- Pharo-dev - lists.pharo.org

Annotating instance variables

by HernÃ¡n Morales Durand

I saw some interesting reflection progress in Pharo 5. I wonder if there is a feature to rank instance variables. For example, from the browser I want to assign a category "important" to an instance variable, or "possibly useless", etc. for later refactorings. Is such thing "easily" possible now? There is API support from Reflectivity maybe? Cheers HernÃ¡n

Sept. 11, 2015

Re: [Pharo-dev] [Reflectivity] Progress with Reifications

by Marcus Denker

> On 11 Sep 2015, at 17:42, Craig Latta <craig(a)netjam.org> wrote: > > > "Reificated" is not a word. The word that means what you want is > "reified". > > Yes, I saw that just after pressing âSendââ¦ should have send a correction. Marcus

Sept. 11, 2015

Re: [Pharo-dev] [Reflectivity] Progress with Reifications

by Marcus Denker

> On 11 Sep 2015, at 17:28, Gustavo Santos <gugajansen(a)gmail.com> wrote: > > Hello Marcus, > > I wanted to modify some event creation in Epicea and I just used this mechanism. > It's pretty cool! > Very good that is was helpful! If you find any problems, let me knowâ¦ there will be many cases where things are not yet working as they should. But it seems to get more stable with each that we fix. Marcus

Sept. 11, 2015

Re: [Pharo-dev] Running a test method in Nautilus

by Marcus Denker

the left hand side of the method name. If the test fail, it does not open a debugger. I need to run it a second time. This is a rather odd behavior. > Am I the only one to experience this? > > Side question: > I try to check on the bug tracker if there is something about that, but I was not able to find it. I entered in the search input field the tree keywords: test button nautilus > The result is https://pharo.fogbugz.com/f/search/?sSearchFor=test+button+nautilus > However, this indicates many closed issues. How can I filter out the close issues? > Yes, it is annoying that the default for the search is to search closed issues. You can, though, filter the search results. Press â+Refine Furtherâ, there are many things, amount them âOpen or Closedâ. Select Open. Marcus

Sept. 11, 2015

Re: [Pharo-dev] Running a test method in Nautilus

by Marcus Denker

> On 11 Sep 2015, at 16:37, Christophe Demarey <Christophe.Demarey(a)inria.fr> wrote: > > I do experience the same thing with Guille today. > Before, Nautilus ran twice the method for a failed test: one to get the status of the test and another one for the debugger. > Maybe in Nautilus cleanup, it was removed. > Yes, I removed that double runnning.. this was not a good idea. Now it is much simpler. And yes, it is better to directly open the debugger when running single tests explicitly. I will fix it. Marcus

Sept. 11, 2015

Re: [Pharo-dev] Catalog does not update to include MessageFlowBrowser

by Esteban Lorenzano

> On 11 Sep 2015, at 16:38, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > > >> On 11 Sep 2015, at 16:29, Torsten Bergmann <astares(a)gmx.de> wrote: >> >> Hi, >> >> ConfigurationOfMessageFlowBrowser is in http://smalltalkhub.com/#!/~Pharo/MetaRepoForPharo50 >> for a week now but still not included in http://catalog.pharo.org >> >> According to the description there is a daily catalog job - but so far no results. It is not >> included although the config includes the necessary catalog methods. >> >> - Is MetaRepoForPharo50 already recognized? >> - Is it possible to "manually" trigger the catalog generation in case a new project is done? >> >> Anyone able to look into this? > > Iâll check the logs. fixed. something went wrong when updating FFI (my fault :P) and then everything stop working. Of course there is a log who tells me that something went wrong, but I forget to check them :D cheers, > > Esteban > >> >> Thx >> T. >

Sept. 11, 2015

Re: [Pharo-dev] [Reflectivity] Progress with Reifications

by Craig Latta

"Reificated" is not a word. The word that means what you want is "reified". -C -- Craig Latta netjam.org +31 6 2757 7177 (SMS ok) + 1 415 287 3547 (no SMS)

Sept. 11, 2015

Re: [Pharo-dev] [Reflectivity] Progress with Reifications

by Gustavo Santos

Hello Marcus, I wanted to modify some event creation in Epicea and I just used this mechanism. It's pretty cool! Cheers, On Thu, Jun 18, 2015 at 2:36 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > Hi, > > in Pharo5 #120 we have lots of progress with Reflectivity regarding to > âreificated valuesâ. > > 1) more reificated value are working. > > #arguments - an array of the arguments of a message send or method > #receiver - receiver of a message (or method executed) > #context for Methods > #sender for method and message > > This means we have now all the info needed to e.g. pass all date needed to > do a message send on the meta level. > (receiver, args, selector), or a variable store (#name, #newValue). > > 2) Reifications as arguments to conditions. When the block used as a > condition has arguments, the name > is matched to the possible reifications and if valid this value is passed > as an argument. Very useful e.g. for > object specific links: > > link := MetaLink new > condition: [: object | object == <the object I am > intersted in>]; > metaObject: <someObject>; > selector: <some selector> > > This link would only be active for this one object. > > 3) Reifications as meta object > ------------- > > Every valid reified value can be the metaobject of the link. e.g. you can > now define a link where the meta is the node where the Link is installed: > > link := MetaLink new > metaObject: #node; > selector: #tagHasBeenExecuted. > > Or the #receiver: > > link := MetaLink new > metaObject: #receiver; > selector: #perform:withArguments:; > arguments: #(selector arguments). > > TODO > > -> #slot, #global. #variable reifications > -> #after needs to wrap in ensure: > -> primitives > -> more tests > -> optimize link preambles: when reifying both #arguments and #receiver, > merge the preamble into one for performance. > > -- Gustavo Santos

Sept. 11, 2015

Re: [Pharo-dev] Catalog does not update to include MessageFlowBrowser

by Esteban Lorenzano

> On 11 Sep 2015, at 16:29, Torsten Bergmann <astares(a)gmx.de> wrote: > > Hi, > > ConfigurationOfMessageFlowBrowser is in http://smalltalkhub.com/#!/~Pharo/MetaRepoForPharo50 > for a week now but still not included in http://catalog.pharo.org > > According to the description there is a daily catalog job - but so far no results. It is not > included although the config includes the necessary catalog methods. > > - Is MetaRepoForPharo50 already recognized? > - Is it possible to "manually" trigger the catalog generation in case a new project is done? > > Anyone able to look into this? Iâll check the logs. Esteban > > Thx > T.

Sept. 11, 2015

Re: [Pharo-dev] Running a test method in Nautilus

by Christophe Demarey

I do experience the same thing with Guille today. Before, Nautilus ran twice the method for a failed test: one to get the status of the test and another one for the debugger. Maybe in Nautilus cleanup, it was removed. Le 11 sept. 2015 Ã 16:34, Peter UhnÃ¡k a Ã©crit : > If the test fail, it does not open a debugger. > > This happens when I click on the button next to the _class_, however I've never experienced such problem with the button next to a _method_. > > Peter > > On Fri, Sep 11, 2015 at 4:21 PM, Alexandre Bergel <alexandre.bergel(a)me.com> wrote: > Hi! > > I can run a test method by pressing the button located on the left hand side of the method name. If the test fail, it does not open a debugger. I need to run it a second time. This is a rather odd behavior. > Am I the only one to experience this? > > Side question: > I try to check on the bug tracker if there is something about that, but I was not able to find it. I entered in the search input field the tree keywords: test button nautilus > The result is https://pharo.fogbugz.com/f/search/?sSearchFor=test+button+nautilus > However, this indicates many closed issues. How can I filter out the close issues? > > Alexandre > -- > _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: > Alexandre Bergel http://www.bergel.eu > ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > > > > >

Sept. 11, 2015