- Pharo-dev - lists.pharo.org

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/40393 Home: https://github.com/pharo-project/pharo-core

Dec. 1, 2014

[pharo-project/pharo-core] 6f6c06: 40393

by GitHub

Branch: refs/heads/4.0 Home: https://github.com/pharo-project/pharo-core Commit: 6f6c06023fa6406371a1e66a33159f03ea6427a2 https://github.com/pharo-project/pharo-core/commit/6f6c06023fa6406371a1e66a… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2014-12-01 (Mon, 01 Dec 2014) Changed paths: M EmbeddedFreeType.package/EmbeddedFreeTypeFontFontDescription.class/class/accessing/fontContents.st M EmbeddedFreeType.package/EmbeddedFreeTypeFontFontDescription.class/class/accessing/installAllFontsIn_.st M EmbeddedFreeType.package/EmbeddedFreeTypeFontFontDescription.class/class/accessing/installFontsIn_.st M EmbeddedFreeType.package/EmbeddedFreeTypeFontFontDescription.class/class/accessing/originalFileName.st M EmbeddedFreeType.package/EmbeddedFreeTypeFontInstaller.class/class/accessing/current.st M EmbeddedFreeType.package/EmbeddedFreeTypeFontInstaller.class/class/extra fonts registration/registerFont_.st R EmbeddedFreeType.package/OpenSansRegular.class/README.md R EmbeddedFreeType.package/OpenSansRegular.class/class/accessing/fontContents.st R EmbeddedFreeType.package/OpenSansRegular.class/class/accessing/originalFileName.st R EmbeddedFreeType.package/OpenSansRegular.class/class/class initialization/initialize.st R EmbeddedFreeType.package/OpenSansRegular.class/definition.st M EmbeddedFreeType.package/SourceCodeFonts.class/class/accessing/defaultFontName.st A EmbeddedFreeType.package/SourceCodeFonts.class/class/accessing/setSourceCodeFontsDefault.st M EmbeddedFreeType.package/SourceCodeFonts.class/class/font registration/registerFonts_.st A EmbeddedFreeType.package/SourceSansProRegular.class/README.md A EmbeddedFreeType.package/SourceSansProRegular.class/class/accessing/fontContents.st A EmbeddedFreeType.package/SourceSansProRegular.class/class/accessing/originalFileName.st A EmbeddedFreeType.package/SourceSansProRegular.class/class/class initialization/initialize.st A EmbeddedFreeType.package/SourceSansProRegular.class/definition.st M FreeType.package/FreeTypeFontProvider.class/instance/loading and updating/updateEmbeddedFreeTypeFonts.st M FreeType.package/FreeTypeFontProvider.class/instance/loading and updating/updateFromSystem.st M FreeType.package/FreeTypeSettings.class/class/startup/startUp_.st A FreeType.package/FreeTypeSettings.class/class/startup/updateFreeType.st A ScriptLoader40.package/ScriptLoader.class/instance/pharo - scripts/script393.st A ScriptLoader40.package/ScriptLoader.class/instance/pharo - updates/update40393.st M ScriptLoader40.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st Log Message: ----------- 40393 14553 FreeType default fonts not correctly loaded when image initializes https://pharo.fogbugz.com/f/cases/14553 http://files.pharo.org/image/40/40393.zip

Dec. 1, 2014

Re: [Pharo-dev] WhatsUp from: 2014-12-01 until: 2014-12-14

by Ben Coman

seaside(a)rmod.lille.inria.fr wrote: > Hi! We're sending this automatic email twice a month, to give the community an opportunity to easily know what's happening and to coordinate efforts. Just answer informally, and feel free to spawn discussions thereafter! > > ### Here's what I've been up to since the last WhatsUp: > * Made my first contribution to the VM, adding <modifier> keypress events for OSX. * Refactored class variables and class-side methods of Delay to instance side of new DelayScheduler. > > ### What's next, until 2014-12-14 (*): > * Change Delays/DelayScheduler base from milliseconds to microseconds, to fix the commandline issue with "super fast delay" * Factor out the mutex interthread synchronisation code and provide some alternatives (e.g. Semaphore & Spinlock) > > (*) we'll be expecting results by then ;) > >

Dec. 1, 2014

14353 Delay refactoring (part 2) - change from milliseconds to microseconds

by Ben Coman

The change of Delay from milliseconds to microseconds is ready for human review. (https://pharo.fogbugz.com/default.asp?14353) With this change from milliseconds to microseconds, clock rollover is eliminated, but it seems Delays run at reduced performance. newCodeEnabled FALSE TRUE 128507 101227 delays scheduled per second 100% 79% relative performance 7.8 9.9 microseconds per scheduled delay 100% 127% relative performance I "guess" this is due to the move from 32bit millisecond immediate SmallIntegers to 64bit microsecond boxed LargePositiveIntegers. It would be interesting to check this on Spur64, but for now, is this performance penalty a reasonable trade for eliminating delay clock rollover? The Delay class comment warns of the critical performance its high priority event loop, and on paper a 21% drop in performance seems significant, but actually the main Morph interCyclePause is 50 milliseconds. So... (9.9 - 7.8) / 50000 = 0.004% difference in performance. (Did I get that right?) Comments? P.S. DelayBenchmark results... =============== DelayBenchmark withNewCodeEnabled: true... #ConcurrentDelays: 1000 . Count: 39018 per second #ConcurrentDelays: 2000 . Count: 71906 per second #ConcurrentDelays: 3000 . Count: 98010 per second #ConcurrentDelays: 4000 . Count: 101227 per second #ConcurrentDelays: 5000 . Count: 64728 per second #ConcurrentDelays: 6000 . Count: 90495 per second #ConcurrentDelays: 7000 . Count: 89220 per second #ConcurrentDelays: 8000 . Count: 89585 per second #ConcurrentDelays: 9000 . Count: 88393 per second #ConcurrentDelays: 10000 . Count: 89073 per second DelayBenchmark withNewCodeEnabled: false... #ConcurrentDelays: 1000 . Count: 39498 per second #ConcurrentDelays: 2000 . Count: 77808 per second #ConcurrentDelays: 3000 . Count: 99889 per second #ConcurrentDelays: 4000 . Count: 117632 per second #ConcurrentDelays: 5000 . Count: 128507 per second #ConcurrentDelays: 6000 . Count: 118729 per second #ConcurrentDelays: 7000 . Count: 114437 per second #ConcurrentDelays: 8000 . Count: 114792 per second #ConcurrentDelays: 9000 . Count: 115137 per second #ConcurrentDelays: 10000 . Count: 113973 per second

Dec. 1, 2014

Re: [Pharo-dev] [squeak-dev] IdentitySet>>collect:

by Eliot Miranda

On Fri, Nov 28, 2014 at 7:42 PM, Ben Coman <btc(a)openinworld.com> wrote: > Levente Uzonyi wrote: > >> The problem with changing #species is that it's overused. There are at >> least three different things it's used for in the Collection hierarchy: >> - the class of the object returned by #collect: (Set's subclasses are the >> exception since 2000 or so, except for WeakSet, but that's really strange, >> and should revisited) >> - the class of the object returned by #select: and friends >> - the class used for comparison of collections >> > > So should there be #collectSpecies ? > No. That's what species was invented for in the first place. > Which defaults to "^self species" and is overidden as required. > > In the case of Sets, you start with an "unordered collection without > duplicates", but since #collect can introduce duplicates, you end up with > "an unordered collection with duplicates" --> Bag (per Dave's suggestion) > > So I think its reasonable that the following two snippets should produce > the same result. > > oc := OrderedCollection newFrom: #( 1 2 3 4 5 ). > (oc collect: [ :e | e even ]) occurrencesOf: true. "--> 2" > > s := Set newFrom: #( 1 2 3 4 5 ). > (s collect: [ :e | e even ]) occurrencesOf: true. "--> 2" > > > which you get with the following change... > > Set>>collectSpecies > ^Bag > > Set>>collect: aBlock > | result | > result := self collectSpecies new: self size. > array do: [:each | each ifNotNil: [result add: (aBlock value: each > enclosedSetElement)]]. > ^ result > > > > >> HashedCollection >> #select: uses #species. Changing IdentitySet >> >>> #species to return a Set would break #select:, because a ~~ b doesn't >>>> >>> imply a ~= b. IdentitySet is the optimal class for the return value of >> IdentitySet >> #select:, because it's guaranteed to be able to hold all >> selected values. I think the same applies to all other kind of sets (and >> collections, because #select: is simply reducing the number of elements). >> >> This is not the first time this discussion comes up. There was one in >> 2003[1], but the thread is hard to follow. And there was one in 2009[2]. >> I think the conclusion was that it's better to leave the method as it is, >> because there's no better way to do it. >> Sure, we could use #species in #collect:, but it would break quite a lot >> of stuff. For example: >> >> | k | >> k := KeyedSet keyBlock: [ :each | each first ]. >> k add: #[1]; add: #[2]. >> k collect: [ :each | each size ] >> > > > That works with the above change to produce --> a Bag(1 1) > > >> Currently this snippet returns "a Set(1)". Using #species in #collect: >> (via #copyEmpty) one would get an error, because SmallInteger does not >> understand #first. Changing the #species of KeyedSet would break #select:. >> >> So IMHO when you #collect: from a set, you should always use >> #collect:as:. Here is a (partial) list of methods which send #collect: to a >> Set: >> >> Categorizer >> #changeFromCategorySpecs: >> ClassFactoryForTestCase >> #createdClassNames >> Dictionary >> #unreferencedKeys >> MCDependencySorterTest >> #assertItems:orderAs:withRequired:toLoad: >> extraProvisions: >> MessageNode >> #analyseTempsWithin:rootNode:assignmentPools: >> MethodNode >> #referencedValuesWithinBlockExtent: >> SetTest >> #testCollect >> SetWithNilTest >> #runSetWithNilTestOf: >> >> There are probably many more, but we should fix all of them. A static >> analyzer could help a lot. >> >> Levente >> >> [1] http://lists.squeakfoundation.org/pipermail/squeak-dev/2003- >> February/052659.html >> [2] http://lists.squeakfoundation.org/pipermail/squeak-dev/2009- >> November/141016.html >> >> >> >> On Wed, 26 Nov 2014, David T. Lewis wrote: >> >> It seems to me that Object>>species is intended to handle this sort of >>> issue. >>> >>> For IdentitySet, it answers what Eliot was expecting: >>> >>> IdentitySet new species ==> IdentitySet >>> Set new species ==> Set >>> >>> However, IdentitySet>>collect: does not make use of this, and it answers >>> a Set for the reasons that Levente explained. >>> >>> Answering an Array or and OrderedCollection would not really make sense, >>> because sets are unordered collections (but Bag might be better). >>> >> > Bag is better. > > >>> Shouldn't we have an implementation of IdentitySet>>species that answers >>> Set (or Bag), with a method comment explaining why this is the case, and >>> with all of the collection methods using #species to answer the right >>> kind of result? >>> >>> I note that IdentitySet>>collect: answers a Set, but IdentitySet>select: >>> sends #species and therefore answers an IdentitySet. >>> >>> So I think that if we want the #species of an IdentitySet to be a Set, >>> then we should make it so, and give it a method comment to explain the >>> rationale. And the #collect: and #select: methods should both answer a >>> result of that #species. >>> >> > I think the species of #collect: and #select: are intrinsically different. > #collect: is a transformation. > > >>> Dave >>> >>> >>> On Thu, Nov 27, 2014 at 01:14:30AM +0100, Levente Uzonyi wrote: >>> >>>> Your example hides the problem of ordering - what Tobias is asking >>>> about - >>>> so here's another: >>>> >>>> (IdentitySet withAll: #(1 1.0)) collect: [ :each | each class ] >>>> >>>> If IdentitySet >> #collect: were returning an Array, then what would be >>>> the >>>> answer? >>>> >>>> { SmallInteger. Float } or { Float. SmallInteger } ? >>>> >>> > --> a Bag(SmallInteger Float) > > >>>> If you really want to have the resulting collection have the same size, >>>> but avoid the problem with ordering, then what you really need is a Bag. >>>> >>> > To me that makes the most sense. > > >>>> On Thu, 27 Nov 2014, Frank Lesser wrote: >>>> >>>> Hi Tobias, >>>>> agree, a problem of "OrderedCollection" >>>>> not to break a lot of other things we could return an Array. >>>>> but for me collecting has priority. >>>>> Frank >>>>> >>>>> -----Urspr?ngliche Nachricht----- >>>>> Von: squeak-dev-bounces(a)lists.squeakfoundation.org >>>>> [mailto:squeak-dev-bounces@lists.squeakfoundation.org] Im Auftrag von >>>>> Tobias >>>>> Pape >>>>> Gesendet: Donnerstag, 27. November 2014 00:48 >>>>> An: The general-purpose Squeak developers list >>>>> Betreff: Re: [squeak-dev] IdentitySet>>collect: >>>>> >>>>> >>>>> On 27.11.2014, at 00:34, Frank Lesser <frank-lesser@lesser-software. >>>>> com> >>>>> wrote: >>>>> >>>>> hmm, not convinced >>>>>> >>>>>> (IdentitySet withAll: #(1 1.0)) collect: [:e| e asInteger ] >>>>>> OrderedCollection(1 1 ) >>>>>> >>>>>> in LSWVST ( one-to-one ), you collect results of evaluating a block on >>>>>> objects. >>>>>> >>>>>> Frank >>>>>> maybe I am wrong ... >>>>>> >>>>> >>>>> Where would the order come from for that _Ordered_Collection? >>>>> >>>> > An unordered OrderedCollection --> Bag > > cheers -ben > > -- best, Eliot

Dec. 1, 2014

Re: [Pharo-dev] [pharo-project/pharo-core] 8be654: 40392

by Ben Coman

> > Log Message: > ----------- > 40392 > > 14546 Failed inherited tests should show home class (in TestRunner, like the debugger) > https://pharo.fogbugz.com/f/cases/14546 > Thanks for this Nicolai. I only logged the idea yesterday, and here it is integrated overnight before I get home from work to review it. The yellow highlight in the attached picture shows the new feature. cheers -ben

Dec. 1, 2014

Re: [Pharo-dev] SmalltalkHub bugs & hangs :-(

by Andreas Wacknitz

Am 01.12.14 12:34, schrieb kilon alios: > No of course not. That would have been a deal braker. > > Personally I prefer using Sourcetree to browser history of commits > than using the pharo on board tools. Pharo is good on that respect, > but sourcetree is better Alas Sourcetree is only available for Windows and Mac OSX. Non-portable programs die with their system. Closed source programs die with their company. That's one of the reasons I like Pharo: it's open source and I was able to port the VM to my prefered OS (Ok, NativeBoost is still missing (not working) but the rest is doing fine). Furthermore I really enjoy Squeak's and Pharo ecosystem. The more tools and frameworks are inside the less they depend on the underlying OS. Andreas > and smartgit I have used in the past. It also has nice visualisation > for branches , merges etc. I have not shown these features in my video > tutorial as I was planning to show them to a separate tutorial. I > would not mind bypassing monticello completely since its usage is > unnecessary at least to me with all these git tools that one can > download and use for free. The same is true for other IDEs out there , > generally people prefer using dedicate tools. > > Usually dedicate tools are hard to beat this is why I believe > modularising Pharo is extremely important as will make integrating > external tools way easier. > > I think however that Roassal can be used to visualise in similar way > the history as Sourcetree does. So pharo is not far from it but as > always its a matter of someone doing the hard work. > > On Mon, Dec 1, 2014 at 12:21 PM, Hilaire <hilaire(a)drgeo.eu > <mailto:hilaire@drgeo.eu>> wrote: > > Le 27/11/2014 11:29, kilon alios a Ã©crit : > > My personal opinion on Github and Pharo is that it already works great > > with Pharo. The workflow with filetree is exactly the same as other > > languages that gives the added advantage that you can use all the > > powerful tools you use with other languages for commiting to git and > > github. I have not experienced any kind of issue of problem > using github > > with Pharo nor my experience has been any worse than other > languages. > > > > With git, don't you lost the capability to browse the history of > changes > in your code? > > Hilaire > > > -- > Dr. Geo - http://drgeo.eu > iStoa - http://istoa.drgeo.eu > > >

Dec. 1, 2014

Re: [Pharo-dev] [Moose-dev] Woden processor occupation and text highlight

by Yuriy Tymchuk

> On 01 Dec 2014, at 16:14, Ronie Salgado <roniesalg(a)gmail.com> wrote: > > It was the case in Roassal2 yes. This happens when the the window was constantly refreshing, > That is not a problem of Woden. It has to be constantly refreshing. > > But it is a problem for Woden-Roassal... > > I guess, that the problem could be worse if I enabled vsync in the SDL2 backend (it blocks). Ok, but can we make this work better? :) > > > 2014-12-01 12:09 GMT-03:00 Alexandre Bergel <alexandre.bergel(a)me.com <mailto:alexandre.bergel@me.com>>: > > Woden and Roassal3D have a problem that when the 3D window is shown, text highlighting is not working. Usually itâs not a big deal, but Rubric writes white text on white background :). I think a problem like that was already in Roassal2 and maybe someone remembers how to solve that? > > It was the case in Roassal2 yes. This happens when the the window was constantly refreshing, thus no time left for the process behind the text hilighting. Maybe there is the same problem. You can have a look at the TRMorph class. > > Cheers, > Alexandre >

Dec. 1, 2014

Re: [Pharo-dev] [Moose-dev] Woden processor occupation and text highlight

by Ronie Salgado

> > It was the case in Roassal2 yes. This happens when the the window was > constantly refreshing, That is not a problem of Woden. It has to be constantly refreshing. But it is a problem for Woden-Roassal... I guess, that the problem could be worse if I enabled vsync in the SDL2 backend (it blocks). 2014-12-01 12:09 GMT-03:00 Alexandre Bergel <alexandre.bergel(a)me.com>: > > Woden and Roassal3D have a problem that when the 3D window is shown, > text highlighting is not working. Usually itâs not a big deal, but Rubric > writes white text on white background :). I think a problem like that was > already in Roassal2 and maybe someone remembers how to solve that? > > It was the case in Roassal2 yes. This happens when the the window was > constantly refreshing, thus no time left for the process behind the text > hilighting. Maybe there is the same problem. You can have a look at the > TRMorph class. > > Cheers, > Alexandre >

Dec. 1, 2014

Re: [Pharo-dev] [Moose-dev] Woden processor occupation and text highlight

by Alexandre Bergel

> Woden and Roassal3D have a problem that when the 3D window is shown, text highlighting is not working. Usually itâs not a big deal, but Rubric writes white text on white background :). I think a problem like that was already in Roassal2 and maybe someone remembers how to solve that? It was the case in Roassal2 yes. This happens when the the window was constantly refreshing, thus no time left for the process behind the text hilighting. Maybe there is the same problem. You can have a look at the TRMorph class. Cheers, Alexandre

Dec. 1, 2014