- Pharo-dev - lists.pharo.org

Re: [Pharo-project] A new GUI visual designer

by StÃ©phane Ducasse

@Tobias: Hi, what is the license because the wiki does not say it or I could not find it. On Dec 1, 2010, at 6:38 PM, Tobias Pape wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Am 2010-12-01 um 16:13 schrieb Igor Stasenko: >> On 1 December 2010 12:48, nullPointer <epicfan(a)gmail.com> wrote: >>> >>> >>> https://www.hpi.uni-potsdam.de/hirschfeld/trac/SqueakCommunityProjects/wiki… >>> >>> It really looks very professional. I made a user interface builder for >>> Polymorph, which I left because I was too complicated for me. @NullPointer: you did a lot of work Now this is not clear how the code is generated even in this new one. >>> I hope that >>> now someone more knowledgeable can really necessary to implement a tool for >>> Squeak / Pharo as a visual designer user interfaces. >>> >> >> +1 >> that's really important stuff. And this project looks very promising. >> Hope, author(s) can shed a light on it, in what stage it is, what >> license etc etc. > > > Its all on that wiki. > Feel free to contact me, anyways. > > So Long > -Tobias > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.9 (Darwin) > > iEYEARECAAYFAkz2iCMACgkQcPVIrP6PLKuiowCeNAFhjRdcFFYlxpaVGqHqBGml > dkQAoKVVzSy6H6ycRuZw0UXIjAW0uHvY > =Pv5I > -----END PGP SIGNATURE----- >

Dec. 4, 2010

Re: [Pharo-project] Out of bounds while loading package

by Guillermo Polito

Maybe it is. I didn't have the last last last version because my image didn't want to update. Another issue for that :P. I used the last prebuild core. On Sat, Dec 4, 2010 at 9:55 AM, Benjamin < benjamin.vanryseghem.pharo(a)gmail.com> wrote: > A fix has been integrated for that ... > Strange :s > > Ben > > > > On Dec 4, 2010, at 1:49 PM, StÃ©phane Ducasse wrote: > > Excellent initiative!!! >> >> On Dec 4, 2010, at 9:12 AM, Guillermo Polito wrote: >> >> Hi! >>> >>> I was trying to load ConfigurationOfPharo in Core 1.2, so we can (slowly) >>> get a Dev image. So It began well, but when it reached AST-Core-lr.84, it >>> raised up an exception: >>> >>> "subscript is out of bounds: -1" >>> >>> It is mainly caused by CommentReference>>parseTimestampFrom: >>> >>> Which has the next comment: >>> >>> "Timestampstrings can be of the form: >>> <authorname><date><time>. >>> <authorname><date> >>> <date><time> >>> <date><time><authorname> >>> <date><authorname> >>> <historical>" >>> >>> The string timestamp is '<historical>' indeed, but it does the follow >>> check: >>> >>> ^ (aString isEmpty or:[aString = 'historical']) >>> ifTrue:[DateAndTime new] >>> ifFalse:[ here comes the error ... >>> >>> Shouldn't that check be >>> >>> ^ (aString isEmpty or:[aString = '<historical>']) >>> >>> ?? >>> >>> Cheers, >>> Guille >>> >> >> >> > >

Dec. 4, 2010

Re: [Pharo-project] Out of bounds while loading package

by Benjamin

A fix has been integrated for that ... Strange :s Ben On Dec 4, 2010, at 1:49 PM, StÃ©phane Ducasse wrote: > Excellent initiative!!! > > On Dec 4, 2010, at 9:12 AM, Guillermo Polito wrote: > >> Hi! >> >> I was trying to load ConfigurationOfPharo in Core 1.2, so we can >> (slowly) get a Dev image. So It began well, but when it reached >> AST-Core-lr.84, it raised up an exception: >> >> "subscript is out of bounds: -1" >> >> It is mainly caused by CommentReference>>parseTimestampFrom: >> >> Which has the next comment: >> >> "Timestampstrings can be of the form: >> <authorname><date><time>. >> <authorname><date> >> <date><time> >> <date><time><authorname> >> <date><authorname> >> <historical>" >> >> The string timestamp is '<historical>' indeed, but it does the >> follow check: >> >> ^ (aString isEmpty or:[aString = 'historical']) >> ifTrue:[DateAndTime new] >> ifFalse:[ here comes the error ... >> >> Shouldn't that check be >> >> ^ (aString isEmpty or:[aString = '<historical>']) >> >> ?? >> >> Cheers, >> Guille > >

Dec. 4, 2010

Re: [Pharo-project] Dollars to documentation

by StÃ©phane Ducasse

yes we should do the same! Stef > Interesting action from the rails project: http://pledgie.com/campaigns/14034 > > Laurent Laffont >

Dec. 4, 2010

Re: [Pharo-project] Out of bounds while loading package

by StÃ©phane Ducasse

Excellent initiative!!! On Dec 4, 2010, at 9:12 AM, Guillermo Polito wrote: > Hi! > > I was trying to load ConfigurationOfPharo in Core 1.2, so we can (slowly) get a Dev image. So It began well, but when it reached AST-Core-lr.84, it raised up an exception: > > "subscript is out of bounds: -1" > > It is mainly caused by CommentReference>>parseTimestampFrom: > > Which has the next comment: > > "Timestampstrings can be of the form: > <authorname><date><time>. > <authorname><date> > <date><time> > <date><time><authorname> > <date><authorname> > <historical>" > > The string timestamp is '<historical>' indeed, but it does the follow check: > > ^ (aString isEmpty or:[aString = 'historical']) > ifTrue:[DateAndTime new] > ifFalse:[ here comes the error ... > > Shouldn't that check be > > ^ (aString isEmpty or:[aString = '<historical>']) > > ?? > > Cheers, > Guille

Dec. 4, 2010

Re: [Pharo-project] static files in a MC package ?

by StÃ©phane Ducasse

no and this is a real extension that we should add to MC. Stef On Dec 4, 2010, at 10:41 AM, laurent laffont wrote: > Hi, > > is it possible to include static files in a MC package (especially images) ? > > Cheers, > > Laurent Laffont > > Pharo Smalltalk Screencasts: http://www.pharocasts.com/ > Blog: http://magaloma.blogspot.com/

Dec. 4, 2010

Re: [Pharo-project] Toolset vs UIManager

by StÃ©phane Ducasse

Ok this sounds like a good idea. So we should do that. Now I would like to have a clear comment in both classes and avoid duplicated code or functionality. Stef > they are related but still a bit different. > a toolset provides an access to standard system tools like browser, > inspector debugger etc. > while UI manager provides an access to things like inform:, progress bars etc > > My proposal remains same as before. i proposed to unify this API and > use idioms like: > > Smalltalk tools browser openOn: aClass > > as well as > > Smalltalk ui inform: 'you banned' > > so, then we don't need to have a distinct classes, but instead a > roles, like 'tools' , 'ui', 'browser' which > should support corresponding protocol. > >> Stef >> > > > > -- > Best regards, > Igor Stasenko AKA sig. >

Dec. 4, 2010

Re: [Pharo-project] [ANN] Swazoo has now WebSocket support

by StÃ©phane Ducasse

good news! On Dec 2, 2010, at 7:07 PM, Janko MivÅ¡ek wrote: > Dear Smalltalkers, > > Past few days I worked on WebSocket implementation in Swazoo Smalltalk > web server [1], in coordination with Philippe Marschall from Seaside [4] > and Nicolas Petton from Iliad [5] devteams, so together with me from > Aida/Web [3] our cooperation ensures that this implementation is ready > to be adapted for all three main web frameworks. > > Why WebSocket? Because it will replace current Comet hacks to enable a > bidirectional messaging between browser and web application, specially > for web app to send something back to the browser without waiting for > user to click something. Read a bit more about WebSocket here: > > http://dev.opera.com/articles/view/introducing-web-sockets/ > > WebSocket as browser supported technology enables opening an always > present channel, so that web app can send commands like update some > element of the webpage at any time. That's why WebSocket is ideal for so > called "real-time" web. > > You are invited to look at the code for Squeak/Pharo as Swazoo2.3beta1.1 > at http://www.squeaksource.org/Swazoo.html > > A bit more about WebSocket in Swazoo: > > http://www.swazoo.org/websocket > > Best regards > Janko > Swazoo maintainer > > [1] Swazoo WebSocket support > http://www.swazoo.org/websocket > [2] Introducing WebSockets > http://dev.opera.com/articles/view/introducing-web-sockets/ > [3] Aida/Web > http://www.aidaweb.si > [4] Seaside > http://www.seaside.st > [5] Iliad > http://www.iliadproject.org >

Dec. 4, 2010

Re: [Pharo-project] Toolset vs UIManager

by Benjamin

+150 !! UIManager is the Utilities of Morphic ^^ I sent an email about restructured it but I haven't got any answer. I think that UIManager was a good idea at the beginning. It's a bag of UI which can be used by every classes to avoid duplications. The maor problem with that is that the implementations of methods are done with too hight tool, and then there is tons of bad dependencies. Jannik and I was discussing about create multiple polymorphic UIManager classes implementd with different level of morphic tools. This way, yo can plug and unplug PolyMorph or things like that. ToolSet is pretty much the same about Tools but with the specificity to inherits from AppRegistry which allow you to switch between classes providing tools. But it's not used. Ben On Dec 3, 2010, at 10:52 PM, Francisco Ortiz PeÃ±aloza wrote: > +2 :) > > On Fri, Dec 3, 2010 at 6:41 PM, Esteban Lorenzano > <estebanlm(a)gmail.com> wrote: > +1 > > El 03/12/2010, a las 6:25p.m., Igor Stasenko escribiÃ³: > > > On 3 December 2010 21:47, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr > > wrote: > >> I was reading the change of benjmain to deprecate > notifyWithLabel: aString from Object. > >> Yes! > >> > >> Now he was putting it on ToolSet and I thought that I would have > put it on UIManager. > >> Now the relationship and responsibility of this two classes are > not clear to me. > >> Does anybody a clear thought on that? > >> > > > > they are related but still a bit different. > > a toolset provides an access to standard system tools like browser, > > inspector debugger etc. > > while UI manager provides an access to things like inform:, > progress bars etc > > > > My proposal remains same as before. i proposed to unify this API and > > use idioms like: > > > > Smalltalk tools browser openOn: aClass > > > > as well as > > > > Smalltalk ui inform: 'you banned' > > > > so, then we don't need to have a distinct classes, but instead a > > roles, like 'tools' , 'ui', 'browser' which > > should support corresponding protocol. > > > >> Stef > >> > > > > > > > > -- > > Best regards, > > Igor Stasenko AKA sig. > > > > >

Dec. 4, 2010

Re: [Pharo-project] Issue 3370 in pharo: NewParagraph and MultiNewParagraph should be merged

by pharo＠googlecode.com

Comment #3 on issue 3370 by aplantec: NewParagraph and MultiNewParagraph should be merged http://code.google.com/p/pharo/issues/detail?id=3370 so asParagraph should generate a NewParagraph ?

Dec. 4, 2010