January 2014 - Pharo-dev - lists.pharo.org

Issue 12601 Can't split Packages

by Sean P. DeNigris

This one really sucks. I have two projects that I can't load nowâ¦ I have ActiveTutor with tags ActiveTutor-Core and ActiveTutor-Screencasts, but the old "delete the working copy, then create MC packages for the sub-categories" doesn't seem to work in 3.0. I deleted the working copy, but when I add the MC packages for ActiveTutor-Core and ActiveTutor-Screencasts, they are empty :( Gofer new smalltalkhubUser: 'SeanDeNigris' project: 'SeansPlayground'; package: 'ActiveTutor'; disablePackageCache; load. Gofer new smalltalkhubUser: 'PharoExtras' project: 'Spotlight2'; package: 'Spotlight2'; disablePackageCache; load. ----- Cheers, Sean -- View this message in context: http://forum.world.st/Issue-12601-Can-t-split-Packages-tp4735288.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

Jan. 8, 2014

Re: [Pharo-dev] Do you want to accept???????? Discard???????

by Nicolai Hess

2014/1/8 Sean P. DeNigris <sean(a)clipperadams.com> > Pharo3.0 > Latest update: #30678 > > This seems reproducible: > 1. In Nautilus, browse a class that has a class comment > 2. Bring up the comment pane > 3. Edit the comment > 4. Click the "Class side" checkbox > > Also (don't know if it's related): > 1. In Nautilus, browse a class that has a class comment > 2. Bring up the comment pane > 3. Edit the comment > 4. Click the "Comment" button > 5. Choose discard > 6. Click the "Comment" button again *** It seems as if the changes were > accepted instead of discarded, butâ¦ > 7. Click the "Class side" checkbox *** Now the comment is reverted to the > original, so it seems the changes were indeed discarded. I guess the > problem > is keeping the comment pane in sync with the modelâ¦ > > Ben, should I open issues for these? If so, separately or as one report? > > > > ----- > Cheers, > Sean > -- > View this message in context: > http://forum.world.st/Do-you-want-to-accept-Discard-tp4735220p4735221.html > Sent from the Pharo Smalltalk Developers mailing list archive at > Nabble.com. > > The same happens on editing code, without comment pane. Select a method, change some code. Click the "Class-side" button. The Question dialog appears. I don't know exaclty why it opens another one if you click any buttons, but maybe it is related to activ focus. If you drag the Question window, the dialog buttons are working!

Jan. 8, 2014

CI server account creation

by Alfredo Sanzo

Hello! I am trying to create an account for the CI server and I have bumped into several links and posts to this list and now that I've asked for help I've finally understood the process (I think). My question is... Can we have this written somewhere? (and corrected, if I'm wrong :) ) "Steps to create a new user: a) Sign up for a user here: https://ci.inria.fr/users/new It gets processed manually, so you have to wait until your account is activated for you. b) Go to: https://ci.inria.fr/project and look for pharo-contribution. Request to join it. (it also gets processed manually) c) Ask in pharo-dev(a)lists.pharo.org to be made an Admin, because if not you won't be able to create a Job. " I see that Pharo-Contribution page redirects here: https://ci.inria.fr/pharo-contribution/job/JobTemplate/ Can we have that step 1 updated with what I wrote here? --------------------- Oh, and I can't get through part a) yet.... Apparently no one is accepting my user... Can somebody help? Thanks in advance :) Cordially, Alf

Jan. 8, 2014

Re: [Pharo-dev] Spotlight Demo

by Sean P. DeNigris

Tudor Girba-2 wrote > I started at Spotlight-SeanDeNigris.25, and then I added some more code Merged. The project depends on ActiveTutor now. A small part of ActiveTutor depends on Spotlight, but I can't seem to split the ActiveTutor package. I have ActiveTutor with tags ActiveTutor-Core and ActiveTutor-Screencasts, but the old "delete the working copy, then create MC packages for the sub-categories" doesn't seem to work in 3.0. I deleted the working copy, but when I add the MC packages for ActiveTutor-Core and ActiveTutor-Screencasts, they are empty :( ----- Cheers, Sean -- View this message in context: http://forum.world.st/Spotlight-Demo-tp4733076p4735284.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

Jan. 8, 2014

Re: [Pharo-dev] Do you want to accept???????? Discard???????

by dimitris chloupis

I am willing to compromise with Growl notifications, only , of course, if everything is undoable.Â Goubier , I am glad to hear that. That means we are closer than I thought to such very nice feature. I would love to implement such feature but currently I am learning Roassal and vector graphics and it would not be much use to jump from one idea to another :) Who knows , maybe in the future, maybe not. Â On Wednesday, 8 January 2014, 13:22, Sean P. DeNigris <sean(a)clipperadams.com> wrote: kilon wrote > The only thing that would make sense for removing confirmation dialogs is > editable action history. Take a look at Photoshop. Some 3d apps also have > something similar. Â Every action is recorded at a history ( a list) and > there is a gui that allow you to remove individual actions without > deleting or affecting other actions. Very cool stuff.Â Yes! That would be a dream. I was thinking last night that undo could be a tree from high level concepts down to individual actions. Something like: - edit #aMethod Â - paste Â - delete Â Â - backspace Â Â - backspace - edit #anotherMethod Â - â¦ kilon wrote > Undo also wont do, because I dont always realise my mistakes until much > later on. And having to undo all my work to fix a stupid mistake is a > nightmare. Certainly it seems too harsh to throw away without any notification. For now, I think the sweet spot for most people would be undo + non-instrusive notification. Maybe an undo list like "Workspace previous contents" and a Growl notification. If someone reaaaally needs the confirmation, we could have a setting, but we're talking about a single method pane's edits - overkill IMHO... ----- Cheers, Sean -- View this message in context: http://forum.world.st/Do-you-want-to-accept-Discard-tp4735220p4735268.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

Jan. 8, 2014

Re: [Pharo-dev] Spotlight Demo

by Tudor Girba

I started at Spotlight-SeanDeNigris.25, and then I added some more code about the themer, so, it is Spotlight-TudorGirba.27 from: Gofer new smalltalkhubUser: 'girba' project: 'GPlay'; package: 'Spotlight'; disablePackageCache; load Doru On Wed, Jan 8, 2014 at 11:13 PM, Sean P. DeNigris <sean(a)clipperadams.com>wrote: > Tudor Girba-2 wrote > > I added the categories now. > > What version of the original Spotlight package did you port, so I can merge > my changes... > > > > ----- > Cheers, > Sean > -- > View this message in context: > http://forum.world.st/Spotlight-Demo-tp4733076p4735281.html > Sent from the Pharo Smalltalk Developers mailing list archive at > Nabble.com. > > -- www.tudorgirba.com "Every thing has its own flow"

Jan. 8, 2014

Re: [Pharo-dev] Spotlight Demo

by Sean P. DeNigris

Tudor Girba-2 wrote > I added the categories now. What version of the original Spotlight package did you port, so I can merge my changes... ----- Cheers, Sean -- View this message in context: http://forum.world.st/Spotlight-Demo-tp4733076p4735281.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

Jan. 8, 2014

Re: [Pharo-dev] Do you want to accept???????? Discard???????

by GOUBIER Thierry

Sean, kilon, in fact Pharo has some of that, through RB and it's ability to undo / redo code change actions and combine them. I did for AltBrowser move to RB actions instead of the usual ones for some code related commands (class remove, class rename) and added a generic "undo" command with the RB related one, but I stopped halfway through it with totally unfriendly names for the undo commands (not fun to have a 100 characters + description of the undo in a menu item !) and lazyness in reworking some operations (method change) so as to integrate them as a RB operation transparently. My knowledge of the RB commands isn't good enough also to understand all the possibilities if we have a GUI associated. At least I do have a class remove without a confirmation dialog ;) and a possible undo. Once we'll have something like EPICEA integrated, I guess we will be able to do that in most cases for code related actions as well. Sean, I like what you described with Growl. This could prove very nice and sleek to have. Thierry ________________________________________ De : Pharo-dev [pharo-dev-bounces(a)lists.pharo.org] de la part de Sean P. DeNigris [sean(a)clipperadams.com] Date d'envoi : mercredi 8 janvier 2014 22:22 Ã : pharo-dev(a)lists.pharo.org Objet : Re: [Pharo-dev] Do you want to accept???????? Discard??????? kilon wrote > The only thing that would make sense for removing confirmation dialogs is > editable action history. Take a look at Photoshop. Some 3d apps also have > something similar. Every action is recorded at a history ( a list) and > there is a gui that allow you to remove individual actions without > deleting or affecting other actions. Very cool stuff. Yes! That would be a dream. I was thinking last night that undo could be a tree from high level concepts down to individual actions. Something like: - edit #aMethod - paste - delete - backspace - backspace - edit #anotherMethod - â¦ kilon wrote > Undo also wont do, because I dont always realise my mistakes until much > later on. And having to undo all my work to fix a stupid mistake is a > nightmare. Certainly it seems too harsh to throw away without any notification. For now, I think the sweet spot for most people would be undo + non-instrusive notification. Maybe an undo list like "Workspace previous contents" and a Growl notification. If someone reaaaally needs the confirmation, we could have a setting, but we're talking about a single method pane's edits - overkill IMHO... ----- Cheers, Sean -- View this message in context: http://forum.world.st/Do-you-want-to-accept-Discard-tp4735220p4735268.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

Jan. 8, 2014

MNU TextModel>>asText while typing in Nautilus code pane

by Sean P. DeNigris

TextModel(Object)>>doesNotUnderstand: #asText TextMorphForEditView(TextMorph)>>newContents: "stringOrText copy asText." where stringOrText is a TextModel. Don't know how to reproduce. Anyone else seeing this? I was putting Halts in Morphic, so maybe I messed something up... ----- Cheers, Sean -- View this message in context: http://forum.world.st/MNU-TextModel-asText-while-typing-in-Nautilus-code-pa… Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

Jan. 8, 2014

Re: [Pharo-dev] [Moose-dev] Re: Feature for request for GraphET?

by Tudor Girba

Hi Daniel, Good work! But, there is still something I do not quite understand. Take a look at the produced image. Why is it that in the first red bar appears behind the gray one, and the rest appear in front of the corresponding gray bars? Related to the GapFraction: Could you describe what makes it such an expensive issue to compute the width of the bar with a bit of scaling? I am asking because this particular diagram does not work without it. Cheers, Doru [image: Inline image 1] On Wed, Jan 8, 2014 at 6:50 PM, Daniel Aviv Notario < daniel_avivnotario(a)hotmail.com> wrote: > So, attending to the request Doru made, here's something similar to the > example he provided in EyeSee... All the methods were already implemented > but most of them didn't have a clear accesor to modify them. > > So, if you just load the last version of GraphET and do this: > > > builder := GETDiagramBuilder new. > grouped := { #item1->{10 . 12 . 11 . 14 } . #item2-> { 1 . 1 . 20 > } > . #item3->{ 1 . 1 . 1 . 1 . 1} }. > diag := (GETVerticalBarDiagram new) > modelLabels: [:each | each key ]; > y: [:each | each value size]; > yAxisLabel: 'Count'; > valueAxis; > height:200; > spacing:20; > yAxisOnRight; > color: Color veryLightGray; > axisColor: Color veryLightGray; > models: grouped; > yourself. > diag2 := (GETVerticalBarDiagram new) > y: [:each | each value sum]; > regularAxis; > height: 200; > spacing:20; > yAxisLabel: 'Sum'; > color: (Color red); > axisColor: (Color red); > models: grouped; > yourself. > (builder unionDiagram) > add: diag; > add: diag2. > builder open > > > You'll see it's very similar. There's two things I couldn't replicate so > far, one of them is the rotating labels, as Roassal can't rotate labels > yet, and the second one was the GapFraction, as axis in GraphET need some > work yet, so implementing that kind of thing would be hard. I think I'll > dedicate to a proper axis in the next couple of weeks. > > Cheers :)! > > > > >>> ---------- Forwarded message ---------- > > > >>> From: Tudor Girba <tudor(a)tudorgirba.com> > > > >>> Date: Sat, Jan 4, 2014 at 10:39 PM > > > >>> Subject: Re: [Pharo-dev] Feature for request for GraphET? > > > >>> To: Pharo Development List <pharo-dev(a)lists.pharo.org> > > > >>> > > > >>> > > > >>> Here is one thing I would want: a correspondent Graph-ET > > > >>> functionality to make something like this EyeSee script work as > > > >>> in > > > >>> the first attached picture. > > > >>> > > > >>> renderer := ESDiagramRenderer new. > > > >>> grouped := { #item1->{10 . 12 . 11 . 14 } . #item2-> { 1 . 1 . 20 > > > >>> } > > > >>> . #item3->{ 1 . 1 . 1 . 1 . 1} }. > > > >>> diag := (ESVerticalBarDiagram new) > > > >>> identifier: [:each | each key ]; > > > >>> y: [:each | each value size]; > > > >>> yAxisLabel: 'Count'; > > > >>> valueAxis; > > > >>> rightYAxis; > > > >>> defaultColor: Color veryVeryLightGray; > > > >>> axisColor: Color veryVeryLightGray; > > > >>> models: grouped; > > > >>> rotatedLabels: true; > > > >>> yourself. > > > >>> diag2 := (ESVerticalBarDiagram new) > > > >>> y: [:each | each value sum]; > > > >>> regularAxis; > > > >>> yAxisLabel: 'Sum'; > > > >>> gapFraction: 2 / 3; > > > >>> defaultColor: (Color red); > > > >>> axisColor: (Color red); > > > >>> models: grouped; > > > >>> yourself. > > > >>> (renderer compositeDiagram) > > > >>> add: diag; > > > >>> add: diag2. > > > >>> renderer open > > > >>> > > > >>> > > > >>> <Screenshot 2014-01-04 22.31.57.png> > > > >>> > > > >>> > > > >>> The best I could produce with Graph-ET is this (I commented out > > > >>> the > > > >>> message to which I did not find a correspondent): > > > >>> > > > >>> builder := GETDiagramBuilder new. > > > >>> grouped := { #item1->{10 . 12 . 11 . 14 } . #item2-> { 1 . 1 . 20 > > > >>> } > > > >>> . #item3->{ 1 . 1 . 1 . 1 . 1} }. > > > >>> diag := (GETVerticalBarDiagram new) > > > >>> "identifier: [:each | each key ];" > > > >>> y: [:each | each value size]; > > > >>> yAxisLabel: 'Count'; > > > >>> valueAxisLine; > > > >>> "rightYAxis;" > > > >>> color: Color veryLightGray; > > > >>> "axisColor: Color veryLightGray;" > > > >>> models: grouped; > > > >>> "rotatedLabels: true;" > > > >>> yourself. > > > >>> diag2 := (GETVerticalBarDiagram new) > > > >>> y: [:each | each value sum]; > > > >>> regularAxis; > > > >>> yAxisLabel: 'Sum'; > > > >>> "gapFraction: 2 / 3;" > > > >>> color: (Color red); > > > >>> "axisColor: (Color red);" > > > >>> models: grouped; > > > >>> yourself. > > > >>> (builder compositeDiagram) > > > >>> add: diag; > > > >>> add: diag2. > > > >>> builder open > > > >>> > > > >>> <Screenshot 2014-01-04 22.36.40.png> > > > >>> > > > >>> Cheers, > > > >>> Doru > -- www.tudorgirba.com "Every thing has its own flow"

Jan. 8, 2014