June 2022 - Pharo-dev - lists.pharo.org

Re: [Compiler API] unifying isScriping / isForScripting / isWorkspace

by Marcus Denker

> On 28 Jun 2022, at 09:29, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > This is not directly the compiler API, but the API used by tools (e.g. requestors): > > There where three methods to test if the requestor operates on whole methods or not. > > - isWorkspace > - isScripting > - isForScripting > > This had lead to odd bugs in the past that something that was tested for one requestor failed if used in another context. > (For that, we added both #isScripting and #isForScripting for some cases already). > > This Issue points to a series of PRs that unifies the situation: > > unify #isScripting and #isForScripting #9921 > https://github.com/pharo-project/pharo/issues/9921 > > > Note: it is not really possible to do this backward compatible in a nice way, but the good news is that this is an API used for tools that should not be used normally outside of the development tools themselves. > There are two PRs to be reviewed: 1) spec https://github.com/pharo-spec/Spec/pull/1289 <https://github.com/pharo-spec/Spec/pull/1289> 2) rest of the system: https://github.com/pharo-project/pharo/pull/11416 <https://github.com/pharo-project/pharo/pull/11416> Marcus

June 28, 2022

[Compiler API] unifying isScriping / isForScripting / isWorkspace

by Marcus Denker

This is not directly the compiler API, but the API used by tools (e.g. requestors): There where three methods to test if the requestor operates on whole methods or not. - isWorkspace - isScripting - isForScripting This had lead to odd bugs in the past that something that was tested for one requestor failed if used in another context. (For that, we added both #isScripting and #isForScripting for some cases already). This Issue points to a series of PRs that unifies the situation: unify #isScripting and #isForScripting #9921 https://github.com/pharo-project/pharo/issues/9921 Note: it is not really possible to do this backward compatible in a nice way, but the good news is that this is an API used for tools that should not be used normally outside of the development tools themselves. Marcus

June 28, 2022

Traits-Based Collections, Streams, Etc

by sean＠clipperadams.com

What ever happened to this work? It was exciting and seemed valuable but never made it into Pharo. The traited collections \[1\] seemed particularly interesting and useful 1 https://www.researchgate.net/profile/Damien-Pollet/publication/45942931_BLO…

June 25, 2022

Re: [Compiler] Variables in DoIts

by Guillermo Polito

> El 25 jun 2022, a las 14:38, Denis Kudriashov <dionisiydk(a)gmail.com> escribiÃ³: > > Now we can play with other interesting options. > We can completely remove reformatting of code for DoIts and make the debugging more "transparent" for users: > > <simpleDoIt.png> > > Notice there is no DoIt header anymore . What is selected for the DebugIt command is what the user really sees in the debugger. > I remember when for the first time I tried the smalltalk environment DoIt methods were a real magic for me. With this view I would have much less questions. > > What do you think? > I like this much much better than what was that before!! Iâd suggest, for further improvement (but not blocking), that Iâd like to see the entire code and context of the do-it context. Just showing the selected text makes one lose context. Instead, it would be super cool to show the entire context + code of where the selection was done, and that the program counter is set where the selection was set. Of course, now that Iâm thinking about it, I would wonder how to resolve when to *stop*. Only showing the selected code solves visually the *stopping*. It is clear that evaluation stops when we arrive at the end. G

June 25, 2022

Re: [Compiler] Variables in DoIts

by Denis Kudriashov

Now we can play with other interesting options. We can completely remove reformatting of code for DoIts and make the debugging more "transparent" for users: [image: simpleDoIt.png] Notice there is no DoIt header anymore . What is selected for the DebugIt command is what the user really sees in the debugger. I remember when for the first time I tried the smalltalk environment DoIt methods were a real magic for me. With this view I would have much less questions. What do you think?

June 25, 2022

[Compiler] Variables in DoIts

by Marcus Denker

Hi, We finally merged a PR form Denis (Sorry for taking too longâ¦) that enables the use of DoitVariable in DoItIn: methods: Denis: #pharo has finally transparent variables in the debugger doIts. No ThisContext readVariable trick And we could even get rid of #DoItIn:, after merging another small cleanup, both methods now is just #DoIt as a selector. How does it work? Denis: It is the power of first class variables. There is DoItVariable fully integrated now which represents variables from different context Marcus: DoitVariable has the original Variable instance and the context (what used to the ThisContext parameter of DoItIn:). It can be read reflectively by sending #read, compiler delegates code generation to the Variables (#emitValue:)

June 24, 2022

This week (25/2022) on the Pharo Issue Tracker

by Marcus Denker

We merged 17 PRs this week. The main news is that DoIts that evaluate against a context (e.g. printing a value of a temp variable in the debugger) now use normal variables names (PR 9588). #DoItIn: is gone, as the parameter is not needed anymore (PR 11402) And PR#11379 enables IM support in SDL2: Some natural languages needs a support agent so-called IM to input letters/symbols including emojis. Compiler ======== - [Compiler] Reify OCSemanticScope's representing the meaning of the code: a method or a doIt expressions #9588 https://github.com/pharo-project/pharo/pull/9588 - Compiler-Cleanup-OldDoItTransformations #11402 https://github.com/pharo-project/pharo/pull/11402 - OpalCompiler-Simplify-interactive #11392 https://github.com/pharo-project/pharo/pull/11392 - Simplify-BytecodeGenerator-AdditionalLiterals #11383 https://github.com/pharo-project/pharo/pull/11383 - Cleanup-IRClosureStackCount #11387 https://github.com/pharo-project/pharo/pull/11387 Reduce Symbol creation ======================= Some more simple changes to avoid creating new symbols during interactive use - 11394-Cursor-classresizeForEdge-should-hard-code-common-cases #11395 https://github.com/pharo-project/pharo/pull/11395 - IconNames-06-21 #11393 https://github.com/pharo-project/pharo/pull/11393 - Symbol-intern-blankIconOfWidth #11385 https://github.com/pharo-project/pharo/pull/11385 - Icon renames #380 https://github.com/pharo-spec/NewTools/pull/380 - IconNames-06-21 #1282 https://github.com/pharo-spec/Spec/pull/1282 Fixes ===== - fix 8661: enables IM support in SDL2 #11379 https://github.com/pharo-project/pharo/pull/11379 - 11388 - show context menu even when no item is selected #11398 https://github.com/pharo-project/pharo/pull/11398 - revert #9765 #11397 https://github.com/pharo-project/pharo/pull/11397 - chore(Fuel): update to 5.1.0 #11384 https://github.com/pharo-project/pharo/pull/11384 Features ======== - TestRunner #11372 Experimental new testrunner model https://github.com/pharo-project/pharo/pull/11372 - Add SpModelChangedPort #1279 https://github.com/pharo-spec/Spec/pull/1279 Documentation ============== - fix download link 32 bit #3 https://github.com/pharo-project/pharo-site/pull/3

June 24, 2022

PHEP: Compile-Time Literal Expression Syntax

by Daniel Slomovits

Hello all, I have submitted a Pharo Enhancement Proposal <https://github.com/pharo-project/pheps/pull/10> for a new compiler/syntax feature, in collaboration with Marcus Denker. This is also serving as a trial run of the PhEP process itself, being the first PhEP not *about* the process. Abstract: This PhEP describes an extension to the compiler that allows arbitrary expressions to be evaluated at compile time, with the results stored in the method's literal frame. This is inspired by and will be compatible with the implementation in Dolphin Smalltalk. I refer to such expressions interchangeably as "compile-time literals", "compile-time expressions", and "optimized expressions". A large portion of the code can be ported directly from Dolphin Smalltalk, and I should be able to complete this part of the process. Marcus said he could help with the integration with OpalCompiler etc., but we can probably break this into several parts (compiler vs. decompiler, other peripheral things) if others are interested. I look forward to hearing what everyone thinks! Thank you, Daniel Slomovits

June 23, 2022

Compiler API: Simplify #interactive

by Marcus Denker

The compiler has a lot of odd features as we wanted to be 100% backward compatible. One of them is the idea of the âinteractiveâ state. -> a compiler compiled with a requestor or without. -> Interactive use needs a requestor - to print error messages - to lookup variables The requestor can implement #interactive to return false. This is used to hook into error handling by implementing a fake requestor. MethodChunkCompilerRequestor is an example of that. But in addition, the compiler added âinteractiveâ as local state of the compilation context: you could make the compiler without a requestor Interactive, or turn off the compiler interactive even though the requestor was interactive. This makes no sense (and I think does not even work). This PR simplifies the idea of #isInteractive: if there is no requestor, we are not interactive. This means we do not need the flag "interactive" on CompilationContext and can simplify. For now we keep the feature that requestors can implement #interactive and return false, to support MethodChunkCompilerRequestor. I think we should simplify that, too, but that is another step. PR is here https://github.com/pharo-project/pharo/pull/11392 (And already merged) Marcus

June 22, 2022

Re: Cheap/free Seaside site hosting (urgent)?

by Noury Bouraqadi

Hi Lucas, Put your image in a docker container hosted by Google cloud. If the container is less than 250MB, you can have free plan for a limited periode of time. Noury On Jun 22 2022, at 1:20 am, Aaron Rosenzweig <aaron(a)chatnbike.com> wrote: > Hi Lucas, > > You didnât say what country you are in. > > Did you try Linode? Maybe Vultr? > > Hereâs a resource that might be helpful for you to find something local: > https://lowendbox.com > > Cheers, > â Aaron > > > On Jun 21, 2022, at 4:05 PM, lucasrodriguesoficial5(a)gmail.com (mailto:lucasrodriguesoficial5@gmail.com) wrote: > > Hello there. Iâve just finished a website which, aside from a simple storage in Mongo/Voyage permanence layer, is otherwise very small and simple. But Seaside Hosting and Pharo Cloud are down, and international transactions are impossible in Digital Ocean. My country doesnât have any of the less complex cloud storages, making it all more difficult. > > At this point, Iâm willing to pay 17 dollars (max I have now) so that someone sets the image up for me at a DO droplet only once, as the app is technically static. This is urgent because the site is an effort to find a job (as a portfolio), and I donât have any additional money for nothing fancier. > > If no one can do that, what are the alternatives?

June 22, 2022