in anybody out there? who uses 2.0?
Locked
12
12
Re: [Pharo-project] in anybody out there? who uses 2.0?
 
|
we should plit that in bug entries and take actions collectively :)
On Mar 28, 2013, at 4:56 AM, Yanni Chiu <[hidden email]> wrote: > On 27/03/13 10:24 AM, Esteban Lorenzano wrote: >> >> Here in Pharo headquarters we are shock that there are just 10 new bugs reported for 2.0 after the release... >> So... I wonder... is that because we made a really cool release, or just because nobody is using it? > > Just before the release I loaded up my project, and did a quick check to find that everything looked fine - except that I would have to migrate to Fuel-1.9. > > I had noticed that package loading seemed extremely slow, but did not look further into it. I think I saw mention that it's due to some usage of #become:, during the compiling of code. Based on build times (of just loading the rough equivalent code), it seems about 3 times slower to do a build on a Pharo-2.0 vs. Pharo-1.4. > > The slowness is not just an annoyance, because I actually compile code in my application - it's just compiling getters and setters. I've not got enough working yet to see whether it's going to adversely affect the usability (it could make startup time too slow). > > Another thing I've noticed is occasional sluggishness in the UI. It's hard to pinpoint, I often feel like my clicks are being lost. > > The behaviour of the TestRunner was odd. Eventually I discovered running tests via the Nautilus browser, but the UI feedback is extremely confusing for "abstract" test cases. I still don't quite understand the results I see there, so I do a final run of the tests in the TestRunner. > > Another strange issue I had with test cases was to do with the interaction of the deprecation warnings. In by build script, I run: > Deprecation raiseWarning: false. > Deprecation showWarning: false. > so the build can run headless. It took me a few hours, and a careful single stepping, to find that the deprecation exceptions were being swallowed. I'm sure the TestRunner did not behave this way before. If you ran a test, you would still see the deprecation exceptions. It was really frustrating to see your test fail, but have the stack cleared out before you could debug the exception that caused the test failure. > > Are these bugs, or just me getting used to the new release? > > |
Re: in anybody out there? who uses 2.0?
|
|
In reply to this post by Yanni Chiu
2013/3/28 Yanni Chiu <[hidden email]>
For me some clicks are counted double (or wrong). From the welcome-workspace hightlight MetacelloConfigurationBrowser open. and use the Mouse and context-menu for "do it". It opens the configuration browser, but it reopens the menu as well. The same happens for other mouse interactions as well. Open a Browser. Click on the codepane, open context menu, use left-mouse button in another browser pane -> the context menu closes and reopens again at the new mouse position. Nicolai |
Re: in anybody out there? who uses 2.0?
|
|
In reply to this post by Marcus Denker-4
On 28/03/13 3:11 AM, Marcus Denker wrote:
> > On Mar 28, 2013, at 4:56 AM, Yanni Chiu <[hidden email]> wrote: >> >> Another thing I've noticed is occasional sluggishness in the UI. >> It's hard to pinpoint, I often feel like my clicks are being lost. >> > I have not seen that one. After some more usage I'll pinpoint where it's happening, since it interrupts the train of thought otherwise. >> The behaviour of the TestRunner was odd. Eventually I discovered >> running tests via the Nautilus browser, but the UI feedback is >> extremely confusing for "abstract" test cases. I still don't quite >> understand the results I see there, so I do a final run of the >> tests in the TestRunner. >> > Odd in which sense? related to the progress bar? The TestRunner was sluggish when selecting test cases. This was the case in Pharo-1.4 too, but a little worse in Pharo-2.0. I had some tests that caused the UI to stop responding (had to kill the image, but the image's cpu usage was normal). Once the test cases passed, I've not seen the problem again. The main problem is that if I see a test failure in the TestRunner. Then when I try to browse that test case, the actual test code is in a superclass. But, it's unclear how to invoke that test case for the subclass through Nautilus, and it's not clear where the test result is shown in Nautilus (i.e. where is the green icon when the actual test method is in the superclass). >> Another strange issue I had with test cases was to do with the >> interaction of the deprecation warnings. In by build script, I >> run: Deprecation raiseWarning: false. Deprecation showWarning: >> false. so the build can run headless. It took me a few hours, and a >> careful single stepping, to find that the deprecation exceptions >> were being swallowed. I'm sure the TestRunner did not behave this >> way before. If you ran a test, you would still see the deprecation >> exceptions. It was really frustrating to see your test fail, but >> have the stack cleared out before you could debug the exception >> that caused the test failure. >> > Can you add a bug tracker entry for that one? https://pharo.fogbugz.com/default.asp?10166 |
Re: in anybody out there? who uses 2.0?
|
|
In reply to this post by smalltalk-4
On 28/03/13 2:32 AM, smalltalk wrote:
> which os do you use? It's MacOSX 10.5.8 (with a VM I built about 6 months ago), but my build/deploy is Debian Linux. |
Re: in anybody out there? who uses 2.0?
|
|
On 28/03/13 11:40 AM, Yanni Chiu wrote:
> On 28/03/13 2:32 AM, smalltalk wrote: >> which os do you use? > > It's MacOSX 10.5.8 (with a VM I built about 6 months ago), but my > build/deploy is Debian Linux. I meant to say that the latest PharoVM from pharo.org is used on Debian. |
Re: in anybody out there? who uses 2.0?
|
|
In reply to this post by Goubier Thierry
I am developing MathsOntologie under Pharo 2.0 beta, and testing it under Pharo 1.4; when the test is OK I distribute MathsOntologie with Pharo 1.4 as "all in one". Don't know if it is the best way to do but I became used to it...
By the way, I am ashamed that I never made a unit test... Also, when I wanted to translate in French the method "add: anObject times: aNumber" to a Collection, I found that it didn't work, so I created it with something like Collection>add: anObject times: aNumber aNumber timesRepeat: [self add: anObject]. I don't know if it was intended or a bug, and and did not find it very difficult to create this method but I asked myself why this method did not work. And I could not test it afterwards because I could not update Pharo 2.0 beta (conflicts between consecutive updates); I have to download Pharo 2.0 and transfer MathsOntologie inside the new version; I just did not have the time... Alain On Thu, Mar 28, 2013 at 12:34 PM, Goubier Thierry <[hidden email]> wrote: Le 27/03/2013 15:24, Esteban Lorenzano a écrit : |
Re: in anybody out there? who uses 2.0?
|
|
Hello alain I am developing MathsOntologie under Pharo 2.0 beta, and testing it under Pharo 1.4; when the test is OK I distribute MathsOntologie with Pharo 1.4 as "all in one". Don't know if it is the best way to do but I became used to it… I would stick with the latest version since 1.4 is one year old. By the way, I am ashamed that I never made a unit test… this is easy. Also, when I wanted to translate in French the method "add: anObject times: aNumber" to a Collection, I found that it didn't work, so I created it with something like Collection>add: anObject times: aNumber^ anObject
What you should always make sure is that you can rebuild your system everyday. This will make your life easier. What kind of problems did you encounter? Stef
|
Re: in anybody out there? who uses 2.0?
|
|
In reply to this post by EstebanLM
On Sun, Mar 31, 2013 at 4:20 PM, stephane ducasse <[hidden email]> wrote:
I am developing with version 2.0, the fact that it works with 1.4 too is a kind of test for me.
This is definitely on my todo list ;-)
I had a "message not understood" but after what you wrote the problem seems to be solved since some time, I was just not aware of this
Apparently, some methods for morphs override older methods and when recompiling Pharo was blocked; I had to restart it at each time. I guess it will be OK when I take a new version with all the methods already ready. But I still did not have time for that (I work on another OS than my usual one) Alain
|
|
|
In reply to this post by EstebanLM
Hi, I know you must be starving for feedback on Pharo 2.0, so here is mine.
I ported my app to it, from Pharo 1.4. The app uses Glorp, Seaside, RFB and HPDF, so it took me an afternoon.
The overall impression is great. Unlike 1.4, it didn't crash a single time. Another great thing is that you didn't make that many backward incompatible changes as I though you would. Of course, I had to patch almost all of the above (their dependencies include FFI, OpenDBX etc.) and rewrite some little bits of my code, but it was worth it. The new FS classes are must better.
From user's point of view there is no much difference to 1.4, which is a good thing. I know you revamped things internally. Keep the good work, and I hope I will be able to provide you with feedback from real production environment soon. Milan Mimica http://sparklet.sf.net |
Re: in anybody out there? who uses 2.0?
|
|
Hi Milan,
On 07 Apr 2013, at 08:43, Milan Mimica <[hidden email]> wrote: > On 27 March 2013 15:24, Esteban Lorenzano <[hidden email]> wrote: > Hi, > > Here in Pharo headquarters we are shock that there are just 10 new bugs reported for 2.0 after the release... > So... I wonder... is that because we made a really cool release, or just because nobody is using it? > > I know you must be starving for feedback on Pharo 2.0, so here is mine. > > I ported my app to it, from Pharo 1.4. The app uses Glorp, Seaside, RFB and HPDF, so it took me an afternoon. > The overall impression is great. Unlike 1.4, it didn't crash a single time. Another great thing is that you didn't make that many backward incompatible changes as I though you would. Of course, I had to patch almost all of the above (their dependencies include FFI, OpenDBX etc.) and rewrite some little bits of my code, but it was worth it. The new FS classes are must better. > From user's point of view there is no much difference to 1.4, which is a good thing. I know you revamped things internally. > > Keep the good work, and I hope I will be able to provide you with feedback from real production environment soon. That is really great to hear. Thanks! Sounds like an impressive application that you got there. We value success stories and/or great use cases a lot, when you are ready. Sven > -- > Milan Mimica > http://sparklet.sf.net -- Sven Van Caekenberghe Proudly supporting Pharo http://pharo.org http://association.pharo.org http://consortium.pharo.org |
|
|
In reply to this post by mmimica
this really cool, thanks!
btw... if you needed to update configurations... could you commit the new configurations to the repository for Pharo2.0? thanks, Esteban On Apr 7, 2013, at 8:43 AM, Milan Mimica <[hidden email]> wrote:
|
Re: in anybody out there? who uses 2.0?
|
|
In reply to this post by mmimica
thanks for the feedback
We would like to migrate all the libraries we all use to SmalltalkHub and start to have a process so that we can validate the project. So if people want to help you are welcome. Stef On Apr 7, 2013, at 8:43 AM, Milan Mimica <[hidden email]> wrote:
|
|
|
In reply to this post by EstebanLM
The only configuration I changed was ConfigurationOfGlorpDBX, it was trivial, just to make it load. I added spec for: #'pharo2.x' version: '2.5'.
to ConfigurationOfGlorpDBX >> stable: Some tests failed, but AFAIR, they also failed on 1.4.
Other changes include fixing Seaside's WAImageStatus, which was also trivial but has been broken for ages. Not sure how to push these changes, especially because these are only for Pharo2.0.
On 7 April 2013 11:47, Esteban Lorenzano <[hidden email]> wrote:
Milan Mimica http://sparklet.sf.net |
Re: in anybody out there? who uses 2.0?
|
|
I think that you should publish the change for DBX in the DBX repo and the metaRepoForPharo20.
Stef
On Apr 7, 2013, at 4:47 PM, Milan Mimica <[hidden email]> wrote:
|
|
In reply to this post by Esteban A. Maringolo
Thanks for the link to the script Esteban. No errors was nice. I don't have the Seaside Control Panel. Is there something else I need to do to get that?
Thanks, Jeff |
«
Return to Pharo Smalltalk Users
|
1 view|%1 views
| Free forum by Nabble | Edit this page |