- Pharo-users - lists.pharo.org

Fwd: Highlighter PetitParser

by Jesus NuÃ±ez

This is a copy from an email I have sent but not sure is indexed in the pharo mailing list. If somebody sees it please let me know it. Best, Jesus NuÃ±ez ====================================================================================== Hello everybody, The following code highlights prolog code statically making use of PPTextHighlighter, I leave the code to share with the community since it might be of help to someone else. Could someone point out how can this be converted to its dynamic counterpart?; The keystroke handler for GLMTextPresentation and what model is currently being used for instance to highlight smalltalk code would be very much appreciated. >From http://bergel.eu/download/Dyla2010/dyla10_submission_4.pdf I know that it might be neccesary a fuzzy highlighter as is written at the end of the code, and it is probably the right object to use but I can not figure out where to plug it. Thank beforehand, Jesus | text textRenderer textShower | text := 'p(Q,R):-q(R), unify_st(X, smalltalk_code), r(T). q(Q):-d(R). q(Q).'. textRenderer := [ PPTextHighlighter new parser: PPPrologParser new; color: 'small_atom' with: Color blue muchDarker; bold: 'small_atom'; color: 'unify_st' with: Color green muchDarker; bold: 'unify_st'; color: 'string' with: Color gray muchDarker; color: 'number' with: Color gray muchDarker; color: 'boolean' with: Color gray muchDarker; highlight: text asText. ]. " __ | <Renders IN> | W GLMTextPresentation | <Renders IN> | W GLMMorphicRenderer " textShower := GLMMorphicRenderer new. (textShower open: ( GLMTextPresentation new display: textRenderer; renderGlamorouslyOn: textShower; yourself )) window title: 'Prolog Editor'. "Here I created a highlighter through the transform method of the grammar" grammar := PPPrologParser new. highlighter := grammar transform: [ :parser | Transcript show: parser. parser class = TokenParser ifTrue: [ parser ==> [ :token | textShower model highlight: token style range: token interval ] ] ifFalse: [ parser ] ]. text := 'p(Q,R):-q(R), unify_st(X, smalltalk_code), r(T). q(Q):-d(R). q(Q).'. pp := highlighter parse: text asText.

March 20, 2013

Re: [Pharo-users] new (learning) project - tiangle with draggable points on a screen

by Hilaire Fernandes

You should look at the Morph hierarchy. In the Pharo Collaborative and Pharo By Examples books there are chapters worth reading. You may also want to learn from DrGeo sour code but it is a hudge beast and probably not the easier path. Hilaire -- Dr. Geo http://drgeo.eu

March 20, 2013

Re: [Pharo-users] Why are SmallTalk projects/source code hosted on SS3 Gemstone instead of Git?

by Stephan Eggermont

Victor Stan wrote: > Why are SmallTalk projects/source code hosted on SS3 Gemstone instead of > Git/GitHub? Because we have been using DVCS since a few years before git. Monticello is not as powerful as git, but it is integrated and the difference is not large enough to force change yet. > I'm coming to SmallTalk from the world of web development with open source > software, primarily Rails, and I'm very familiar with the amazing social > network/source code repository that is GitHub. It is truly an industry > defining entity, so many open source projects have been able to harness the > ease of use, features and community around Git and GitHub. Our perspective is a bit different. We use Moose and see the work of Veronica Uquillas Gomez on Torch (http://soft.vub.ac.be/torch/Home.html) and ask ourselves why people continue using bad abstractions. > At the moment, as I'm trying to learn more about SmallTalk and Pharo > especially, (my primary interest in Pharo is to use it as a web development > platform), I am a bit shocked, if I may be frank, at the tooling used for > source code and open source project management. I see that the popular > trend now is to move to SS3/Gemstone, and I appreciate anyone that helps > open source development/projects, but I can't see how they can even come > close to the functionality of GitHub for source code hosting and OS project > management, so I pose the question: is there an effort, why or why not, to > start integrating with GitHub and Git for source code management? Because it isn't that good. It is better than what we now have widely deployed, but it doesn't come close to what we can do when we find the time to integrate all the nice research tools. GitHub is centralized and a single point of failure. We will see it replaced by a distributed system. So it is a question of how much effort we want to put into what for us can only be an intermediate tool chain and way to connect to non-smalltalk parts of the systems we develop. Stephan Eggermont

March 20, 2013

Re: [Pharo-users] suggestion for book code snippets

by stephane ducasse

not yet :) On Mar 19, 2013, at 10:23 PM, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > stephane ducasse wrote >> normally code snippet in the book that are tagged with a special tag are >> extracted and run as unit tests. >> This is long time that I did not try but this was working. > > Is there a Jenkins job? > > > > ----- > Cheers, > Sean > -- > View this message in context: http://forum.world.st/suggestion-for-book-code-snippets-tp4677338p4677420.h… > Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com. >

March 20, 2013

Re: [Pharo-users] Keymappings 101 tutorial

by stephane ducasse

cool :) I like it. Stef On Mar 19, 2013, at 6:56 PM, Guillermo Polito <guillermopolito(a)gmail.com> wrote: > Hi guys, > > I've written a little tutorial on Keymappings for Pharo 2.0. > > http://playingwithobjects.wordpress.com/2013/03/19/keymappings-101-for-phar… > > Hope it's useful :), > Guille

March 20, 2013

Re: [Pharo-users] Why are SmallTalk projects/source code hosted on SS3 Gemstone instead of Git?

by Camillo Bruni

On 2013-03-20, at 10:50, Davorin Rusevljan <davorin.rusevljan(a)gmail.com> wrote: > On Wed, Mar 20, 2013 at 2:58 AM, Victor Stan <victor.stan(a)gmail.com> wrote: > >> Why are SmallTalk projects/source code hosted on SS3 Gemstone instead of >> Git/GitHub? >> >> I'm coming to SmallTalk from the world of web development with open source >> software, primarily Rails, and I'm very familiar with the amazing social >> network/source code repository that is GitHub. It is truly an industry >> defining entity, so many open source projects have been able to harness the >> ease of use, features and community around Git and GitHub. >> >> At the moment, as I'm trying to learn more about SmallTalk and Pharo >> especially, (my primary interest in Pharo is to use it as a web development >> platform), I am a bit shocked, if I may be frank, at the tooling used for >> source code and open source project management. I see that the popular >> trend now is to move to SS3/Gemstone, and I appreciate anyone that helps >> open source development/projects, but I can't see how they can even come >> close to the functionality of GitHub for source code hosting and OS project >> management, so I pose the question: is there an effort, why or why not, to >> start integrating with GitHub and Git for source code management? >> >> I know that historical precent and the tools built into Pharo/SmallTalk >> images, like Monticello are predecessors to GUI source control, but given >> the leaps that Git has managed to take, in distributed source code >> management, how does the existing SmallTalk community feel about it's >> current tooling in this regard? >> >> > > First welcome! > > As others have said, there are efforts to use git as source control and I > am looking forward to their progress. As for slow adoption of git, I think > the following might be some of the reasons: > > 1) file based VCS, are a bit rude for Smalltalk, which operates on smaller > source granularity, and typicall Smalltalk based VCS handle that in more > natural way not true... and just for the record, Monticello is file-based too, it does not store any high-level information. If you store each method as separate file as done with filetree you get exactly the same granularity as monticello. see: https://github.com/pharo-project/pharo-core > 2) Monticello is distributed VCS, and I am not so sure that it compared to > git lacks many features that would be esseintial to Smalltalk developer. > (but since I am only causal user of both, I might be wrong on this) as a model monticello and git overlap pretty well, however many things are much more implicit in monticello than in git (namely branches).

March 20, 2013

new (learning) project - tiangle with draggable points on a screen

by sminni

Hi I have a small project (a real requirement) and I want a little guidance to do it (myself) - pl. assume I am new to pharo though I can navigate the IDE & have a background in O-O development. 1. Draw a triangle with draggable handles on points 2. The sides length and angles will be displayed (in editable fields) 3. We can lock any of the sides / angles 4. If the side / angle is edited then an action can be initiated to recomplete the triangle by redrawing the side clicked next Thanks Sanjay ----- --- Regards, Sanjay -- View this message in context: http://forum.world.st/new-learning-project-tiangle-with-draggable-points-on… Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

March 20, 2013

Re: [Pharo-users] Why are SmallTalk projects/source code hosted on SS3 Gemstone instead of Git?

by Davorin Rusevljan

On Wed, Mar 20, 2013 at 2:58 AM, Victor Stan <victor.stan(a)gmail.com> wrote: > Why are SmallTalk projects/source code hosted on SS3 Gemstone instead of > Git/GitHub? > > I'm coming to SmallTalk from the world of web development with open source > software, primarily Rails, and I'm very familiar with the amazing social > network/source code repository that is GitHub. It is truly an industry > defining entity, so many open source projects have been able to harness the > ease of use, features and community around Git and GitHub. > > At the moment, as I'm trying to learn more about SmallTalk and Pharo > especially, (my primary interest in Pharo is to use it as a web development > platform), I am a bit shocked, if I may be frank, at the tooling used for > source code and open source project management. I see that the popular > trend now is to move to SS3/Gemstone, and I appreciate anyone that helps > open source development/projects, but I can't see how they can even come > close to the functionality of GitHub for source code hosting and OS project > management, so I pose the question: is there an effort, why or why not, to > start integrating with GitHub and Git for source code management? > > I know that historical precent and the tools built into Pharo/SmallTalk > images, like Monticello are predecessors to GUI source control, but given > the leaps that Git has managed to take, in distributed source code > management, how does the existing SmallTalk community feel about it's > current tooling in this regard? > > First welcome! As others have said, there are efforts to use git as source control and I am looking forward to their progress. As for slow adoption of git, I think the following might be some of the reasons: 1) file based VCS, are a bit rude for Smalltalk, which operates on smaller source granularity, and typicall Smalltalk based VCS handle that in more natural way 2) Monticello is distributed VCS, and I am not so sure that it compared to git lacks many features that would be esseintial to Smalltalk developer. (but since I am only causal user of both, I might be wrong on this) 1+2 = there are a bit less reasons to switch than it appears on the first look, and since every switch requires energy, it did not happen yet. I am certainly looking forward to it :) , since Github visibility would be very beneficial. Once again welcome! Davorin Rusevljan http://www.cloud208.com

March 20, 2013

Re: [Pharo-users] Glorp + Postgres array type

by CHRIS BAILEY

Hi Guillermo, thanks for that. I'm using DBX for legacy reasons but there's no reason why it couldn't use native. I managed to get it working by looking at how some other DatabaseTypes worked but am not sure whether I'm doing it completely right. It won't perform a full toManyMapping so I'm just making do with directly mapping the id's and then doing another read in the post fetch to get at the joined objects. On a similar note, I noticed that Glorp isn't retrieving two objects with a composite SQL statement when you specify a one to one mapping in the descriptor. It is fetching the first and then leaving a Proxy in the second; which will only go to the database with a separate query when you actually ask for it. I have a few cases where having the objects fully resolved at startup might be better than doing each one on demand. Is there any better way to do this than a loop in the post fetch? It also caused me a little confusion because as soon as you do something such as put them into a Set, the = matching of a real Object against a new Proxy that comes along won't work unless you've explicity defined equality (Proxy = Object does in fact work because it does a getValue first) I was trusting Glorp to only allow one instance of a particular object and therefore expecting identity equality to be sufficient !PostgreSQLPlatform methodsFor: '*glorp-extensions' stamp: 'user 3/19/2013 05:13'! arrayConverter Â Â Â ^DelegatingDatabaseConverter Â Â Â Â Â Â named: #array Â Â Â Â Â Â hostedBy: self Â Â Â Â Â Â fromStToDb: #toArray:for: Â Â Â Â Â Â fromDbToSt: #readArray:for:.! ! !PostgreSQLPlatform methodsFor: '*glorp-extensions' stamp: 'user 3/19/2013 05:14'! arrayOf: elementType Â Â Â ^self typeNamed: #arrayOf, elementType asString ifAbsentPut: [GlorpPGArrayType new elementType: elementType ].! ! !PostgreSQLPlatform methodsFor: '*glorp-extensions' stamp: 'user 3/19/2013 05:13'! readArray: anObject for: aType Â Â Â anObject ifNil: [^nil]. Â Â Â ^((anObject copyFrom: 2 to: anObject size - 1) subStrings: ',') collect: [:each | aType elementType fromString: each]! ! !PostgreSQLPlatform methodsFor: '*glorp-extensions' stamp: 'user 3/19/2013 05:13'! toArray: anObject for: aType Â Â Â anObject ifNil: [^nil]. Â Â Â ^'{', anObject asCommaString, '}'! ! Â Â Â GlorpDatabaseType subclass: #GlorpPGArrayType Â Â Â instanceVariableNames: 'elementType' Â Â Â classVariableNames: '' Â Â Â poolDictionaries: '' Â Â Â category: 'Glorp-Extensions'! !GlorpPGArrayType methodsFor: 'as yet unclassified' stamp: 'user 3/19/2013 05:03'! converterForStType: aClass Â Â Â ^self platform converterNamed: #array.! ! !GlorpPGArrayType methodsFor: 'accessing' stamp: 'user 3/19/2013 05:05'! elementType Â Â Â ^ elementType! ! !GlorpPGArrayType methodsFor: 'accessing' stamp: 'user 3/19/2013 05:05'! elementType: anObject Â Â Â elementType := anObject! ! "-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- "! GlorpPGArrayType class Â Â Â instanceVariableNames: ''! !GlorpPGArrayType class methodsFor: 'as yet unclassified' stamp: 'user 3/19/2013 05:05'! of: elementType Â Â Â ^self new elementType: elementType! ! Â --- On Mon, 18/3/13, Guillermo Polito <guillermopolito(a)gmail.com> wrote: From: Guillermo Polito <guillermopolito(a)gmail.com> Subject: Re: [Pharo-users] Glorp + Postgres array type To: "A friendly place where any question about pharo is welcome" <pharo-users(a)lists.gforge.inria.fr> Cc: dbxtalk(a)googlegroups.com, glorp-group(a)googlegroups.com Date: Monday, 18 March, 2013, 11:41 Hi Chris, I'm not really sure Glorp supports that... but at least I can try to give a look at how the driver responds to it, so: - Which driver are you using? OpenDBX or the native postgres one?Â - can you provide me some script to setup the problem? I'm not really an expert on postgresql specifics :) On Sun, Mar 17, 2013 at 11:46 AM, Chris <cpmbailey(a)btinternet.com> wrote: Hi, does anyone know if Glorp supports the array type in Postgres? I did a basicExecuteSQLString: and it came back with '{1,2,3}' so as if it might not be too complicated to make work for my scenario at least. I really want to do a toManyMapping using these integers in the array as a join to another table and have it resolve their objects, but wouldn't know where to start! I'll forward this question to the glorp and DBXtalk lists also, probably someone else has more idea.Â Thanks Chris Tx,Guille

March 20, 2013

Re: [Pharo-users] Why are SmallTalk projects/source code hosted on SS3 Gemstone instead of Git?

by Camillo Bruni

On 2013-03-20, at 02:58, Victor Stan <victor.stan(a)gmail.com> wrote: > Why are SmallTalk projects/source code hosted on SS3 Gemstone instead of > Git/GitHub? > > I'm coming to SmallTalk from the world of web development with open source > software, primarily Rails, and I'm very familiar with the amazing social > network/source code repository that is GitHub. It is truly an industry > defining entity, so many open source projects have been able to harness the > ease of use, features and community around Git and GitHub. > > At the moment, as I'm trying to learn more about SmallTalk and Pharo > especially, (my primary interest in Pharo is to use it as a web development > platform), I am a bit shocked, if I may be frank, at the tooling used for > source code and open source project management. I see that the popular > trend now is to move to SS3/Gemstone, and I appreciate anyone that helps > open source development/projects, but I can't see how they can even come > close to the functionality of GitHub for source code hosting and OS project > management, so I pose the question: is there an effort, why or why not, to > start integrating with GitHub and Git for source code management? > > I know that historical precent and the tools built into Pharo/SmallTalk > images, like Monticello are predecessors to GUI source control, but given > the leaps that Git has managed to take, in distributed source code > management, how does the existing SmallTalk community feel about it's > current tooling in this regard? there are efforts being made, albeit slow ones, but eventually we will get there. 1. There is dale henrich's fileout format filetree so we can actually run any file-based versioning system ontop: https://github.com/dalehenrich/filetree This is already used in a couple of projects, for instance a regular fileout of the pharo core: https://github.com/pharo-project/pharo-core 2. There is a working (read and write) implementation of a git repository which can be loaded as a FileSystem in the image: FilesystemGit http://smalltalkhub.com/#!/~FileSystemGitDev/FileSystem-Git and a working image form here https://ci.inria.fr/pharo-contribution/job/FileSystem-Git/ What is missing? Mostly time to finish the full git-protocol so we can solve the current lack of communication in the implementation. If you are motivated, please join!

March 20, 2013