February 2018 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] printString of fileReference sucks!

by Stephane Ducasse

Tx I will reread the threads! On Sat, Feb 3, 2018 at 4:07 AM, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > Stephane Ducasse-3 wrote >> What do you think? > > If it helps, here is some background materialâ¦ > > - The original issue where we (I) created this behavior, along with the > rationale: > https://pharo.fogbugz.com/f/cases/7366/FileReferences-have-a-confusing-wron… > - A subsequent ML thread where we discussed this fairly extensively: > http://forum.world.st/Why-aFileReference-asString-aFileReference-fullName-i… > > FTR I am not attached to the current solution and defer to the majorityâ¦ > > > > ----- > Cheers, > Sean > -- > Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html >

Feb. 3, 2018

Re: [Pharo-dev] printString of fileReference sucks!

by Sean P. DeNigris

Stephane Ducasse-3 wrote > What do you think? If it helps, here is some background materialâ¦ - The original issue where we (I) created this behavior, along with the rationale: https://pharo.fogbugz.com/f/cases/7366/FileReferences-have-a-confusing-wron… - A subsequent ML thread where we discussed this fairly extensively: http://forum.world.st/Why-aFileReference-asString-aFileReference-fullName-i… FTR I am not attached to the current solution and defer to the majorityâ¦ ----- Cheers, Sean -- Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html

Feb. 3, 2018

Re: [Pharo-dev] MCGitlabRepository Missing

by Sean P. DeNigris

Pavel Krivanek-3 wrote > The PR was against a wrong branch. The fixed one is being validated. The class has appeared in the latest built image! The only thing left for this to work is to update Iceberg (dev-0.6 branch) where the other half of the patch lives :) ----- Cheers, Sean -- Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html

Feb. 3, 2018

Re: [Pharo-dev] Disabling a MenuGroupModel

by HernÃ¡n Morales Durand

https://pharo.fogbugz.com/f/cases/21233/Disabling-a-MenuGroupModel-bug 2018-02-02 10:10 GMT-03:00 HernÃ¡n Morales Durand <hernan.morales(a)gmail.com>: > Hi guys, > > If you develop an application window with multiple grouped menu items, > you will certainly be able to enable/disable menu groups at once. > Currently it is not possible with MenuGroupModel (Morphic). > > - A menu (item or group) model manages its state using enabledHolder > instance variable (the enabledHolder is configured in its superclass' > initialize) > -- Being a value holder, one might think there is a propagation of > state (enabled/disabled) changes to subwidgets, but it does not happen > currently. > --- Longer explanation: On static menu building, MenuGroupModel does > not propagate #enable: state changes to its widget because "widget" > i.v. is nil. The reason is a MenuGroupModel #enabled: is evaluated > BEFORE its widget is built. > > How to reproduce: > | mm | > mm := MenuModel new > addGroup: [ :group | > group > addItem: [ :item | > item > name: 'File'; > icon: #openIcon asIcon; > subMenu: (MenuModel new > addGroup: [ : gp | > gp > disable; " <------ HERE we > disable the group " > addItem: [ : it | > it > name: 'Open'; > icon: #openIcon asIcon; > shortcut: $o meta; > action: [ self inform: 'Open' ] ]; > addItem: [ : it | > it > name: 'Save'; > icon: #smallSaveIcon asIcon; > shortcut: $s meta; > action: [ self inform: 'Save' ] ] ]) ] ]; > yourself. > mm extent: 200 @ 100. > mm openWithSpec. > mm inspect. > > (it doesn't matter if #disable is sent after the #addItem: block evaluation) > > - Even after all MenuGroupModel widgets were built, if you try to > disable a menu group dinamically: > > mm menuGroups anyOne menuItems anyOne subMenu menuGroups anyOne disable. > > you would get "Instance of OrderedCollection did not understand > #enabled:" (the widget of a MenuGroupModel is an OrderedCollection!!). > > - However disabling/enabling MenuItemModel works: > > mm menuGroups anyOne menuItems anyOne subMenu menuGroups anyOne > menuItems first disable. > > - MenuItemModel gets actually disabled in the > ToggleMenuItemMorph/UpdatingMenuMorph (#isEnabled:) when the Canvas > draws it. > > - Incredibly, all disabled groups are re-enabled from several places > in the image. > > -- When the menu is clicked, a ToolDockingBarMorph triggers > re-enablement of the MenuMorph in > #removeMatchString/#displayFiltered:. This is presumably the code > triggered when a menu is displayed and you type to filter the menu > items. Something like this: "As initially there is no matchString in > the activeSubMenu, then enable all menu items". Obviously it cleans > disabled items. > > The message flow is: > > DockingBarMorph>>activeSubmenu: -> MenuMorph>>removeMatchString -> > MenuMorph>>displayFiltered: --> m isEnabled: isMatch > > - Another place where menu items are automatically disabled is this flow: > > FormCanvas>>draw: -> ToggleMenuItemMorph>>drawOn: --> > ToggleMenuItemMorph>>isEnabled --> ToggleMenuItemMorph>>isEnabled: !!! > --> MorphicMenuItemAdapter>>enabled (This actually CHANGES the > enabled state on the Model, disconnecting enabled state between the > morph and the model adapter) > > This seem to have multiple paths of resolution. I'm going to propose a > fix that I tested in Pharo 6.1 and Pharo 7. > I don't know yet how to fix the menu search auto-enable on key-press > but would be glad to read a solution to distinguish between both uses. > My idea for now is to have a Morphic property which sets > #doNotAutoReEnableItems. > > However I will open an issue and let people check. > > You can test it with > > " Disable the group " > mm menuGroups anyOne menuItems anyOne subMenu menuGroups anyOne disable. > > " Enable the group " > mm menuGroups anyOne menuItems anyOne subMenu menuGroups anyOne enable. > > Cheers, > > HernÃ¡n

Feb. 2, 2018

Re: [Pharo-dev] Weird ZnClient benchmarking result

by Sven Van Caekenberghe

https://pharo.fogbugz.com/f/cases/21232/BenchmarkResult-printFrequencyOn-ca… https://github.com/pharo-project/pharo/pull/794 > On 12 Dec 2017, at 07:22, Stephane Ducasse <stepharo.self(a)gmail.com> wrote: > > sven did you submit a fix to get a clearer printout? > > Stef > > > On Sun, Dec 10, 2017 at 7:14 PM, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: >> @ Henrik >> >> Arg, arg, arg, yes of course. >> I actually did think about that, but still did not look carefully enough !! >> Really stupid of me. >> >> It is even worse because I wrote that code ... >> >> Now, the #printOn: of BenchmarkResult is much clearer, less confusing: >> >> ZnClient new in: [ :client | >> [ client get: 'http://localhost:8080' ] benchFor: 5 seconds ]. >> >> a BenchmarkResult(14,732 iterations in 5 seconds. 2,946 per second) >> >> vs >> >> ZnClient new in: [ :client | >> client loggingOff. >> [ client get: 'http://localhost:8080' ] benchFor: 5 seconds ]. >> >> a BenchmarkResult(42 iterations in 5 seconds 88 milliseconds. 8.255 per second) >> >> Still the same issue with $, and $. in frequency, but the iteration count is crystal clear. >> >> The reason #bench works like that is backwards compatibility at the time we introduced that. >> >> @ Ben >> >> Yes, I should have looked at the other side as well, to confirm things actually happened as I imagined them (the did not). >> >> Thx and sorry for the noise. >> >> Sven >> >>> On 10 Dec 2017, at 18:00, Henrik-Nergaard <draagren(a)outlook.com> wrote: >>> >>> Hi Sven, >>> >>> What you are seeing is most likely a $, vs $. issue. >>> See BenchmarkResult >>#printFrequenceOn: it uses both decimal and thousand >>> separators. >>> >>> If i run this code: >>> ----------------------------------------- >>> | counter | >>> counter := >>> ZnClient new in: [ :client | >>> client loggingOff. >>> [ client get: 'http://localhost:8080'. counter ] bench >>> ]. >>> counter. >>> ----------------------------------------- >>> >>> Then i get 9035 ('1,773 per second') when no inspector is open, and only 22 >>> ('5.995 per second') when inspecting the logs. >>> >>> Best regards, >>> Henrik >>> >>> >>> >>> >>> >>> >>> >>> >>> -- >>> Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html >>> >> >> >

Feb. 2, 2018

Re: [Pharo-dev] Change Pharo window icon

by Eliot Miranda

Hi Stef, On Fri, Feb 2, 2018 at 11:32 AM, Stephane Ducasse <stepharo.self(a)gmail.com> wrote: > Would it be possible to change the icon without having to compile a new VM? > Through an AppleScript: e.g. https://www.sethvargo.com/replace-icons-osx/ > To me the current setup looks so monolithic and dated. > I would expect that somebody can deploy a application with its own > logo by just providing some new resources. > And that's indeed possible. Take the .app bundle and change the .icns file as described above. > > Stef > > On Wed, Jan 31, 2018 at 11:17 PM, Eliot Miranda <eliot.miranda(a)gmail.com> > wrote: > > Hi Vincent, > > > > On Wed, Jan 31, 2018 at 12:07 PM, <Vincent.Blondeau(a)lamresearch.com> > wrote: > >> > >> Hi, > >> > >> I would like to know how to change the main Pharo window icon. I saw > that > >> there is a primitive for the title: > >> DisplayScreen class >> primitiveWindowTitle:string: ; primitive: > >> 'primitiveHostWindowTitle' module:'HostWindowPlugin' > >> but I cannot find one for the icon. > >> Should another primitive need to be implemented? > > > > > > The icon is baked into the VM executable. So to change it one has to > build > > a VM with a different icon. > > > > On Windows it is in > > {build.win32x86,build.win64x64}/pharo.cog.spur/Pharo.ico > > and the file that specifies to use Pharo.ico is > > {build.win32x86,build.win64x64}/pharo.cog.spur/Pharo.rc > > > > On Mac OS X it is in > > platforms/iOS/vm/OSX/Pharo.icns > > (alongside three others such as PharoImage.icns) > > and the file that specifies to use Pharo.icns is > > {build.macos32x86,build.macos64x64}/pharo.cog.spur/Makefile > > in setting the VM variable. The file that associates the other icons > with > > specific file types is > > platforms/iOS//vm/OSX/Pharo-Info.plist > > > > If you're changing the Pharo icon let me suggest you update the icon > files > > themselves in the opensmalltalk-vm source tree. > > > > If you're creating a new variant of the VM for some new purpose (say a > > special Lam VM) then let me suggest you add the icons to the > > opensmalltalk-vm source tree, creating special build directories for > these > > VMs, such as {build.macos32x86,build.macos64x64}/lam.pharo.cog.spur/ > > > > If you want to do this privately, then take either of the approaches > above > > and simply don't publish the edits. You can write a script that takes an > > updated checked-out opensmalltalk-vm source tree and edits it with files > > from a specific repository. I have such scripts and can help you with > them. > > Hint, pax is a very convenient directory hierarchy copying tool > available at > > least on Mac OS X. pax -rwlk will copy the trees under a sequence of > > directories into their corresponding places in a target tree. > > > >> > >> Thanks! > >> > >> Vincent > >> > >> > > > > > > > > -- > > _,,,^..^,,,_ > > best, Eliot > > -- _,,,^..^,,,_ best, Eliot

Feb. 2, 2018

Re: [Pharo-dev] another file reference question

by Alistair Grant

Hi Stef, On 2 February 2018 at 20:58, Stephane Ducasse <stepharo.self(a)gmail.com> wrote: > Then I do not get why / is expecting a string and cannot accept a path. > > Then I do not get why we have paths. > Stef Paths are intended to be internal, and not something you ever deal with directly. I'm not sure I understand what you want to do, but maybe try replacing #/ with #resolve:, e.g.: '/home' asFileReference resolve: 'user' asFileReference " File @ /home/user" or: '/home' asFileReference resolve: 'user' asFileReference path " File @ /home/user" Cheers, Alistair > On Fri, Feb 2, 2018 at 8:22 PM, Stephane Ducasse > <stepharo.self(a)gmail.com> wrote: >> HI >> >> I have the following scenario: >> >> '/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk-Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result' >> asFileReference / >> 'book.pillar' >> >>>>> >> "File @ /Users/ducasse/Workspace/FirstCircle/MyBooks/Bk-Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result/book.pillar" >> >> ok it works. >> >> I want to update the contents of a target folder only if a file is >> missing or too old compared to a source. >> >> The idea is that I have a target folder and I need to know if a file >> located in the source tree should be copied under the target. >> So I compute the difference from the root of the source to the file >> and I want to apply this path to the target to check some file >> properties. >> >> now I do not understand how I can get the following working: >> >> '/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk-Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result' >> asFileReference >> / >> >> ('/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk-Writing/PharoBooks/Booklet-AMiniSchemeInPharo/book.pillar' >> asFileReference path) >> relativeTo: ('/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk-Writing/PharoBooks/Booklet-AMiniSchemeInPharo' >> asFileReference path) >> >> >> Stef >

Feb. 2, 2018

Re: [Pharo-dev] another file reference question

by Denis Kudriashov

Hi Can you provide more simple example with shorter paths? I am lost in these long strings :). 2018-02-02 20:22 GMT+01:00 Stephane Ducasse <stepharo.self(a)gmail.com>: > HI > > I have the following scenario: > > '/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result' > asFileReference / > 'book.pillar' > > >>> > "File @ /Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result/book.pillar" > > ok it works. > > I want to update the contents of a target folder only if a file is > missing or too old compared to a source. > > The idea is that I have a target folder and I need to know if a file > located in the source tree should be copied under the target. > So I compute the difference from the root of the source to the file > and I want to apply this path to the target to check some file > properties. > > now I do not understand how I can get the following working: > > '/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result' > asFileReference > / > > ('/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/book.pillar' > asFileReference path) > relativeTo: ('/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo' > asFileReference path) > > > Stef > >

Feb. 2, 2018

Re: [Pharo-dev] printString of fileReference sucks!

by Denis Kudriashov

2018-02-02 20:39 GMT+01:00 Esteban A. Maringolo <emaringolo(a)gmail.com>: > If for some reason the option chosen is to keep the current > printString then the printString should be: > >>> 'File @ ''/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result''' > So I am for this approach

Feb. 2, 2018

Re: [Pharo-dev] printString of fileReference sucks!

by Denis Kudriashov

Hi. I do not think that getting simple string as result is a good idea. It is important to easily distinguish class of object using printIt operation and as a field in raw inspector. During code migration to FileSystem it is classic way to check variables in debugger to see if they are already file references or they are still strings. printIt is very handy in such cases. Best regards, Denis 2018-02-02 20:39 GMT+01:00 Esteban A. Maringolo <emaringolo(a)gmail.com>: > Hi Stef, all, > > I've seen this "inconsistency" as well. The current printSting tries > to be descriptive but isn't handy to work with. > > I would go one step further and have no #asFileReference suffix nor > File @ prefix, it is: > > '/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result' > asFileReference printString > >>> '/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result' > > > If for some reason the option chosen is to keep the current > printString then the printString should be: > >>> 'File @ ''/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result''' > > It is, a string with three tokens, the class name, the @ symbol and a > string as an argument, that way you can evaluate the result back. > > I prefer the first one, it enables you to work with files on > workspaces, if you're using inspectors the printString doesn't matter > much, since you're dealing with a specialized tool and there is no way > you can confuse a FileReference with a regular String instance > > Best regards, > Esteban A. Maringolo > > > 2018-02-02 16:12 GMT-03:00 Stephane Ducasse <stepharo.self(a)gmail.com>: > > Hi > > > > '/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result' > > asFileReference printString > >>>> "File @ /Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result" > > > > so either we add @ one File or we redefine printString one reference to > be > > > > '/Users/ducasse/Workspace/FirstCircle/MyBooks/Bk- > Writing/PharoBooks/Booklet-AMiniSchemeInPharo/_result' > > asFileReference so that we can do something with the printString > > result. > > > > What do you think? > > Guille is it what we talked together? > > > > Setf > > > >

Feb. 2, 2018