Re: [sage-release] Re: Sage 10.4.beta3 released
I forget to write en english , chat gpt translation : I updated to WSL2 as soon as it became available (around October/November). I ran |update-manager -d| because it works now, and then switched to SageMath dev with |git checkout dev|. I haven't had any issues so far. I created an icon so I can start it like a regular application. Also, JupyterLab is available in the Jupyter Notebook menu, which is great! However, I have Ubuntu in dual boot and I can't build Sage-10.4.xx there. There's an error with |cypari| and |numpy|, so I gave up until Ubuntu 24.04 is released. In the meantime, I'm using Sage in Conda version 10.3. I installed Manjaro on an external drive. Since I have a new fast USB port (Ryzen 9 7900x 12 cores), my USB SSD is very fast, and I don't see any difference compared to my internal SSDs. So, I'm keeping it on the external USB. With Manjaro, SageMath 10.4.3 beta compiled without any problems. It's so easy that you don't need to be an expert—just follow the SageMath tutorial to build it from source! Awesome. Le 14/04/2024 à 15:46, Emmanuel Charpentier a écrit : On Ubuntu 24.04 (upgraded from 22.04) running under WSL2 on Windows 11, a few things may worth mention : - The upgrade entailed a new version of `gcc`, which entailed failures with system's `brial` (as already reported). Reconfiguration excluding system's `brial` and forced installation of Sage's `brial` fixed the problem. - Problems with gap (see below) led me to cleanup my tree and do a fresh recompilation. In this process, `fricas` was somehow misconfigured : `configure` (correctly) detected system's `fricas`, but somehow misplaced it ; the resulting Sage failed all `fricas`-related tests. Reconfiguring `sage` not to use system's `fricas` and forcing fricas compilation fixed the problem. - A `DeprecationWarning` causes (apparently wrongly) a failure in the doctests of `real-mpfr.pyx`. - `src/sage/libs/gap/element.pyx` doctest systematically crashes the underlying Sage. This is reproducible, and has survived reconfiguration and recompilation from a fresh tree (i. e. after `make distclean`). Logs of re-execution of the latter two permanent failures attached. Suggestions more than welcome... HTH, -- You received this message because you are subscribed to the Google Groups "sage-release" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-release+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-release/09a34406-6c74-408d-a5cc-294310d36437%40gmail.com.
Re: [sage-release] Re: Sage 10.4.beta3 released
J'ai mis à jour wsl2 dès la mise à disposition de celle ci (octobe/novembre) je fais update-manager -d puisque ça marche maintenant et je suis passé à sagemath dev par git checkout dev J'ai eu aucun problème à ce jour, j'ai fait une icone donc je peux le démarrer comme une application normal , d'ailleurs avec julpyter lab qui est proposé dans le menu de jupyter-notebook, ok :) Par contre j'ai ubuntu en dual boot et là je n'arrive pas à construire sage-10.4...xx, une erreur avec cypari et numpy donc j'ai laissé tomber jusqu'à la sortie de ubuntu de 24.04. En attendant j'utilise sage dans conda version 10.3.. J'ai installé manjaro sur un disque externe : Comme j'ai une nouvelle sortie usb rapide (ryzen 9 7900x 12 cores) mon ssd sur usb est très rapide je ne vois aucune différence avec mes ssd internes, si bien que je le garde sur usb externe , avec manajaro sagemath10.4.3 beta c'est compilé sans aucun probrlème en plus c'est si facile qu'il n'y a pas besoin d'être un spécialiste juste regarder le tutorial de sagemath pour le faire à partir des sources ! Super Le 14/04/2024 à 15:46, Emmanuel Charpentier a écrit : On Ubuntu 24.04 (upgraded from 22.04) running under WSL2 on Windows 11, a few things may worth mention : - The upgrade entailed a new version of `gcc`, which entailed failures with system's `brial` (as already reported). Reconfiguration excluding system's `brial` and forced installation of Sage's `brial` fixed the problem. - Problems with gap (see below) led me to cleanup my tree and do a fresh recompilation. In this process, `fricas` was somehow misconfigured : `configure` (correctly) detected system's `fricas`, but somehow misplaced it ; the resulting Sage failed all `fricas`-related tests. Reconfiguring `sage` not to use system's `fricas` and forcing fricas compilation fixed the problem. - A `DeprecationWarning` causes (apparently wrongly) a failure in the doctests of `real-mpfr.pyx`. - `src/sage/libs/gap/element.pyx` doctest systematically crashes the underlying Sage. This is reproducible, and has survived reconfiguration and recompilation from a fresh tree (i. e. after `make distclean`). Logs of re-execution of the latter two permanent failures attached. Suggestions more than welcome... HTH, -- You received this message because you are subscribed to the Google Groups "sage-release" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-release+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-release/0d03d7f7-113c-4ce4-84ed-5c3592d4777a%40gmail.com.
[sage-release] Re: Sage 10.4.beta3 released
On Ubuntu 24.04 (upgraded from 22.04) running under WSL2 on Windows 11, a few things may worth mention : - The upgrade entailed a new version of `gcc`, which entailed failures with system's `brial` (as already reported). Reconfiguration excluding system's `brial` and forced installation of Sage's `brial` fixed the problem. - Problems with gap (see below) led me to cleanup my tree and do a fresh recompilation. In this process, `fricas` was somehow misconfigured : `configure` (correctly) detected system's `fricas`, but somehow misplaced it ; the resulting Sage failed all `fricas`-related tests. Reconfiguring `sage` not to use system's `fricas` and forcing fricas compilation fixed the problem. - A `DeprecationWarning` causes (apparently wrongly) a failure in the doctests of `real-mpfr.pyx`. - `src/sage/libs/gap/element.pyx` doctest systematically crashes the underlying Sage. This is reproducible, and has survived reconfiguration and recompilation from a fresh tree (i. e. after `make distclean`). Logs of re-execution of the latter two permanent failures attached. Suggestions more than welcome... HTH, -- You received this message because you are subscribed to the Google Groups "sage-release" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-release+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-release/cdf48550b6417f53be0c2b9a6e2d585bf03535b2.camel%40free.fr. errors.log.gz Description: GNU Zip compressed data