June 2008 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] Who is working on which tests? Why are there new, broken tests?

by Oscar Nierstrasz

Hi Stef, I loaded a fresh pharo-dev image and updated everything. Then the testHex tests failed, so I fixed them. Can you please tell me which implementors of hex you find in your image? I only find Character>>hex and Float>>hex. I had to implement Integer>>hex myself. - on On Jun 10, 2008, at 6:38 PM, StÃ©phane Ducasse wrote: > testHex > self assert: $a hex '61' = '61'. > self assert: Character space hex = '20' > > is green on my image > > as well as this one > > testHex > self assert: 0 hex = '0'. > self assert: 12 hex = 'C'. > self assert: 1234 hex = '4D2'. > > They are red in yours? > > Stef > > > On Jun 10, 2008, at 4:00 PM, Oscar Nierstrasz wrote: > >> >> Well, I thought we would start to fix the existing broken tests >> before adding new broken tests. >> >> I would also be very happy to see some comments when someone adds a >> new broken test, so I know what the plan is. >> >> Now I added an implementation of Integer>>hex, for example, but I >> am not so sure why we need this. >> Ain't too many senders. An easier fix would have been to remove the >> broken test. >> >> Cheers, >> - on >> >> On Jun 10, 2008, at 15:53, Bill Schwab wrote: >> >>> Is your concern that the new tests are broken (which seems normal >>> to me >>> that they would start out that way), or are you objecting to their >>> being >>> included in the image? >>> >>> I can see arguments either way on the latter. If code in the >>> image is >>> wrong and a clearly correct but currently failing test highlights >>> it, >>> I'd rather be alerted to it. You are also correct that having them >>> included draws attention that could easily encourage duplication of >>> effort. >>> >>> Comments? >>> >>> Bill >>> >>> >>> >>> >>> Wilhelm K. Schwab, Ph.D. >>> University of Florida >>> Department of Anesthesiology >>> PO Box 100254 >>> Gainesville, FL 32610-0254 >>> >>> Email: bschwab(a)anest.ufl.edu >>> Tel: (352) 846-1285 >>> FAX: (352) 392-7029 >>> >>>>>> oscar(a)iam.unibe.ch 06/10/08 8:31 AM >>> >>> >>> I looked at the wiki page >>> http://code.google.com/p/pharo/wiki/CurrentlyWorkingOn >>> but it does not say much. >>> >>> I would like to work through some tests, but I cannot find an >>> overview. >>> >>> The tests CharacterTest>>testHex and IntegerTest>>testHex are broken >>> but seem to be very new. Stef, are you working on these? You >>> seem to >>> have written the tests ... >>> >>> - on >>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

June 10, 2008

Re: [Pharo-project] Who is working on which tests? Why are there new, broken tests?

by Oscar Nierstrasz

Hi Bill, That's not what I mean. I agree fully with you that it is good to add tests that show in which way the system is broken. Adding tests for methods that don't even exist doesn't make much sense to me right now. testHex is (was) testing something that doesn't exist. OK if you are doing test-driven development. Not good if you want a system to be coherent and understandable. At the very least such tests need comments to explain why they have been added. Tests that are fragile and meaningless is also bad. The testHash tests that test that a hash returns a specific value are not good. I prefer to get rid of them. On Jun 10, 2008, at 5:34 PM, Bill Schwab wrote: > But if the test is revealing a true flaw in the system, better to keep > it than to remove it and pat ourselves on the back about "all tests > passing." > > Bill > > > > > Wilhelm K. Schwab, Ph.D. > University of Florida > Department of Anesthesiology > PO Box 100254 > Gainesville, FL 32610-0254 > > Email: bschwab(a)anest.ufl.edu > Tel: (352) 846-1285 > FAX: (352) 392-7029 > >>>> oscar(a)iam.unibe.ch 06/10/08 10:00 AM >>> > > Well, I thought we would start to fix the existing broken tests before > adding new broken tests. > > I would also be very happy to see some comments when someone adds a > new broken test, so I know what the plan is. > > Now I added an implementation of Integer>>hex, for example, but I am > not so sure why we need this. > Ain't too many senders. An easier fix would have been to remove the > broken test. > > Cheers, > - on > > On Jun 10, 2008, at 15:53, Bill Schwab wrote: > >> Is your concern that the new tests are broken (which seems normal to >> me >> that they would start out that way), or are you objecting to their >> being >> included in the image? >> >> I can see arguments either way on the latter. If code in the image >> is >> wrong and a clearly correct but currently failing test highlights it, >> I'd rather be alerted to it. You are also correct that having them >> included draws attention that could easily encourage duplication of >> effort. >> >> Comments? >> >> Bill >> >> >> >> >> Wilhelm K. Schwab, Ph.D. >> University of Florida >> Department of Anesthesiology >> PO Box 100254 >> Gainesville, FL 32610-0254 >> >> Email: bschwab(a)anest.ufl.edu >> Tel: (352) 846-1285 >> FAX: (352) 392-7029 >> >>>>> oscar(a)iam.unibe.ch 06/10/08 8:31 AM >>> >> >> I looked at the wiki page >> http://code.google.com/p/pharo/wiki/CurrentlyWorkingOn >> but it does not say much. >> >> I would like to work through some tests, but I cannot find an >> overview. >> >> The tests CharacterTest>>testHex and IntegerTest>>testHex are broken >> but seem to be very new. Stef, are you working on these? You seem >> to >> have written the tests ... >> >> - on >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

June 10, 2008

Re: [Pharo-project] Who is working on which tests? Why are there new, broken tests?

by StÃ©phane Ducasse

testHex self assert: $a hex '61' = '61'. self assert: Character space hex = '20' is green on my image as well as this one testHex self assert: 0 hex = '0'. self assert: 12 hex = 'C'. self assert: 1234 hex = '4D2'. They are red in yours? Stef On Jun 10, 2008, at 4:00 PM, Oscar Nierstrasz wrote: > > Well, I thought we would start to fix the existing broken tests > before adding new broken tests. > > I would also be very happy to see some comments when someone adds a > new broken test, so I know what the plan is. > > Now I added an implementation of Integer>>hex, for example, but I am > not so sure why we need this. > Ain't too many senders. An easier fix would have been to remove the > broken test. > > Cheers, > - on > > On Jun 10, 2008, at 15:53, Bill Schwab wrote: > >> Is your concern that the new tests are broken (which seems normal >> to me >> that they would start out that way), or are you objecting to their >> being >> included in the image? >> >> I can see arguments either way on the latter. If code in the image >> is >> wrong and a clearly correct but currently failing test highlights it, >> I'd rather be alerted to it. You are also correct that having them >> included draws attention that could easily encourage duplication of >> effort. >> >> Comments? >> >> Bill >> >> >> >> >> Wilhelm K. Schwab, Ph.D. >> University of Florida >> Department of Anesthesiology >> PO Box 100254 >> Gainesville, FL 32610-0254 >> >> Email: bschwab(a)anest.ufl.edu >> Tel: (352) 846-1285 >> FAX: (352) 392-7029 >> >>>>> oscar(a)iam.unibe.ch 06/10/08 8:31 AM >>> >> >> I looked at the wiki page >> http://code.google.com/p/pharo/wiki/CurrentlyWorkingOn >> but it does not say much. >> >> I would like to work through some tests, but I cannot find an >> overview. >> >> The tests CharacterTest>>testHex and IntegerTest>>testHex are broken >> but seem to be very new. Stef, are you working on these? You seem >> to >> have written the tests ... >> >> - on >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

June 10, 2008

Re: [Pharo-project] Why does Installer package stay dirty?

by StÃ©phane Ducasse

for the installer I had some problem at one point so damien hardcoded the version we load. Stef On Jun 10, 2008, at 4:20 PM, Oscar Nierstrasz wrote: > > When I update a pharo-dev image, I see that Installer still has > updates. > > However I cannot update this consistently. > > If I load Installer-sd.4.mcz then Installer-dc.4.mcz is > highlighted. One of these two should be deleted, I think. > Which one? > > - on > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

June 10, 2008

Re: [Pharo-project] Who is working on which tests? Why are there new, broken tests?

by StÃ©phane Ducasse

I do not remember what we did with lukas about hex I know that we want hex but I forgot. :( Stef On Jun 10, 2008, at 2:31 PM, Oscar Nierstrasz wrote: > > I looked at the wiki page http://code.google.com/p/pharo/wiki/CurrentlyWorkingOn > but it does not say much. > > I would like to work through some tests, but I cannot find an > overview. > > The tests CharacterTest>>testHex and IntegerTest>>testHex are broken > but seem to be very new. Stef, are you working on these? You seem > to have written the tests ... I think that these tests are from serge or harvested from a mantis entry. > > > - on > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

June 10, 2008

Re: [Pharo-project] KernelTests green

by StÃ©phane Ducasse

We checked with lukas what is strange is that the same tests are green in 3.10. Stef On Jun 10, 2008, at 3:48 PM, Oscar Nierstrasz wrote: > > I have posted some patches to the inbox to make the Kernel Number > tests run, and in the Issues page I propose to delete a bunch of > nonsensical testHash tests. With these changes the Kernel tests > should all be green. > > - on > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

June 10, 2008

Re: [Pharo-project] self halt in RxParserTest>>testMatchesInwW

by StÃ©phane Ducasse

Yes go ahead. Stef On Jun 10, 2008, at 2:06 PM, Oscar Nierstrasz wrote: > > Stef, > > RxParserTest>>testMatchesInwW > > has a self halt in it dating from 2006, apparently added by you. > > I guess you should remove it ... The test runs fine otherwise. > > - on > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

June 10, 2008

Re: [Pharo-project] Who is working on which tests? Why are there new, broken tests?

by Bill Schwab

But if the test is revealing a true flaw in the system, better to keep it than to remove it and pat ourselves on the back about "all tests passing." Bill Wilhelm K. Schwab, Ph.D. University of Florida Department of Anesthesiology PO Box 100254 Gainesville, FL 32610-0254 Email: bschwab(a)anest.ufl.edu Tel: (352) 846-1285 FAX: (352) 392-7029 >>> oscar(a)iam.unibe.ch 06/10/08 10:00 AM >>> Well, I thought we would start to fix the existing broken tests before adding new broken tests. I would also be very happy to see some comments when someone adds a new broken test, so I know what the plan is. Now I added an implementation of Integer>>hex, for example, but I am not so sure why we need this. Ain't too many senders. An easier fix would have been to remove the broken test. Cheers, - on On Jun 10, 2008, at 15:53, Bill Schwab wrote: > Is your concern that the new tests are broken (which seems normal to > me > that they would start out that way), or are you objecting to their > being > included in the image? > > I can see arguments either way on the latter. If code in the image is > wrong and a clearly correct but currently failing test highlights it, > I'd rather be alerted to it. You are also correct that having them > included draws attention that could easily encourage duplication of > effort. > > Comments? > > Bill > > > > > Wilhelm K. Schwab, Ph.D. > University of Florida > Department of Anesthesiology > PO Box 100254 > Gainesville, FL 32610-0254 > > Email: bschwab(a)anest.ufl.edu > Tel: (352) 846-1285 > FAX: (352) 392-7029 > >>>> oscar(a)iam.unibe.ch 06/10/08 8:31 AM >>> > > I looked at the wiki page > http://code.google.com/p/pharo/wiki/CurrentlyWorkingOn > but it does not say much. > > I would like to work through some tests, but I cannot find an > overview. > > The tests CharacterTest>>testHex and IntegerTest>>testHex are broken > but seem to be very new. Stef, are you working on these? You seem to > have written the tests ... > > - on > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list Pharo-project(a)lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

June 10, 2008

[Pharo-project] Why does Installer package stay dirty?

by Oscar Nierstrasz

When I update a pharo-dev image, I see that Installer still has updates. However I cannot update this consistently. If I load Installer-sd.4.mcz then Installer-dc.4.mcz is highlighted. One of these two should be deleted, I think. Which one? - on

June 10, 2008

Re: [Pharo-project] Who is working on which tests? Why are there new, broken tests?

by Oscar Nierstrasz

Well, I thought we would start to fix the existing broken tests before adding new broken tests. I would also be very happy to see some comments when someone adds a new broken test, so I know what the plan is. Now I added an implementation of Integer>>hex, for example, but I am not so sure why we need this. Ain't too many senders. An easier fix would have been to remove the broken test. Cheers, - on On Jun 10, 2008, at 15:53, Bill Schwab wrote: > Is your concern that the new tests are broken (which seems normal to > me > that they would start out that way), or are you objecting to their > being > included in the image? > > I can see arguments either way on the latter. If code in the image is > wrong and a clearly correct but currently failing test highlights it, > I'd rather be alerted to it. You are also correct that having them > included draws attention that could easily encourage duplication of > effort. > > Comments? > > Bill > > > > > Wilhelm K. Schwab, Ph.D. > University of Florida > Department of Anesthesiology > PO Box 100254 > Gainesville, FL 32610-0254 > > Email: bschwab(a)anest.ufl.edu > Tel: (352) 846-1285 > FAX: (352) 392-7029 > >>>> oscar(a)iam.unibe.ch 06/10/08 8:31 AM >>> > > I looked at the wiki page > http://code.google.com/p/pharo/wiki/CurrentlyWorkingOn > but it does not say much. > > I would like to work through some tests, but I cannot find an > overview. > > The tests CharacterTest>>testHex and IntegerTest>>testHex are broken > but seem to be very new. Stef, are you working on these? You seem to > have written the tests ... > > - on > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

June 10, 2008