June 2018 - Pharo-users - lists.pharo.org

Re: [Pharo-users] Calypso - can we improve the bar of radio buttons?

by Ben Coman

On 25 June 2018 at 21:49, Tim Mackinnon <tim(a)testit.works> wrote: > Hi - I donât want to case a UI war, particularly as I was very impressed > with Denisâ presentation about Calypso at the PharoDays conference last > year. > I'm also very much enjoying Calypso. Especially the multiple tabs for the code pane and being able to peak at the class definition half way through editing a method. > Now that Iâve made a concerted effort to try and use Calypso, and in > general like it, I do however find that the bar of Radio Buttons above the > tabs is a bit overwhelming and Iâm wondering if there might be something we > can do to help make it less so? > > Iâm sure the problem is the age old observation that a checkbox or button > that toggles between states is difficult to understand whether its on or > off and what the alternate state is - so this is why we now have a row of 8 > radio buttons. > > But 8 is a lot of radio buttonsâ¦ which I find visually hard to parse and > often find the order of the buttons the opposite to what I expect - eg. > When I want to see class methods I keep pausing over that group to see > which one to click (for some reason I keep thinking the class one should be > before the instance one - not sure if anyone else notices this). > > Iâm also not overly keen on the terminology of âsideâ (as in class side, > vs inst. side) and equally I question abbreviations like: Hier. Inst. Refs. > > Can we not just get rid of that bar completely - and put little icon > buttons next to the filter field for pkg/project and flat/hierarchy ? > It would be interesting to see an experiment that replaced that bar with tabs above each pane: Pane 1: [Packages][Projects] tabs Pane 2: [Flat][Hierarchy] tabs Pane 3: [Instance][Class] tabs** Pane 4: [Method][Variables] tabs If 3[Instance] > 4[Variables] would show class-variables... instance-variables... and 3[Class] > 4[Variables] would show class-variables... class-instance-variables... this could help familiarize newcomers with the scope of these terms. cheers -ben > OR given the method pane is now a tree and it already has an entry > âinstance sideâ - why not just have entries for âinstance methodsâ and > âclass methodsâ and just expand accordingly? We also now have a tab âInst. > side methodâ - why not 2 tabs â+instance methodâ and â+class methodâ. It > breaks a bit with regular smalltalk browser conventions - but I think > Calypso has already pushed the boat out on all this anyway in favour of > improved navigation and ease of typing code. > > My understanding is that this is all pretty doable with the mechanism > Denis has put in place. Iâm tempted to have a look, but thought Iâd see > what that general feeling is here, as maybe Iâm on my own here. > > Tim >

June 25, 2018

Some facts and figures about pharo gaining momentum?

by Andrei Stebakov

Hi! Do we as a community have some proofs in numbers that the popularity of smalltalk grows? I saw some articles that have some proofs that smalltalk is on the rise, but for my presentation I'd like to back it up with some numbers if possible. You know, it's one thing when you preach to the choir but quite another when you talk to some die-hard Java or JavaScript devs! Any ideas about some great convincing material is also very welcome!

June 25, 2018

Re: [Pharo-users] Calypso - how do you revert unsaved changes?

by Tim Mackinnon

Ahhhh - mouse blindness - glad you pointed that out, it was driving me batty. > On 25 Jun 2018, at 14:03, Cyril Ferlicot D. <cyril.ferlicot(a)gmail.com> wrote: > > On 25/06/2018 14:48, Tim Mackinnon wrote: >> Hi - in Calypso in Pharo 7, there is a neat concept of method changes that are yet saved (and the top right of the editor has the normal orange smudge indicator). >> >> I like the idea (which lets you naviaged elsewhere to check something before making your change) - however if you want to revert that change and just go back to the original source - how do you do this? >> >> There is a little X indicator in the status bar - but that seems to do nothing (?). >> > > Hi, > > In the context menu there is a "cancel" option with the Ctrl + L shortcut. > >> Do you have to use the versions menu and pick revert? But then when I do that - I see a Red smudge on the text pane now, and the method source hasnât reverted? So Iâm a bit confused what you are supposed to do? The closes I can get to - is closing the tab in the browser and then reclick on the method in question - but Iâm wondering if Iâm missing an obvious alternative? >> >> Tim >> >> > > > -- > Cyril Ferlicot > https://ferlicot.fr >

June 25, 2018

Calypso - can we improve the bar of radio buttons?

by Tim Mackinnon

Hi - I donât want to case a UI war, particularly as I was very impressed with Denisâ presentation about Calypso at the PharoDays conference last year. Now that Iâve made a concerted effort to try and use Calypso, and in general like it, I do however find that the bar of Radio Buttons above the tabs is a bit overwhelming and Iâm wondering if there might be something we can do to help make it less so? Iâm sure the problem is the age old observation that a checkbox or button that toggles between states is difficult to understand whether its on or off and what the alternate state is - so this is why we now have a row of 8 radio buttons. But 8 is a lot of radio buttonsâ¦ which I find visually hard to parse and often find the order of the buttons the opposite to what I expect - eg. When I want to see class methods I keep pausing over that group to see which one to click (for some reason I keep thinking the class one should be before the instance one - not sure if anyone else notices this). Iâm also not overly keen on the terminology of âsideâ (as in class side, vs inst. side) and equally I question abbreviations like: Hier. Inst. Refs. Can we not just get rid of that bar completely - and put little icon buttons next to the filter field for pkg/project and flat/hierarchy ? OR given the method pane is now a tree and it already has an entry âinstance sideâ - why not just have entries for âinstance methodsâ and âclass methodsâ and just expand accordingly? We also now have a tab âInst. side methodâ - why not 2 tabs â+instance methodâ and â+class methodâ. It breaks a bit with regular smalltalk browser conventions - but I think Calypso has already pushed the boat out on all this anyway in favour of improved navigation and ease of typing code. My understanding is that this is all pretty doable with the mechanism Denis has put in place. Iâm tempted to have a look, but thought Iâd see what that general feeling is here, as maybe Iâm on my own here. Tim

June 25, 2018

Re: [Pharo-users] Iceberg - finding deleted classes, reverting versions?

by Ben Coman

On 25 June 2018 at 19:41, Tim Mackinnon <tim(a)testit.works> wrote: > Iâd be really interested if someone with lower level GIT knowledge might > try a: > > git checkout <gitid> src/<proj>/<class>.class.st > <http://prismicblock.class.st/> > > For their project - as I donât understand what Iâm doing wrong - and Iâd > like the comfort of knowing that our source is in a place/state where we > can rely on normal git tools in a case of emergency. At the moment, Iâm a > bit nervous that we are corrupting something . > > Tim > I'm not sure if this is what you wanted, but I found a test case for for Pharo 7, a deleted class "MultiByteFileStreamTest.class.st" https://github.com/pharo-project/pharo/pull/1031/files#diff-750a25fb99d29cd… >From Windows 10 cmd.exe I tried the following (I can't remember which tool installed `git`)... > mkdir C:\temp\test > cd C:\temp\test > git clone git@github.com:pharo-project/pharo.git > cd pharo\src\Deprecated70 > dir Multi* no result > git checkout e74308e67d9f84 MultiByteFileStreamTest.class.st > dir Multi* MultiByteFileStreamTest.class.st Then I compared the file I checked out to the raw file on github and they were identical... https://www.diffchecker.com/gVCpJzFe btw, I did get a momentary error "error: pathspec 'src/Deprecated70/ MultiByteFileStreamTest.class.st' did not match any file(s) known to git." when I incorrectly did... > cd pharo\src\Deprecated70 > git git checkout e74308e67d9f84 src/Deprecated70/ MultiByteFileStreamTest.class.st This worked with the longer path... > cd pharo > git git checkout e74308e67d9f84 src/Deprecated70/ MultiByteFileStreamTest.class.st cheers -ben

June 25, 2018

Re: [Pharo-users] Calypso - how do you revert unsaved changes?

by Cyril Ferlicot D.

On 25/06/2018 14:48, Tim Mackinnon wrote: > Hi - in Calypso in Pharo 7, there is a neat concept of method changes that are yet saved (and the top right of the editor has the normal orange smudge indicator). > > I like the idea (which lets you naviaged elsewhere to check something before making your change) - however if you want to revert that change and just go back to the original source - how do you do this? > > There is a little X indicator in the status bar - but that seems to do nothing (?). > Hi, In the context menu there is a "cancel" option with the Ctrl + L shortcut. > Do you have to use the versions menu and pick revert? But then when I do that - I see a Red smudge on the text pane now, and the method source hasnât reverted? So Iâm a bit confused what you are supposed to do? The closes I can get to - is closing the tab in the browser and then reclick on the method in question - but Iâm wondering if Iâm missing an obvious alternative? > > Tim > > -- Cyril Ferlicot https://ferlicot.fr

June 25, 2018

Calypso - how do you revert unsaved changes?

by Tim Mackinnon

Hi - in Calypso in Pharo 7, there is a neat concept of method changes that are yet saved (and the top right of the editor has the normal orange smudge indicator). I like the idea (which lets you naviaged elsewhere to check something before making your change) - however if you want to revert that change and just go back to the original source - how do you do this? There is a little X indicator in the status bar - but that seems to do nothing (?). Do you have to use the versions menu and pick revert? But then when I do that - I see a Red smudge on the text pane now, and the method source hasnât reverted? So Iâm a bit confused what you are supposed to do? The closes I can get to - is closing the tab in the browser and then reclick on the method in question - but Iâm wondering if Iâm missing an obvious alternative? Tim

June 25, 2018

Re: [Pharo-users] Iceberg - finding deleted classes, reverting versions?

by Tim Mackinnon

Iâd be really interested if someone with lower level GIT knowledge might try a: git checkout <gitid> src/<proj>/<class>.class.st <http://prismicblock.class.st/> For their project - as I donât understand what Iâm doing wrong - and Iâd like the comfort of knowing that our source is in a place/state where we can rely on normal git tools in a case of emergency. At the moment, Iâm a bit nervous that we are corrupting something . Tim > On 22 Jun 2018, at 01:39, Tim Mackinnon <tim(a)testit.works> wrote: > > Just to add more information to this - I did try what was suggested here (it refers to a branch and not a commit id - but heck worth a shot) https://stackoverflow.com/questions/5989592/git-cannot-checkout-branch-erro… <https://stackoverflow.com/questions/5989592/git-cannot-checkout-branch-erro…> - but nah /refs/heads/srcâ¦ or /origin/srcâ¦ didnât seem to give a recognised path. > > Grasping at straws I also tried --ignore-skip-worktree-bits > > So Iâm really curious if anyone else has managed to restore a deleted Pharo class in git? > > Tim > >> On 21 Jun 2018, at 16:07, Tim Mackinnon <tim(a)testit.works <mailto:tim@testit.works>> wrote: >> >> Guille - just following up on this thread as Iâd like to get more confident with this stuff. >> >> You mentioned a Calypso plugin for versions - where is that? I loaded a recentish P7 image and I donât see those icons? Is this something I can easily load to try out? >> >> As I donât have that plugin, Iâve followed up on your next suggestion (just to learn how to do this) about looking up a commit having used the command line git history to list all of my deleted classes (I think this might be a useful thing to add to Iceberg BTW). >> >> Not sure how to get a repository but I cobbled together the following? >> >> (IceLibgitRepository registry detect: [ :r | r name beginsWith: 'Prismic' ]) lookupCommit: '65363adâ. >> >> This give me an IceGitCommit - but then what can I easily do with this to try and iterate over deleted classes to try and get the one Iâve identified back? >> >> Failing all of this - I went back to the command line again (which should work) - and Iâm really wondering if we do something weird with files or paths when checking in? >> >> Having got a valid commitID (I used: git log --diff-filter=D --summary âpretty="format:%cd|%h|%cn%n%s%nâ ) >> >> if I simply run the following (with the id and pathname taken from the output of my previous command >> >> git checkout 65363ad src/PrismicDemo/PrismicBlock.class.st <http://prismicblock.class.st/> >> >> I keep getting an error : error: pathspec âsrc/PrismicDemo/PrismicBlock.class.st <http://prismicblock.class.st/>' did not match any file(s) known to git. >> >> Are you able to confirm if you can restore a deleted class from a commit? Iâve tried it on 2 different OSX machines and neither of them works - making me think we do something odd. >> >> Tim >> >>> On 14 Jun 2018, at 08:52, Guillermo Polito <guillermopolito(a)gmail.com <mailto:guillermopolito@gmail.com>> wrote: >>> >>> Hi, >>> >>> Regarding history, right now we have a history browser implemented as a Calypso plugin, that is open using context menu => history or the little box button on the top right of the method pane (second button from the left in the picture): >>> >>> <Screen Shot 2018-06-14 at 09.34.24.png> >>> >>> That button will nowadays only be shown if the method's package is linked to an iceberg repository. >>> Once you click it, you will have the entire history of the method. >>> >>> <Screen Shot 2018-06-14 at 09.47.46.png> >>> >>> With the possibility to install that version of the method (among others). >>> >>> Now, regarding the recovery of deleted classes/methods, have you tried the repository browser? >>> Go to Iceberg, right click on a repository and select "Repository". >>> You can there select a commit in history and then in the tabs below see the diff between >>> - your current version and the selected commit >>> - the selected commit and its main parent >>> >>> Of course any of these can be improved, but if you have concrete requests, it is much easier :) >>> >>> https://github.com/pharo-vcs/iceberg <https://github.com/pharo-vcs/iceberg> >>> >>> Guille >>> >>> On Wed, Jun 13, 2018 at 11:22 PM Tim Mackinnon <tim(a)testit.works <mailto:tim@testit.works>> wrote: >>> Hi Sean - I tried it again, and it worked: >>> >>> git log --full-history -- */PrismicBlock.class* >>> >>> (I got my wildcard slightly wrong for tonel format) - although the beauty of the one I gave was that it shows you all deleted classes (in case you donât know the name) >>> >>> Iâm still confused why I canât checkout the deleted class though - damn you git, for the cryptic error: : error: pathspec 'PrismicDemo/PrismicBlock.class.st <http://prismicblock.class.st/>' did not match any file(s) known to git. >>> >>> I was hoping a quick hack to iceberg might be to OSProcess a few choice git commands to help us along while we work out better ways to do things. >>> >>> Tim >>> >>>> On 13 Jun 2018, at 21:53, Sean P. DeNigris <sean(a)clipperadams.com <mailto:sean@clipperadams.com>> wrote: >>>> >>>> Tim Mackinnon wrote >>>>> I didnât quite get your fu to work >>>> >>>> Interesting. >>>> >>>> When I searched for commits affecting the deleted class SuDebianKey: >>>> >>>> $ git log --full-history -- */SuDebianKey.class/properties.json >>>> >>>> I got back a list of commits (obviously with the last chronologically being >>>> the deletion): >>>> >>>> commit a38fbced4abec59ff9879d4c607da80dc89b6637 >>>> Author: Sean DeNigris <sean(a)Seans-MBP.home <mailto:sean@Seans-MBP.home>> >>>> Date: Mon Jan 30 17:13:58 2017 -0500 >>>> >>>> Extract Lots to ComputerWorld, Absorb Rest of Old ScriptingBase Project >>>> >>>> commit 61175745d57c60a1707d5e2f9a2fc92e6c19a6ea >>>> Author: Sean DeNigris <sean(a)clipperadams.com <mailto:sean@clipperadams.com>> >>>> Date: Sat Aug 20 17:14:12 2016 -0400 >>>> >>>> Basket O' Enhancements >>>> ... >>>> >>>> >>>> >>>> ----- >>>> Cheers, >>>> Sean >>>> -- >>>> Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html <http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html> >>>> >>> >>> >>> >>> -- >>> >>> Guille Polito >>> Research Engineer >>> >>> Centre de Recherche en Informatique, Signal et Automatique de Lille >>> CRIStAL - UMR 9189 >>> French National Center for Scientific Research - http://www.cnrs.fr <http://www.cnrs.fr/> >>> >>> Web: http://guillep.github.io <http://guillep.github.io/> >>> Phone: +33 06 52 70 66 13 >> >

June 25, 2018

Re: [Pharo-users] Pharo 7 streams API

by Herbert VojÄÃk

Peter UhnÃ¡k wrote on 23. 6. 2018 15:39: > Hi, > > I'm starting to familiarize myself with new streams, and one thing I've > noticed is the removal of #lineEndConvention (which I use all the time). > > So a statement like this > > aFile writeStreamDo: [ :stream | > stream lineEndConvention: #lf. > stream << '...'> ]. > > has to be written like so > > aFile writeStreamDo: [ :rawStream | |stream| > stream := (ZnNewLineWriterStream on: rawStream) forLf. > stream << '...' > ]. > > which feels very messy because I am mixing writing with the > configuration. And I don't even take account for buffered/encoded > decorators. Plus it increases the incidental complexity -- I need > another variable, and I can accidentally write to the wrong stream, etc. > > Would a method like #writeStream:do: (or #writeStreamTransform:do:) make > sense? E.g. > > aFile writeStreamTransform: [ :stream | (ZnNewLineWriterStream on: > stream) ] do: [ :stream | > stream << '...' > ] aFile writeStreamDo: [ :rawStream | (ZnNewLineWriterStream on: rawStream) in: [ :stream | stream << '...' ] ]. As for transformation, I'd go for some more generic (functional?) approach like: aFile writeStreamDo: ([:x | ZnNewLineWriterStream on: x] pipe: [ :stream | stream << '...' ]). Herby > To separate the composition from the usage? > > Thanks, > Peter

June 25, 2018

Re: [Pharo-users] Pharo 7, Iceberg, Proxy and Windows

by Guillermo Polito

Hi Vitor, I've created an issue with your problem. https://github.com/pharo-vcs/iceberg/issues/885 I encourage you to do so too, because following the mailing list is sometimes hard :) About your second issue, you should know that libgit ( libgit2.github.com/libgit2) will mostly bypass git and reimplement most of its features. So, most of the times it will read git's settings, and some other times not. And also there is the possibility that Iceberg overrides those settings too. For example, your fix *self prim_fetch_opts prim_proxy_opts prim_type: LGitProxyTypeEnum git_proxy_auto* does enable automatic proxy verification and forbids people to set their own proxy. Not saying that this is bad but the contrary: I think it is a really good default value :). Similar to that there are the ssh verifications on Iceberg. So far - we do not check the certificate stores from the machine (we could through FFI, the sub-project is listed in here https://github.com/pharo-vcs/iceberg/wiki/How-to-help-us,-What-you-could-co… ) - I don't think so far there is a hook to "disable ssh verification". This could be dangerous... I've googled the error a bit, and found this. Maybe you can try it and see wether the solution belongs to iceberg, or that we should enhance the documentation? https://stackoverflow.com/questions/48985995/gitkraken-and-github-failed-to… Tx, Guille On Thu, Jun 21, 2018 at 5:49 PM Vitor Medina Cruz <vitormcruz(a)gmail.com> wrote: > I am sorry, the correct message is "LGit_GIT_ERROR: failed to get server > certificate: The handle is in the wrong state for the requested operation". > > > I tried to add corporate certificates to the git curl-ca-bundle.crt, I > tried to git config --global http.sslVerify false also, no success... :( > > On Thu, Jun 21, 2018 at 9:31 AM, Vitor Medina Cruz <vitormcruz(a)gmail.com> > wrote: > >> Ok, I forgot to mention I have changed >> LGitRepository>>clone:url:local_path:options:, but for some odd reason it's >> arguments names were switched to arg 1, arg2, arg3 etc, so the out error >> was because was renamed to one of those generic arg names. I figure that >> out looking at the versions of the methods, and it apers that when I first >> access it this change is made. >> >> Anyways, I fixed that and I still get "IceGenericError: failed to get >> server certificate: The identifier is not in the correct state for the >> resquested operation" >> >> On Wed, Jun 20, 2018 at 4:49 PM, Vitor Medina Cruz <vitormcruz(a)gmail.com> >> wrote: >> >>> Hello, >>> >>> Iceberg fail to function behind a proxy on Pharo 7 32bits Windows. When >>> I try to clone I get a: >>> >>> IceGenericError: failed to get server certificate: The identifier is not >>> in the correct state for the resquested operation >>> >>> I had a similar problem some time ago with Pharo 6, in which I made a >>> local hot fix after questioning here. The fix was to the LGitCloneOptions >>> and LGitFetchOptions both at the initializeWithDefaults, currently this >>> method on LGitCloneOptions is like this: >>> >>> >>> initializeWithDefaults >>> self withReturnHandlerDo: [ >>> self >>> clone_init_options: self >>> version: LGitOptionsVersionsEnum >>> git_clone_options_version_1 ]. >>> >>> >>> >>> * self prim_fetch_opts prim_proxy_opts prim_type: LGitProxyTypeEnum >>> git_proxy_auto* >>> The bold is the fix, but it should come first in the method: >>> >>> initializeWithDefaults >>> >>> >>> * self prim_fetch_opts prim_proxy_opts prim_type: LGitProxyTypeEnum >>> git_proxy_auto* self withReturnHandlerDo: [ >>> self >>> clone_init_options: self >>> version: LGitOptionsVersionsEnum >>> git_clone_options_version_1 ]. >>> >>> >>> That way worked on Pharo 6, but now I got a: >>> >>> "FFIVariableNameNotFound: Could not find accessor for variable named >>> 'out'" >>> >>> Any clues? >>> >>> Regards, >>> Vitor >>> >> >> > -- Guille Polito Research Engineer Centre de Recherche en Informatique, Signal et Automatique de Lille CRIStAL - UMR 9189 French National Center for Scientific Research - *http://www.cnrs.fr <http://www.cnrs.fr>* *Web:* *http://guillep.github.io* <http://guillep.github.io> *Phone: *+33 06 52 70 66 13

June 25, 2018