- Pharo-users - lists.pharo.org

Re: [Pharo-users] singleton trait

by Sven Van Caekenberghe

> On 21 Jan 2017, at 11:59, Peter Uhnak <i.uhnak(a)gmail.com> wrote: > > On Sat, Jan 21, 2017 at 11:19:24AM +0100, stepharong wrote: >> Pay attention that Singleton should not be about access but time. >> >> If you can by adding on instance variable avoid to need a singleton >> then it means that it was not a singleton. >> >> A singleton is often: >> you **CANNOT** have two instances at the same time. >> >> Many people do not use well singleton so watch out. > > We have several variations of singletons in the Pharo codebase, some of which do allow multiple instances, some of which don't. But they are differentiated by names. > > https://www.peteruhnak.com/blog/2015/12/06/singleton-variations/ That's a good summary, thanks Peter. > Peter > >> >> Stef >> >> On Thu, 19 Jan 2017 11:23:21 +0100, Siemen Baader >> <siemenbaader(a)gmail.com> wrote: >> >>> Is there a trait to quickly create singletons in Pharo, along the >>> way of the Ruby Singleton mixin? >>> >>> https://ruby-doc.org/stdlib-1.9.3/libdoc/singleton/rdoc/Singleton.html >>> >>> (and ideally also observer/observable, which I am going to need next..) >>> >>> -- Siemen >> >> >> >> -- >> Using Opera's mail client: http://www.opera.com/mail/ >

Jan. 21, 2017

Re: [Pharo-users] singleton trait

by Peter Uhnak

On Sat, Jan 21, 2017 at 11:19:24AM +0100, stepharong wrote: > Pay attention that Singleton should not be about access but time. > > If you can by adding on instance variable avoid to need a singleton > then it means that it was not a singleton. > > A singleton is often: > you **CANNOT** have two instances at the same time. > > Many people do not use well singleton so watch out. We have several variations of singletons in the Pharo codebase, some of which do allow multiple instances, some of which don't. But they are differentiated by names. https://www.peteruhnak.com/blog/2015/12/06/singleton-variations/ Peter > > Stef > > On Thu, 19 Jan 2017 11:23:21 +0100, Siemen Baader > <siemenbaader(a)gmail.com> wrote: > > >Is there a trait to quickly create singletons in Pharo, along the > >way of the Ruby Singleton mixin? > > > >https://ruby-doc.org/stdlib-1.9.3/libdoc/singleton/rdoc/Singleton.html > > > >(and ideally also observer/observable, which I am going to need next..) > > > >-- Siemen > > > > -- > Using Opera's mail client: http://www.opera.com/mail/

Jan. 21, 2017

Re: [Pharo-users] Tools for easy subtext extraction from text

by Henrik Nergaard

>* I have to manually remove the 0x, even though it is a very common way of expressing hex numbers Adding a few lines to NumberParser>>#nextNumber enables it to parse 0x... ----------------------------------- ((sourceStream peekFor: $r)) ifTrue: ["<base>r<integer>" ... ] ifFalse: [ (sourceStream peekFor: $x) ifTrue: [ "0x<integer>" (integerPart isZero and: [ numberOfTrailingZeroInIntegerPart = 1]) ifFalse: [ sourceStream skip: -1. ^ self expected: 'one leading 0 before x' ]. ^ self nextUnsignedIntegerBase: 16 ] ]. ---------------------------------- 0x10. "16" 0xFF + 16rFF = (2 * 0xff). "true" Best regards, Henrik -----Opprinnelig melding----- Fra: Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] PÃ¥ vegne av Peter Uhnak Sendt: 20 January 2017 16:16 Til: pharo-users(a)lists.pharo.org Emne: [Pharo-users] Tools for easy subtext extraction from text Hi, what are the tools available from easier text extraction? The input is unstructured text, but I want to extract portion from it. I am not looking for an engineered approach (writing a parser or something), but something that can be done quickly by hand (i.e. interactively). For example I have string str = ' Temperature 0 37C (98F) [0x25] (TMPIN0)' Now I want to extract '0x25' from it and convert it into integer In Ruby it is dead simple: str[/\[(.*)\]/,1].hex # "=> 37" , or .to_i(16) In Pharo I have to break my fingers first: rx := '.*\[0x(.*)\].*' asRegex. rx matches: str. Integer readFrom: (rx subexpression: 2) base: 16 "=>37". * I have to know that to get subexpression I have to match first to manipulate the internal state * I have to store the matcher to access the subexpression * I need to explicitly use some global variable Integer as a conversion utility to convert hex to dec * I have to manually remove the 0x, even though it is a very common way of expressing hex numbers So the question is: do we have a better way to do these things? And as I've mentioned - the use case is interactive coding where you often throw the code away when you are done; so dead easy to write and use. Thanks, Peter

Jan. 21, 2017

Re: [Pharo-users] is PetitParser eager by default?

by stepharong

Hi doru where do we see it? Stef On Fri, 20 Jan 2017 16:57:51 +0100, Tudor Girba <tudor(a)tudorgirba.com> wrote: > Hi, > > When you have questions like these, you can also use the built-in > debugging facilities. For example, in your case, you can see that the > #any parser consumed everything like this: > > > > Cheers, > Doru > > >> On Jan 20, 2017, at 4:38 PM, Norbert Hartl <norbert(a)hartl.name> wrote: >> >> >>> Am 20.01.2017 um 15:24 schrieb Peter Uhnak <i.uhnak(a)gmail.com>: >>> >>> Is PetitParser eager by default? >>> >>> I've used PetitParser countless times so I am really baffled why this >>> doesn't work >>> >>> str := 'a0b'. >>> #any asParser star, #digit asParser, #any asParser star parse: str. >>> >>> -> PPFailure (input expected at: 3) >>> >> >> PetitParser is not greedy per default. But back tracking only works if >> a parser fails. Using , creates a sequence of combined parsers. If one >> fails the whole sequence fails. As a star >>parser always succeeds it >> would be huge luck if your rule would succeed. The probability that the >> parser consumes exactly one character is not high. Maybe negating the >> first >>sequence part is what you want >> >> Norbert >> >> > > -- > www.tudorgirba.com > www.feenk.com > > "What we can governs what we wish." > > > > -- Using Opera's mail client: http://www.opera.com/mail/

Jan. 21, 2017

Re: [Pharo-users] About Git

by stepharong

On Sun, 15 Jan 2017 14:53:31 +0100, Peter Uhnak <i.uhnak(a)gmail.com> wrote: > I find these endless git vs monticello discussions confusing and > pointless. + 1 :) > > Maybe we can hang Q&A list somewhere on pharo website to point to? > Because git is getting increased traction in Pharo, so the same > questions and endless discussions will popup over and over again. > > 1. Some people here are acting like STHub and Monticello is going away > tomorrow. I don't see how that can happen unless Pharo decides to > fundamentally redo its object model. (STHub may die to free some > resources, but I don't think that's a foreseeable future plan.) > 2. Git is not "new", or "cool", or "bandwagon" technology; it has been > here for a decade and more than proven that its value (in fact git is > older than Pharo ;)). > 3. If you consider the nomenclature confusing because you don't > understand or use some concepts, that's fine. That's what tools are for > (e.g. I don't use `git add` with Pharo, because I can pick what changes > I want with Komitter ( > https://www.peteruhnak.com/blog/2016/08/12/fine-grained-committing-and-exte… > ) similarly Iceberg should be capable of this too (I am on Pharo 5 so > iirc)). > 4. Some comparisons between online git managers (github/bitbucket/...) > and sthub just show that people don't understand > e.g. > "i download a program, add something, upload my addition. lets take an > upload step, _one_ simple step with monticello > "i upload something once (git add), i upload it a second time (git > commit), i upload it a third time (git push), i try to upload it a > fourth time (pull request)" > > The correct version for monticello is (from what I've experienced): I > download a program, add something locally, send mail to maintainer, > wait, get permission. Now I have complete control over someone else's > repository. I release a new version but make a mistake in configuration; > now every downstream project is broken and CI informs me only > afterwards). > If said person doesn't want to give me full access, I have to fileout my > changes (or copy paste them), send them by email, wait more. > I don't find that a good workflow. > > If you are working on your own project and want to ignore many things > git has to offer (which I do regularly for many projects where it makes > sense), then you can have literally the same "one click" with very > little work (that can be part of Iceberg/GitFileTree for anyone else to > use). You click commit in pharo; now it's on github (or wherever). > > My views are certainly biased as I've been using git for long time > before I came to Monticello, which I still find very limiting. > > If I am needlessly unfair towards monticello then please correct me --- > which is also part of my point; this knowledge shouldn't be hidden in > mailing lists (as they are usually disorganized by nature), but in some > easily discoverable and structured form (which I guess also loops this > discussion back into the original Google visibility topic ;)). > > Peter > > On Sun, Jan 15, 2017 at 01:12:37PM +0100, Norbert Hartl wrote: >> >> >> > Am 15.01.2017 um 10:29 schrieb "itlists(a)schrievkrom.de" >> <itlists(a)schrievkrom.de>: >> > >> > I would give you many more "+" - if you wish - for your opinion: >> > >> > ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ >> > >> > >> > Another point I would like to mention: keep the source code locally or >> > at least have an easy option to hold all source code locally. >> > >> > I've seen so many times, that git or one of the other git repositories >> > are down - that's a very critical point. >> > >> But....that's always the case with git. You have always all the code >> plus history locally. Plus you can commit while being offline. Which is >> exactly one point where git excels. Even envy cannot do that if I >> understood it correctly. >> >> >> > I would like to have a local repository, where I can import external >> > packages and I work locally only. >> > >> You can do that with metacello. >> >> > I also have thought several times that it would be nice to have a >> SQLite >> > database holding all sources/packages. How much easier life would be >> > with that. >> > >> I cannot see how using an in-memory sql database can make your life >> easier. To do what? >> >> Norbert >> > Other than that: I prefer my simple server directory holding >> monticello >> > packages - in the Gemstone/S area. >> > >> > But I have to admit, that all the source code management stuff I've >> seen >> > since ENVY in the Smalltalk community are pretty poor stuff - and even >> > considering integrating git is not a step forward in terms of >> technical >> > improvements, but only in terms of marketing and mainstream >> technology. >> > >> > >> > Marten >> > >> > >> > >> >> Am 14.01.2017 um 18:44 schrieb werner kassens: >> >> Hi Dimitris, >> >> i as a simple user tend to think about these things in simple terms: >> i >> >> download a program, add something, upload my addition. lets take an >> >> upload step, _one_ simple step with monticello: i upload something >> once >> >> (git add), i upload it a second time (git commit), i upload it a >> third >> >> time (git push), i try to upload it a fourth time (pull request), >> only >> >> then i'm done (yes, there are possible shortcuts but so what). i'm >> sure >> >> all this makes sense for the seasoned coder, but i certainly won't >> learn >> >> that. <friendly grin> just as a small example how a user like me >> thinks >> >> about this. >> >> werner >> >> >> > >> > >> > >> > -- >> > Marten Feldtmann >> > >> >> > -- Using Opera's mail client: http://www.opera.com/mail/

Jan. 21, 2017

Re: [Pharo-users] super super

by stepharong

No and I'm happy that this is like that. I think that it means that you need to create special hooks in the super superclass. > Hi, > > Is there a way from an instance to access a method situated 2 layers up > in the class hierarchy? > > A kind of super super. > > Thanks > > Hilaire -- Using Opera's mail client: http://www.opera.com/mail/

Jan. 21, 2017

Re: [Pharo-users] singleton trait

by stepharong

Pay attention that Singleton should not be about access but time. If you can by adding on instance variable avoid to need a singleton then it means that it was not a singleton. A singleton is often: you **CANNOT** have two instances at the same time. Many people do not use well singleton so watch out. Stef On Thu, 19 Jan 2017 11:23:21 +0100, Siemen Baader <siemenbaader(a)gmail.com> wrote: > Is there a trait to quickly create singletons in Pharo, along the way of > the Ruby Singleton mixin? > > https://ruby-doc.org/stdlib-1.9.3/libdoc/singleton/rdoc/Singleton.html > > (and ideally also observer/observable, which I am going to need next..) > > -- Siemen -- Using Opera's mail client: http://www.opera.com/mail/

Jan. 21, 2017

Re: [Pharo-users] Tools for easy subtext extraction from text

by Dimitris Chloupis

> > > In Ruby it is dead simple: > str[/\[(.*)\]/,1].hex # "=> 37" , or .to_i(16) > and dead unreadable Pharo way is both dead simple and dead readable str:= ' Temperature 0 37C (98F) [0x25] (TMPIN0)'. ('16r',(text copyWithRegex: '(.*\[0x)|(\].*)' matchesReplacedWith: '')) asNumber .

Jan. 21, 2017

Re: [Pharo-users] Couple grafoscopio notes

by Offray Vladimir Luna CÃ¡rdenas

Hi Peter, :-) On 19/01/17 11:49, Peter Uhnak wrote: > Hi, > > I've started playing around with grafoscopio, I am happy that I can finally organize all those little scripts and not lose them all the time. :) Thanks for using Grafoscopio. It started to explore moldability by bridging documentation and scripting with live coding and objects. Pharo and the communities around it have been really empowering for such exploration, so I'm glad you find it useful. > > However there are couple of issues and notes I've ran into (don't worry about fixing them in any timely fashion, it is not blocking me) > > 1. For saving `UITheme builder` is used, but the correct approach is via `UIManager default` > * this has obvious impact if the UIManager is different (UIManager has reference to UITheme, but not the other way around) > > To see all the calls, you can do this (or find the uses manually) > > calls := UITheme allCallsOn select: [ :each | each package name = #Grafoscopio ]. > SystemNavigation default browseMessageList: calls name: 'Users of class UITheme' autoSelect: 'UITheme builder'. > > 2. Typo "Add nodo" -> "Add node" > > 3. "Move up|down node" -> "Move node up|down" (to be consistend with the rest of the naming) Items 1 to 3 have been fixed. Please update to Grafoscopio-OffrayLuna.250 or later. > > 4. During installation (from Catalog) a 'Grafoscopio' folder is created in documents and documentation is downloaded > > This is a big no. Please don't do this. > Not only it violates the isolation of images, but it actually breaks the installation itself when you install Grafoscopio in two different images. > > If you have to download the docs and you cannot just store them inside the image (which would be logical choice for read-only), then please use the image's directory by default, and only in e.g. development mode use some other location. This was for historical reasons. At the beginning documentation was installed on the same image directory, but, as a developer, I need a fixed place to refer to the documentation from different images and also, in the Data Week workshops, participants need such fixed place to share scripts more easily. Now all documentation is located in the same image directory, so we don't violate images isolation, but the new menu item 'Update > Documents place' is available, if users need some fixed shared location to work on shared notebooks (this will be improved, but works for now). In the future I plan to make the notebook tree available from the playgrounds embedded in it, so you can write stuff like: 'thisNode children computeAll; exportAsHTML", to bring self-referential programable notebooks to Pharo and data storytelling :-), but that would be theme for its own mail thread. > > 5. Node-related bugs > > Imho this is somehow related to improper use and updates of tree. I remember I had similar issues some time ago, so maybe I can take a look at this in couple of weeks. > > These bugs sometimes autoresolve themselves (so sometimes it is usable and sometimes not) > > * 1. Launch > new notebook > * 2. Add node > error > > * 1. Launch > new notebook > * 2. Select Node 1 > * 3. Add node > * 4. Select newNode > * 5. rename node to something else > * 6. Click on previous "Node 1" > error Yes. I don't understand pretty well the Spec GUI, its events, and its relation with GT-Tools. I would like to have shorcuts for saving the notebook, move nodes in the tree, creation, selection and edition, so I will be interested in your recommendations in a couple of weeks. Let me know, also, if you need any particular permission in the repository. > > To summarize: > > Most are just minor things or annoyances, but #4 is breaking --- as I have to remember to delete the folder every time I want to download Grafoscopio in new image. (If it works for you, then maybe it's platform related (I tested this on Windows 10); in either case, the data related to the image should stay with the image (either in the image, or in the image's folder), and not somewhere in the system). > > Thanks! > > Peter > Thanks, Offray

Jan. 21, 2017

Google summer of code

by Alexandre Bergel

Hi everyone- Google Summer of Code 2017 has officially begun! Organization applications open today, Thursday January 19 and are open through Thursday, February 9th. Please see our program site, <http://g.co/gsoc> official timeline <https://developers.google.com/open-source/gsoc/timeline> and FAQ <https://developers.google.com/open-source/gsoc/faq> for more details. What makes a good organization application? Take a look at our manuals <https://developers.google.com/open-source/gsoc/resources/manual> for tips and best practices. We look forward to seeing each of your applications and kicking off another great year of GSoC. If you have any questions, donât hesitate to reach out to gsoc-support(a)google.com <mailto:gsoc-support@google.com>. Weâre here to help! Google Open Source Programs Office -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

Jan. 20, 2017