- Pharo-dev - lists.pharo.org

Re: [Pharo-dev] ConfigurationOf and Git

by Ben Coman

On Thu, Oct 22, 2015 at 1:30 AM, Dale Henrichs <dale.henrichs(a)gemtalksystems.com> wrote: > > > On 10/21/2015 05:55 AM, Ben Coman wrote: > > > > On Wed, Oct 21, 2015 at 2:49 AM, Dale Henrichs > <dale.henrichs(a)gemtalksystems.com> wrote: >> >> Here's the documentation for the Metacello github:// repository >> description[1]. ... >> >> In recent versions of Metacello I have made it possible for you to use >> pattern matching in the <version identifier> to provide for >> symbolic-version-like facility for github references. >> >> Instead of stableForPharo4 you would associate a semantic version with the >> code that is "stableForPharo4" for example `4.0.0` and use the tag `v4.0.0` >> to mark the commit that is "stableForPharo4". >> >> Similarly you'd use the tag `v5.0.0` to mark the commit that is >> "stableForPharo5". >> >> Then you'd use the following in your baseline: method (not in a >> symbolicVersion method). >> >> spec for: #'pharo4.0.x' >> do: [spec >> baseline: 'Project' >> with: [ spec repository: 'github://username/Project:v4.?/']]. >> spec for: #'pharo5.0.x' >> do: [spec >> baseline: 'Project' >> with: [ spec repository: 'github://username/Project:v5.?/']]. >> >> If you end up with a new commit that patches a Pharo4 problem, you'd tag >> that commit as `v4.0.1` and so on .... The above `v4.?` pattern will match >> `v4.0.1` and you'll pick up that tag the next time you refresh your build >> (i.e., do a `get` on the Project baseline ... which causes a new download >> from Github) ... > > > Intuitively I would have thought that #'pharo4.0.x' > would correspond to repository: 'github://username/Project:v4.0.?/' > I seem to be missing something. > > I guess it's not intuitive to me, because I'm not sure what point you are > making:) My point was that these two 4.0.x 4.0.? seem to correspond better than these two 4.0.x 4.? but maybe I looked at this the wrong way. So v4.? would also match pharo4.1.0 ? cheers -ben > > In general terms I'm not very intuitive, so this is my personality flaw:) > > Dale

Oct. 22, 2015

Re: [Pharo-dev] binary serialization

by stepharo

> > > Which kind of migration example you have in mind that would not be > > supported? An example would help. > > Well, my pics will demonstrate. I am interested in doing more than > mappping ivars or a class rename. I want to do a total substitution, > then a further substitution on the receiving, import side: > > Vat1: anObject (Class A) ---> On wire: desc (Descriptor) ---> Vat2: > aProxy (Class FarERef) > > A desc is substituted for a PassByProxy object, then a FarERef is > substituted for the desc. > >> >> #fuelAccept: is a serialization side method. >> >> If Fuel supports substitution on serialization, I don't understand >> why no substitution support on materialization. >> >> >> There was a reason, which I cannot remember completely. Maybe Martin or >> Max can remember. > > It seems your focus was pickling to disk then back. My focus is > distributed proxy graphs, which has different use cases. I'm interesting in knowing more :) Especially Vat and others. Stef

Oct. 22, 2015

Re: [Pharo-dev] How to get Pharo into Ubuntu Software center

by Peter UhnÃ¡k

I don't use Ubuntu, but I do use Debian and I see at least these problems 1) Can the AptUrl ( https://wiki.ubuntu.com/AptUrl ) add a PPA repository? >From the link it seems like it can only enable/refresh them? 2) This works out of the box only for Firefox (so you still need instructions for Chrome, but maybe Ubuntu users don't use chrome :)) 3) You need to specify 32bit architecture I have a feeling that the thrid step must be done by user, but maybe there could be 64bit package that would execute script that would add the package and then it would have 32bit package as dep? Peter On Thu, Oct 22, 2015 at 9:32 AM, Torsten Bergmann <astares(a)gmx.de> wrote: > Hi, > > in Ubuntu one can click on a link with apt:// protocol > to easily install a piece of software right from a webpage: > > <a class="apt" href="apt://chromium-browser">Install Now</a> > > It just opens a link to the Software center with the software preselected. > > Unfortunately Pharo is not part of the Software center :( > > Can we improve this situation? What is necessary so that > Pharo (or ideally the Pharo launcher) gets directly added to the > Software center and we just need a link on the webpage? > > Just want to understand if it is a technical or time issue. > > Thanks > T. > >

Oct. 22, 2015

MethodClassifier pragmas?

by Stephan Eggermont

I've committed a version of MethodClassifier that can classify based on pragmas. I've only added <example>->#examples Which ones should be added? Stephan

Oct. 22, 2015

How to get Pharo into Ubuntu Software center

by Torsten Bergmann

Hi, in Ubuntu one can click on a link with apt:// protocol to easily install a piece of software right from a webpage: <a class="apt" href="apt://chromium-browser">Install Now</a> It just opens a link to the Software center with the software preselected. Unfortunately Pharo is not part of the Software center :( Can we improve this situation? What is necessary so that Pharo (or ideally the Pharo launcher) gets directly added to the Software center and we just need a link on the webpage? Just want to understand if it is a technical or time issue. Thanks T.

Oct. 22, 2015

Re: [Pharo-dev] Smalltalkhub projects not indexed?

by stepharo

> So the #! URL syntax was especially geared for sites that got the > fundamental web development best practices horribly wrong, and gave > them a lifeline to getting their content seen by Googlebot. > > And today, this emergency rescue package seems to be regarded as the > One True Way of web development by engineers from Facebook, Twitter, > and now Lifehacker. Makes me cry :( Le 19/10/15 16:27, Levente Uzonyi a Ã©crit : > This is because for some unknown reason the creators of Smalltalkhub > decided to use hashbang urls at a time when it was already considered > bad practice to do so[1]. Crawlers can't crawl such pages, unless > there are non-hashbang versions provided as well[2], but I strongly > doubt Smalltalkhub has anything like that. > > Levente > > [1] http://isolani.co.uk/blog/javascript/BreakingTheWebWithHashBangs > [2] > https://developers.google.com/webmasters/ajax-crawling/docs/learn-more > > On Mon, 19 Oct 2015, Jan KurÅ¡ wrote: > >> Hi, >> >> I noticed that my projects on Smalltalkhub cannot be googled :( I >> have project set up as a public, I actively develop and update the >> overview >> page for several months, but even the query: smalltalkhub <myname> >> <nameOfMyProject> does not find anything. >> >> Is there a way to help this? >> >> Cheers, >> Jan >> >>

Oct. 22, 2015

Re: [Pharo-dev] FastTable question

by Thierry Goubier

2015-10-22 8:54 GMT+02:00 Esteban Lorenzano <estebanlm(a)gmail.com>: > Hi, > > Sorry late response but yesterday I was not around. > > > > On 20 Oct 2015, at 18:42, stepharo <stepharo(a)free.fr> wrote: > > > > for the mooc videos I must use larger fonts. > > So I ended up hacking and changing the hardcoded number in default > > > > defaultRowHeight > > ^ 24 > > > > rowHeight > > "This is the row height your rows will have. Cells answered in > dataSource will be forced to have > > this height number... We force it instead allowing lists to have any > height because the logic to > > calculate rows becomes complicated. Possible, but complicated :)" > > > > ^ rowHeight ifNil: [ self class defaultRowHeight ] > > > > > > Esteban two questions: > > > > - when I read the code of FTmorph I was thinking that a lot of it has > not much to do with UI but looks like a model :). > > this is not a question :) > in the case of rowHeight, it determines how the row looks (and not what it > shows) then is UI (IMO)â¦ anyway Iâm aware this is not 100% MVC but knowing > the height of rows is very important to calculate in a fast way how many > rows fit and get the rows from datasource (an operation that happens a lot > so is convenient to have it as fast as possible (at least in a âfast > tableâ). > this do not supports variable height but that can be achieved by extending > the table without compromising the performance of the parent(just very few > use-cases need a variable height table). > Hum, from practical results playing with the FastTable core code: - With varying height, if you know the start index, then you can fill the container fast by streaming over the data source, building the row morphs and adding the heights, stopping when you reach the end of the container. - Current FastTable core code is very inefficient if #at: is slow on the data source - What is hard is getting the vertical scrollbar right. - Current FastTable core code does multiple passes on some of the row building / coordinates stuff for no clear reasons. > > > > > - if I were about to manage the height of a row based on its actual > contents where should I put it. > > as described, you need to extend the functionality, this is not supported > in current version (reason already explained). > For me, to support this you need to extend: > > - rowHeight should be rowHeightFor: index > - visibleRows and around need to take first visible row height, then > iterate one by one asking for the each row size. > - datasource needs to be extended with ârowHeight: indexâ > It is not the only way to do it. It at: is slow on the data source, then rowheight: will be slow. > > > FTTableContainerMorph is containing what is visible but I wonder why it > was asking this to the FastTable and to the elements displayed? > > explained above: speed needs. > > > > > Why the ContainerMorph does not ask the FTTableRowMorph for its height > and it could be computed based on the font. > > I still did not get where the strings built. > > I didnât thought about, thatâs why is not calculating font size :) > Alsoâ¦ I cannot know your table will contain textâ¦ thatâs a task for the > datasource, then (to answer appropriate row height)â¦ but it needs to be > fast! :) > Thierry > > Esteban > > > > > > > may be here > > > > updateExposedRows > > > > ... > > > > | cell | > > cell := (self owner dataSource > > cellColumn: (columns at: columnIndex) > > row: rowIndex). > > cell width: (columnWidths at: columnIndex). > > > > We should add more comments to methods. > > > > > > Stef > > > > > > > > drawOn: canvas > > | x y cellWidth cellHeight rowsToDisplay rowSubviews > highligtedRowIndexes primarySelectionIndex | > > > > self bounds ifNil: [ ^ self ]. "Nothing to show yet" > > self owner ifNil: [ ^ self ]. > > > > x := self left + self class rowLeftMargin. > > y := self top. > > cellWidth := self width - self class rowLeftMargin. > > cellHeight := self owner rowHeight. > > ^^^^^^^^^^^^^^^^^^^^^^^ > > > > highligtedRowIndexes := > > self owner selectedRowIndexes, > > self owner highlightedRowIndexes. > > primarySelectionIndex := self owner selectedRowIndex. > > > > "For some superweird reason, calling #calculateExposedRows here > instead in #changed (where > > it should be called) is 10x faster. Since the whole purpose of this > component is speed, for > > now I'm calling it here and adding the #setNeedRecalculateRows > mechanism. > > History, please forgive me." > > self updateAllRows. > > > > rowsToDisplay := self exposedRows. > > rowSubviews := OrderedCollection new: rowsToDisplay size + 1. > > headerRow ifNotNil: [ > > headerRow bounds: (x@y extent: cellWidth@cellHeight). > > y := y + cellHeight + self owner intercellSpacing. > > rowSubviews add: headerRow ]. > > > > rowsToDisplay keysAndValuesDo: [ :rowIndex :row | | visibleHeight | > > visibleHeight := cellHeight min: (self bottom - y). > > row bounds: (x@y extent: cellWidth@visibleHeight). > > y := y + visibleHeight + self owner intercellSpacing. > > rowSubviews add: ((highligtedRowIndexes includes: rowIndex) > > ifTrue: [ > > "IMPORTANT: I need to set owner to nil because otherwise > it will trigger an > > invalidation of the owner when adding morph to > selectionMorph, causing an > > infinite loop" > > self > > toSelectionRow: (row privateOwner: nil) > > primary: primarySelectionIndex = rowIndex ] > > ifFalse: [ row ]) ]. > > > > submorphs := rowSubviews asArray. > > super drawOn: canvas. > > needsRefreshExposedRows := false > > > > >

Oct. 22, 2015

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/50398 Home: https://github.com/pharo-project/pharo-core

Oct. 22, 2015

[pharo-project/pharo-core] 696b64: 50398

by GitHub

Branch: refs/heads/5.0 Home: https://github.com/pharo-project/pharo-core Commit: 696b64506d81e4e4d4cfbc6b5ec018847eac34d5 https://github.com/pharo-project/pharo-core/commit/696b64506d81e4e4d4cfbc6b… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2015-10-22 (Thu, 22 Oct 2015) Changed paths: M Polymorph-Widgets.package/MorphicUIManager.class/instance/debug/debugProcess_context_title_full_.st M Reflectivity.package/HookGenerator.class/instance/private/hookFor_.st A Reflectivity.package/RFASTTranslator.class/instance/reflectivity/visitLiteralVariableNode_.st A Reflectivity.package/RFLiteralVariableNode.class/README.md A Reflectivity.package/RFLiteralVariableNode.class/class/instance creation/value_.st A Reflectivity.package/RFLiteralVariableNode.class/definition.st A Reflectivity.package/RFLiteralVariableNode.class/instance/accessing/binding.st A Reflectivity.package/RFLiteralVariableNode.class/instance/accessing/binding_.st A Reflectivity.package/RFLiteralVariableNode.class/instance/accessing/start.st A Reflectivity.package/RFLiteralVariableNode.class/instance/accessing/stop.st A Reflectivity.package/RFLiteralVariableNode.class/instance/testing/isFaulty.st A Reflectivity.package/RFLiteralVariableNode.class/instance/visiting/acceptVisitor_.st A Reflectivity.package/extension/TRBProgramNodeVisitor/instance/visitLiteralVariableNode_.st R ScriptLoader50.package/ScriptLoader.class/instance/pharo - scripts/script50397.st A ScriptLoader50.package/ScriptLoader.class/instance/pharo - scripts/script50398.st R ScriptLoader50.package/ScriptLoader.class/instance/pharo - updates/update50397.st A ScriptLoader50.package/ScriptLoader.class/instance/pharo - updates/update50398.st M ScriptLoader50.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st Log Message: ----------- 50398 16808 inspect decompiled method with breakpoint https://pharo.fogbugz.com/f/cases/16808 16771 Morphic "debug drawing error" broken https://pharo.fogbugz.com/f/cases/16771 http://files.pharo.org/image/50/50398.zip

Oct. 22, 2015

Re: [Pharo-dev] Smalltalkhub projects not indexed?

by stepharo

sadly enough... Le 19/10/15 16:27, Levente Uzonyi a Ã©crit : > This is because for some unknown reason the creators of Smalltalkhub > decided to use hashbang urls at a time when it was already considered > bad practice to do so[1]. Crawlers can't crawl such pages, unless > there are non-hashbang versions provided as well[2], but I strongly > doubt Smalltalkhub has anything like that. > > Levente > > [1] http://isolani.co.uk/blog/javascript/BreakingTheWebWithHashBangs > [2] > https://developers.google.com/webmasters/ajax-crawling/docs/learn-more > > On Mon, 19 Oct 2015, Jan KurÅ¡ wrote: > >> Hi, >> >> I noticed that my projects on Smalltalkhub cannot be googled :( I >> have project set up as a public, I actively develop and update the >> overview >> page for several months, but even the query: smalltalkhub <myname> >> <nameOfMyProject> does not find anything. >> >> Is there a way to help this? >> >> Cheers, >> Jan >> >>

Oct. 22, 2015