- Pharo-dev - lists.pharo.org

[Pharo-project] dirty package
by Pavel Krivanek July 29, 2012

July 29, 2012

Hi, in the image 20231 on Jenkins there is a dirty package KernelTests. -- Pavel

5 9

[Pharo-project] could it be possible to have a setting to get the old shortcut for
by StÃ©phane Ducasse July 29, 2012

July 29, 2012

implementor, sender, users? Because I spent 10 minutes and I still cannot find how I get all the references to a class :( I'm I know I'm old and stupid but we should not forget that what makes the success of pharo is happy users. Stef

4 8

[Pharo-project] markdown to latext and html
by StÃ©phane Ducasse July 29, 2012

July 29, 2012

Hi To support the publication in pdf and html directly, plus let people not to bother about latex, I would really to experiment the following for the next chapters of books. I do not know well markdown but I would like to know if the following makes sense. I would like to see if we can use markdown (plus some extensions for in text annotation, figure definition) as pivot language and to be able to generate latex and html from it. >From a simple analysis of current chapter here are the needs: structure title (with ref) section (with ref) subsection paragraph itemize/enumerate code snippet (with ref) figure with caption and ref inside a text area link code index comment for editor citation reference I probably miss some of them but this is the core and we can reduce the number from the latex stand point. What do you think? Does markdown support the item above. Is the petitMarkdown parser working? Because I would like to write a visitor that generates the latex we use for the book. I would like to use Zinc documentation as a test. Stef

5 11

[Pharo-project] Pharo in OpenShift
by garduino July 28, 2012

July 28, 2012

Hi People: I'm doing some efforts to make Pharo work in OpenShift (the cloud initiative of Red Hat) using the DIY (do it yourself) cartridge. I started trying with Swazoo, but looks like SwazooServer is ignoring the ip address specified and binding to all interfaces (0.0.0.0) (Was an advice of a RedHat engineer). Then tried with Zinc (thanks Sven by the off list help [1]) and seems to work now (Using #bindingAddress:). I need still to polish a ton of things (the code itself, that not always work ok reading the RedHat environment (I'm doing via OSProcess), also I need to discover some random issues opening the authorized port, and finally develop the automation scripts to start/stop/restart). But I'm happy because is proved that the cog vm 2540 + Pharo 1.4 with Zinc works ok in OpenShift, as you can see at: http://zinc-garduino.rhcloud.com/ When I have more progress I will comment and document, my goal is that Smalltalk get included as an option in OpenShift (as currently are php, python, java, ruby). We are not second class citizens :) I must also say a big thanks to several people of Red Hat because they have showed your interest in Pharo and helped me a lot. The interesting part of OpenShift is that in the Express version 1.5 GB RAM and 3 GB disk space are availables for free! Well, it's all for now, cheers! [1] Sven, the latest mcz published today of ZincHTTPComponents seems to have some issues with the newest method #bindingAddress: (I don't have time now, but will try to reproduce in the next days if you need). -- View this message in context: http://forum.world.st/Pharo-in-OpenShift-tp4560365p4560365.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

4 6

[Pharo-project] a test
by StÃ©phane Ducasse July 28, 2012

July 28, 2012

3 2

[Pharo-project] a little pearl from SystemProgressMorph
by Tudor Girba July 27, 2012

July 27, 2012

Just for fun read the below piece of code :) SystemProgressMorph>>label: shortDescription min: startMinValue max: startMaxValue "Answer the block that updates the progress bar." "some fun stuff added (by kph) - bar value: #label. - tell me my current label. - bar value: 'newLabel'. - enable changing the label from within the workBlock - bar value: #increment. - enable progress by one without keeping a counter - bar value: #decrement. - go backwards (if you really have to, useful for an abort, or rollback)! - bar value: newBigNum. - change the max on the fly - when you find there is more to do. - bar value: (bar value: #setMax) + 20 - change the max on the fly - when you find there is more/less to do. - bar value: (bar value: #setMin) - 20 - change the min on the fly - not sure why you would want to. " | slot range barSize lastRefresh maxValue minValue bar| maxValue := startMaxValue. minValue := startMinValue. ((range := maxValue - minValue) <= 0 or: [(slot := self nextSlotFor: shortDescription) = 0]) ifTrue: [^[:barVal| 0 ]]. self recenter. self openInWorld. barSize := -1. "Enforces a inital draw of the morph" lastRefresh := Time millisecondClockValue. bar := bars at: slot. ^[:barValArg | | barVal return newBarSize | barVal := barValArg. return := nil. bar := bars at: slot. "new fun stuff here" barVal == #current ifTrue: [ return := barSize ]. barVal == #label ifTrue:[ return := (labels at: slot) contents ]. barVal == #setMax ifTrue: [ return := maxValue. maxValue := minValue ]. barVal == #setMin ifTrue: [ return := minValue. minValue := maxValue ]. barVal == #increment ifTrue: [return := barVal := barSize + 1]. barVal == #decrement ifTrue: [ return := barVal := barSize - 1]. (barVal isString and: [barVal isSymbol not]) ifTrue: [ (labels at: slot) contents: barVal. barVal := #refresh]. barVal == #refresh ifTrue: [self currentWorld displayWorld. return := true]. (barVal == SmallInteger maxVal or: [ barVal == #finished ]) ifTrue: [return := slot]. return ifNil: [ barVal > maxValue ifTrue: [return := maxValue := barVal]. barVal < minValue ifTrue: [return := minValue := barVal]. (barVal between: minValue and: maxValue) ifTrue: [newBarSize := (barVal - minValue / range * BarWidth) truncated. newBarSize = barSize ifFalse: [barSize := newBarSize. (Time millisecondsSince: lastRefresh) > 25 ifTrue: [barVal := #refresh ]]]. barVal == #refresh ifTrue: [ ((bar valueOfProperty: #useStripes) ifNil: [false]) ifTrue: [bar fillStyle origin: bar position - ((Time millisecondClockValue // 50 \\ 30) @ 0)]. bar barSize: barSize. self currentWorld displayWorld. lastRefresh := Time millisecondClockValue]]. return] Cheers, Doru -- www.tudorgirba.com "Every thing has its own flow"

4 7

[Pharo-project] [update 2.0] #20235
by Esteban Lorenzano July 27, 2012

July 27, 2012

20235 ----- Issue 6467: Remove SystemChangeNotifier 2 - All events using SystemAnnouncer http://code.google.com/p/pharo/issues/detail?id=6467 (just loading a method needed, still not finished)

1 0

[Pharo-project] Strange behavior of Metacello in Pharo 2.0
by Alexandre Bergel July 27, 2012

July 27, 2012

Hi! I've just downloaded a configuration using monticello browser. In a workspace I am doing: -=-=-=-=-=-=-=-=-= ConfigurationOfVersionner project -=-=-=-=-=-=-=-=-= and apparently Pharo tries to get something from http://www.squeaksource/MetacelloRepository , which utterly fail. I am using the last Pharo from the forge. Is there something obvious I am missing? Am I the only one to experience this? Cheers, Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

5 7

[Pharo-project] Jenkins home folder almost full
by Sean P. DeNigris July 27, 2012

July 27, 2012

>From https://ci.lille.inria.fr/pharo/manage : Your Jenkins data directory "/net/www/ci/pharo/.jenkins" (AKA JENKINS_HOME) is almost full. You should act on it before it gets completely full. -- View this message in context: http://forum.world.st/Jenkins-home-folder-almost-full-tp4641876.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

1 0

[Pharo-project] Google Code labels/templates
by Sean P. DeNigris July 27, 2012

July 27, 2012

Someone had suggested to remove the "Pharo 1.4 Enhancement" template, since we're not accepting any. That makes sense. Also, the Nautilus template labels the issue "Nautilus", but the label in the list is "Type-Nautilus". Which brings me to my next question... The other Types are Bug, Enhancement, Cleanup, Feature [1]. Nautilus and Jenkins [2] really don't fit into this framework. Maybe because there will be many fewer issues, we don't need the level of granularity of [1] for [2]. But when we get close to a release, it'd be nice to know that there is 1 Nautilus bug and 10 Enhancements pending (i.e. probably only one critical thing) vs. 11 Type-Nautilus. So, now that I said all that, it seems the most powerful/flexible approach would be to leave the "nautilus" tag, remove the leading "Type-" from "Type-[Jenkins|Spec|Seed]", and then to make templates for Jenkins/Spec/Seed like the one for Nautilus. What do you think? Sean -- View this message in context: http://forum.world.st/Google-Code-labels-templates-tp4636767.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

3 5