December 2019 - Pharo-users

how to change data on real data but not on test data
by Roelof Wobben Dec. 28, 2019

Dec. 28, 2019

1 0

Re: [Pharo-users] uses or instead of a searching literal
by Sven Van Caekenberghe Dec. 28, 2019

Dec. 28, 2019

I would go for 'Portable Network Graphics' findTokens: ' -_'. > On 28 Dec 2019, at 09:35, Roelof Wobben via Pharo-users <pharo-users(a)lists.pharo.org> wrote: > > > From: Roelof Wobben <r.wobben(a)home.nl> > Subject: Re: [Pharo-users] uses or instead of a searching literal > Date: 28 December 2019 at 09:35:17 GMT+1 > To: pharo-users(a)lists.pharo.org > > > yep. Im aware of this tool but on this case nothing pops up. > > Roelof > > > > > Op 28-12-2019 om 09:22 schreef Kasper Ãsterbye: >> Are you aware of the âfinderâ tool?, in particular the âexamples' mode is useful. >> >> try: âaaa_bbb-cccâ. â_-â. #(âaaaâ âbbbâ âcccâ) >> >> >> Best, >> >> Kasper >> >> On 28 December 2019 at 08.12.03, Roelof Wobben via Pharo-users (pharo-users(a)lists.pharo.org) wrote: >> >>> Op 27-12-2019 om 23:33 schreef Richard O'Keefe: >>> > aString splitOn: ' -_' asSet >>> >>> Hello Richard, >>> >>> Thanks again , I find this "aString splitOn: '_- ` asSet " much >>> cleaner but on some way I does not split for example 'Portable Network >>> Graphics' into " #(Portable, Network, Graphics) ". When I debug it , >>> it seems there is no splitting at all. >>> >>> So today time to dive into it why it does not work and how to solve it. >>> >>> Roelof >>> >>> > > >

2 1

Re: [Pharo-users] uses or instead of a searching literal
by Kasper Ãsterbye Dec. 28, 2019

Dec. 28, 2019

Are you aware of the âfinderâ tool?, in particular the âexamples' mode is useful. try: âaaa_bbb-cccâ. â_-â. #(âaaaâ âbbbâ âcccâ) Best, Kasper On 28 December 2019 at 08.12.03, Roelof Wobben via Pharo-users ( pharo-users(a)lists.pharo.org) wrote: Op 27-12-2019 om 23:33 schreef Richard O'Keefe: > aString splitOn: ' -_' asSet Hello Richard, Thanks again , I find this "aString splitOn: '_- ` asSet " much cleaner but on some way I does not split for example 'Portable Network Graphics' into " #(Portable, Network, Graphics) ". When I debug it , it seems there is no splitting at all. So today time to dive into it why it does not work and how to solve it. Roelof

2 1

Re: [Pharo-users] uses or instead of a searching literal
by Richard O'Keefe Dec. 28, 2019

Dec. 28, 2019

You know, this is an area where style opinions differ a lot. What the message is suggesting you try is aString splitOn: [:each | ' -_' includes: each] That's all. No need for anything more complicated. But there is a trade-off. Clarity vs efficiency. You should make your code clear and correct before you worry about efficiency, true. But you should not go out of your way to make it *less* efficient. The test you have is perfectly readable and is as fast as it gets. Using #includes: would make the code shorter, but not clearer, and certainly slower. You could also use aString splitOn: '[-_ ]' asRegex because #splitOn: is documented as treating strings as regular expressions. This is the briefest, the most confusing to read, and the slowest. It is also by far the most error-prone: - forget "asRegex" and you are in trouble - forget the square brackets and you are in trouble - put the hyphen-minus second instead of first and ... If only #split:indicesDo: were defined for Set. I cannot imagine why it is not. Here is a method definition that goes in the (new) 'splitting' category of Set: split: aSequenceableCollection indicesDo: aBlock "Perform an action specified as aBlock (with a start and end argument) to each of the indices of aSequenceableCollection that have been identified by taking the receiver as a splitter." | position | position := 1. aSequenceableCollection withIndexDo: [:element :idx | (self includes: element) ifTrue: [ aBlock value: position value: idx - 1. position := idx + 1 ]]. aBlock value: position value: aSequenceableCollection size With that in place you can use aString splitOn: ' -_' asSet On Sat, 28 Dec 2019 at 10:24, Roelof Wobben via Pharo-users <pharo-users(a)lists.pharo.org> wrote: > > Hello, > > > How can I get rid of the above error message with this code : > > > abbreviatePhrase: aString > | splitted | > splitted := aString > splitOn: [ :char | char = Character space or: [ char = $- or: [ char = $_ ] ] ]. > ^ String > streamContents: [ :stream | > splitted > reject: [ :word | word isEmpty ] > thenDo: [ :word | stream nextPut: word first uppercase ] ] > > > Regards, > > Roelof >

2 1

Re: [Pharo-users] can I write this without the three if then;s
by Richard O'Keefe Dec. 27, 2019

Dec. 27, 2019

Myself, I see only one issue in your code, and that is the pointless use of sqrt. scoreX: anInteger y: anInteger2 | radiusSquared | radiusSquared := anInteger squared + anInteger2 squared. radiusSquared > 100 ifTrue: [ ^ 0 ]. radiusSquared > 25 ifTrue: [ ^ 1 ]. radiusSquared > 1 ifTrue: [ ^ 5 ]. ^ 10 Make that two issues. The names anInteger and anInteger2. How about scoreHorizontal: horizontal vertical: vertical ... The OO dogma about "if' that you refer to is about not using "if" to make decisions based on TYPE. Making decisions based on NUMERIC RANGES is perfectly OK. You could introduce a 'HalfOpenRangeDictionary" class [K1,K2) -> V1 [K2,K3) -> V2 ... [Kn,infinity) -> Vn But "You Ain't Gonna Need It", so don't. On Sat, 28 Dec 2019 at 08:18, Roelof Wobben via Pharo-users <pharo-users(a)lists.pharo.org> wrote: > > Hello, > > Im trying to solve a challenge from exercism where I have to calculate the points somehow gets on a very simple darts board. > > I solved it like this : > > > scoreX: anInteger y: anInteger2 > | distance | > distance := (anInteger squared + anInteger2 squared) sqrt. > distance > 10 > ifTrue: [ ^ 0 ]. > distance > 5 > ifTrue: [ ^ 1 ]. > distance > 1 > ifTrue: [ ^ 5 ]. > ^ 10 > > > but now I use three if then and I think it's ugly code. > > Is there a way I can make it more the smalltalk way ? > > > Regards, > > Roelof >

2 1

Re: [Pharo-users] uses or instead of a searching literal
by tbrunz Dec. 27, 2019

Dec. 27, 2019

Since you're splitting a String, why not use a Regex to do this? -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

1 0

uses or instead of a searching literal
by Roelof Wobben Dec. 27, 2019

Dec. 27, 2019

1 0

Re: [Pharo-users] can I write this without the three if then;s
by tbrunz Dec. 27, 2019

Dec. 27, 2019

I would probably set up an array of associations whose keys are the boundary limits and whose values are the target values (ordered, in this example, from outer to inner). Then loop over the array's associations using inject:into:, testing for the point where (each < key), at which point you return the previous association's value. -t -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

1 2

Re: [Pharo-users] can I write this without the three if then;s
by Kasper Ãsterbye Dec. 27, 2019

Dec. 27, 2019

Hi First, I think your code is readable and easy to understand. I think the problem does not lend itself too much towards having âan object oriented solutionâ, as such. If you are looking for a one-liner using smalltalk libraries, you could do: ({10 -> 0. 5 -> 1. 1 -> 5. 0 -> 10} detect: [ :dist | dist key < distance ]) value but I really think your code is easier to read, which is what I appreciate the most. Best, Kasper On 27 December 2019 at 20.18.38, Roelof Wobben via Pharo-users ( pharo-users(a)lists.pharo.org) wrote: Hello, Im trying to solve a challenge from exercism where I have to calculate the points somehow gets on a very simple darts board. I solved it like this : scoreX: anInteger y: anInteger2 | distance | distance := (anInteger squared + anInteger2 squared) sqrt. distance > 10 ifTrue: [ ^ 0 ]. distance > 5 ifTrue: [ ^ 1 ]. distance > 1 ifTrue: [ ^ 5 ]. ^ 10 but now I use three if then and I think it's ugly code. Is there a way I can make it more the smalltalk way ? Regards, Roelof

2 1

Re: [Pharo-users] can I write this without the three if then;s
by tbrunz Dec. 27, 2019

Dec. 27, 2019

You might try Magnitude >> between: min and: max -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

2 1