- Pharo-dev - lists.pharo.org

Re: [Pharo-project] [update 2.0] #20046

by Marcus Denker

On May 4, 2012, at 1:30 PM, Sean P. DeNigris wrote: > > Marcus Denker-4 wrote >> >> Issue 5795: Merge System-BreakPoints with Tools-Debugger >> http://code.google.com/p/pharo/issues/detail?id=5795 >> > > Does this, by chance, take care of the problem where you add a breakpoint in > the Senders browser, and then it is not recognized by the Debugger? > No, I just moved the class around... Marcus -- Marcus Denker -- http://marcusdenker.de

May 4, 2012

Re: [Pharo-project] [update 2.0] #20046

by Sean P. DeNigris

Sean P. DeNigris wrote > > ... then it is not recognized by the Debugger? > I meant browser... probably OB... don't know about the debugger... -- View this message in context: http://forum.world.st/update-2-0-20046-tp4608340p4608412.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

May 4, 2012

Re: [Pharo-project] [update 2.0] #20046

by Sean P. DeNigris

Marcus Denker-4 wrote > > Issue 5795: Merge System-BreakPoints with Tools-Debugger > http://code.google.com/p/pharo/issues/detail?id=5795 > Does this, by chance, take care of the problem where you add a breakpoint in the Senders browser, and then it is not recognized by the Debugger? Sean -- View this message in context: http://forum.world.st/update-2-0-20046-tp4608340p4608404.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

May 4, 2012

Re: [Pharo-project] [update 2.0] #20045

by Sean P. DeNigris

Marcus Denker-4 wrote > > Issue 5758: Improve CommandLine handling > http://code.google.com/p/pharo/issues/detail?id=5758 > Cool! This deserves some documentation... -- View this message in context: http://forum.world.st/update-2-0-20045-tp4608123p4608400.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

May 4, 2012

[Pharo-project] [update 2.0] #20047

by Marcus Denker

20047 ----- Issue 5782: Close ALL windows, no matter if they have unchanged changes http://code.google.com/p/pharo/issues/detail?id=5782 Issue 5792: Fix tempNamed and tempNamedPut http://code.google.com/p/pharo/issues/detail?id=5792 Issue 5790: Merge ExpandedSourceFileArray into SourceFileArray (part2 install and fix tests) http://code.google.com/p/pharo/issues/detail?id=5790 -- Marcus Denker -- http://marcusdenker.de

May 4, 2012

[Pharo-project] [update 2.0] #20046

by Marcus Denker

20046 ----- Issue 5789: Merge ExpandedSourceFileArray into SourceFileArray (part1 copy methods) http://code.google.com/p/pharo/issues/detail?id=5789 Issue 5804: Failing Test:testCommandLineHandlerCondition http://code.google.com/p/pharo/issues/detail?id=5804 Issue 5795: Merge System-BreakPoints with Tools-Debugger http://code.google.com/p/pharo/issues/detail?id=5795 -- Marcus Denker -- http://marcusdenker.de

May 4, 2012

Re: [Pharo-project] #getSource is broken for temp names

by Marcus Denker

On May 4, 2012, at 11:19 AM, Mariano Martinez Peck wrote: > > I don't know. In Fuel we have tests for each kind of trailer, included those which store temp names. > If this kind of trailer is not supported any more, I can remove the tests. > Yes! > If you want to remove temp names, then I think we still need to remove: > > #hasTempNames > #encodeTempsNamesQCompress > #encodeTempsNamesZip > #encodeUsingQCompress > #encodeUsingZip > #tempNames > #tempNames: > #decodeTempsNamesQCompress > #decodeTempsNamesZip > > remove the trailer type from #trailerKinds > Yes. One step at a time. Marcus -- Marcus Denker -- http://marcusdenker.de

May 4, 2012

Re: [Pharo-project] #getSource is broken for temp names

by Mariano Martinez Peck

On Fri, May 4, 2012 at 11:09 AM, Marcus Denker <marcus.denker(a)inria.fr>wrote: > > On May 4, 2012, at 11:03 AM, Mariano Martinez Peck wrote: > > > | trailer materializedCompiledMethod method | > > > > trailer := CompiledMethodTrailer new tempNames: 'param1 param2 temp1 > temp2'. > > self assert: trailer kind = #TempsNamesQCompress. > > self assert: trailer kindAsByte = 8. > > method := (self class >> #fooWith:and:) copyWithTrailerBytes: > trailer. > > self assert: (method getSource includesSubString: 'param1'). > > > > > > this last assert fails. > > > > Problem is that #getSource is broken and somehow it looks like if the > part of tempNames was removed. I propose to put back its original verison > or just add at the beginning: > > > > trailer tempNames ifNotNil: [:namesString | > > "Magic sources -- decompile with temp names" > > ^ ((class decompilerClass new withTempNames: namesString) > > decompile: selector in: class method: self) > > decompileString]. > > > > > > agree? if true I open an issue. > > No, the idea was to remove temp name embedding because it is not needed. > > Why do you need it? > > I don't know. In Fuel we have tests for each kind of trailer, included those which store temp names. If this kind of trailer is not supported any more, I can remove the tests. If you want to remove temp names, then I think we still need to remove: #hasTempNames #encodeTempsNamesQCompress #encodeTempsNamesZip #encodeUsingQCompress #encodeUsingZip #tempNames #tempNames: #decodeTempsNamesQCompress #decodeTempsNamesZip remove the trailer type from #trailerKinds and maybe more stuff. We should analyze this with care. Cheers > Marcus > > > -- > Marcus Denker -- http://marcusdenker.de > > > -- Mariano http://marianopeck.wordpress.com

May 4, 2012

Re: [Pharo-project] [update 2.0] #20043

by Marcus Denker

On May 4, 2012, at 11:04 AM, Mariano Martinez Peck wrote: > > > On Thu, May 3, 2012 at 10:44 AM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > 20043 > ----- > > Issue 5744: Clean up Temp Embedding 2): DoIts > http://code.google.com/p/pharo/issues/detail?id=5744 > > > - Doits are now compiled with source embedded > - embedding sources for doits pretty prints from AST as doits are compiled specially (with a return added) > > CompiledMethod: > > - remove blockExtendsToTempsMap: Not needed anymore, all calls go the MethodNode > - remove #copyWitTempsFromMethodNode: All senders use #copyWithSource > - remove #holdsTempNames. Temp name embedding not needed > > > I didn't understand. I have a test in fuel where I serialize/materialize a compiled method with an empty trailer. so after the materialization one of the things I test are: > > self assert: method sourcePointer = 0. > self deny: method holdsTempNames. > > of course I can still do: > > self deny: method trailer hasTempNames Just remove the test. The idea is to remove the implementation from the Trailer, too (later). Marcus -- Marcus Denker -- http://marcusdenker.de

May 4, 2012

Re: [Pharo-project] #getSource is broken for temp names

by Marcus Denker

On May 4, 2012, at 11:03 AM, Mariano Martinez Peck wrote: > | trailer materializedCompiledMethod method | > > trailer := CompiledMethodTrailer new tempNames: 'param1 param2 temp1 temp2'. > self assert: trailer kind = #TempsNamesQCompress. > self assert: trailer kindAsByte = 8. > method := (self class >> #fooWith:and:) copyWithTrailerBytes: trailer. > self assert: (method getSource includesSubString: 'param1'). > > > this last assert fails. > > Problem is that #getSource is broken and somehow it looks like if the part of tempNames was removed. I propose to put back its original verison or just add at the beginning: > > trailer tempNames ifNotNil: [:namesString | > "Magic sources -- decompile with temp names" > ^ ((class decompilerClass new withTempNames: namesString) > decompile: selector in: class method: self) > decompileString]. > > > agree? if true I open an issue. No, the idea was to remove temp name embedding because it is not needed. Why do you need it? Marcus -- Marcus Denker -- http://marcusdenker.de

May 4, 2012