March 2016 - Pharo-dev - lists.pharo.org

Some bugs I'm experiencing in latest 50
by stepharo March 13, 2016

March 13, 2016

- I got nautilus blocked and irresponsive (but I could close it). - When I saved my image the vm crashed but the image was fully saved. Stef Image ----- /Users/ducasse/Library/Application Support/Pharo/images/OPAX/OPAX.image Pharo5.0 Latest update: #50636 Unnamed Virtual Machine --------------- /Applications/Pharo/SpurPharo.app/Contents/MacOS/Pharo CoInterpreter VMMaker.oscog-eem.1664 uuid: 9e676eea-f638-4da0-bd70-920a096bda28 Jan 23 2016 StackToRegisterMappingCogit VMMaker.oscog-eem.1664 uuid: 9e676eea-f638-4da0-bd70-920a096bda28 Jan 23 2016 https://github.com/estebanlm/pharo-vm.git Commit: 8872c82ba7e8a56d06a95fb473501bde1b6ec807 Date: 2016-01-23 03:23:40 +0100 By: Esteban Lorenzano <estebanlm(a)gmail.com> Jenkins build #533 Mac Cocoa Cog 5.8b12 21-Sep-10 >1B0534FA-246C-47C5-AB29-7A76C81CCDCB< VMMaker versionString https://github.com/estebanlm/pharo-vm.git Commit: 8872c82ba7e8a56d06a95fb473501bde1b6ec807 Date: 2016-01-23 03:23:40 +0100 By: Esteban Lorenzano <estebanlm(a)gmail.com> Jenkins build #533 CoInterpreter VMMaker.oscog-eem.1664 uuid: 9e676eea-f638-4da0-bd70-920a096bda28 Jan 23 2016 StackToRegisterMappingCogit VMMaker.oscog-eem.1664 uuid: 9e676eea-f638-4da0-bd70-920a096bda28 Jan 23 2016

3 10

Bug in code formatter
by stepharo March 13, 2016

March 13, 2016

Hi {'film'. 'FILM'. 'Film'} is formatted as {'film'. 'FILM'. 'Film'. } https://pharo.fogbugz.com/f/cases/17811/film-FILM-is-formatted-as-film-FILM Stef

1 1

Giving lectures all week at Prague...
by stepharo March 13, 2016

March 13, 2016

And coding there and preparing next ESUG :) Stef

1 0

Comparison of actual and expected Delay length on small durations
by Aliaksei Syrel March 13, 2016

March 13, 2016

Hi At some point I got a feeling that actual Delay time is longer than expected. It is especially visible on small delays less than 100ms (otherwise difference is < 1%). [image: Inline image 1] Documentation says that *Delay waits approximately* for specified amount of time. However, according to experimental data an actual difference is always positive with the value around 1.5-2 seconds. It is actually strange where such a constant difference, independent from delay length, comes from. [image: Inline image 2] To conclude, if you want to wait more precisely you need to create delays with duration 1.5 seconds less than expected. Also it would be interesting to test on different hardwares to make sure that difference stays constant (around 1.5-2s) I used the following script to gather data and save in .csv file http://ws.stfx.eu/DSCLG2ON0VBI Tested on: Pharo: #50636 Spur-VM: #463 OS: Mac OS Yosemite CPU: Intel Core i7 2.39 GHz Cheers, Alex

5 9

problems with the default font
by Tudor Girba March 12, 2016

March 12, 2016

Hi, It looks to me as if in the latest Pharo, the default font is rendered with the DejaVu bitmap font instead of Source Sans Pro. Here is how the font looks like in Nautilus (look at the fonts in the top part): When I look in Settings, it says that it uses Source Sans Pro. However, if I reload the fonts and set it again to Source Sans Pro, I get this picture with the correct rendering (look at the category name âannouncing"): Interestingly, the code font looks the same in both situations. Can you confirm this effect? Cheers, Doru -- www.tudorgirba.com www.feenk.com "From an abstract enough point of view, any two things are similar."

3 6

Doc tutorial about uFFI
by stepharo March 12, 2016

March 12, 2016

Hi esteban Where did you publish the doc/tutorial of uFFI? Stef

2 1

Improving doc of XMLParser
by stepharo March 12, 2016

March 12, 2016

Hi monty Yesterday one student ask me to import in Moose an XML document. I spent more than 15 min trying.... What would have really help is the following (could you add it into the class comment): Many of the comment do not help because are scarce and not precise enough. To use a SaxHandler subclass it SAXHandler subclass: #MySAXHandler instanceVariableNames: '' classVariableNames: '' category: 'MyXMLImporter' and redefine the method corresponding the notification you want for example: MySAXHandler>>startElement: aQualifiedName attributes: aDictionary self halt. Then to use it simply do | h | h := MySAXHandler new on: FileSystem workingDirectory / 'FILMS.XML'. h parseDocument. ^ h Your method startElement: aQualifiedName attributes: aDictionary should get invoked. I really think that having a tutorial on XML parser ***use*** is cruelly needed. Stef

3 4

Fwd: How to access XML tag name?
by stepharo March 11, 2016

March 11, 2016

Hi Yesterday I started to hack a smart SAX handler. The idea is that I want to just specify the tags I want to visit and the SAX handler should invoke (generated) visit methods. Like that I can easily get visitors on XML domain. Here is an example of what I did. | h | h := SmartSAXHandler new visitor: (MyFilmVisitor new visitTags: #(FILM ROLE)); on: FileSystem workingDirectory / 'FILMS.XML'. h parseDocument. ^ h SmartSAXHAndler >> startElement: aQualifiedName attributes: aDictionary (visitor shouldVisit: aQualifiedName) ifTrue: [ visitor perform: (visitor createdVisitSelector: aQualifiedName) with: aQualifiedName with: aDictionary ] Object subclass: #GenericTagSAXVisitor instanceVariableNames: 'visitTag visitTags' classVariableNames: '' category: 'SmartXMLHandler' GenericTagSAXVisitor>>visitTags: aCollection "set the tags that will lead to a call to a visitTag:with: method in the visitor" visitTags := aCollection collect: [ :each | each asLowercase ]. self createVisitMethods. and in a subclass the visit* methods automatically generated Now I could not get when I have a <TITRE>Vertigo</TITRE> where I can get the Vertigo information. I redefined several methods of SAXhandler but without success. <?xml version="1.0" encoding="iso-8859-1"?> <FILMS> <FILM annee="1958"> <TITRE>Vertigo</TITRE> <GENRE>Drame</GENRE> <PAYS>USA</PAYS> <MES idref="3"/> <ROLES> <ROLE> <PRENOM>James</PRENOM> <NOM>Stewart</NOM> <INTITULE>John Ferguson</INTITULE> </ROLE> <ROLE> <PRENOM>Kim</PRENOM> <NOM>Novak</NOM> <INTITULE>Madeleine Elster</INTITULE> </ROLE> </ROLES> <RESUME>Scottie Ferguson, ancien inspecteur de police, est sujet au vertige depuis qu'il a vu mourir son collegue. Elster, son ami, le charge de surveiller sa femme, Madeleine, ayant des tendances suicidaires. Amoureux de la jeune femme Scottie ne remarque pas le piege qui se trame autour de lui et dont il va etre la victime... </RESUME> </FILM>

1 0

New Cog VMs available
by Eliot Miranda March 11, 2016

March 11, 2016

... at http://www.mirandabanda.org/files/Cog/VM/VM.r3643 CogVM binaries as per VMMaker.oscog-eem.1722/r3643 General: Change vm parameter 65 to be a set of flags, not merely the boolean for MULTIPLE_BYTECODE_SETS, and make the flags answer MULTIPLE_BYTECODE_SETS & IMMUTABILITY. Nuke unused named prim primitiveOriginalMillisecondClock so VM can eliminate ioOldMSecs. Nuke ioOldMSecs and ioLowResMSecs to simplify the tiume management and avoid mistakes were we use different versions of clocks for different parts of the same basic io operations. Nuke SqueakUIFlushUseHighPercisionClock preference. Make the leak checker operational in the non-assert (& non-debug) VM. Cogit: Revise the register saving convention for trampoline calls. Instead of passing in a boolean to indicate if all registers should be saved or not (an aproach Ryan correctly notes is bogus), pass in a register mask of registers to be saved. If the trampoline call has a machine code continuation then the argument is usually callerSavedRegMask, and on MIPS and ARM this is empty. Use the above to move the saving of registers in the store-check into the store-check trampoline and out of the call of the trampoline. Further, use genJumpImmediate: in the immediate test, which saves an instruction. 64-bit Cogit: Fix primtive 171 (primitiveImmediateAsInteger) in the Cogit for SmallFloat. The old code only dealt with Characters. The new code replicates the rotatedFloatBitsOf: scheme in Spur64BitMemoryManager. Spur: Resolve the conflict between 32-bit and 64-bit tag assignments. In 32-bits we have 1=even SmallIntegers, 2=Characters, 3=odd SmallIntegers, and in 64-bits we had 1=SmallIntegers, 2=Characters, 3=SmallFloats. Hence we would want SmallFloat64's identityHash to be 3, which conflicts with 32 bits' odd SmallIntegers. Change is for 64-bits to use 1=SmallIntegers, 2=Characters, 4=SmallFloats. This also means single-bit tests in the Cogit, which produces better code, and no scratch registers to hold masked tags. Hence roll the 64-bit Spur image format version number from 68019 to 68021. Delegate to the object memories to determine the image format version number. Fix the bug in markAndTraceClassOf: that causes the first class table page to become a visible array. Plugins: Fix primitiveFFIIntegerAt for signed 4-byte results. Wuth the support for 64-bit results, signed 4-byte results must be sign-extended to 64-bits. Update primitiveFileRead for Spur. It doesn't need the PrimErrObjectMayMove handling since Spur has pinning. Don't use cCode: pointer aliasing for storing/fetching float/double in FFI primitives. Replace with a call to mem:cp:y: - the first one is dangerous with modern C compilers - while the second is optimized away (no call to memcpy) Fix a missing guard in SmartSyntaxPlugin like explained at http://forum.world.st/VM-safety-missing-failing-guards-in-SmartSyntaxPlugin… Fix two missing guards in VM profiling primitives: one should not use the result of stackObjectValue: before checking for success, otherwise a null pointer exception will occur in case of failure. _,,,^..^,,,_ best, Eliot

1 0

[pharo-project/pharo-core] cd0546: 50636
by GitHub March 11, 2016

March 11, 2016

Branch: refs/heads/5.0 Home: https://github.com/pharo-project/pharo-core Commit: cd0546cf067bf729331ded6385f3f1bb58df8a30 https://github.com/pharo-project/pharo-core/commit/cd0546cf067bf729331ded63… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2016-03-11 (Fri, 11 Mar 2016) Changed paths: M DebuggerActions.package/DebugAction.class/instance/actions/execute.st A DebuggerActions.package/DebugAction.class/instance/actions/notifyDebuggingActionAboutToExecute.st A DebuggerActions.package/DebugAction.class/instance/actions/preAction.st A DebuggerActions.package/DebugActionAboutToExecute.class/README.md A DebuggerActions.package/DebugActionAboutToExecute.class/class/as yet unclassified/forDebuggingAction_.st A DebuggerActions.package/DebugActionAboutToExecute.class/definition.st A DebuggerActions.package/DebugActionAboutToExecute.class/instance/accessing/action.st A DebuggerActions.package/DebugActionAboutToExecute.class/instance/accessing/action_.st M Kernel.package/Context.class/instance/private/isFailToken_.st M Kernel.package/Context.class/instance/private/isPrimFailToken_.st M MonticelloFileTree-Core.package/extension/MCFileRepositoryInspector/instance/refresh.st R ScriptLoader50.package/ScriptLoader.class/instance/pharo - scripts/script50635.st A ScriptLoader50.package/ScriptLoader.class/instance/pharo - scripts/script50636.st R ScriptLoader50.package/ScriptLoader.class/instance/pharo - updates/update50635.st A ScriptLoader50.package/ScriptLoader.class/instance/pharo - updates/update50636.st M ScriptLoader50.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st A Text-Core.package/TextClassLink.class/instance/comparing/=.st A Text-Core.package/TextClassLink.class/instance/comparing/hash.st A Text-Core.package/TextLink.class/instance/comparing/=.st A Text-Core.package/TextLink.class/instance/comparing/hash.st M Text-Core.package/TextMethodLink.class/definition.st A Text-Core.package/TextMethodLink.class/instance/accessing/browseSenders.st A Text-Core.package/TextMethodLink.class/instance/comparing/=.st A Text-Core.package/TextMethodLink.class/instance/comparing/hash.st A Text-Core.package/TextVariableLink.class/instance/as yet unclassified/=.st A Text-Core.package/TextVariableLink.class/instance/as yet unclassified/hash.st Log Message: ----------- 50636 17504 TextLink + subclasses lacks implementation of #= and #hash https://pharo.fogbugz.com/f/cases/17504 17796 Debugger step over basicNew of proxies is not working https://pharo.fogbugz.com/f/cases/17796 17769 The debugger should trigger an event before executing a debugging action https://pharo.fogbugz.com/f/cases/17769 17655 Changes in package loading behavior https://pharo.fogbugz.com/f/cases/17655 http://files.pharo.org/image/50/50636.zip

1 0