September 2013 - Pharo-dev - lists.pharo.org

Pharo Sprint in Buenos Aires
by Camillo Bruni Sept. 3, 2013

Sept. 3, 2013

I spend a month in Buenos Aires working together with Guido Chari @ the UBA, we take this opportunity to organize a Pharo sprint outside France ;) We are in early preparation phase and the date nor location isn't fixed yet, so if you are motivated to join, can mark possible dates on this doodle: http://doodle.com/qey7iieqm5yr4wwy cheers, camillo

2 1

Re: [Pharo-dev] Should Job use SystemAnnouncer?
by Stephan Eggermont Sept. 2, 2013

Sept. 2, 2013

Camillo wrote: >the main problem with the current progress bar morph design is that updates are forced whenever >the underlying model (here job) changes. This is *very* costly for high-frequency updates, which >is exactly the moment when you want to use the progress bar ;) > >It should work the other way around, the progress bar morph polls the underlying model with a given >frequency (~10/s) and then refreshes the UI if the model changed. I'm not in favor of polling. Most models are not thread-safe. The progress bar knows how much precision it has (1% or so) and so knows at which boundaries it needs to refresh. In addition the progress bar should show that something is happening a few times a second. For the first, it would be enough to only call self changed in ProgressBarMorph>value: when the result would be visible, i.e. when the completedWidth changes. ProgressBarMorph>value: aNumber |testWidth| value := aNumber. testWidth := self completedWidth. testWidth ~= cachedWidth ifTrue: [ cachedWidth := testWidth. self changed] and setting the cachedWidth to 0 on initialize. Hmm, where should I commit the 3.0 slice? Getting an error. Stephan

2 1

Smacc question
by StÃ©phane Ducasse Sept. 2, 2013

Sept. 2, 2013

Hi thierry I imported the generated parsers (a student did a while ago on VW using Smacc). The generated parser parse apparently the tests. Now when I regenerate the parser with Pharo I cannot get the Smacc compuling the parser. I get a SmaCCParserError: Token not expected do you have any idea what would not be supported in the Pharo/Squeak version # # Expression Parser #============================================= goal: expression* ; expression: relation ( ( and | and then | or | or else | xor ) relation )*; relation: simple_expression sndPart?; sndPart: relational_operator simple_expression | ( not )? ( in range | in type_mark); simple_expression: (unary_adding_operator)? term (binary_adding_operator term)*; term: factor (multiplying_operator factor)*; factor: primary ( ** primary)? | abs primary | not primary; #<identifier> instead of name primary: <numeric_lit> | null | <string_lit> | <number> | <identifier> | <identifier> '' <identifier> | function_call | ( expression ) ; #simplified binary_adding_operator: + | - | & ; unary_adding_operator: + | - ; multiplying_operator: * | / | mod | rem ; relational_operator: = | /= | < | <= | > | >= ; #============================================================================= function_call: <identifier> ''identifier'' function_arguments ''args'' { ADFunctionCall withElements: args withName: identifier } ; function_arguments: ( expression ( , expression)* ) ; #==== Declarations =========================================================== discrete_range: range | subtype_indication; #discrete_? subtype_indication: type_mark (constraint)? ; type_mark: type; constraint: range_constraint; range_constraint: range range; range: range_attribute | (simple_expression .. simple_expression); range_attribute: <identifier> '' range ; #cf. 4.1.4 #==== Names ================================================================== name: simple_name | <char_lit> | indexed_component | slice | selected_component | attribute; #operator_symbol prefix: name | function_call; simple_name: <identifier>; indexed_component: prefix ( expression ( , expression )* ) ; slice: prefix ( discrete_range ) ; selected_component: prefix . selector; selector: simple_name | <char_lit> | all ; #operator_symbol attribute: prefix '' attribute_designator; attribute_designator: simple_name ( ( <universal_static_expression> ) )? ; #==== Types: appendix c: package STANDARD plus more ========================================================= type: BOOLEAN | REAL | INTEGER | FLOAT | CHARACTER | DURATION | STRING | RAINBOW ; '

2 3

Suggestion about Spec
by Erwan Douaille Sept. 2, 2013

Sept. 2, 2013

Hi, I'm working on the UIPainter importer. I'm able to import already existing spec windows but now i need to retrieve widgets properties. For example, on a ButtonModel i would like to retrieve the Icon. Actually if i want to retrieve it it means than i have to send #iconHolder message. The thing is, to set it we used #icon: Why not uniformize it ? For me, #iconHolder should be #icon. It will be more easier for me to retrieve informations, and for users to use it. -- Best regards, Douaille Erwan <douaille.erwan(a)gmail.com>

2 1

Re: [Pharo-dev] Should Job use SystemAnnouncer?
by Stephan Eggermont Sept. 2, 2013

Sept. 2, 2013

Stef wrote: >did you notice a different between SystemAnnouncer and replacing it by another one? Yes, it made a significant difference. It did not remove enough overhead though. Job>value: was called over 100K times, doing that only every 100 increments helped enough. Job>>jobAnnouncer ^jobAnnouncer ifNil: [ jobAnnouncer := Announcer new ] Job>announce: anAnnouncementClass | announcement | announcement := anAnnouncementClass on: self. self announcer announce: announcement. Job>announcer ^Job jobAnnouncer SystemProgressMorph>>enable Job jobAnnouncer on: JobStart send: #startJob: to: self. Job jobAnnouncer on: JobEnd send: #endJob: to: self. Job jobAnnouncer on: JobChange send: #updateJob: to: self. SystemProgressMorph>>disable Job jobAnnouncer unsubscribe: self. Morphic-StephanEggermont.1375.mcz Kernel-StephanEggermont.1351.mcz

2 1

"self foo: (var := â¦)" vs "var := â¦. self foo: var"
by Alexandre Bergel Sept. 2, 2013

Sept. 2, 2013

Hi! I am facing a strange situation: The following expression raises an error: -=-=-=-=-=-=-=-=-= rawView add: (helpLabel := (ROElement on: 'Click here!') + ROLabel). -=-=-=-=-=-=-=-=-= Apparently, what is sent as argument to #add: is an integer. I have to update my code with : -=-=-=-=-=-=-=-=-= (helpLabel := (ROElement on: 'Click here!') + ROLabel). rawView add: helpLabel. -=-=-=-=-=-=-=-=-= To not have an error. This is really strange! By the way, this looks like the issue with Metacello? Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

6 6

[update 3.0] #30363
by Marcus Denker Sept. 2, 2013

Sept. 2, 2013

30363 ----- 11524 MessageBrowser new protocol method https://pharo.fogbugz.com/f/cases/11524 11523 SystemNavigation small refactoring https://pharo.fogbugz.com/f/cases/11523 11498 Be able to set an initial filter in ListDialogWindow https://pharo.fogbugz.com/f/cases/11498 11522 NewList separator support https://pharo.fogbugz.com/f/cases/11522 Diff information: http://smalltalkhub.com/mc/Pharo/Pharo30/main/Tools-MarcusDenker.1233.diff http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Tools-MarcusDenker.133.d… http://smalltalkhub.com/mc/Pharo/Pharo30/main/Polymorph-Widgets-MarcusDenke… http://smalltalkhub.com/mc/Pharo/Pharo30/main/NewList-MarcusDenker.56.diff

1 0

Aida 6.6 OneClicks for Pharo now ready
by Janko MivÅ¡ek Sept. 2, 2013

Sept. 2, 2013

Dear Pharoers, OneClicks for Mac, Win and Linux are now prepared from three Pharo 2.0 distros. I checked Linux and Win, but for Mac I'm not sure if I prepared distribution correctly. Can someone check it? All three OneClicks are available at http://www.aidaweb.si/download . Best regards Janko -- Janko MivÅ¡ek Aida/Web Smalltalk Web Application Server http://www.aidaweb.si

2 5

WhatsUp from: 2013-09-02 until: 2013-09-15
by seaside＠rmod.lille.inria.fr Sept. 2, 2013

Sept. 2, 2013

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: - $HEROIC_ACHIEVEMENTS_OR_DISMAL_FAILURES_OR_SIMPLE_BORING_NECESSARY_TASKS ### What's next, until 2013-09-15 (*): - $NEXT_STEPS_TOWARDS_WORLD_DOMINATION (*) we'll be expecting results by then ;)

3 2

[regression reporter]regression occurred
by no-reply＠ci.inria.fr Sept. 2, 2013

Sept. 2, 2013

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2-Validation/label=win/… 1 regressions found. Zinc.Zodiac.ZnHTTPSTests.testGetPharoVersion

1 0