April 2017 - Pharo-users - lists.pharo.org

Re: [Pharo-users] Sending byte array in a JSON format

by Sven Van Caekenberghe

I am puzzled by how they actually encoded the PNG as a String, I tried a couple of alternatives but I could not get binary data out of it so that it parsed successfully as PNG. If I would have to encode binary data in JSON I would use Base64 encoding (but alternatives exist). > On 24 Apr 2017, at 20:36, Juraj Kubelka <juraj.kubelka(a)icloud.com> wrote: > > Hi, > > I was playing with GitHub Gist API and I have queried the following Gist: https://gist.github.com/mbostock/5503544 > I was interested how the PNG image is returned: https://gist.github.com/mbostock/5503544#file-thumbnail-png > > I can obtain the whole Gist executing: > > STONJSON fromString: > (ZnClient new > url: 'https://api.github.com/gists/5503544'; > accept: 'application/vnd.github.v3+json'; > get). > > I can get PNG contents executing: > > pngData := (ZnEasy get: > ((((STONJSON fromString: > (ZnClient new > url: 'https://api.github.com/gists/5503544'; > accept: 'application/vnd.github.v3+json'; > get)) at: 'files') at: 'thumbnail.png') at: 'raw_url')) contents. > PNGReadWriter formFromStream: rawPng readStream. > > But the PNG image is part of the Gist query and can be retrieved by: > > pngContent := ((((STONJSON fromString: > (ZnClient new > url: 'https://api.github.com/gists/5503544'; > accept: 'application/vnd.github.v3+json'; > get)) at: 'files') at: 'thumbnail.png') at: 'content'). > > "As pngContent is a WideString, I cannot use:" > PNGReadWriter formFromStream: pngContent readStream. > > How can I read the PNG image from the pngContent? Any idea? > And the reverse question: How can I send the PNG bytes using JSON format? > > Thanks! > Juraj >

April 26, 2017

Re: [Pharo-users] Hot to retrieve values from Nested Dictionaries

by Sven Van Caekenberghe

> On 26 Apr 2017, at 01:01, PBKResearch <peter(a)pbkresearch.co.uk> wrote: > > Browsing in Pharo, I found that Dictionary has a subclass KeyedTree, which is in effect a set of nested dictionaries. Retrieval is by quoting a path, which is simply an array of keys. So in your case it would read: > dict1 atPath: #(âkey1â âkey2â âkey3â). > This might be more efficient than repeated use of the DNU mechanism. It is not clear (to me, at least) how to construct the nested dictionaries as a KeyedTree in the first place, but this might be an alternative approach. I saw this too and I thought it was interesting, hence I added it to NeoJSONObject. Name: Neo-JSON-Core-SvenVanCaekenberghe.40 Author: SvenVanCaekenberghe Time: 26 April 2017, 4:35:49.433269 pm UUID: c336587e-740a-0d00-b51a-dacd0088e41c Ancestors: Neo-JSON-Core-SvenVanCaekenberghe.39 Name: Neo-JSON-Tests-SvenVanCaekenberghe.38 Author: SvenVanCaekenberghe Time: 26 April 2017, 4:36:04.583272 pm UUID: 73623f7f-740a-0d00-b51b-f4000088e41c Ancestors: Neo-JSON-Tests-SvenVanCaekenberghe.37 >From the class comment: Additionally, I support path access for nested instances of me, using #atPath: and #atPath:put: The first is special because it returns nil as soon as a key is missing. The second is special because it creates extra levels (instances of me) as needed to follow the path of keys. NeoJSONObject new atPath: #(one two three) put: 42; yourself. NeoJSONObject new atPath: #(one two three) put: 42; atPath: #(one two three). For the examples given above, the actual JSON looks like this: {"one":{"two":{"three":42}}} Sven > Hope this helps > > Peter Kenny > > From: Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] On Behalf Of Markus BÃ¶hm > Sent: 24 April 2017 18:42 > To: pharo-users(a)lists.pharo.org > Subject: [Pharo-users] Hot to retrieve values from Nested Dictionaries > > May I ask: > What's the idiomatic way to retrieve values from nested dictionaries? > > (((dict1 at: 'key1') at: 'key2') at: 'key3') > > Are all the brackets really necessary? > > BR Mike

April 26, 2017

Re: [Pharo-users] ss3 and gemsource planned downtime today at 1:30 pm PDT (-0700 UTC)

by Stephane Ducasse

tx dale On Wed, Apr 26, 2017 at 2:00 AM, Dale Henrichs < dale.henrichs(a)gemtalksystems.com> wrote: > server is back online... > > > > On 04/25/2017 05:28 AM, Dale Henrichs wrote: > >> The server hosting http://ss3.gemtalksystems.com/ and >> http://seaside.gemtalksystems.com/ss will be down for a couple of hours >> today starting at 1:30 pm PDT for maintenance. >> >> Dale >> >> > >

April 26, 2017

Re: [Pharo-users] little "challenge"

by Stephane Ducasse

splitEvery: n ^ (n to: self size by: n) collect: [:index | (self copyFrom: index - n + 1 to: index)] is my solution On Wed, Apr 26, 2017 at 11:34 AM, phil(a)highoctane.be <phil(a)highoctane.be> wrote: > So the last 20 is left out. > > Correct? > > Phil > > On Wed, Apr 26, 2017 at 11:20 AM, Stephane Ducasse < > stepharo.self(a)gmail.com> wrote: > >> Now I fully got the implementation >> >> >> " >> #(16 17 17 16 18 17 18 19 19 19 18 19 19 20 19 20 19 20 20 20 19 >> 20) groupsOf: 7 atATimeCollect: [ :x | x ] >> >>> #(#(16 17 17 16 18 17 18) #(19 19 19 18 19 19 20) #(19 20 19 20 20 20 >> 19)) >> " >> " >> #(1 1 1 10 10 10 100 100 100) >> groupsOf: 3 atATimeCollect: [ :x :y :z | x + y + z] >> >>> #(3 30 300) >> " >> >> " >> #(1 1 1 10 10 10 100 100 100) >> groupsOf: 3 atATimeCollect: [ :x | x ] >> >>> #(#(1 1 1) #(10 10 10) #(100 100 100)) >> " >> >> On Wed, Apr 26, 2017 at 8:38 AM, Stephane Ducasse < >> stepharo.self(a)gmail.com> wrote: >> >>> we do not have nextOrLess: in Pharo. >>> >>> On Wed, Apr 26, 2017 at 6:19 AM, Richard Sargent <rsargent(a)5x5.on.ca> >>> wrote: >>> >>>> In 2003, I implemented the following method. It is something that comes >>>> up from time to time. >>>> >>>> !SequenceableCollection publicMethods ! >>>> splitEvery: chunkSize >>>> "Answer a collection of sub-collections each of the specified chunk >>>> size. >>>> If the receiver size is not a multiple of chunkSize, the last >>>> sub-collection will be shorter." >>>> >>>> | chunks stream | >>>> chunkSize < 1 ifTrue: [ ^self splitEvery: 1 ]. >>>> >>>> stream := ReadStream on: self. >>>> chunks := OrderedCollection new. >>>> [stream atEnd] whileFalse: [ >>>> chunks add: (stream nextOrLess: chunkSize). >>>> ]. >>>> >>>> ^chunks. >>>> ! ! >>>> >>>> ------------------------------ >>>> *From:* Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] *On >>>> Behalf Of *Stephane Ducasse >>>> *Sent:* April 25, 2017 13:08 >>>> *To:* Any question about pharo is welcome >>>> *Subject:* [Pharo-users] little "challenge" >>>> >>>> #(16 17 17 16 18 17 18 19 19 19 18 19 19 20 19 20 19 20 20 20 >>>> 19 20) >>>> >>>> -> >>>> >>>> { #(17 17 16 18 17 18 19) . #(19 19 18 19 19 20 19) . #(20 19 20 20 20 >>>> 19 20)} >>>> >>>> any code? >>>> >>> >>> >> >

April 26, 2017

Updated tech talk archive

by marcus.denker＠inrira.fr

http://pharo.org/TechTalk <http://pharo.org/TechTalk> Now has a link to the recoding of yesterday Marcus

April 26, 2017

Re: [Pharo-users] little "challenge"

by phil＠highoctane.be

So the last 20 is left out. Correct? Phil On Wed, Apr 26, 2017 at 11:20 AM, Stephane Ducasse <stepharo.self(a)gmail.com> wrote: > Now I fully got the implementation > > > " > #(16 17 17 16 18 17 18 19 19 19 18 19 19 20 19 20 19 20 20 20 19 20) > groupsOf: 7 atATimeCollect: [ :x | x ] > >>> #(#(16 17 17 16 18 17 18) #(19 19 19 18 19 19 20) #(19 20 19 20 20 20 > 19)) > " > " > #(1 1 1 10 10 10 100 100 100) > groupsOf: 3 atATimeCollect: [ :x :y :z | x + y + z] > >>> #(3 30 300) > " > > " > #(1 1 1 10 10 10 100 100 100) > groupsOf: 3 atATimeCollect: [ :x | x ] > >>> #(#(1 1 1) #(10 10 10) #(100 100 100)) > " > > On Wed, Apr 26, 2017 at 8:38 AM, Stephane Ducasse <stepharo.self(a)gmail.com > > wrote: > >> we do not have nextOrLess: in Pharo. >> >> On Wed, Apr 26, 2017 at 6:19 AM, Richard Sargent <rsargent(a)5x5.on.ca> >> wrote: >> >>> In 2003, I implemented the following method. It is something that comes >>> up from time to time. >>> >>> !SequenceableCollection publicMethods ! >>> splitEvery: chunkSize >>> "Answer a collection of sub-collections each of the specified chunk >>> size. >>> If the receiver size is not a multiple of chunkSize, the last >>> sub-collection will be shorter." >>> >>> | chunks stream | >>> chunkSize < 1 ifTrue: [ ^self splitEvery: 1 ]. >>> >>> stream := ReadStream on: self. >>> chunks := OrderedCollection new. >>> [stream atEnd] whileFalse: [ >>> chunks add: (stream nextOrLess: chunkSize). >>> ]. >>> >>> ^chunks. >>> ! ! >>> >>> ------------------------------ >>> *From:* Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] *On >>> Behalf Of *Stephane Ducasse >>> *Sent:* April 25, 2017 13:08 >>> *To:* Any question about pharo is welcome >>> *Subject:* [Pharo-users] little "challenge" >>> >>> #(16 17 17 16 18 17 18 19 19 19 18 19 19 20 19 20 19 20 20 20 19 >>> 20) >>> >>> -> >>> >>> { #(17 17 16 18 17 18 19) . #(19 19 18 19 19 20 19) . #(20 19 20 20 20 >>> 19 20)} >>> >>> any code? >>> >> >> >

April 26, 2017

Re: [Pharo-users] little "challenge"

by Stephane Ducasse

Now I fully got the implementation " #(16 17 17 16 18 17 18 19 19 19 18 19 19 20 19 20 19 20 20 20 19 20) groupsOf: 7 atATimeCollect: [ :x | x ] >>> #(#(16 17 17 16 18 17 18) #(19 19 19 18 19 19 20) #(19 20 19 20 20 20 19)) " " #(1 1 1 10 10 10 100 100 100) groupsOf: 3 atATimeCollect: [ :x :y :z | x + y + z] >>> #(3 30 300) " " #(1 1 1 10 10 10 100 100 100) groupsOf: 3 atATimeCollect: [ :x | x ] >>> #(#(1 1 1) #(10 10 10) #(100 100 100)) " On Wed, Apr 26, 2017 at 8:38 AM, Stephane Ducasse <stepharo.self(a)gmail.com> wrote: > we do not have nextOrLess: in Pharo. > > On Wed, Apr 26, 2017 at 6:19 AM, Richard Sargent <rsargent(a)5x5.on.ca> > wrote: > >> In 2003, I implemented the following method. It is something that comes >> up from time to time. >> >> !SequenceableCollection publicMethods ! >> splitEvery: chunkSize >> "Answer a collection of sub-collections each of the specified chunk size. >> If the receiver size is not a multiple of chunkSize, the last >> sub-collection will be shorter." >> >> | chunks stream | >> chunkSize < 1 ifTrue: [ ^self splitEvery: 1 ]. >> >> stream := ReadStream on: self. >> chunks := OrderedCollection new. >> [stream atEnd] whileFalse: [ >> chunks add: (stream nextOrLess: chunkSize). >> ]. >> >> ^chunks. >> ! ! >> >> ------------------------------ >> *From:* Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] *On >> Behalf Of *Stephane Ducasse >> *Sent:* April 25, 2017 13:08 >> *To:* Any question about pharo is welcome >> *Subject:* [Pharo-users] little "challenge" >> >> #(16 17 17 16 18 17 18 19 19 19 18 19 19 20 19 20 19 20 20 20 19 >> 20) >> >> -> >> >> { #(17 17 16 18 17 18 19) . #(19 19 18 19 19 20 19) . #(20 19 20 20 20 19 >> 20)} >> >> any code? >> > >

April 26, 2017

Epicea vs RB refactorings

by Peter Uhnak

Hi, I'm under the impression that Epicea is supposed to replace RB refactorings at some point. If that is true, is it currently possible to build refactorings by hand? For example in RB I often do: ``` model := RBNamespace new. model defineClass: ('<1s> subclass: #<2s><n><t>instanceVariableNames: <3p><n><t>classVariableNames: ''''<n><t>poolDictionaries: ''''<n><t>category: <4p>' expandMacrosWith: 'Object' with: 'MySomething' with: 'var1 var2' with: 'MyPackage'). cls := model classNamed: #MySomething. cls compile: 'var1 ^ var1' classified: 'accessing'. (ChangesBrowser changes: model changes changes) open ``` or (removing a method from a class) ``` model := RBNamespace new. model removeMethod: #testTrimRight from: StringTest. (ChangesBrowser changes: model changes changes) open ``` Thanks, Peter

April 26, 2017

Re: [Pharo-users] little "challenge"

by phil＠highoctane.be

I implemented it for fun and noticed that in your test data, the first 16 is gone. So, groupings are wrong with all methods proposed in a sense. With Richard's version, we get this in Pharo #(16 17 17 16 18 17 18 19 19 19 18 19 19 20 19 20 19 20 20 20 19 20) splitEvery: 7 "an OrderedCollection(#(16 17 17 16 18 17 18) #(19 19 19 18 19 19 20) #(19 20 19 20 20 20 19) #(20))" splitEvery: chunkSize "Answer a collection of sub-collections each of the specified chunk size. If the receiver size is not a multiple of chunkSize, the last sub-collection will be shorter." | chunks stream | chunkSize < 1 ifTrue: [ ^self splitEvery: 1 ]. stream := ReadStream on: self. chunks := OrderedCollection new. [stream atEnd] whileFalse: [ chunks add: (stream next: chunkSize). ]. ^chunks. Phil On Wed, Apr 26, 2017 at 8:38 AM, Stephane Ducasse <stepharo.self(a)gmail.com> wrote: > we do not have nextOrLess: in Pharo. > > On Wed, Apr 26, 2017 at 6:19 AM, Richard Sargent <rsargent(a)5x5.on.ca> > wrote: > >> In 2003, I implemented the following method. It is something that comes >> up from time to time. >> >> !SequenceableCollection publicMethods ! >> splitEvery: chunkSize >> "Answer a collection of sub-collections each of the specified chunk size. >> If the receiver size is not a multiple of chunkSize, the last >> sub-collection will be shorter." >> >> | chunks stream | >> chunkSize < 1 ifTrue: [ ^self splitEvery: 1 ]. >> >> stream := ReadStream on: self. >> chunks := OrderedCollection new. >> [stream atEnd] whileFalse: [ >> chunks add: (stream nextOrLess: chunkSize). >> ]. >> >> ^chunks. >> ! ! >> >> ------------------------------ >> *From:* Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] *On >> Behalf Of *Stephane Ducasse >> *Sent:* April 25, 2017 13:08 >> *To:* Any question about pharo is welcome >> *Subject:* [Pharo-users] little "challenge" >> >> #(16 17 17 16 18 17 18 19 19 19 18 19 19 20 19 20 19 20 20 20 19 >> 20) >> >> -> >> >> { #(17 17 16 18 17 18 19) . #(19 19 18 19 19 20 19) . #(20 19 20 20 20 19 >> 20)} >> >> any code? >> > >

April 26, 2017

Re: [Pharo-users] little "challenge"

by phil＠highoctane.be

Doesn't matter as next returns if there is less than required. groups := OrderedCollection new. input := (String loremIpsum: 20). input readStream in: [ :s | [s atEnd] whileFalse: [ groups add: (s next: 7) ]]. groups. Interestingly loremIpsum: adds Character cr to from paragraphs (output looked strange at first glance). Phil On Wed, Apr 26, 2017 at 8:38 AM, Stephane Ducasse <stepharo.self(a)gmail.com> wrote: > we do not have nextOrLess: in Pharo. > > On Wed, Apr 26, 2017 at 6:19 AM, Richard Sargent <rsargent(a)5x5.on.ca> > wrote: > >> In 2003, I implemented the following method. It is something that comes >> up from time to time. >> >> !SequenceableCollection publicMethods ! >> splitEvery: chunkSize >> "Answer a collection of sub-collections each of the specified chunk size. >> If the receiver size is not a multiple of chunkSize, the last >> sub-collection will be shorter." >> >> | chunks stream | >> chunkSize < 1 ifTrue: [ ^self splitEvery: 1 ]. >> >> stream := ReadStream on: self. >> chunks := OrderedCollection new. >> [stream atEnd] whileFalse: [ >> chunks add: (stream nextOrLess: chunkSize). >> ]. >> >> ^chunks. >> ! ! >> >> ------------------------------ >> *From:* Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] *On >> Behalf Of *Stephane Ducasse >> *Sent:* April 25, 2017 13:08 >> *To:* Any question about pharo is welcome >> *Subject:* [Pharo-users] little "challenge" >> >> #(16 17 17 16 18 17 18 19 19 19 18 19 19 20 19 20 19 20 20 20 19 >> 20) >> >> -> >> >> { #(17 17 16 18 17 18 19) . #(19 19 18 19 19 20 19) . #(20 19 20 20 20 19 >> 20)} >> >> any code? >> > >

April 26, 2017