February 2011 - Pharo-dev - lists.pharo.org

[Pharo-project] [COTDC] 20 - Loader
by laurent laffont Feb. 27, 2011

Feb. 27, 2011

Today: Loader Comment Of The Day Contest - One Day One Comment Rules: #1: Each day a not commented class is elected. Each day the best comment will be integrated with name of the author(s). #2: If you cannot comment it, deprecate it. Results: http://code.google.com/p/pharo/wiki/CommentOfTheDayContest Laurent

3 2

[Pharo-project] PharoCasts has a new skin
by laurent laffont Feb. 27, 2011

Feb. 27, 2011

Thanks to Claire & Patrick Barocca, PharoCasts now has its own skin and a nice logo ! http://www.pharocasts.com/ Laurent

1 0

Re: [Pharo-project] 1.2 one-click OSX and Linux: more failing tests
by laurent laffont Feb. 27, 2011

Feb. 27, 2011

1.2 one-click on (Arch) Linux (build #160) there's errors too (see below). So I wonder which VM is used in Hudson ? 10773 run, 10669 passes, 69 expected failures, 6 failures, 29 errors, 0 unexpected passes Failures: FileDirectoryTest>>#testRecursiveMatching ProcessSpecificTest>>#testDynamicVariable FileStreamTest>>#testPositionPastEndIsAtEnd MethodContextTest>>#testClosureRestart BecomeTest>>#testBecomeIdentityHash ProcessSpecificTest>>#testLocalVariable Errors: PharoCorePackageDependenciesTests>>#testMonticello SocketStreamTest>>#testNextIntoClose SocketStreamTest>>#testNextIntoCloseNonSignaling SocketStreamTest>>#testUnCategorizedMethods SocketStreamTest>>#testUnCategorizedMethods SocketStreamTest>>#testUpTo SocketStreamTest>>#testUpTo SocketStreamTest>>#testUpToAfterCloseNonSignaling SocketStreamTest>>#testUpToAfterCloseNonSignaling SocketStreamTest>>#testUpToAfterCloseSignaling SocketStreamTest>>#testUpToAfterCloseSignaling SocketStreamTest>>#testUpToAll SocketStreamTest>>#testUpToAll SocketStreamTest>>#testUpToAllAfterCloseNonSignaling SocketStreamTest>>#testUpToAllAfterCloseNonSignaling SocketStreamTest>>#testUpToAllAfterCloseSignaling SocketStreamTest>>#testUpToAllAfterCloseSignaling SocketStreamTest>>#testUpToAllLimit SocketStreamTest>>#testUpToAllLimit SocketStreamTest>>#testUpToAllTimeout SocketStreamTest>>#testUpToAllTimeout SocketStreamTest>>#testUpToEndClose SocketStreamTest>>#testUpToEndClose SocketStreamTest>>#testUpToEndCloseNonSignaling SocketStreamTest>>#testUpToEndCloseNonSignaling SocketStreamTest>>#testUpToMax SocketStreamTest>>#testUpToMax SocketStreamTest>>#testUpToTimeout SocketStreamTest>>#testUpToTimeout Laurent Laffont - @lolgzs <http://twitter.com/#!/lolgzs> Pharo Smalltalk Screencasts: http://www.pharocasts.com/ Blog: http://magaloma.blogspot.com/ On Sun, Feb 27, 2011 at 6:06 PM, laurent laffont <laurent.laffont(a)gmail.com>wrote: > I take the last 1.2 one-click on OSX (build #160), run all tests. > > => tests don't finish because > of SocketStreamTest>>#testUpToAfterCloseNonSignaling which wait infinitely > (however, if you run only SocketStreamTests it works). > > So I deselect SocketStreamTests and run again. > > Result: > > 10743 run, 10669 passes, 67 expected failures, 6 failures, 1 errors, 0 > unexpected passes > Failures: > FileStreamTest>>#testPositionPastEndIsAtEnd > MethodContextTest>>#testClosureRestart > ProcessSpecificTest>>#testDynamicVariable > FileDirectoryTest>>#testRecursiveMatching > ProcessSpecificTest>>#testLocalVariable > AllocationTest>>#testOneGigAllocation > > Errors: > PharoCorePackageDependenciesTests>>#testMonticello > > > > Laurent Laffont - @lolgzs <http://twitter.com/#!/lolgzs> > > Pharo Smalltalk Screencasts: http://www.pharocasts.com/ > Blog: http://magaloma.blogspot.com/ >

1 0

Re: [Pharo-project] we cannot add attachment anymore on the bug tracker
by Marcus Denker Feb. 27, 2011

Feb. 27, 2011

On Feb 25, 2011, at 8:33 PM, Tudor Girba wrote: > Isn't it because you went passed the quota? The bad thing about attachments in Google Code is that you have only 50MB for it. > > You can find the current status in Administer/Advanced. > Maybe this is the hint to move to something else... what I really want (need) is a bugtracker that allows better sorting of bug reports. Tags are great. But tags do not replace sub-projects. I want a bug tracker just for Cog. And one just for "the nuts", or one for morphic... This way we could actually get structure in the 560 reports... Redmine (http://www.redmine.org/) supports sub-projects. Or maybe https://www.chiliproject.org/ (a fork of redmine). Marcus -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

8 13

[Pharo-project] 1.2 one-click OSX: more failing tests
by laurent laffont Feb. 27, 2011

Feb. 27, 2011

I take the last 1.2 one-click on OSX (build #160), run all tests. => tests don't finish because of SocketStreamTest>>#testUpToAfterCloseNonSignaling which wait infinitely (however, if you run only SocketStreamTests it works). So I deselect SocketStreamTests and run again. Result: 10743 run, 10669 passes, 67 expected failures, 6 failures, 1 errors, 0 unexpected passes Failures: FileStreamTest>>#testPositionPastEndIsAtEnd MethodContextTest>>#testClosureRestart ProcessSpecificTest>>#testDynamicVariable FileDirectoryTest>>#testRecursiveMatching ProcessSpecificTest>>#testLocalVariable AllocationTest>>#testOneGigAllocation Errors: PharoCorePackageDependenciesTests>>#testMonticello Laurent Laffont - @lolgzs <http://twitter.com/#!/lolgzs> Pharo Smalltalk Screencasts: http://www.pharocasts.com/ Blog: http://magaloma.blogspot.com/

1 0

Re: [Pharo-project] Good reference on time on unit testing?
by laurent laffont Feb. 27, 2011

Feb. 27, 2011

On Sun, Feb 27, 2011 at 5:35 PM, Schwab,Wilhelm K <bschwab(a)anest.ufl.edu>wrote: > Laurent, > > *I* get it - my audience does not. > :) What I've learned: - people think it will take time to write tests because they don't know what is testing, they don't know what is TDD, because it's hard (yes, it's hard). - showing them your fully tested project, 100% test coverage, happy customer is not enough (but can help to bring confidence) - showing statistics rarely works (may be for your boss, but not developers) - buying them good books can help, but for example here in France it's hard: developers rarely read, even less when the book is technical, even less if it's for their work, even less if written in english (and only one book about TDD in French :) What have worked for me: - the best (by far) is pair programming: but hard to convince people to do pair-programming - except when you are the manager of your team. So you can pair-program with every member and do TDD with them. - regular internal coding-dojo (randori mode is better): easy acceptance. But that doesn't mean people will test when alone. So it have to be often (every week). - reading group (as described in Pragmatic Thinking and Learning - Andrew Hunt): take a book, a copy for each developer. Each week (or more often if possible) meet for 1 hour and do the summary of last read chapter, discuss, see how it can be applied. It seems that to spread an idea / practice, we have to help and do things together (see Your brain at work - David Rock) Hope it can help. Laurent. > > Bill > > > > > ________________________________________ > From: laurent laffont [laurent.laffont(a)gmail.com] > Sent: Sunday, February 27, 2011 10:19 AM > To: Pharo-project(a)lists.gforge.inria.fr > Cc: Schwab,Wilhelm K > Subject: Re: [Pharo-project] Good reference on time on unit testing? > > On Sun, Feb 27, 2011 at 2:12 PM, Schwab,Wilhelm K <bschwab(a)anest.ufl.edu > <mailto:bschwab@anest.ufl.edu>> wrote: > Understood. But, doing it correctly, how much of your time do you spend on > the test writing/maintenance side? > > None. Seriously none. I never pause and say "OK now it's time to write > tests and maintain existing ones". It's part of the TDD cycle: write a test, > make it green, refactor. write a test, make it green, refactor. The cycle > can go from 1mn to 10mn, rarely more. And if you use the Pomodoro > technique, you feel it even better. Test writing/maintenance is part of > programming. > > Sometimes you can say "OK before implementing this new story, I would like > to refactor here and here (tests + code) because I don't like it". We also > do this stuff on legacy code when tests are not here. But it's far easier > when we have good test coverage. > > > I found some blog posts, but I was hoping for some peer-reviewed papers > that go on the record. > > I am trying to make the point you are making, and it will go over better if > I can point to literature and some numbers that I suspect will be seen as > sobering. Tests ARE being "written after" in this case, but assuming they > change to a proper process, they still need to understand that tests take > thought and work. > > > When tests are written after, it's another kind of story. Then "how much > time we should spend writing test" is valid. I don't have any answer. > > TDD is not easy. Learning to write good tests takes time (years), > investment. If someone in your team have good experience with it, he should > pair program continuously with other programmers. Participating in > coding-dojo , code retreats with experienced developers is great too. > > Laurent. > > > > > > ________________________________________ > From: pharo-project-bounces(a)lists.gforge.inria.fr<mailto: > pharo-project-bounces(a)lists.gforge.inria.fr> [ > pharo-project-bounces(a)lists.gforge.inria.fr<mailto: > pharo-project-bounces(a)lists.gforge.inria.fr>] On Behalf Of Peter > Hugosson-Miller [oldmanlink(a)gmail.com<mailto:oldmanlink@gmail.com>] > Sent: Sunday, February 27, 2011 2:58 AM > To: Pharo-project(a)lists.gforge.inria.fr<mailto: > Pharo-project(a)lists.gforge.inria.fr> > Subject: Re: [Pharo-project] Good reference on time on unit testing? > > +100 > > Couldn't have said it better myself! > > -- > Cheers, > Peter. > > On 27 feb 2011, at 08:27, laurent laffont <laurent.laffont(a)gmail.com > <mailto:laurent.laffont@gmail.com><mailto:laurent.laffont@gmail.com > <mailto:laurent.laffont@gmail.com>>> wrote: > > On Sun, Feb 27, 2011 at 4:52 AM, Schwab,Wilhelm K <<mailto: > bschwab(a)anest.ufl.edu<mailto:bschwab@anest.ufl.edu>>bschwab(a)anest.ufl.edu > <mailto:bschwab@anest.ufl.edu><mailto:bschwab@anest.ufl.edu<mailto: > bschwab(a)anest.ufl.edu>>> wrote: > Hello all, > > Can anyone recommend a good reference on the amount of time one should > expect to spend writing tests? I will have to be the messenger (will be > wearing running shoes just in case...), but I want the message to come from > a solid source. > > If you develop the TDD way (which I always do at work) then I would say > this question is not valid. Your production code will be there because of > tests. In legacy system, new and refactored code will be there because of > tests. > > IMHO the questions "how much time should I spend writing tests?" or "how > many tests should I write ?" are raised when tests are written after. And > sadly tests often become a burden in this case and doesn't support the > pressure of releases / last minute changes / ... (I've done it this way > almost 10 years ago, not good :) > > In TDD the question is "which stories / scenarios do we have to implement > ?". It's more a Product owner / customer point of view. > > The best book on TDD I've read is Dave Astels TDD: a practical guide > (Indeed, this one has really changed my mind on how testing should be done, > and other books I've read since don't). > > Laurent > > > > Bill > > > > > > > >

1 0

[Pharo-project] Issue 3707 in pharo: Syntax highlighting looks strange
by pharo＠googlecode.com Feb. 27, 2011

Feb. 27, 2011

Status: Accepted Owner: marcus.d...(a)gmail.com Labels: Milestone-1.2-DevImage New issue 3707 by marcus.d...(a)gmail.com: Syntax highlighting looks strange http://code.google.com/p/pharo/issues/detail?id=3707 ... especially using BOLD for variables (both global and local), but arguments use different color... messages are colored (and the same as arguments). I attach old and new png All very strange. Especially the bold is horrible. The only bold thing in a method should be the method signature at the top. Attachments: old.png 20.9 KB new.png 25.5 KB

1 23

[Pharo-project] [update 1.1] #11421
by Marcus Denker Feb. 27, 2011

Feb. 27, 2011

11421 ----- Issue 3575: [Failing Tests] Obsolete classes after tests (workaround) -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

1 0

[Pharo-project] Issue 3484 in pharo: Windows VM in One click image for 1.2 not yet customized
by pharo＠googlecode.com Feb. 27, 2011

Feb. 27, 2011

Status: Accepted Owner: Torsten....(a)astares.de CC: marcus.denker, stephane.ducasse Labels: Milestone-1.2 New issue 3484 by Torsten....(a)astares.de: Windows VM in One click image for 1.2 not yet customized http://code.google.com/p/pharo/issues/detail?id=3484 Hi, the windows VM included in the PharoOneClick1.2.zip on the Hudson server is not customized for Pharo. It still has - the croquet icon - a bad *.ini file so that "Croquet" is displayed instead of Pharo when one quits with ALT+F4 - no resource section

1 7

Re: [Pharo-project] seaside.gemstone.com down?
by Marcus Denker Feb. 27, 2011

Feb. 27, 2011

On Feb 27, 2011, at 1:35 PM, Tobias Pape wrote: > Am 2011-02-27 um 13:16 schrieb Marcus Denker: >> Hi, >> >> I tried to load the glamour based tools: >> >> Gofer new squeaksource: 'glamoroust'; package: 'ConfigurationOfGlamoroust'; load. >> (Smalltalk at: #ConfigurationOfGlamoroust) perform: #loadDefault. >> >> And got this error: >> >> NameLookupFailure: Could not resolve the server named: seaside.gemstone.com > > > hmm, > $ host seaside.gemstone.com > seaside.gemstone.com has address 66.192.188.209 > > works fine for me, probably a DNS failure elsewhere? > Marcus-MacBook-Pro:~ denker$ host seaside.gemstone.com Host seaside.gemstone.com not found: 2(SERVFAIL) strange. -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

1 0