- Pharo-users - lists.pharo.org

Re: [Pharo-users] Usability issues with Calypso

by kmo

Thanks for the information. I'm sure Calypso is going to be brilliant. -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Sept. 2, 2017

Re: [Pharo-users] Usability issues with Calypso

by Denis Kudriashov

Hi 2017-09-01 14:26 GMT+02:00 kmo <voxkmp(a)gmail.com>: > I have been keen to use Calypso (I love the way it shows inherited methods) > as my browser but I am finding some usability issues. It may be just me but > some things I find off-putting. > Any feedback is very important. > > I don't really like the way that adding a new class is done by a new tab > that appears now and then in the editing area. To my mind the Nautilus way > of launching a dialog to create a class seems more natural - and opens the > way in future for a class creation wizard - something that can't be done in > a code window. If we have to do it in a code tab then it would be better if > the tab remained there all the time - I don't like the way it appears and > disappears. It takes a while to learn what to click to do this. > In fact current behaviour is not replacement of old context menu command. It is just rethinking of classic single source code pane of old browsers. So we will add missing command to context menu. But I think the command should open the current "New class" tab instead of old dialog request. Also old browser behaviour is still working. You can create new class from the class editor tab. Just type new class name in the definition of selected class. > > Also, all class creation presumes a subclass of Object in Calypso. i miss > the ability in Nautilus to select a class and immediately create a subclass > of it. Am I missing something here? Is there a way? > You are right. It is missing command like previous one. We will add it. > > On the topic of my ignorance - what is a /project /in Calypso? How do you > create one? > Now it is more like a stub. There is integration with new package management Cargo (Christophe is working on it). But it is probably not ready yet. I think we should also implement Metacello backend. Anyway project mode will become default view in future. > And what does realize do? > It creates stub implementation of abstract methods (found in superclasses). I think in Calypso this command is not needed anymore because there is "should be implemented" method group which shows all these abstract methods. > > Also I think that the pop up menus could be subdivided into categories > rather than present all the options in an unbroken list. And the order of > options seems odd. To my mind/ senders /and /implementers /are two menu > options that should always go together - not be separated by/ rename/ and > /move to class side/ et cetera. > Yes, I will add it. You can report issues directly at github project https://github.com/dionisiydk/Calypso/issues. Thank's for you feedback. > > Just my two cents. > > > > > > -- > Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html > >

Sept. 2, 2017

Re: [Pharo-users] Pharo 6.0 Generating Ruby-on-Rails Applications

by Brad Selfridge

Hi Peter, Really good work. I'm impressed. Pharo is really gaining some cutting edge options. I would like to know a lot more about your framework. Thanks for the effort. ----- Brad Selfridge -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Sept. 2, 2017

[ann] moldable editor - documenter and examples without pragma magic

by Tudor Girba

Hi, When we announced the GT Documenter as an application for the moldable editor, one of the remarks was that GT Examples implementation relies too much on pragma magic. We are now happy to announce that this is no longer the case because Andrei did a great job and rebuilt examples without the need for static pragmas for specifying dependencies. So, now we can example dependencies simply through normal calls: emptyView <gtExample> ^ GtMondrian new twoNodes <gtExamples> view := self emptyView view nodes with: {1 . 2} ... and the engine can figure out the static dependencies. This means that examples can be used both as plain unary methods, and as replacement for tests. We think this should address the worry of examples being too heavy (the current implementation is ~350 lines). With a bit of extra effort, we also adapted GT Documenter to expand the example dependencies as seen below. This means that examples can now be used directly as units of documentation as well. We think this is significant. Cheers, The feenk team -- www.tudorgirba.com www.feenk.com "We are all great at making mistakes."

Sept. 1, 2017

Small background picture extension for DesktopManager

by Torsten Bergmann

Hi, In latest DesktopManager (available from catalog) package you can now also easily choose a background picture from the "Bing.com" collection in a custom spotter and apply it to the current desktop. See the attached screenshot for an example. The magic key combination is to hit CTRL+ D (like Desktop) followed by CTRL+B (like Background) There is also CTRL + D followed by CTRL + DââââDesktop Displayâ to display all desktops CTRL + D followed by CTRL + AââââDesktop Addâ to add a new desktop CTRL + D followed by CTRL + NââââDesktop Nextâ to go to next desktop CTRL + D followed by CTRL + PââââDesktop Previousâ to go to previous desktop Have fun T.

Sept. 1, 2017

Re: [Pharo-users] Pharo 6.0 Generating Ruby-on-Rails Applications

by H. Hirzel

Thank you for sharing, Peter. Good that you 'rediscovered' Smalltalk! Most interesting point of what you write <citation> Ruby-on-Rails, of course, has built-in generators that can create data models, controllers, views, etc. And there are also more sophisticated application generators such as Yeoman. These generators are based on templates and are great as far as they go. Where my Smalltalk approach is different is the level of abstraction. The Smalltalk models are not bound to any particular target language and could just as easily be used to generate PHP Laravel, NodeJS, or Java applications. </citation> This is similar to a Pillar document object model. A object model (an app or a document) may be inspected in the dynamic Smalltalk environment and rendered in different ways. Regards Hannes On 9/1/17, Peter Fisk <peter.fisk(a)gmail.com> wrote: > Hi everyone,The latest version of my "Rails Express" rapid application > development environment is built in Pharo 6.0.Here is a blog post with a > few > screenshots and some explanation. > https://railsexpress.quora.com/Pharo-6-Generating-Ruby-on-Rails-Applications > <https://railsexpress.quora.com/Pharo-6-Generating-Ruby-on-Rails-Applications> > > I mostly programmed in Smalltalk from 1987 until 1997 using ParcPlace > VisualWorks and then IBM VisualAge.Pharo is now better than either of those > two frameworks.Congratulations to the Pharo team for all your great > work!Regards,-- Peter Fisk > > > > -- > Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Sept. 1, 2017

Pharo 6.0 Generating Ruby-on-Rails Applications

by Peter Fisk

Hi everyone,The latest version of my "Rails Express" rapid application development environment is built in Pharo 6.0.Here is a blog post with a few screenshots and some explanation. https://railsexpress.quora.com/Pharo-6-Generating-Ruby-on-Rails-Applications <https://railsexpress.quora.com/Pharo-6-Generating-Ruby-on-Rails-Applications> I mostly programmed in Smalltalk from 1987 until 1997 using ParcPlace VisualWorks and then IBM VisualAge.Pharo is now better than either of those two frameworks.Congratulations to the Pharo team for all your great work!Regards,-- Peter Fisk -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Sept. 1, 2017

Re: [Pharo-users] Usability issues with Calypso

by Stephane Ducasse

Hi KMO thanks for your feedback. Calypso is still an enfant and this is IMPORTANT that people tell what they do not like so that we can improve. Stef On Fri, Sep 1, 2017 at 2:26 PM, kmo <voxkmp(a)gmail.com> wrote: > I have been keen to use Calypso (I love the way it shows inherited methods) > as my browser but I am finding some usability issues. It may be just me but > some things I find off-putting. > > I don't really like the way that adding a new class is done by a new tab > that appears now and then in the editing area. To my mind the Nautilus way > of launching a dialog to create a class seems more natural - and opens the > way in future for a class creation wizard - something that can't be done in > a code window. If we have to do it in a code tab then it would be better if > the tab remained there all the time - I don't like the way it appears and > disappears. It takes a while to learn what to click to do this. > > Also, all class creation presumes a subclass of Object in Calypso. i miss > the ability in Nautilus to select a class and immediately create a subclass > of it. Am I missing something here? Is there a way? > > On the topic of my ignorance - what is a /project /in Calypso? How do you > create one? > > And what does realize do? > > Also I think that the pop up menus could be subdivided into categories > rather than present all the options in an unbroken list. And the order of > options seems odd. To my mind/ senders /and /implementers /are two menu > options that should always go together - not be separated by/ rename/ and > /move to class side/ et cetera. > > Just my two cents. > > > > > > -- > Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html >

Sept. 1, 2017

Re: [Pharo-users] Code mystery

by Richard Sargent

On Fri, Sep 1, 2017 at 11:39 AM, PBKResearch <peter(a)pbkresearch.co.uk> wrote: > But surely all the code critic does is to ask a question â it is actually > displayed with the question mark. In many cases the answer is âYes, it is > OKâ â e.g. if the comment is âExcessive number of methods?â In the case > under discussion, it seems legitimate to ask: âDid you really mean to > declare the argument and then not use it?â The programmer will presumably > know if the complex situation you describe applies here, and then ignore > the message. > I don't know if you are writing as if the code critic does exist and doesn't work as you would wish or doesn't exist. However, from observation, I've concluded that Brant and Roberts tried to avoid too low of signal to noise ratio. They may have never considered this scenario. Obviously, Pharo is free to do as they wish. Feel free to add the missing code critic. As you suggest, you might be able to use the existing Unused Temporary Variable code critic as your basis. (But do create a new one, rather than make the existing one responsible for two different things.) > > Peter Kenny > > > > *From:* Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] *On > Behalf Of *Richard Sargent > *Sent:* 01 September 2017 15:04 > *To:* Any question about pharo is welcome <pharo-users(a)lists.pharo.org> > *Subject:* Re: [Pharo-users] Code mystery > > > > I think the answer my lie in the general case and no one tried to address > the complexity of the problem. > > > > Consider an API that's implemented in a number of classes, classes which > may not be in a hierarchy. Let's use #ifNil: as an example. > > > > Some implementers use the argument and others don't, so you cannot > criticise the methods that don't. > > > > It's much more difficult to recognize whether methods with the same name > actually implement the same API or are entirely independent of each other. > > > > That being said, one could quite easily create a criticized for the > special case of a single method, lacking any other implementations, that > doesn't use it's argument. > > > > On Aug 31, 2017 11:55, "PBKResearch" <peter(a)pbkresearch.co.uk> wrote: > > Stef > > Done - https://pharo.fogbugz.com/f/cases/20363/ThemeIcons- > downloadTo-has-an-argument-which-is-never-referenced-in-code > > What really puzzled me is that this situation was not picked up by a code > critic. 'Local variable declared and not referenced' is a common critic > message, and this looks like a parallel situation. But I don't think this > can be called a bug in the code critic system - can it? > > Peter Kenny > > -----Original Message----- > From: Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] On Behalf > Of Stephane Ducasse > Sent: 31 August 2017 19:16 > To: Any question about pharo is welcome <pharo-users(a)lists.pharo.org> > Subject: Re: [Pharo-users] Code mystery > > Can you open a bug entry? > I can tell you that we clean a lot but human activity will always leads to > glitches. > > On Tue, Aug 29, 2017 at 1:23 AM, Dimitris Chloupis <kilon.alios(a)gmail.com> > wrote: > > You are not going crazy that's an ugly method. Probably the author > > intended to use dir and the forgot about it and instead hard coded the > > dir path inside a class method. A mistake that can happen to anyone. > > > > The method will have to be updated anyway because name is to be > > removed, so it won't work. > > > > As always too much code too few people. If you think that's bad > > embrace yourself if you try to read Morphic code. Huge suffering for > > me when I tried to learn how the task bar works and apparently using > > it wrong it freezes the image. > > > > Tons of Pharo code needs a clean but needs also a lot more people. > > On Tue, 29 Aug 2017 at 01:52, PBKResearch <peter(a)pbkresearch.co.uk> > wrote: > >> > >> Hello All > >> > >> > >> > >> Following the discussion on dark mode, I was browsing the code on > >> themes (in Moose 6.1 = Pharo 6.0, Latest update: #60486). In Class > >> ThemeIcons, I found this method: > >> > >> > >> > >> downloadTo: dir > >> > >> | zipArchive | > >> > >> > >> > >> zipArchive := self class destinationPath / (self name, > >> '.zip'). > >> > >> zipArchive exists > >> > >> ifFalse: [ > >> > >> ZnClient new > >> > >> url: self > >> url; > >> > >> downloadTo: > >> zipArchive ]. > >> > >> > >> > >> ^ zipArchive > >> > >> > >> > >> The mystery is that the argument dir is not referred to anywhere in > >> the code. It probably works, because the only invocation of the > >> method is from > >> ThemeIcons>>downloadFromUrl, which sets the argument from self class > >> destinationPath, and the code above recreates this as the path to > >> zipArchive. > >> > >> > >> > >> I thought I understood Smalltalk coding fairly well, but this really > >> puzzles me. Why would anyone code like this? Shouldnât it be picked > >> up by a code critic? Or am I going crazy? > >> > >> > >> > >> Any help gratefully received > >> > >> > >> > >> Peter Kenny > >

Sept. 1, 2017

Re: [Pharo-users] Code mystery

by PBKResearch

But surely all the code critic does is to ask a question â it is actually displayed with the question mark. In many cases the answer is âYes, it is OKâ â e.g. if the comment is âExcessive number of methods?â In the case under discussion, it seems legitimate to ask: âDid you really mean to declare the argument and then not use it?â The programmer will presumably know if the complex situation you describe applies here, and then ignore the message. Peter Kenny From: Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] On Behalf Of Richard Sargent Sent: 01 September 2017 15:04 To: Any question about pharo is welcome <pharo-users(a)lists.pharo.org> Subject: Re: [Pharo-users] Code mystery I think the answer my lie in the general case and no one tried to address the complexity of the problem. Consider an API that's implemented in a number of classes, classes which may not be in a hierarchy. Let's use #ifNil: as an example. Some implementers use the argument and others don't, so you cannot criticise the methods that don't. It's much more difficult to recognize whether methods with the same name actually implement the same API or are entirely independent of each other. That being said, one could quite easily create a criticized for the special case of a single method, lacking any other implementations, that doesn't use it's argument. On Aug 31, 2017 11:55, "PBKResearch" <peter(a)pbkresearch.co.uk <mailto:peter@pbkresearch.co.uk> > wrote: Stef Done - https://pharo.fogbugz.com/f/cases/20363/ThemeIcons-downloadTo-has-an-argume… What really puzzled me is that this situation was not picked up by a code critic. 'Local variable declared and not referenced' is a common critic message, and this looks like a parallel situation. But I don't think this can be called a bug in the code critic system - can it? Peter Kenny -----Original Message----- From: Pharo-users [mailto:pharo-users-bounces@lists.pharo.org <mailto:pharo-users-bounces@lists.pharo.org> ] On Behalf Of Stephane Ducasse Sent: 31 August 2017 19:16 To: Any question about pharo is welcome <pharo-users(a)lists.pharo.org <mailto:pharo-users@lists.pharo.org> > Subject: Re: [Pharo-users] Code mystery Can you open a bug entry? I can tell you that we clean a lot but human activity will always leads to glitches. On Tue, Aug 29, 2017 at 1:23 AM, Dimitris Chloupis <kilon.alios(a)gmail.com <mailto:kilon.alios@gmail.com> > wrote: > You are not going crazy that's an ugly method. Probably the author > intended to use dir and the forgot about it and instead hard coded the > dir path inside a class method. A mistake that can happen to anyone. > > The method will have to be updated anyway because name is to be > removed, so it won't work. > > As always too much code too few people. If you think that's bad > embrace yourself if you try to read Morphic code. Huge suffering for > me when I tried to learn how the task bar works and apparently using > it wrong it freezes the image. > > Tons of Pharo code needs a clean but needs also a lot more people. > On Tue, 29 Aug 2017 at 01:52, PBKResearch <peter(a)pbkresearch.co.uk <mailto:peter@pbkresearch.co.uk> > wrote: >> >> Hello All >> >> >> >> Following the discussion on dark mode, I was browsing the code on >> themes (in Moose 6.1 = Pharo 6.0, Latest update: #60486). In Class >> ThemeIcons, I found this method: >> >> >> >> downloadTo: dir >> >> | zipArchive | >> >> >> >> zipArchive := self class destinationPath / (self name, >> '.zip'). >> >> zipArchive exists >> >> ifFalse: [ >> >> ZnClient new >> >> url: self >> url; >> >> downloadTo: >> zipArchive ]. >> >> >> >> ^ zipArchive >> >> >> >> The mystery is that the argument dir is not referred to anywhere in >> the code. It probably works, because the only invocation of the >> method is from >> ThemeIcons>>downloadFromUrl, which sets the argument from self class >> destinationPath, and the code above recreates this as the path to >> zipArchive. >> >> >> >> I thought I understood Smalltalk coding fairly well, but this really >> puzzles me. Why would anyone code like this? Shouldnât it be picked >> up by a code critic? Or am I going crazy? >> >> >> >> Any help gratefully received >> >> >> >> Peter Kenny

Sept. 1, 2017