Re: [PATCH] genbc: Deal with multiple definition errors

Hi Jean-Marc! Any chance you could try the patch from the previous email? It seems to do the right thing. In the long run we probably want to move to a reentrant bison parser and avoid playing with global state. thank you holger > On 16. Mar 2024, at 21:10, Holger Hans Peter Freyther >

Re: libgst/Makefile.am diff

this happen as well? match.h dropping names... #define MATCH_BYTECODES_ \ MATCH_BYTECODES__: \ \ goto MATCH_BYTECODES_SWITCH_; \ MATCH_BYTECODES__: \ > On 8. Mar 2024, at 02:49, Holger Freyther wrote: > > Thank you! Let me AX_CHECK_LINK_FLAG and land this quickly! > >&g

Re: libgst/Makefile.am diff

Thank you! Let me AX_CHECK_LINK_FLAG and land this quickly! > On 26. Feb 2024, at 13:10, Jean-Marc Farinas wrote: > > I've finally found a way (quite accidentally) to add linker flags when > compiling genbc. Modifing grammar file did not work for me either (to heavy > code modification

Re: Problem to resolve 3.3 release; regarding point 3.) bison, genbc and modern linkers

> On 23. Feb 2024, at 11:55, Jean-Marc Farinas wrote: > > But with this modification I was able to build a "gst" program, and running > 'make check' seem to give adequate results, but not 100% > (it will be left for another session). I see sporadic crashes on resume too. I didn't have much

Re: Problem to resolve 3.3 release; regarding point 3.) bison, genbc and modern linkers

> On 24. Feb 2024, at 03:52, Stefan Monnier wrote: > > I suspect the less hacky way involves looking at those multiple > definition warnings/errors, trying to find their origin and removing > the duplication. > Usually it's not very hard. I am happy to take either patch. For the additional

Interesting problems to resolve towards the 3.3 release

Hi folks, I could use a hand with some interesting problems with some of the basics of our VM and its support code. 1.) Upgrade of (parts) of the embedded gnulib. lib-src is part gnulib and part custom. We should update the gnulib parts to address compiler warnings to work better with the

Re: gnu-smalltalk was dropped from debian

Hi bill! > On 29. Oct 2023, at 05:20, bill-auger wrote: > ... > i think the highest priority would be to repair or drop anything that is > broken > in 3.2.91 as it is now (eg: the gst-browser, the blox browser, etc), even if > that means yanking GTK entirely; and make a release of what

Re: gnu-smalltalk was dropped from debian

Thanks for the heads up. Do we have any volunteers to re-activate the development of GNU Smalltalk, move towards GTK4? > On 27. Sep 2023, at 13:21, bill-auger wrote: > > reasons given for removal: > * no debian maintainer > * depends on GTK2 > * inactive upstream > > For details on the

Re: GNU Smalltalk test suite failure with lightning 2.2.2

Hi Andy, Do you pass --enable-jit in your build at all? Looking at the test failures it seems it is crashing when resuming an image. What is my easiest way to reproduce this? best regards holger > On 24. May 2023, at 04:17, Andy Tai wrote: > > reference: comments from GNU Lightning

Re: Is GNU Smalltalk dead?

Indeed. We should promote 3.2.91 to 3.3.0 and cope with VisualGST being slightly broken (it's based on GTK+ 2.0 anyway). Let me aim to do this before the end of the year. holger > On 14. Nov 2021, at 09:11, bill-auger wrote: > > FWIW, the history of smalltralk is lined with arguments such

Re: ProcessVariable>>valueIfAbsent: bug (and fix)

Lovely find! Can you think of a testcase? thank you! holger > On 8. Jun 2021, at 23:06, Piotr Klibert wrote: > > In the previous thread I mentioned a problem in ProcessVariable I encountered > while trying out VisualGST. Turns out I was mistaken, the problem was > triggered by

Re: UDP Receive Socket Spontaneously Unblocks Every 30 Seconds - 3.2.91

Looking at the the >>#receive selector of DatagramSocket shows that there is some timeout logic. The method will return false when it ran into a timeout and true when there was a datagram. IIRC you want to use >>#port: on the outgoing Datagram to select where it should be sent. > On 16. Feb

Re: How to build GNU Smalltalk 3.2.91

> On 27. Jan 2021, at 07:43, Mark Bratcher wrote: > > Hello > > Previously when I pulled 3.2.91 (a couple of years ago), it was a fairly > standard source tree with a ./configure, etc. autoreconf --install --force will create the configure script. > > The latest one seems to start with

Re: Processes

Hi Gary! How did you get GNU Smalltalk 3.2.5? On the surface this sounds like a problem already fixed[1] in the development branch of GNU Smalltalk? Could you try to cherry-pick this[1] and see if the problem persists? thank you holger [1]

Re: Network Communications Information Missing From Docs

For 3.3 I had hoped to solve the deadlock we experienced in the VisualGST debugger with the new event loop. GTK+-2.0 stopped getting maintained and I think we should release 3.3 as is... What do you think? holger > On 6. Jan 2021, at 04:02, s...@pandora.be wrote: > > > I'm sorry if

Re: smalltalk.gnu.org is down

Not abandoned. Let me chase the GNU sysadmins to be able to point this to a static page. Now that there is interest. Anyone of you can help to create a static site? :) cheers holger > On 2. Apr 2020, at 03:35, bill-auger wrote: > > it is true - nagging helps >

Re: smalltalk.gnu.org is down

> On 25. Jan 2020, at 01:46, Andrew Engelbrecht wrote: > Hi Andrew, > I'm Andrew, a sysadmin at the FSF. Someone brought smalltalk.gnu.org to > our attention, and it looks like the site's currently down. If you know > who manages this site, could you please let them know? I added you to a

Re: [Help-smalltalk] gst-browser crashes when loading modified images.

> On 14. Sep 2019, at 08:24, Colton Lewis wrote: Hi! > As best as I can tell, the problem occurs because loading the program > image resumes GTK's never-shut-down event loop and attempts to continue > processing events. The stale events from the last session cause > operations on

Re: [Help-smalltalk] gst-browser crashes when loading modified images.

> On 3. Sep 2019, at 01:18, Colton via help-smalltalk > wrote: > Hi, sorry for experiencing a crash. I don't remember if this part ever worked. I think what happens is that after the image resumes pointers to the old GTK+ are dangling. It would be great if you want to see what happens.

Re: [Help-smalltalk] Process & Fork

> On 26. Jul 2019, at 20:07, Quenio dos Santos > wrote: > > Hello Holger, Hey! > Sorry for the delay on replying. I got really busy the last couple of days. > > I was able to finally get the counter.st running continuously on my Mac with > the master branch: > > $

Re: [Help-smalltalk] Process & Fork

> On 22. Jul 2019, at 20:52, Quenio dos Santos > wrote: > > Hello Holger, > > I've run into more problems as seen below. > > At this point, I'm ready to give up on getting the build working on my Mac. I have Macos too and we can share notes. > can you use make V=1 to show me the link

Re: [Help-smalltalk] GNU Smalltalk 3.3 & VisualGST

> On 19. Jul 2019, at 08:25, Quenio dos Santos > wrote: > > Hello Holger, Hey! > I also noticed that I don't have the "cairo" package in my local system: > > $ ls /usr/local/share/smalltalk > Announcements.starLibSDL_image.star VFSAddOns.star > BLOXBrowser.star

Re: [Help-smalltalk] Process & Fork

> On 21. Jul 2019, at 00:59, Quenio dos Santos > wrote: Hi! > I haven't been able to compiler master on MacOS. I installed GCC and > configured as follows, but it still seems to be using CLANG: You can use CC=/usr/loca.../gcc on configure and make ti pick GCC but either clang or gcc

Re: [Help-smalltalk] Process & Fork

> On 20. Jul 2019, at 21:36, Quenio dos Santos > wrote: Hi! > > And then I have to interrupt it. > > I'm wondering if there's something basic I'm missing here. Maybe I don't > understand how "Process yield" works, but I thought it was just going to > suspend the current process and let

Re: [Help-smalltalk] GNU Smalltalk 3.3 & VisualGST

> On 19. Jul 2019, at 00:01, Quenio dos Santos > wrote: > > Hello, Hi! > Just started trying out GNU Smalltalk. I'm finding it easier to use than > the alternatives, because it doesn't require you to program from within a > Smalltalk image. this is how I got started with GNU Smalltalk. I

Re: [Help-smalltalk] All tests passed using the git head

> On 26. May 2019, at 05:50, Brian E. Lavender wrote: > > I cloned the source from the Git repository and it compiles and tests > pass. This is on Fedora 30, 64bit. This is the latest stable release (3.2.9) vs. master. I will probably make a quick 3.2.10 release for distributions. And

Re: [Help-smalltalk] 1e-4 printing bug

> On 1. Apr 2019, at 10:06, Derek Zhou wrote: > > dont worry, i have assignment form filled a long time ago. also this is a > simple patch, and cannot.assert copyright anyway. Thanks for the patches! I will try to apply them this week. AFAIK unfortunately the last statement is not

Re: [Help-smalltalk] ZeroDivide bug in float printing

> On 29. Mar 2019, at 10:15, Derek Zhou wrote: > > >> http://git.savannah.gnu.org/cgit/smalltalk.git/commit/libgst/interp.inl?id=72ada189aba0283c551ead16635c1983968080b8 > That's good; however there are poor poeple still stuck with old > compilers Is this true for mainstream OS and CPU

Re: [Help-smalltalk] ZeroDivide bug in float printing

> On 29. Mar 2019, at 00:07, Derek Zhou wrote: > > list, > thanks for looking into it! > There were discussion on this list about this bug in the debian package: Are you aware of:

Re: [Help-smalltalk] adding smalltalk-mode to ELPA?

> On 26. Mar 2019, at 22:46, bill-auger wrote: > > On Tue, 26 Mar 2019 12:51:34 -0400 Stefan wrote: >> Anyway, from what I understand, you have no intention of moving >> smalltalk-mode.el elsewhere nor to spend your time with a 2-way sync. >> In that case, I think it's best not to add

Re: [Help-smalltalk] Max/min value of any number class (maxVal/minVal) in gst?

> On 16. Oct 2018, at 15:30, Garreau, Alexandre wrote: > > Hi, Hi! > > It seems other platforms such as Squeak or Pharo have a message “maxVal” > which gives the max value of a given class, such as “SmallInteger > minVal” or “3 class minVal”, for instance. > > Does GNU Smalltalk have

Re: [Help-smalltalk] 'ZeroDivide' error using debian package

> On 12. Aug 2018, at 00:31, bill-auger wrote: > Hi, > `gcc -v` and `g++ -v` both show --enable-default-pie; but i can not > expose the 'ZeroDivide' problem compiling myself - compiling with all of > the following flags produce the same (non-error) result for 1e-4 and 0.05: you might

Re: [Help-smalltalk] [GNU Smalltalk 3.2.5] testsuite: 7 8 32 41 70 118 122 failed

> On 18. May 2018, at 15:51, Vy Pham wrote: I think GCC 7.x is optimizing away the overflow checks. How does master behave? Do you have time to help us to make the 3.3 release? holger ___ help-smalltalk mailing list

Re: [Help-smalltalk] gst-browser startup crash

> On 11. Mar 2018, at 11:19, Andreas Rosenberg wrote: > > gst-gtk is a standard shared library that can be examined with ldd. > ldd output shows all dependend shared libs and contains the link to > libgtk-x11-2.0.so. > I did not examine the makefile for gst-gtk in

Re: [Help-smalltalk] Help with emacs

> On 23. Jan 2018, at 16:40, Andrei Stebakov wrote: > Hi! > I downloaded 3.2.90 gst to my Windows system and found gst-mode.el and > smalltalk-mode.el. > Both are pretty old. > When I run M-x gst it prints "GNU Smalltalk ready" line, but when I > evaluate code with

Re: [Help-smalltalk] gst-browser startup crash

> On 9. Mar 2018, at 18:43, Andreas Rosenberg wrote: > > Hi! Hey! > Digging a bit further, I found that the module "gst-gtk-3.2.92.so" is > linked against > "libgtk-x11-2.0.so", which is (now?) only provided by package: gtk2-devel. "known" packaging issue. With

Re: [Help-smalltalk] small fixups to the tutorial

> On 5. Feb 2018, at 13:30, bill-auger wrote: > Hi, > i was hoping someone would respond to this as i have a rather large set > of corrections and suggestions for the manual - i sent a message to this > list about a year ago but no one has responded if it is

Re: [Help-smalltalk] small fixups to the tutorial

> On 23. Jan 2018, at 20:02, Mathias Laurin wrote: > > Hi list, Hi! > Zecke told me that I should send my patches to the mailing list, > so here I am. sorry for the delay. The new day job took most of my time and energy. > I am half through the online tutorial

Re: [Help-smalltalk] [GNU Smalltalk 3.2.91] testsuite: 7 8 31 40 59 63 69 70 72 74 79 83 86 89 90 92 94 104 118 124 failed

> On 3. Dec 2017, at 22:47, Patrick O'Brien wrote: > > Holger, Hi! it does look like the compiler "optimizing" the overflow check away. Please try my patch and make sure that your clang is new enough and that configure is ran. Otherwise go to the git version and test

Re: [Help-smalltalk] [PATCH] Making smalltalk-mode.el available on Emacs package archives

> On 29. Nov 2017, at 22:01, Wilfred Hughes wrote: > > Dear Holger, Hi! > Thanks for your patience and support! applied. thank you holger ___ help-smalltalk mailing list help-smalltalk@gnu.org

Re: [Help-smalltalk] [GNU Smalltalk 3.2.91] testsuite: 7 8 31 40 59 63 69 70 72 74 79 83 86 89 90 92 94 104 118 124 failed

> On 28. Nov 2017, at 19:00, Patrick O'Brien wrote: > > Holger, thanks for the reply. Hey! > Let me know if you need more information (as I may be making invalid > assumptions based on what I have written here.) could you try "master". As you are using "3.2.91" tarball

Re: [Help-smalltalk] [PATCH v2 1/4] libgst: Fix comments

> On 30. Nov 2017, at 17:31, Lee Duhem wrote: > Hey! > /* Grows the allocated memory to SPACESIZE bytes, if it's not there > - already. > - the memory could not be allocated. Should be called after the > - sweep has occurred so that things are contiguous.

Re: [Help-smalltalk] [PATCH v2 2/4] libgst: Miscellaneous improving to code style

> On 30. Nov 2017, at 17:31, Lee Duhem wrote: > Hi again! > @@ -2025,7 +2023,7 @@ _gst_to_wide_cstring (OOP stringOOP) > string = (gst_unicode_string) OOP_TO_OBJ (stringOOP); > len = oop_num_fields (stringOOP); > result = (wchar_t *) xmalloc (len + 1); > - if

Re: [Help-smalltalk] [PATCH v2 0/4] Some fixes to libgst

> On 30. Nov 2017, at 18:11, Blake McBride wrote: > Hey! > Looks good. I hope your changes get integrated. > > Can you tell me what "This version 2 have no change since v1, only rebased to > top of master > branch." means? I had asked Lee to rebase the changes so I

Re: [Help-smalltalk] [PATCH] gst-tool: Fix ASAN issue on comparing options

> On 12. Feb 2017, at 13:16, Holger Freyther <hol...@freyther.de> wrote: > Dear Paolo, >> Wouldn't the '\0' mismatch first? > > "Both strings are assumed to be n bytes long". I guess an optimized memcmp > will fetch 32/64 bytes at a time (manual loop u

Re: [Help-smalltalk] [GNU Smalltalk 3.2.91] testsuite: 7 8 31 40 59 63 69 70 72 74 79 83 86 89 90 92 94 104 118 124 failed

> On 19. Nov 2017, at 06:26, Patrick O'Brien wrote: Hey! > An update to this - that won't surprise you! > > I downloaded 3.2.91 source to my old Ubuntu 14.04 system and compiled it > there, and 132 of the tests ran successfully to completion, so no issue > with the released

Re: [Help-smalltalk] [PATCH] Making smalltalk-mode.el available on Emacs package archives

> On 13. Aug 2017, at 18:59, Wilfred Hughes wrote: > > Hi Dear Wilfred, > You can see my patch as commits here: > https://github.com/Wilfred/gnu-smalltalk/compare/master...refactor_emacs_mode?expand=1 sorry for the delay! Patches are merged! holger

Re: [Help-smalltalk] [PATCH] Making smalltalk-mode.el available on Emacs package archives

> On 13. Aug 2017, at 12:59, Wilfred Hughes wrote: Hi Wilfred, > You can see my patch as commits here: > https://github.com/Wilfred/gnu-smalltalk/compare/master...refactor_emacs_mode?expand=1 sorry for the delay. Is this still the most recent branch? I would like to

Re: [Help-smalltalk] [PATCH] Making smalltalk-mode.el available on Emacs package archives

> On 27. Aug 2017, at 03:12, Wilfred Hughes wrote: > Dear Wilfred, > I've received some more feedback from Emacs people about the header > formatting on my patch, so here's an updated version. Would anyone be > able to apply this? thank you for your contribution. I was

Re: [Help-smalltalk] [PATCH] Making smalltalk-mode.el available on Emacs package archives

> On 27. Aug 2017, at 03:12, Wilfred Hughes wrote: > Dear Wilfred, > I've received some more feedback from Emacs people about the header > formatting on my patch, so here's an updated version. Would anyone be > able to apply this? thank you for your contribution. I was

Re: [Help-smalltalk] gst-browser on Ubuntu/Debian dlopening the library

> On 18 Aug 2013, at 15:30, Paolo Bonzini wrote: > >> >> Do you have an idea on how that could be solved? > > The Debian package is supposed to provide the proper symbolic links from > *.so to *.so.VERSION, in a directory that is private to GNU Smalltalk > (maybe the modules

Re: [Help-smalltalk] [PATCH] build-aux/tcltk.m4: Simplify tclConfig.sh searching

> On 27 Feb 2017, at 09:19, Lee Duhem wrote: > > Hi, > > Even if my building system already has tclConfig.sh, configure script > still cannot find it, > because it is looking at the wrong places. > > According to Tcl wiki article [1], the following Tcl code can give the >

[Help-smalltalk] ASAN error in testsuite

Hi Paolo, maybe you could give me a hand with this. I was chasing a problem of sporadic test failures and tried ASAN once more and the only failure right now is the compiler test. Does any of this ring a bell? I will probably start having a look tomorrow. holger Execution begins... 'abc'

Re: [Help-smalltalk] Some fixes/improvements to libgst

> On 7 Feb 2017, at 15:43, Lee Duhem wrote: > >> >> is what I am pushing through travis-ci right now. Can you give it a try as >> well? > > I tested on Ubuntu 14.04, x86-64. No regression observed, looks good. > great and I pushed the wrong (not rebased/edited) branch

Re: [Help-smalltalk] crashing browser

> On 4 Feb 2017, at 19:54, Paolo Bonzini wrote: > > > The header files need not. Perhaps another file... can you post a list > of files included in libgtk2.0-dev? most likely it is the .so file to be able to dlopen gtk... :( ___

Re: [Help-smalltalk] Some fixes/improvements to libgst

> On 14 Jan 2017, at 12:27, Lee Duhem wrote: > Hi! another flight, another time to look at it. :) >> if (free < (to - from)) >> { >> - memcpy (_gst_cur_bytecodes->ptr, from, free); >> - _gst_cur_bytecodes->ptr += free; >> - from += free; >>

Re: [Help-smalltalk] Some fixes/improvements to libgst

> On 30 Dec 2016, at 07:44, Lee Duhem wrote: > > Hi! sorry for the delay. > * byte.c (_gst_compile_bytecodes): Remove unnecessary copy. diff --git a/libgst/byte.c b/libgst/byte.c index 50ff07c..a90a139 100644 --- a/libgst/byte.c +++ b/libgst/byte.c @@ -450,9 +450,6 @@

Re: [Help-smalltalk] Fix a few typos in comments of libgst

> On 30 Dec 2016, at 07:42, Lee Duhem wrote: > > Hi, Hi! > The attached patch fixes a few typos in comments of libgst. You may > want to take a look. looks good and I am back from 33C3. I might split it up a bit. holger

Re: [Help-smalltalk] Error build - can't find gnutls

> On 17 Oct 2016, at 11:27, Blake McBride wrote: > > 17.1 > > Attached. Thank you! GST tries to use pkg-config --exists --print-errors "gnutls" to find GNUtls but that fails. Could you use dpkg -L package to see where the gnutls.pc is put? You can wokaround by modifying

Re: [Help-smalltalk] Error build - can't find gnutls

> On 17 Oct 2016, at 11:16, Blake McBride wrote: > > LinuxMint 64 bit Which release? 18? > libtool: link: gcc -g -O2 -Wall -Wno-format -Wdeclaration-after-statement > -Wno-format -Wpointer-arith -Wno-pointer-sign -Wno-switch -Wwrite-strings > -Wno-strict-aliasing

Re: [Help-smalltalk] Error build - can't find gnutls

> On 17 Oct 2016, at 11:02, Blake McBride wrote: > > Hi, Hi! > I am building the latest git master on 64 bit Linux using the following: which OS? > One of the configure output lines reads: > checking for gnutls_global_init in -lgnutls... yes so autoconf figured out

Re: [Help-smalltalk] make check fails

> On 16 Oct 2016, at 16:10, Wolfgang Dann wrote: > > Hi! > 6: exceptions.st ok > 7: intmath.st FAILED > (testsuite.at:33) > 8: floatmath.st > ^t > > linux-4.8.1 > gcc-6.2.0 > gmp-6.1.1 >

[Help-smalltalk] [PATCH] libgst: Do not re-define yylineno with newer versions of flex

From: Holger Hans Peter Freyther With flex 2.6.0 the minor version is < 31 and one would end with redefining the yylineno. Make the check more specific. 2016-09-18 Holger Hans Peter Freyther * genbc-scan.l: Compile with flex 2.6.0. *

Re: [Help-smalltalk] further tests Re: Float printString?

> On 07 Jul 2016, at 12:33, Robert Kuszinger wrote: > > > Dear Holger, Paolo > Hi! > 3.2.5 works well meanwhile the GMP update arrived this morning into Arch. So, > this is not a smalltalk/gmp divergence or whatever. > > Thanks for your support and ideas! > > > As

Re: [Help-smalltalk] further tests Re: Float printString?

> On 06 Jul 2016, at 13:49, Robert Kuszinger wrote: > > > Dear Holger, Hi! > > So, -13 - 54 = -67 > > 2^(-67) with calculator: 2^(-67) = 6.7762636e-21 > > > Evaluating the following leads to the same: > > st> 2 raisedToInteger: -67 > Object: 1 error: The

Re: [Help-smalltalk] Float printString?

> On 06 Jul 2016, at 10:44, Robert Kuszinger wrote: > > Hello! > > I've made a script to perform GRASS GIS operations automatically. > It was running for weeks nicely. > > Today it emerged: > > Object: 1 error: The program attempted to divide a number by zero >

Re: [Help-smalltalk] Looking for a database repord band editor...

> On 30 Jun 2016, at 20:55, anonyme wrote: > > Hi, > Hi! > Do you know a database report band editor with gnu smalltalk ? I am not aware of any. In the past i created my reports by using DBI to make queries and using Iliad to display them. holger

Re: [Help-smalltalk] file copying, etc

> On 22 May 2016, at 18:30, Robert Kuszinger wrote: > > Smalltalk system: 'cp ',from,' ',to. > > Also simple :) and unless you control from and to maybe a bit insecure. We have 'withShellEscapes' (e.g. use like '; rm -rf /' withShellEscapes).

Re: [Help-smalltalk] file copying, etc

> On 22 May 2016, at 18:05, Robert Kuszinger wrote: > > Hello! Hi! > I'm new to this list. > After years of mixed awk-bash scripting I came to the question: :) > > "Why don't I script in smalltalk since I'm doing everything else in > Smalltak?" > > (Pharo, formerly

Re: [Help-smalltalk] Trying to get Blox or Browser to run

> On 28 Mar 2016, at 20:52, dougt wrote: Dear Doug, > correction: after cleaning up the results of other attempts at installing > gnu-smalltalk and browser, the Ubuntu packages gnu-smalltalk and > gnu-smalltalk-browser loaded, and the browser ran ok. > > The Ubuntu

Re: [Help-smalltalk] Problem when using NCurses

> On 28 Mar 2016, at 12:19, olivier auverlot wrote: > > I have dug in the mailing list archive and I found a similar problem. It > seems that I must write the following code: > > PackageLoader fileInPackage: 'NCurses'. > > DLD addLibrary: 'libncurses.so.5'. >

Re: [Help-smalltalk] [PATCH] blox: Fix fall-out from calling >>#initialize as part of new

> On 28 Jan 2016, at 18:49, > wrote: > > I'm just using Emacs/mu4e for email so I'd be surprised if it did > anything odd to the text. > > What's the relationship between your git repo and the one on github: > >

Re: [Help-smalltalk] [PATCH] blox: Fix fall-out from calling >>#initialize as part of new

> On 25 Jan 2016, at 22:48, tho...@2worthingtons.net wrote: > > I've created a patch with just this mod, not including your previous > patch and it seems to work. Hopefully this is the correct format: thanks applied and pushed. For some reason I had issues with taking it out of the mail

Re: [Help-smalltalk] [PATCH] blox: Fix fall-out from calling >>#initialize as part of new

> On 24 Jan 2016, at 17:16, tho...@2worthingtons.net wrote: > > That's helped a lot. Blox starts up and is a useful browser, although > actually adding code thought it seems impossible at the moment. Oh. does it give another routine? I had fixed one blox issue of the workspace from within blox

Re: [Help-smalltalk] [PATCH] blox: Fix fall-out from calling >>#initialize as part of new

> On 24 Jan 2016, at 22:09, tho...@2worthingtons.net wrote: > > > > It's not that there is a broken function, just that there's no obvious > way to create methods. There's an "add" option for classes but I'm not > seeing any way to add a method. If I try editing in the lower pane and >

Re: [Help-smalltalk] Blox and browser

> On 21 Jan 2016, at 21:21, Holger Freyther <hol...@freyther.de> wrote: > > >> On 21 Jan 2016, at 20:48, Holger Freyther <hol...@freyther.de> wrote: >> > > >> PList doesn't seem to have a parentView when >>#initialize is called. This >>

Re: [Help-smalltalk] Blox and browser

> On 21 Jan 2016, at 23:13, tho...@2worthingtons.net wrote: > > Hi, Hi! > Thanks for the quick reply. I will try your suggestions here and in > the other email over the weekend and let you know. I'm not that > worried about blox if I can get the other browser working. my pleasure. >

Re: [Help-smalltalk] Blox and browser

> On 21 Jan 2016, at 18:54, tho...@2worthingtons.net wrote: Hi! > I'm trying to use the current master branch (I've tried the 3.2.92 tag > and the 3.2.5 tag as well). 3.2.92 and master are very close to each other. In fact the main reason to wait for 3.3 is gst-browser (deadlocks and crashes)

Re: [Help-smalltalk] Blox and browser

> On 21 Jan 2016, at 20:48, Holger Freyther <hol...@freyther.de> wrote: > > PList doesn't seem to have a parentView when >>#initialize is called. This > sounds like a fallout from my changes to >>#new changes I made in commit: > 4a224a329b5e63ab6ed432f437eb822e

[Help-smalltalk] Release of GNU Smalltalk 3.2.91

Hi, this has been long overdue. I have released 3.2.91 and try to continue and make a release. On my todo list for version 3.3 I need: * Use __builtin_mul_overflow to fix overflow code for FreeBSD and OSX. Newer clangs have the __builtin_mul_overdlow code * Integrate the lightning 2.0 patches

Re: [Help-smalltalk] [PATCH] i18n: Fix printing of integers and floats

> On 18 Oct 2015, at 22:20, Holger Hans Peter Freyther > wrote: > > +numeric := I18N.LcNumeric language: 'en_US'. On Ubuntu this doesn't lead a good locale. So I change this to en_US.UTF-8 and the test result now looks better. holger

[Help-smalltalk] i18n numeric printing/formating

Dear Mark, I finally had time to look into the issue and it is mostly about calculating the groups. E.g. with: lang := I18N.LcNumeric language: 'en_US'. lang computeNumberOfGroups: 3 => 2 lang computeNumberOfGroups: 7 => 2 it explains why "123" ends up as ",123" and why

Re: [Help-smalltalk] Make check on FreeBSD 10.x - Intel

> On 22 May 2015, at 05:43, Holger Freyther <hol...@freyther.de> wrote: Hi, >> And see a fix in >> http://smallissimo.blogspot.fr/2015/04/removing-ub-in-bytecodeprimmultiply.html. > > oh, I think my preference is to go with __builtin_mul_overflow (or for c

Re: [Help-smalltalk] DBI.Connection call yields "Object: nil error: did not understand #atEnd" in ST 3.2.91-b98173d

> On 31 Aug 2015, at 15:11, Mark Bratcher wrote: > > Holger > > I think that using `basicNew` is reasonable pattern for a solution in this > case. Thank you! > did you try the patch with basicNew? Did it work? I try to squeeze out some time to understand why the failure

Re: [Help-smalltalk] Creating new instance variables in a dynamically created class in gst-remote

> On 05 Sep 2015, at 19:24, Mark Bratcher wrote: > > Hello Hi! > I narrowed it down specifically to the instance variable declarations. If I > simply add these declarations and nothing else, I get the error. I'm adding > instance variables using the common

Re: [Help-smalltalk] DBI.Connection call yields "Object: nil error: did not understand #atEnd" in ST 3.2.91-b98173d

> On 31 Aug 2015, at 04:51, Mark Bratcher wrote: > > Holger Hi! > I believe I found the problem. oops! > The socket is connecting fine. But digging into the connection code, I found > that a `class new` operation now automatically calls the `initialize` method > in

Re: [Help-smalltalk] DBI.Connection call yields "Object: nil error: did not understand #atEnd" in ST 3.2.91-b98173d

> On 31 Aug 2015, at 12:28, Mark Bratcher wrote: > > Hi Holger Hi, > Thanks again for being so responsive. > > After doing some admittedly light research on the `new` method versus > `initialize`, it looks like Pharo (and probably, therefore, Squeak) is the > only

Re: [Help-smalltalk] DBI.Connection call yields "Object: nil error: did not understand #atEnd" in ST 3.2.91-b98173d

> On 31 Aug 2015, at 14:08, Mark Bratcher wrote: > > Thank you, Holger. > > I understand the pragmatics of the decision. One could argue that, in the > current bug case, the `(self new) stream: aStream ; initialize` statement is > a bit "off-pattern". :) I think it's a

Re: [Help-smalltalk] DBI.Connection call yields Object: nil error: did not understand #atEnd in ST 3.2.91-b98173d

On 29 Aug 2015, at 23:53, Mark Bratcher mdbra...@gmail.com wrote: The exact same connection works fine under Smalltalk 3.2.5. So it's not an authentication issue in my case. I tried it on a couple of different databases, each time working in 3.2.5, but the same error using

Re: [Help-smalltalk] DBI.Connection call yields Object: nil error: did not understand #atEnd in ST 3.2.91-b98173d

On 29 Aug 2015, at 21:31, Mark Bratcher mdbra...@gmail.com wrote: Hello Hi! I've been using GNU Smalltalk 3.2.5 with DBI without any serious issues. However, when I tried version 3.2.91-b98173d, I get an error attempting to connect to a simple MySQL database: |st PackageLoader

Re: [Help-smalltalk] Make check on FreeBSD 10.x - Intel

On 22 May 2015, at 04:00, nicolas cellier nicolas.cellier.aka.n...@gmail.com wrote: That's exactly it. Overflow is undefined behavior. The C compiler has a license to ignore undefined behavior because you are not suppose to rely on it. Yes, but I couldn’t cite the relevant

Re: [Help-smalltalk] Make check on FreeBSD 10.x - Intel

On 16 Apr 2015, at 01:14, Holger Freyther hol...@freyther.de wrote: Our mul_with_check doesn’t properly detect the overflow on a multiplication like (380536542838076625 * 576687883419648000) printNl. The result is negative. I don’t have a fix yet but that is easy now. We need to review

Re: [Help-smalltalk] [PATCH] libgst: Fix small amount of sleeps on OSX

On 20 Apr 2015, at 04:35, Paolo Bonzini bonz...@gnu.org wrote: int64_t deltaMicro = deltaNano 1000 ? 1 : deltaNano / 1000; use = 1000 so we avoid the “1000 / 1000” multiplication that gcc 4.9.2 doesn’t appear to optimize away. but yes, nice simplificiation

Re: [Help-smalltalk] Sporadic process issue on Magritte tests

On 16 Apr 2015, at 04:01, Paolo Bonzini bonz...@gnu.org wrote: The point of using absolute deadline is exactly to match deadlines as close as possible (perfect if the syscall ultimately lets you specify an absolute deadline). Relative deadlines become less and less accurate the more

Re: [Help-smalltalk] Make check on FreeBSD 10.x - Intel

On 04 Feb 2015, at 05:28, Stephen Woolerton direct...@gmail.com wrote: I got the same results when running “make check” on both FreeBSD 10.0 and FreeBSD 10.1 (x86_64). Interestingly, the arm install only failed one test, while the x86_64 install fails a number of them. I finally had

Re: [Help-smalltalk] Sporadic process issue on Magritte tests

On 15 Apr 2015, at 20:07, Holger Freyther hol...@freyther.de wrote: it does hang from time to time :(. The first I wonder if we couldn’t change the way VMpr_Processor_signalAtNanosecondClockValue and family is implemented. Couldn’t we do the _gst_get_ns_time at a later stage? E.g

[Help-smalltalk] setitimer, small delta and XNU

Dear Paolo, I am currently switching my development environment from GNU/Linux to XNU (OSX) and there appears a funny “scheduling” or time issue that leads to a race-condition. (Delay forMilliseconds: 500) wait Can get stuck forever (e.g. the process.st testcase fails because of that). From

Re: [Help-smalltalk] Which parser to use to parse the gnu-smalltalk code

On 15 Apr 2015, at 14:21, Junsong Li ljs.darkf...@gmail.com wrote: Thanks. Using STClassLoader to load the class and then walking the AST of the class is the way to go! Done! What are you implementing? holger ___ help-smalltalk mailing list

Re: [Help-smalltalk] Which parser to use to parse the gnu-smalltalk code

Hi, Can anyone point to me some resources about how to parse/traverse the gnu-smalltalk code? There are many Parsers in the STInST package. I assume you mean programatically? The Parser package of STInST is the way to go. And the content in packages/sttools has an example on how to use it.

Re: [Help-smalltalk] Which parser to use to parse the gnu-smalltalk code

On 12 Apr 2015, at 16:56, Junsong Li ljs.darkf...@gmail.com wrote: Thanks. Basically what I want is just a method that compile the .st file, give me the root node of the AST, then I can traverse the AST and print out each kind of expression. I have tried _various_ classes in STInST, none

  1   2   >