July 2014 - Pharo-users - lists.pharo.org

Re: [Pharo-users] Add external package to metacello using versionner

by Sean P. DeNigris

NorbertHartl wrote > I try to give Neo-Caching a shot in a project of mine. Neo-Caching does > not have a metacello configuration The more religious MetaC practitioners will tell you never to to this and to just make a configuration yourself ;) I've found this can be overkill in simple cases. IIRC you can add a repository to the spec for just that package i.e. something like: spec package: 'Neo-Caching' with: [ spec repository: ... ----- Cheers, Sean -- View this message in context: http://forum.world.st/Add-external-package-to-metacello-using-versionner-tp… Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

July 28, 2014

Re: [Pharo-users] [Tip] Image init scripts when working with PharoLauncher

by Damien Cassou

On Fri, Jul 25, 2014 at 1:54 PM, Torsten Bergmann <astares(a)gmx.de> wrote: > I've found the following script very useful: here is what I do for MC credentials: credentials := Dictionary new. MCRepository classVarNamed: 'Settings' put: credentials. credentials at: 'account smalltalkhub' put: '*smalltalkhub.com* DamienCassou:mypassword'. credentials at: 'account squeaksource' put: '*squeaksource.com* dc:mypassword'. credentials at: 'account ss3' put: '*ss3.gemstone.com* dc:mypassword'. credentials at: 'account source.squeak' put: '*source.squeak.org* DamienCassou:mypassword'. -- Damien Cassou http://damiencassou.seasidehosting.st "Success is the ability to go from one failure to another without losing enthusiasm." Winston Churchill

July 28, 2014

Re: [Pharo-users] Reference documentation on Trait

by Damien Cassou

On Fri, Jul 25, 2014 at 9:47 AM, Hilaire <hilaire(a)drgeo.eu> wrote: > Do we have such document(s)? http://scg.unibe.ch/research/traits -- Damien Cassou http://damiencassou.seasidehosting.st "Success is the ability to go from one failure to another without losing enthusiasm." Winston Churchill

July 28, 2014

Add external package to metacello using versionner

by Norbert Hartl

I try to give Neo-Caching a shot in a project of mine. Neo-Caching does not have a metacello configuration. If I add the package via versionner the package is added but without repository information. That fails on loading? How to do? Bug? Norbert

July 28, 2014

Re: [Pharo-users] [Moose-dev] Tree/Outliners of playgrounds, markdown inside comments and some quick medium size dreams for Pharo/Smalltalk

by Alexandre Bergel

Hi Offray, Something that I usually do, is to simply subclass MooseEntity to create the hierarchy of my domain. I can then enjoy the moose panel to browse my model. Cheers, Alexandre On Jul 27, 2014, at 10:59 AM, Offray Vladimir Luna CÃ¡rdenas <offray(a)riseup.net> wrote: > Hi, > > Answering to myself: I have solved the code that selects the headers of the main tree. The key is to create a new collection containing only node names. Here is the code: > > "*************************" > | mainTree node1 node2 explorer headers | > > mainTree := UbakyeNode > header: 'Arbol raÃz' > body: ''. > > node1 := UbakyeNode > header: 'Nodo 1' > body: 'Texto 1'. > > node2 := UbakyeNode > header: 'Nodo 2' > body: 'Texto 2'. > > mainTree > addNode: node1; > addNode: node2. > > explorer := GLMTabulator new > title: (mainTree header). > explorer column: #tree; > column: #body. > > headers := (mainTree children) > collect: [:node | node header]. > > explorer transmit to: #tree; andShow: [:a | > a tree > display: headers]. > > explorer openOn: mainTree. > > "*************************" > > Now I need to make the children sellectable, and that all the contents of the tree can be updated with a shortcut. > > I will keep you posted. > > Cheers, > > Offray > > On 07/26/2014 09:01 PM, Offray Vladimir Luna CÃ¡rdenas wrote: >> Hi again, >> >> I will be using this thread to update my advances and questions about >> how to build an outliner in Pharo Smalltalk. If there is a better method >> like starting a new thread for particular questions, or a less narrative >> style, please let me know. >> >> The idea is to use the tools provided by Moose to build a quick outliner >> that can be extended to suit my needs on academical writing. This is >> kind of a strange approach in the sense that I'm not following the >> tutorials with a predefined problems (make a game and so) but trying to >> start with a real (in the sense of closer) problem (making an outliner) >> and to see which knowledge I need to solve this necessity. In that sense >> is more like the Freire's alphabetization of adults in Brazil. >> >> So, the things I have done so far was to search for a good model to >> start with. Something already done that can be used as scaffolding for >> my outliner. The Help System seems like a good start for an outliner (in >> fact it is already one), so I have taken the Help-Core system and start >> to use it as a base for my project. >> >> After that I have used the Moose browsers to build a simple interface, >> as seen here: >> >> http://mutabit.com/offray/static/blog/output/galleries/objetos/ubakye-brows… >> >> >> The part I want to deal with is this: >> >> =============== >> >> explorer := GLMTabulator new >> title: (mainTree header). >> explorer column: #tree; >> column: #body. >> >> explorer transmit to: #tree; andShow: [:a | >> a tree >> display: mainTree children ]. >> >> explorer openOn: mainTree. >> >> =============== >> >> So, instead of "display: mainTree children" I need something that takes >> the get names (headers) of the two nodes and the contents in the right >> panel. For that I think that I need to learn some iterators. I have >> already a "header" method for the nodes. Any clue would be appreciated >> and I will keep you posted on my advances. >> >> Cheers, >> >> Offray >> >> >> On 07/21/2014 12:58 PM, Offray Vladimir Luna CÃ¡rdenas wrote: >>> Hi Damien, >>> >>> Thanks for your answer. Comments below. >>> >>> On 07/21/2014 11:09 AM, Damien Cassou wrote: >>>> On Sat, Jul 19, 2014 at 2:47 AM, Offray Vladimir Luna CÃ¡rdenas >>>> <offray(a)riseup.net> wrote: >>>>> The first idea that comes to mind is using STON for storage nodes and >>>>> tree >>>>> information, so I can interchange it with the flatland files world >>>>> and keep >>>>> it readable. Sounds that reasonable? >>>> >>>> >>>> without more information, it is hard to stay. Try with STON and change >>>> if that does not work :-). We have XML and JSON generators as well. >>>> >>> >>> >>> This is a kind of raw preview of I'm talking about: >>> >>> http://www.enlightenment.org/ss/e-53cd4f36f021e9.68569046.jpg >>> >>> Of course in this case, it is just a Help browser with a Playground >>> window over it, but I would like to have something like Playgrounds >>> inside the help browser. I was trying to build a custom browser with >>> Glamour, but seems that Help Browser already has the machinery I'm >>> looking for. >>> >>> So my first question is how to use the Help Browser class as a template >>> for my outliner class? And can I put a Playground where the plain text >>> is located right now? >>> >>>> >>>>> The second thing I would like to do is to add pandoc's markdown inside >>>>> comments, but I don't like the syntax of comments in Smalltalk because >>>>> single quotes are fairly easy to find in light markup language like >>>>> markdown. Is difficult to change it to create something more like >>>>> python >>>>> (with """) or Lua (with -[]- )? >>>> >>>> >>>> There is only one syntax for comments in Pharo. Instead of Markdown, >>>> you might want to have a look at Pillar which is implemented in Pharo >>>> and can generate Markdown (and html, and pdf) : >>>> https://github.com/pillar-markup/pillar-documentation/ >>>> >>>> >>> >>> I have seen Pillar. Seems really interesting, but Pandocs markdown >>> support academic citation in several formats and I have already long >>> docs wrote on that format integrated in my workflow from Zotero and even >>> there is a growing community working on Scholarly Markdown[1][2] so I >>> would like to stick with it as much as I can for my own writing. >>> That being said. I would like also a better integration between >>> Smalltalk outliners and all the academic publication work flow, >>> including working better with pandoc as a external library. >>> >>> [1] https://github.com/scholmd/scholmd/wiki >>> [2] >>> http://blog.martinfenner.org/2013/06/29/metadata-in-scholarly-markdown/ >>> [3] >>> http://programminghistorian.org/lessons/sustainable-authorship-in-plain-tex… >>> >>> >>> >>> Thanks again, this conversation with people in the community is very >>> valuable to me, >>> >>> Offray >>> >>> >> >> >> > > > _______________________________________________ > Moose-dev mailing list > Moose-dev(a)iam.unibe.ch > https://www.iam.unibe.ch/mailman/listinfo/moose-dev -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

July 27, 2014

Re: [Pharo-users] Roassal2 orthogonal lines, attach points and self-lines.

by Alexandre Bergel

Hi Peter! Even if trivial to add them, orthogonal lines are indeed missing in Roassal2. Having a way to indicate lines from an element to itself is also important. My first guess is that we need a class hierarchy to describe combination of control points. I can work on these point mid-next week. Else, you are very welcome to give a try. Cheers, Alexandre On Jul 26, 2014, at 3:57 PM, Peter UhnÃ¡k <i.uhnak(a)gmail.com> wrote: > Hi, > > is there any support for orthogonal lines in Roassal2? I've seen something similar (ROOrthoVerticalLineShape), however nothing of that sort in Roassal2. I would imagine it to behave similarly to RTBezierLine - having controllingElements to specify the corners. (While ROOrtholines did it automatically I was looking for something more manual, so it can go in both directions - thus the controllingElements). > > Second thing I wasn't able to figure out is whether it is possible to create line from element to itself. All *AttachPoint classes aren't really able to cope with that. The best result I got was from Vertical/Horizontal AttachPoints (combined with BezierLine), however I would like to be able to specify which side it would start on and end on. Ideally not just middle of sides, but arbitrary part of the shape (I imagine that might be hard for non-rectangular shapes). > > Thanks, > Peter -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

July 27, 2014

Re: [Pharo-users] Tree/Outliners of playgrounds, markdown inside comments and some quick medium size dreams for Pharo/Smalltalk

by Offray Vladimir Luna CÃ¡rdenas

Hi again, I will be using this thread to update my advances and questions about how to build an outliner in Pharo Smalltalk. If there is a better method like starting a new thread for particular questions, or a less narrative style, please let me know. The idea is to use the tools provided by Moose to build a quick outliner that can be extended to suit my needs on academical writing. This is kind of a strange approach in the sense that I'm not following the tutorials with a predefined problems (make a game and so) but trying to start with a real (in the sense of closer) problem (making an outliner) and to see which knowledge I need to solve this necessity. In that sense is more like the Freire's alphabetization of adults in Brazil. So, the things I have done so far was to search for a good model to start with. Something already done that can be used as scaffolding for my outliner. The Help System seems like a good start for an outliner (in fact it is already one), so I have taken the Help-Core system and start to use it as a base for my project. After that I have used the Moose browsers to build a simple interface, as seen here: http://mutabit.com/offray/static/blog/output/galleries/objetos/ubakye-brows… The part I want to deal with is this: =============== explorer := GLMTabulator new title: (mainTree header). explorer column: #tree; column: #body. explorer transmit to: #tree; andShow: [:a | a tree display: mainTree children ]. explorer openOn: mainTree. =============== So, instead of "display: mainTree children" I need something that takes the get names (headers) of the two nodes and the contents in the right panel. For that I think that I need to learn some iterators. I have already a "header" method for the nodes. Any clue would be appreciated and I will keep you posted on my advances. Cheers, Offray On 07/21/2014 12:58 PM, Offray Vladimir Luna CÃ¡rdenas wrote: > Hi Damien, > > Thanks for your answer. Comments below. > > On 07/21/2014 11:09 AM, Damien Cassou wrote: >> On Sat, Jul 19, 2014 at 2:47 AM, Offray Vladimir Luna CÃ¡rdenas >> <offray(a)riseup.net> wrote: >>> The first idea that comes to mind is using STON for storage nodes and >>> tree >>> information, so I can interchange it with the flatland files world >>> and keep >>> it readable. Sounds that reasonable? >> >> >> without more information, it is hard to stay. Try with STON and change >> if that does not work :-). We have XML and JSON generators as well. >> > > > This is a kind of raw preview of I'm talking about: > > http://www.enlightenment.org/ss/e-53cd4f36f021e9.68569046.jpg > > Of course in this case, it is just a Help browser with a Playground > window over it, but I would like to have something like Playgrounds > inside the help browser. I was trying to build a custom browser with > Glamour, but seems that Help Browser already has the machinery I'm > looking for. > > So my first question is how to use the Help Browser class as a template > for my outliner class? And can I put a Playground where the plain text > is located right now? > >> >>> The second thing I would like to do is to add pandoc's markdown inside >>> comments, but I don't like the syntax of comments in Smalltalk because >>> single quotes are fairly easy to find in light markup language like >>> markdown. Is difficult to change it to create something more like python >>> (with """) or Lua (with -[]- )? >> >> >> There is only one syntax for comments in Pharo. Instead of Markdown, >> you might want to have a look at Pillar which is implemented in Pharo >> and can generate Markdown (and html, and pdf) : >> https://github.com/pillar-markup/pillar-documentation/ >> >> > > I have seen Pillar. Seems really interesting, but Pandocs markdown > support academic citation in several formats and I have already long > docs wrote on that format integrated in my workflow from Zotero and even > there is a growing community working on Scholarly Markdown[1][2] so I > would like to stick with it as much as I can for my own writing. > That being said. I would like also a better integration between > Smalltalk outliners and all the academic publication work flow, > including working better with pandoc as a external library. > > [1] https://github.com/scholmd/scholmd/wiki > [2] http://blog.martinfenner.org/2013/06/29/metadata-in-scholarly-markdown/ > [3] > http://programminghistorian.org/lessons/sustainable-authorship-in-plain-tex… > > > Thanks again, this conversation with people in the community is very > valuable to me, > > Offray > >

July 27, 2014

Roassal2 orthogonal lines, attach points and self-lines.

by Peter UhnÃ¡k

Hi, is there any support for orthogonal lines in Roassal2? I've seen something similar (ROOrthoVerticalLineShape), however nothing of that sort in Roassal2. I would imagine it to behave similarly to RTBezierLine - having controllingElements to specify the corners. (While ROOrtholines did it automatically I was looking for something more manual, so it can go in both directions - thus the controllingElements). Second thing I wasn't able to figure out is whether it is possible to create line from element to itself. All *AttachPoint classes aren't really able to cope with that. The best result I got was from Vertical/Horizontal AttachPoints (combined with BezierLine), however I would like to be able to specify which side it would start on and end on. Ideally not just middle of sides, but arbitrary part of the shape (I imagine that might be hard for non-rectangular shapes). Thanks, Peter

July 26, 2014

Re: [Pharo-users] [Tip] Image init scripts when working with PharoLauncher

by stepharo

> Cool, thanks! My script was getting too complicated, so I made it into a > class with a MetaC config, and the script is now just a Gofer load. This has > the additional benefit that the start up scripts for the Pharo loader never > change (they're just Gofer loads) and so avoid the caching problems that can > otherwise arise. Yes this is the right way to manage scripts: turn them into classes :) > > > Torsten Bergmann wrote >> "Set password for Squeaksource" >> (MCRepositoryGroup default repositories >> select: [:each | (each isKindOf: MCHttpRepository) >> and: [((each locationWithTrailingSlash includesSubstring: >> 'www.squeaksource.com') >> or: [each locationWithTrailingSlash includesSubstring: >> 'http://ss3.gemstone.com/ss/'])] >> ]) >> do: [:each | each user: 'squeaksourceuser'; password: >> 'secretsqueaksourcepassword']. >> ... > I do something like this too. The shortcoming is that newly created repos > are not authenticated. This should be in core built into the repo objects. I > guess that's what the KeyChain stuff is about? > > > > > ----- > Cheers, > Sean > -- > View this message in context: http://forum.world.st/Tip-Image-init-scripts-when-working-with-PharoLaunche… > Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com. > >

July 26, 2014

Re: [Pharo-users] Commandline

by Dale Henrichs

Oooooh! That might just do the trick for me... Turns out that I've got one more problem in that the gsDevKit pharo vm and/or gsdevkit image files can be installed in arbitrary locations ... for this one I guess I need to consider installing pharo and pharo-ui in /usr/local/ ... will work for Linux and Mac ... for Windows, I'm already jumping through some ugly hoops... Thanks Thierry and Sven! Dale On Fri, Jul 25, 2014 at 1:53 PM, Thierry Goubier <thierry.goubier(a)gmail.com> wrote: > You can try to get the current working directory this way: > > argumentsAsFileReferences > |pwd | > pwd := (Smalltalk os environment at: 'PWD' ifAbsent: [ Smalltalk os > environment at: 'CD' ]) asFileReference. > ^ self argumentsWithoutOptions collect: [ :each | pwd resolveReference: > each asFileReference ] > > This works when passing arbitrary files as argument to a shell script > doing an exec on arbitraryPath/pharo afterwards. > > (This is how GitFileTree-MergeDriver handles being called by git with file > paths). > > Now, it would be necessary to fix that on the default .st loading :) > > Thierry > > Le 25/07/2014 22:05, Sven Van Caekenberghe a Ã©crit : > > Yes, for some reason, the shebang (#!) machinery passes ./test2.st as >> argument to pharo, which it seems can only be resolved if it is in the >> image directory (as opposed to the current shell directory). >> >> I have no time to look at this now, but I vaguely remember that pharo >> mixed the concept of current working directory and image directory (at >> least in some places). >> >> On 25 Jul 2014, at 21:43, Dale Henrichs <dale.henrichs(a)gemtalksystems.com> >> wrote: >> >> Sven, >>> Without putting a test.st file in /Users/daleh/gsDevKitHome/pharo, I >>> get the following (it works if I `cp ./test.st >>> /Users/daleh/gsDevKitHome/pharo`): >>> >>> nehalem:junk daleh$ pwd >>> /Users/daleh/junk >>> nehalem:junk daleh$ cat test.st >>> #!/Users/daleh/gsDevKitHome/pharo/pharo-vm/Pharo.app/Contents/MacOS/Pharo >>> --headless /Users/daleh/gsDevKitHome/pharo/todeClient.image st --quit >>> FileStream stdout nextPutAll: 42 factorial asString; cr. >>> nehalem:junk daleh$ ./test.st >>> FileDoesNotExist: File @ /Users/daleh/gsDevKitHome/pharo/test.st >>> FileHandle>>streamError >>> FileHandle>>readStream >>> FileSystem>>readStreamOn: >>> FileReference>>readStream >>> FileReference(AbstractFileReference)>>readStreamDo: >>> STCommandLineHandler>>installSourceFile: >>> STCommandLineHandler>>installSourceFiles in Block: [ :reference | self >>> installSourceFile: reference ]...etc... >>> OrderedCollection>>do: >>> STCommandLineHandler>>installSourceFiles in Block: [ sourceFiles do: [ >>> :reference | self installSourc...etc... >>> BlockClosure>>ensure: >>> STCommandLineHandler>>installSourceFiles >>> STCommandLineHandler>>activate >>> STCommandLineHandler class(CommandLineHandler class)>>activateWith: >>> PharoCommandLineHandler(BasicCommandLineHandler)>>activateSubCommand: >>> in Block: [ aCommandLinehandler activateWith: commandLine ] >>> BlockClosure>>on:do: >>> PharoCommandLineHandler(BasicCommandLineHandler)>>activateSubCommand: >>> PharoCommandLineHandler(BasicCommandLineHandler)>>handleSubcommand >>> PharoCommandLineHandler(BasicCommandLineHandler)>>handleArgument: >>> PharoCommandLineHandler(BasicCommandLineHandler)>>activate in Block: [ >>> self handleArgument: (self arguments ifEmpty: [ ...etc... >>> BlockClosure>>on:do: >>> PharoCommandLineHandler(BasicCommandLineHandler)>>activate >>> PharoCommandLineHandler>>activate >>> PharoCommandLineHandler class(CommandLineHandler class)>>activateWith: >>> PharoCommandLineHandler class>>activateWith: in Block: [ super >>> activateWith: aCommandLine ] >>> WorldState>>runStepMethodsIn: >>> WorldMorph>>runStepMethods >>> WorldState>>doOneCycleNowFor: >>> WorldState>>doOneCycleFor: >>> WorldMorph>>doOneCycle >>> MorphicUIManager>>spawnNewProcess in Block: [ ... >>> >>> >>> >>> On Fri, Jul 25, 2014 at 12:29 PM, Sven Van Caekenberghe <sven(a)stfx.eu> >>> wrote: >>> >>> On 25 Jul 2014, at 21:25, Dale Henrichs <dale.henrichs@gemtalksystems. >>> com> wrote: >>> >>> Sven, >>>> >>>> I'm using Pharo3.0 and it appears that the .st scripts have to be >>>> located in the same directory as the image and changes files ... I see >>>> that you are using Pharo4.0 ... >>>> >>>> So is it possible to get Pharo3.0 to work on .st files in arbitary >>>> directores? >>>> >>> Are you using the shebang or direct version ? >>> In any case, I am always using absolute paths. >>> Does that not work for you ? >>> >>> Dale >>>> >>>> >>>> On Thu, Jul 24, 2014 at 7:42 AM, Sven Van Caekenberghe <sven(a)stfx.eu> >>>> wrote: >>>> >>>> On 24 Jul 2014, at 16:34, phil(a)highoctane.be wrote: >>>> >>>> One question I have is how fast the load of an image and processing by >>>>> an image is when compared with bash. >>>>> >>>> Obviously, it is slower, there is a whole image that needs to be >>>> loaded, etc. >>>> >>>> $ cat test.st >>>> #!/Users/sven/tmp/pharo4/pharo-vm/Pharo.app/Contents/MacOS/Pharo >>>> --headless /Users/sven/tmp/pharo4/Pharo.image st --quit >>>> FileStream stdout nextPutAll: 42 factorial asString; cr. >>>> >>>> $ time ./test.st >>>> 1405006117752879898543142606244511569936384000000000 >>>> >>>> real 0m0.644s >>>> user 0m0.516s >>>> sys 0m0.067s >>>> >>>> So about half a second overhead. >>>> >>>> Sven >>>> >>>> >>> >>> >> >> >> > >

July 25, 2014