- Pharo-users - lists.pharo.org

Personal Programming

by Trygve

Dear all, The final draft of my magnum opus about Personal Programming is now ready for review: http://folk.uio.no/trygver/themes/Personal/PP-019%20-%20Copy%20(17).pdf The article's main theme is Personal Programming for everybody with /Loke/, a personal object computer. Its first purpose is to empower laypeople to take control over their corner of the Net with its IoT. I have created a proof-of-concept implementation as a non-intrusive extension of Squeak version 3.10.2, and have used it to demonstrate how a novice uses the Loke IDE to create a small and intuitive program. The article describes the concepts behind Loke .The current Squeak implementation should be ported to Pharo and can grow into the preferred Pharo-based IDE for laypeople taking control over their information environment. I will appreciate your possible comments before Aug. 31. Enjoy --Trygve The article's 43 pages has several high points, I have included one of them here: ----------- begin extract ------------------ C.7.We need a paradigm shift The history of Western astronomy shows a series of paradigm shifts from the geocentric paradigm with its stationary Earth as the center of the Universe with its epicycles and other bizarre explanations of what appeared to be essential complexities. Astronomy evolved via the heliocentric to the current distributed paradigm with its chunks of mass connected by gravity. What appeared as essential complexity in one paradigm was easily resolved in the next. It is tempting to look for similar paradigm shifts in computing. Mainstream programming has based much of its theory and practice on the CPU-centric paradigm exemplified by the von Neumann machine. A memory-centric paradigm came in 1960 with the Autokon CAC/ CAM system and its central database (Reenskaug, 1973). The solution was obvious, and there must have been many similar initiatives without me being aware of them. It is time to realize that the first two paradigms do not meet our current challenges: We are plagued with immensely large, complex, and insecure systems that long ago left the realm of human understanding. A recent example: Customers found that their bank charged them twice for the same transaction. Several weeks after the problem was discovered, the bank publicly admitted that they still didn't understand how the problem could arise: The complexity of their system was clearly beyond human comprehension. The bank has a staff of very competent experts, but they need a better foundation for modeling and implementing their sophisticated requirements. Computers can transform, store, and communicate data, (Figure below). The essence of the CPU-centric paradigm is that computers are primarily used to transform data; they compute. The essence of the memory-centric paradigm is that computers are used primarily to store data; they organize applications around a shared database. The essence of the communication-centric paradigm is that computers are primarily used to exchange messages with other computers to make them collaborate to achieve a common goal. /The three paradigms of computing // /// It is time to heed Tony Hoare's plea for simplicity and achieve a better way of separating concerns. Mainstream programming should shift to the communication-centric paradigm exemplified by the object computer that is the foundation for this article. The communication-centric paradigm has been on the horizon for many years. I first met it in Prokon's idea of distributed computers (Reenskaug, 1977), but there must have been many other initiatives. A newer example is Service-Oriented Architectures (SOA) that, in essence, is communication-centric. It didn't meet with immediate success, possibly because people tried to apply it within the CPU-centric paradigm where it doesn't belong. There are many other examples such as distributed computing. And of course, DCI and the IoT itself are, by definition, communication-centric. ----------- end extract ------------------

July 25, 2019

Re: [Pharo-users] OrderedCollection as an instance variable

by Richard O'Keefe

Comment 1. You probably meant to write tracks: aCollection ^tracks := aCollection Comment 2. This is a poor design. As it is, any object can replace the tracks of an artist with *anything*. And even without doing that, any object can add and remove items to an artist's tracks, even if the added items are not Tracks. There are a number of OO design principles, notably the Law of Demeter and of course the GRASP patterns, which basically say "never mutate another object's parts, ask IT to do the mutation." I have come to regard the self-encapsulation "pattern" as an anti-pattern. This is Smalltalk heresy, admittedly, but it doesn't so much break (real) encapsulation as set it on fire, defenestrate it, and urinate on the smouldering remains. So this is the way I'd do it. Object subclass: #Artist instanceVariableNames: 'tracks' methods for: 'initialising' initialize tracks := OrderedCollection new. ^self methods for: 'checking' invariant ^(tracks isMemberOf: OrderedCollection) and: [ tracks allSatisfy: [:each | each isKindOf: Track]] methods for: 'accessing tracks' addTrack: aTrack. tracks add: aTrack. ^track tracksDo: doBlock tracks do: doBlock with no way for outside code to accesss 'tracks' directly. On Thu, 25 Jul 2019 at 02:47, sergio ruiz <sergio.rrd(a)gmail.com> wrote: > I'm implementing an instance variable as an OrderedCollection, and am > doing something idiotic. > > I have an Artist class. > > An Artist can have many tracks: > > tracks > ^ tracks ifNil: [ self tracks: OrderedCollection new ] > > tracks: anObject > tracks := anObject > > Doing something like: > > a := Artist new. > t := Track new. > a tracks add: t > > Gives me the error: > > Artist(Object)>>doesNotUnderstand: #add: > > Once I get past that error, which will evaluate `a tracks`, I can run this. > > I get why it works the second time, but I don't get why it thinks the > instance of Artist is the receiver of add: on the first run. > > Thanks! > ---- > peace, > sergio > photographer, journalist, visionary > > Public Key: http://bit.ly/29z9fG0 > #BitMessage BM-NBaswViL21xqgg9STRJjaJaUoyiNe2dV > http://www.codeandmusic.com > http://www.twitter.com/sergio_101 > http://www.facebook.com/sergio101 > >

July 25, 2019

Re: [Pharo-users] [OrderedCollection] add a reference or a copy?

by Richard O'Keefe

To a very good approximation, Smalltalk doesn't copy anything unless you ask it to. In this respect it's just like Java, Python, Ruby, ECMAScript, and most OO languages. C++ *does* like to copy things, but it is unusual. On Thu, 25 Jul 2019 at 00:46, sergio ruiz <sergio.rrd(a)gmail.com> wrote: > I think my understanding of OrderedCollections has been incorrect for a > very long time. I have never had a situation where it mattered, but in > modeling a current project, I think I have been approaching it incorrectly. > > Since I have been using Smalltalk, I assumed that a copy of an object is > added to an OrderedCollection. A quick test shows that a reference to that > object is added to the OrderedCollection. > > I just wanted a sanity check on this. Is this true with all collections? > > Thanks! > > > ---- > peace, > sergio > photographer, journalist, visionary > > Public Key: http://bit.ly/29z9fG0 > #BitMessage BM-NBaswViL21xqgg9STRJjaJaUoyiNe2dV > http://www.codeandmusic.com > http://www.twitter.com/sergio_101 > http://www.facebook.com/sergio101 > >

July 25, 2019

Re: [Pharo-users] P7 regression

by Sven Van Caekenberghe

That is correct, sk. #writeStreamDo: is enough though, #utf8 is the default encoding > On 24 Jul 2019, at 22:16, sk via Pharo-users <pharo-users(a)lists.pharo.org> wrote: > > > From: sk <snoobabk(a)yahoo.ie> > Subject: Re: [Pharo-users] P7 regression > Date: 24 July 2019 at 22:16:50 GMT+2 > To: Any question about pharo is welcome <pharo-users(a)lists.pharo.org> > > > Hi Hilaire > > I downloaded latest Dr Geo and took a quick look at how the files are saved. > It looks like the xml contents are not encoded (in UTF-8) but simply written to a binary write stream, which means that when a WideString is written you should consistently get a badly encoded file. > > You could test this out by changing this code > DrGDirectoryLocal>>put: stream into: filename > (location asFileReference / filename) ensureDelete binaryWriteStreamDo: [ :fileStream | > fileStream nextPutAll: stream contents] > > > to rather be: > DrGDirectoryLocal>>put: stream into: filename > (location asFileReference / filename) ensureDelete > writeStreamEncoded: 'utf-8' > do: [ :fileStream | fileStream nextPutAll: stream contents ] > > When I made this modification it worked for me i.e. the file was readable and correctly encoded. > > Regards > Carlo > > On 24 Jul 2019, at 21:07, Hilaire <hilaire(a)drgeo.eu> wrote: > > [...] > stream := WriteStream on: (String new: 4000). > DrGeoXml new > app: self app; > saveOn: stream. > [...] > > DrGeoXml>>saveOn: stream > | doc writer root | > doc := XMLDocument new version: '1.0'. > writer := XMLWriter on: stream. > root := XMLElement named: #drgenius. > self writeFigureAsXmlTo: root. > doc addElement: root. > doc printXMLOn: writer. > > The stream is passed to the XMLWriter, then it goes down to each items > of the sketch to write its own XML node. > > For example, to write its basic information as its name, where I have issue: > > writeAsXmlTo: aNode > "return the newly created element" > | node | > self rehash. > node := XMLElement named: self basicType attributes: Dictionary new. > node attributeAt: #type put: self nodeType; > attributeAt: #name put: (name ifNil: ['']); > attributeAt: #id put: self hash asString. > self writeParentsAsXmlTo: node. > aNode addElement: node. > ^node > > This code is as is since about 2005-2007 > > I will dig in it but it is strange. > > > Le 24/07/2019 Ã 18:45, Sven Van Caekenberghe a Ã©crit : >> But how do you write it to a file then ? >> >> FileLocator desktop / 'hilaire.xml' writeStreamDo: [ :out | out << 'Les Ã©lÃ¨ves FranÃ§aises ont 10 â¬' ]. >> >> (FileLocator desktop / 'hilaire.xml') contents. >> >> ">> 'Les Ã©lÃ¨ves FranÃ§aises ont 10 â¬'" > > -- > Dr. Geo > http://drgeo.eu > > > > > > >

July 24, 2019

Re: [Pharo-users] P7 regression

by sk

Hi Hilaire I downloaded latest Dr Geo and took a quick look at how the files are saved. It looks like the xml contents are not encoded (in UTF-8) but simply written to a binary write stream, which means that when a WideString is written you should consistently get a badly encoded file. You could test this out by changing this code DrGDirectoryLocal>>put: stream into: filename (location asFileReference / filename) ensureDelete binaryWriteStreamDo: [ :fileStream | fileStream nextPutAll: stream contents] to rather be: DrGDirectoryLocal>>put: stream into: filename (location asFileReference / filename) ensureDelete writeStreamEncoded: 'utf-8' do: [ :fileStream | fileStream nextPutAll: stream contents ] When I made this modification it worked for me i.e. the file was readable and correctly encoded. Regards Carlo On 24 Jul 2019, at 21:07, Hilaire <hilaire(a)drgeo.eu> wrote: [...] stream := WriteStream on: (String new: 4000). DrGeoXml new app: self app; saveOn: stream. [...] DrGeoXml>>saveOn: stream | doc writer root | doc := XMLDocument new version: '1.0'. writer := XMLWriter on: stream. root := XMLElement named: #drgenius. self writeFigureAsXmlTo: root. doc addElement: root. doc printXMLOn: writer. The stream is passed to the XMLWriter, then it goes down to each items of the sketch to write its own XML node. For example, to write its basic information as its name, where I have issue: writeAsXmlTo: aNode "return the newly created element" | node | self rehash. node := XMLElement named: self basicType attributes: Dictionary new. node attributeAt: #type put: self nodeType; attributeAt: #name put: (name ifNil: ['']); attributeAt: #id put: self hash asString. self writeParentsAsXmlTo: node. aNode addElement: node. ^node This code is as is since about 2005-2007 I will dig in it but it is strange. Le 24/07/2019 Ã 18:45, Sven Van Caekenberghe a Ã©crit : > But how do you write it to a file then ? > > FileLocator desktop / 'hilaire.xml' writeStreamDo: [ :out | out << 'Les Ã©lÃ¨ves FranÃ§aises ont 10 â¬' ]. > > (FileLocator desktop / 'hilaire.xml') contents. > > ">> 'Les Ã©lÃ¨ves FranÃ§aises ont 10 â¬'" -- Dr. Geo http://drgeo.eu

July 24, 2019

Re: [Pharo-users] P7 regression

by Hilaire

[...] stream := WriteStream on: (String new: 4000). DrGeoXml new Â Â app: self app; Â Â saveOn: stream. [...] DrGeoXml>>saveOn: stream Â Â Â | doc writer root | Â Â Â doc := XMLDocument new version: '1.0'. Â Â Â writer := XMLWriter on: stream. Â Â Â root := XMLElement named: #drgenius. Â Â Â self writeFigureAsXmlTo: root. Â Â Â doc addElement: root. Â Â Â doc printXMLOn: writer. The stream is passed to the XMLWriter, then it goes down to each items of the sketch to write its own XML node. For example, to write its basic information as its name, where I have issue: writeAsXmlTo: aNode Â Â Â "return the newly created element" Â Â Â | node | Â Â Â self rehash. Â Â Â node :=Â XMLElement named: self basicType attributes: Dictionary new. Â Â Â node attributeAt: #type put: self nodeType; Â Â Â Â Â Â attributeAt: #name put: (name ifNil: ['']); Â Â Â Â Â Â attributeAt: #id put: self hash asString. Â Â Â self writeParentsAsXmlTo: node. Â Â Â aNode addElement: node. Â Â Â ^node This code is as is since about 2005-2007 I will dig in it but it is strange. Le 24/07/2019 Ã 18:45, Sven Van Caekenberghe a Ã©critÂ : > But how do you write it to a file then ? > > FileLocator desktop / 'hilaire.xml' writeStreamDo: [ :out | out << 'Les Ã©lÃ¨ves FranÃ§aises ont 10 â¬' ]. > > (FileLocator desktop / 'hilaire.xml') contents. > > ">> 'Les Ã©lÃ¨ves FranÃ§aises ont 10 â¬'" -- Dr. Geo http://drgeo.eu

July 24, 2019

Save the date: October 22nd in Ghent for a Smalltalk WebCamp

by Johan Brichau

Hi Seaside users, We are organising a Smalltalk WebCamp day in the Yesplan offices in Ghent on October 22nd. The idea originated in the Seaside channel of the Pharo discord group. However, itâs not supposed to be Seaside only. All Smalltalkers doing web development, possibly involving other technologies on top of Smalltalk, are encouraged to join. Letâs meet and exchange ideas. The topics are to be determined, following the idea of an unconference [1], except that, for practical reasons, we will ask you to confirm your attendance beforehand. More info to follow asap. Save the date and respond already to the group if you are interested to attend. Cheers Johan [1] https://en.wikipedia.org/wiki/Unconference <https://en.wikipedia.org/wiki/Unconference>

July 24, 2019

Re: [Pharo-users] OrderedCollection as an instance variable

by Sebastian Heidbrink

A book that looks explores implementations like that and that always had a place on my desk is this https://www.amazon.ca/Smalltalk-Style-Suzanne-Skublics/dp/0131655493 I can highly recommend it. Sebastian On 2019-07-24 8:32 a.m., Marcus Denker wrote: > >> On 24 Jul 2019, at 17:30, sergio ruiz <sergio.rrd(a)gmail.com> wrote: >> >> hmm??? >> >> maybe this is cleaner.. >> >> tracks >> tracks ifNil: [ self tracks: OrderedCollection new ]. >> ^ tracks >> > I write these methods as > > > tracks > ^tracks ifNil: [ tracks := OrderedCollection new ]. > > - one line > - does not mix using accusers and not using accessors. > > > Marcus >

July 24, 2019

Re: [Pharo-users] P7 regression

by Sven Van Caekenberghe

> On 24 Jul 2019, at 17:32, Hilaire <hilaire(a)drgeo.eu> wrote: > > Hi Sven, > > Le 24/07/2019 Ã 17:05, Sven Van Caekenberghe a Ã©crit : >> Can you include the file that you are trying to load ? > > Attached. > > It looks like issue come when saving the file. The UTF8 characters are > not saved properly. The unix file command was indicating it is an utf8 > but Emacs shows it was different. > >> >> Also, the stack trace is very short. > > Yes, I saw that... I don't know why. I did not cut it. I got a longer > one this time. But it is likely not the cause. > >> Are you sure XML parser is ready for Pharo 8 (i.e. are its tests green) ? > > My bad, I am using P7. But I don't know either about the full > compatibility of the XMLParser version I am using, the one before > compatibility break. > > It is now clear the is issue when saving. To write, I use WriteStream on > String. Should it be different? But how do you write it to a file then ? FileLocator desktop / 'hilaire.xml' writeStreamDo: [ :out | out << 'Les Ã©lÃ¨ves FranÃ§aises ont 10 â¬' ]. (FileLocator desktop / 'hilaire.xml') contents. ">> 'Les Ã©lÃ¨ves FranÃ§aises ont 10 â¬'" > Hilaire > > > -- > Dr. Geo > http://drgeo.eu > > <test.fgeo><PharoDebug.log>

July 24, 2019

Re: [Pharo-users] OrderedCollection as an instance variable

by Esteban Maringolo

I agree with that, but in some cases, like reading objects from an ORM, you would be creating lots of instances of objects that are going to be discarded right away. This would put no stress on GC since it would happen in the "new" space, but would consume more memory (although I never measured how much more). Esteban A. Maringolo On Wed, Jul 24, 2019 at 1:14 PM Richard Sargent <richard.sargent(a)gemtalksystems.com> wrote: > > I will add that I prefer the static model to tell the truth about its data types. So, I prefer having an #initialize method which ensures the new object is created in a self-consistent way and anyone looking at it in an inspector will see useful and correct information. Leaving instance variables uninitialized is a excellent way to generate Does Not Understand errors (perhaps not in the short-term, but definitely in the long-term). > > > On Wed, Jul 24, 2019 at 8:39 AM sergio ruiz <sergio.rrd(a)gmail.com> wrote: >> >> >> tracks >> ^tracks ifNil: [ tracks := OrderedCollection new ]. >> >> - one line >> - does not mix using accusers and not using accessors. >> >> >> Okay, I totally get the subtlety now. I donât even need an accessor. >> >> Thanks! >> >> ---- >> peace, >> sergio >> photographer, journalist, visionary >> >> Public Key: http://bit.ly/29z9fG0 >> #BitMessage BM-NBaswViL21xqgg9STRJjaJaUoyiNe2dV >> http://www.codeandmusic.com >> http://www.twitter.com/sergio_101 >> http://www.facebook.com/sergio101 >>

July 24, 2019