly" "Display reverse; reverse." "<-- So we can catch the suspend bug" Processor terminateActive It means that the following code is probably the shortest way to kill your image (save first!) nil suspend It terminates a UI process if evaluated from anywhere using CMD+d :D Cheers, Alex

5 4

Esteban's ChangeLog week of 29 January 2018
by estebanlm＠gmail.com Feb. 5, 2018

Feb. 5, 2018

Hello! This is my weekly ChangeLog, from 29 January 2018 to 4 February 2018. You can see it in a better format by going here: http://log.smallworks.eu/web/search?from=29/1/2018&to=4/2/2018 ChangeLog ========= 29 January 2018: ---------------- * I spent the morning adding certificates to the VM to sign it :) Next batch of VMs should come signed... now it rest to see if we need to sign the images also. cheers! Esteban

2 1