May 2020 - Pharo-users - lists.pharo.org

Re: [Pharo-users] Package Browser (AKA 6-paned Browser) and Pharo package organization

by Stephen Pope

As a side note, the Cuis fork of Squeak simplifies and reorganizes the core system categories in the style of Smalltalk-80, see the screen shot below. stp -- Stephen Travis Pope Santa Barbara, California, USA http://HeavenEverywhere.com <http://heaveneverywhere.com/> http://FASTLabInc.com <http://fastlabinc.com/> https://vimeo.com/user19434036/videos <https://vimeo.com/user19434036/videos> http://heaveneverywhere.com/Reflections -- > On Apr 30, 2020, at 12:00 PM, stephen(a)heaveneverywhere.com wrote: > > > Hello friends, > > Iâm getting started with Pharo after decades using VisualWorks and Squeak; itâs pretty wonderful what you all have assembled! > > My question is related to what we used to teach as the first law of software reuse: âYou canât reuse it if you canât find it,â and the related software engineering "principle of least astonishment." > > When I fire up Pharo, the system browser presents me with a list of several hundred categories (from AST to Zodiac) in a system with over 8000 classes. The system categorization makes no sense since I donât know the naming conventions and so many packages have cute but quite non-descriptive names (Zinc? Metacello? Calypso?). > > In Smalltalk-80, the class category names were organized as a 2-level hierarchy where the top-level were items such as Magnitudes, Collections, Streams, Graphics, Text, System, Tools, Files, etc. This made it easy to find (e.g.,) the browser source code by looking in the Tools package for the class category Tools-Browser. Even packages with cute names (like my own âSirenâ), were categorized for ease of finding; e.g., the Siren classes were in class categories like Music-Events and Music-Magnitudes. > > Parsing the class category names on the first instance of $- made it possible to build 6-paned Browsers (called package pane browser in Squeak). (We acknowledged that this violates the âzero/one/infinity" rule.) Is something like this available for Pharo? I looked through the Calypso browser code and itâs so over-engineered (IMHO) that itâd take me several days to figure out how to implement this (it was about 1.5 pages of code in Smalltalk-80). > > If Pharo had a browser that scaled better and a reorganization/simplification of the class categories to use names that were more self-explanatory, it would be *much* easier for new users (in fact, for all users) to find their way around. > > I apologize for the stepping on toes... > > Stephen Pope > > > -- > > Stephen Travis Pope Santa Barbara, California, USA > <pastedGraphic.tiff> http://HeavenEverywhere.com <http://heaveneverywhere.com/> http://FASTLabInc.com <http://fastlabinc.com/> > https://vimeo.com/user19434036/videos <https://vimeo.com/user19434036/videos> http://heaveneverywhere.com/Reflections <http://heaveneverywhere.com/Reflections> > > -- > > >

May 1, 2020

Re: [Pharo-users] Package Browser (AKA 6-paned Browser) and Pharo package organization

by Sean P. DeNigris

Stephen Pope wrote > When I fire up Pharo, the system browser presents me with a list of > several hundred categories (from AST to Zodiac) in a system with over 8000 > classes. The system categorization makes no sense since I donât know the > naming conventions and so many packages have cute but quite > non-descriptive names (Zinc? Metacello? Calypso?). Even after using Pharo (and this probably applies somewhat to current Squeak too, but I haven't opened an image in a while so forgive me if I'm wrong) for over a decade, I still feel your pain on this issue! The problem IMHO is the ugly Monticello SCM hack hijacking *logical* categories to use for packaging, which is an orthogonal concern. Here is a a fuller description from a previous thread [1]: Sean P. DeNigris wrote >> we have overloaded system categories to package code for SCM. System >> categories should be tags (preferably multiple allowed) >> which offer a logical view of the system. Packages, the POV we show now, >> are orthogonal and much less useful for users. > (edited) > and another: >> I feel more and more that the standard "Package" pane is only useful >> for... packaging, and when one takes off the dependency management hat >> and >> puts the user hat on (i.e. most of the time), what you really want there >> is a logical view of the system. So I see three use cases: > - Logical view of the system - I guess this was the original intention of > Categories, but has been hijacked by Monticello > - By project - which, as you just showed, we have now, yay! > - By package - the least useful, but primary (up til now), view > (edited) > and regarding Nautilus' tree package pane (when it first arrived): > I noticed that right now, separate packages within the same project are > not > collapsed. E.g. if I have MyProject-Core and MyProject-Platform, they will > be siblings in the tree, instead of both under MyProject. It seems like it > would be more useful to have > - MyProject > - Core > - Platform > in the tree > >> it seems that the tree is primarily about chunking information into >> manageable pieces. > > A primary difficulty here is that packages are often divided for reasons > that have nothing to do with the domain model, e.g. the ubiquitous > MyPackage-Platform, which is an artifact of Metacello that is not all that > relevant to a user wanting to understand the system. > > From the naive user perspective, if I'm exploring from the top level of > the > system, I want to see things like: > - CodeImport > - Collections > - Compiler > > From this perspective, the 14 entries for Collections, multiplied by a few > dozen top-level categories make the list unwieldy and only marginally less > daunting than the flattened list we used to have (see > http://en.wikipedia.org/wiki/The_Magical_Number_Seven,_Plus_or_Minus_Two ) 1. http://forum.world.st/Why-can-t-we-use-in-protocol-for-package-extension-tp… ----- Cheers, Sean -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

May 1, 2020

Re: [Pharo-users] Package Browser (AKA 6-paned Browser) and Pharo package organization

by Sven Van Caekenberghe

http://books.pharo.org https://github.com/topics/pharo > On 1 May 2020, at 14:37, Noury Bouraqadi <bouraqadi(a)gmail.com> wrote: > > Hi Stefen, > > Welcome to Pharo :-) > > Here are 2 tips that whould help you find your way : > - Spotter (open it with Shift+Enter). It searches the whole image for names (classes, methods...) that include the given substring > - Finder (Menu Tools) : Allows various kinds of searches. Searching with examples does allow finding a message that provides a given outcome given a receiver, and parameters. > > Please note that the image does include only a small subset of what you can do with Pharo. There's much more out there. One way to discover cool stuff, is to visit this catalog: > https://github.com/pharo-open-documentation/awesome-pharo > > Cheers, > Noury > >> On 30 Apr 2020, at 21:00, stephen(a)heaveneverywhere.com wrote: >> >> >> Hello friends, >> >> Iâm getting started with Pharo after decades using VisualWorks and Squeak; itâs pretty wonderful what you all have assembled! >> >> My question is related to what we used to teach as the first law of software reuse: âYou canât reuse it if you canât find it,â and the related software engineering "principle of least astonishment." >> >> When I fire up Pharo, the system browser presents me with a list of several hundred categories (from AST to Zodiac) in a system with over 8000 classes. The system categorization makes no sense since I donât know the naming conventions and so many packages have cute but quite non-descriptive names (Zinc? Metacello? Calypso?). >> >> In Smalltalk-80, the class category names were organized as a 2-level hierarchy where the top-level were items such as Magnitudes, Collections, Streams, Graphics, Text, System, Tools, Files, etc. This made it easy to find (e.g.,) the browser source code by looking in the Tools package for the class category Tools-Browser. Even packages with cute names (like my own âSirenâ), were categorized for ease of finding; e.g., the Siren classes were in class categories like Music-Events and Music-Magnitudes. >> >> Parsing the class category names on the first instance of $- made it possible to build 6-paned Browsers (called package pane browser in Squeak). (We acknowledged that this violates the âzero/one/infinity" rule.) Is something like this available for Pharo? I looked through the Calypso browser code and itâs so over-engineered (IMHO) that itâd take me several days to figure out how to implement this (it was about 1.5 pages of code in Smalltalk-80). >> >> If Pharo had a browser that scaled better and a reorganization/simplification of the class categories to use names that were more self-explanatory, it would be *much* easier for new users (in fact, for all users) to find their way around. >> >> I apologize for the stepping on toes... >> >> Stephen Pope >> >> >> -- >> >> Stephen Travis Pope Santa Barbara, California, USA >> <pastedGraphic.tiff> http://HeavenEverywhere.com http://FASTLabInc.com >> https://vimeo.com/user19434036/videos http://heaveneverywhere.com/Reflections >> >> -- >> >> >> >

May 1, 2020

Re: [Pharo-users] Package Browser (AKA 6-paned Browser) and Pharo package organization

by Noury Bouraqadi

Hi Stefen, Welcome to Pharo :-) Here are 2 tips that whould help you find your way : - Spotter (open it with Shift+Enter). It searches the whole image for names (classes, methods...) that include the given substring - Finder (Menu Tools) : Allows various kinds of searches. Searching with examples does allow finding a message that provides a given outcome given a receiver, and parameters. Please note that the image does include only a small subset of what you can do with Pharo. There's much more out there. One way to discover cool stuff, is to visit this catalog: https://github.com/pharo-open-documentation/awesome-pharo Cheers, Noury > On 30 Apr 2020, at 21:00, stephen(a)heaveneverywhere.com wrote: > > > Hello friends, > > Iâm getting started with Pharo after decades using VisualWorks and Squeak; itâs pretty wonderful what you all have assembled! > > My question is related to what we used to teach as the first law of software reuse: âYou canât reuse it if you canât find it,â and the related software engineering "principle of least astonishment." > > When I fire up Pharo, the system browser presents me with a list of several hundred categories (from AST to Zodiac) in a system with over 8000 classes. The system categorization makes no sense since I donât know the naming conventions and so many packages have cute but quite non-descriptive names (Zinc? Metacello? Calypso?). > > In Smalltalk-80, the class category names were organized as a 2-level hierarchy where the top-level were items such as Magnitudes, Collections, Streams, Graphics, Text, System, Tools, Files, etc. This made it easy to find (e.g.,) the browser source code by looking in the Tools package for the class category Tools-Browser. Even packages with cute names (like my own âSirenâ), were categorized for ease of finding; e.g., the Siren classes were in class categories like Music-Events and Music-Magnitudes. > > Parsing the class category names on the first instance of $- made it possible to build 6-paned Browsers (called package pane browser in Squeak). (We acknowledged that this violates the âzero/one/infinity" rule.) Is something like this available for Pharo? I looked through the Calypso browser code and itâs so over-engineered (IMHO) that itâd take me several days to figure out how to implement this (it was about 1.5 pages of code in Smalltalk-80). > > If Pharo had a browser that scaled better and a reorganization/simplification of the class categories to use names that were more self-explanatory, it would be *much* easier for new users (in fact, for all users) to find their way around. > > I apologize for the stepping on toes... > > Stephen Pope > > > -- > > Stephen Travis Pope Santa Barbara, California, USA > <pastedGraphic.tiff> http://HeavenEverywhere.com <http://heaveneverywhere.com/> http://FASTLabInc.com <http://fastlabinc.com/> > https://vimeo.com/user19434036/videos <https://vimeo.com/user19434036/videos> http://heaveneverywhere.com/Reflections <http://heaveneverywhere.com/Reflections> > > -- > > >

May 1, 2020

Re: [Pharo-users] Automating release announcements

by Ben Coman

On Fri, 1 May 2020 at 00:26, Gabriel Cotelli <g.cotelli(a)gmail.com> wrote: > I finally figure out how to make it work. For this mailing list you have > to send the e-mail in CC and not BCC, because if not it gets retained > awaiting for moderator approval. > In case anyone is interested here is the GitHub action making the magic: > https://github.com/ba-st-actions/email-release-notification > Cool !! Thanks for sharing Gabriel. cheers -ben

May 1, 2020

Re: [Pharo-users] mentor question 4

by Roelof Wobben

Op 1-5-2020 om 02:51 schreef Richard O'Keefe: > (oddSum + evenSum) dividedBy: 10 > > You previously had _ isDivisibleBy: 10 > which certainly works. > > Squeak, Pharo, and ST/X have #isDivisibleBy: > VisualWorks. Dolphin, and GNU Smalltalk do not. > > Here's the code from Number.st in ST/X. > isDivisibleBy:aNumber > "return true, if the receiver can be divided by the argument, > aNumber without a remainder. > Notice, that the result is only worth trusting, if the receiver > is an integer." > > aNumber = 0 ifTrue: [^ false]. > aNumber isInteger ifFalse: [^ false]. > ^ (self \\ aNumber) = 0 > > The comment is wrong: the question makes sense for any combination > of exact numbers. > When, as in this case, aNumber is a literal integer, all > #isDivisibleBy: really adds is overhead. > > (oddSum + evenSum) \\ 10 = 0 > > is quite clear, and completely portable. > > > > On Fri, 1 May 2020 at 02:16, Roelof Wobben <r.wobben(a)home.nl> wrote: >> Op 30-4-2020 om 16:06 schreef Richard O'Keefe: >>> This sounds very much like the Luhn test task at RosettaCode. >>> https://rosettacode.org/wiki/Luhn_test_of_credit_card_numbers >>> except that there it is described as working on the digits of an >>> integer. >>> >>> (1) There are two approaches to traversing a sequence in reverse. >>> (A) Reverse the sequence, then traverse the copy forward. >>> aString reverse do: [:each | ...] >>> (B) Just traverse the sequence in reverse >>> aString reverseDo: [:each | ...] >>> My taste is for the second. >>> >>> (2) There are two approaches to deleting spaces. >>> (A) Make a copy of the string without spaces. >>> x := aString reject: [:each | each = Character space]. >>> x do: ... >>> (B) Ignore spaces as you go: >>> (i) aString do: [:each | each = Character space ifFalse: [...]] >>> (ii) aString select: [:each | each ~= Character space] thenDo: >>> [:each | ...] >>> >>> Combining (1A) and (2A) you get very obvious code: >>> (aString reject: [:each | each = Character space]) reverse do: >>> [:digit } ...] >>> Combining (1B) and (2Bi) you get more efficient code: >>> aString reverseDo: [:digit | >>> digit = Character space ifFalse: [ ...]] >>> >>> By the way, let's start by checking that the character in the string *are* >>> digits or spaces: >>> (aString allSatisfy: [:each | each isDigit or: [each = Character s[ace]]) >>> ifFalse: [^false], >>> >>> (3) There are two approaches to doubling the even digits. >>> (A) Make a new string that starts as a copy and change every second >>> digit from the right. >>> (B) Simply *act* as if this has been done; keep track of whether the >>> current digit position is even or odd and multiply by 1 or 2 as >>> appropriate. >>> nextIsOdd := true. >>> aString reverseDo: [:digit | >>> digit = Character space ifFalse: [ >>> nextIsOdd >>> ifTrue: [oddSum := ...] >>> ifFalse: [evenSum := ...]. >>> nextIsOdd := nextIsOdd not]]. >>> >>> I *like* code that traverses a data structure exactly once and >>> allocates no intermediate garbage, so I'd be making (B) choices. >>> >>> >> For me , I use this to practice solving problems and doing the "right" >> steps. >> So I love it , that so many people share there way of solving it. >> I can learn a lot from it >> Expecially when they explain there thinking process so detailed. >> >> I like this code also a lot. >> Am I correct for testing if it is a valid string by doing this ^ >> (oddSum + evenSum) dividedBy: 10 >> >> Roelof >> oke, so this is better cardNumber := '8273 1232 7352 0569'. oddSum := 0. evenSum := 0. nextIsOdd := false. Â Â Â Â cardNumber reverseDo: [:character | Â Â Â Â Â Â Â Â digit := character digitValue. Â Â Â Â Â Â Â Â character = Character space ifFalse: [ Â Â Â Â Â Â Â Â nextIsOdd Â Â Â Â Â Â Â Â Â Â Â Â ifFalse:Â [oddSum := oddSum + digit ] Â Â Â Â Â Â Â Â Â Â Â Â ifTrue: [(digit >= 5 ) Â Â Â ifTrue: [evenSum := evenSum + (digit * 2) - 9 ] Â Â Â ifFalse: [ evenSum := evenSum + (digit * 2) ]]. Â Â Â Â Â Â Â Â Â Â Â nextIsOdd := nextIsOdd not]]. ^ evenSum + oddSum // 10 == 0. where I could even make a seperate method of the ifTrue branch when the digit is greater then 5.

May 1, 2020

Re: [Pharo-users] mentor question 4

by Richard O'Keefe

(oddSum + evenSum) dividedBy: 10 You previously had _ isDivisibleBy: 10 which certainly works. Squeak, Pharo, and ST/X have #isDivisibleBy: VisualWorks. Dolphin, and GNU Smalltalk do not. Here's the code from Number.st in ST/X. isDivisibleBy:aNumber "return true, if the receiver can be divided by the argument, aNumber without a remainder. Notice, that the result is only worth trusting, if the receiver is an integer." aNumber = 0 ifTrue: [^ false]. aNumber isInteger ifFalse: [^ false]. ^ (self \\ aNumber) = 0 The comment is wrong: the question makes sense for any combination of exact numbers. When, as in this case, aNumber is a literal integer, all #isDivisibleBy: really adds is overhead. (oddSum + evenSum) \\ 10 = 0 is quite clear, and completely portable. On Fri, 1 May 2020 at 02:16, Roelof Wobben <r.wobben(a)home.nl> wrote: > > Op 30-4-2020 om 16:06 schreef Richard O'Keefe: > > This sounds very much like the Luhn test task at RosettaCode. > > https://rosettacode.org/wiki/Luhn_test_of_credit_card_numbers > > except that there it is described as working on the digits of an > > integer. > > > > (1) There are two approaches to traversing a sequence in reverse. > > (A) Reverse the sequence, then traverse the copy forward. > > aString reverse do: [:each | ...] > > (B) Just traverse the sequence in reverse > > aString reverseDo: [:each | ...] > > My taste is for the second. > > > > (2) There are two approaches to deleting spaces. > > (A) Make a copy of the string without spaces. > > x := aString reject: [:each | each = Character space]. > > x do: ... > > (B) Ignore spaces as you go: > > (i) aString do: [:each | each = Character space ifFalse: [...]] > > (ii) aString select: [:each | each ~= Character space] thenDo: > > [:each | ...] > > > > Combining (1A) and (2A) you get very obvious code: > > (aString reject: [:each | each = Character space]) reverse do: > > [:digit } ...] > > Combining (1B) and (2Bi) you get more efficient code: > > aString reverseDo: [:digit | > > digit = Character space ifFalse: [ ...]] > > > > By the way, let's start by checking that the character in the string *are* > > digits or spaces: > > (aString allSatisfy: [:each | each isDigit or: [each = Character s[ace]]) > > ifFalse: [^false], > > > > (3) There are two approaches to doubling the even digits. > > (A) Make a new string that starts as a copy and change every second > > digit from the right. > > (B) Simply *act* as if this has been done; keep track of whether the > > current digit position is even or odd and multiply by 1 or 2 as > > appropriate. > > nextIsOdd := true. > > aString reverseDo: [:digit | > > digit = Character space ifFalse: [ > > nextIsOdd > > ifTrue: [oddSum := ...] > > ifFalse: [evenSum := ...]. > > nextIsOdd := nextIsOdd not]]. > > > > I *like* code that traverses a data structure exactly once and > > allocates no intermediate garbage, so I'd be making (B) choices. > > > > > > For me , I use this to practice solving problems and doing the "right" > steps. > So I love it , that so many people share there way of solving it. > I can learn a lot from it > Expecially when they explain there thinking process so detailed. > > I like this code also a lot. > Am I correct for testing if it is a valid string by doing this ^ > (oddSum + evenSum) dividedBy: 10 > > Roelof >

May 1, 2020