Pike-devel

pike-devel@lists.lysator.liu.se

3041 discussions

Commit access to the git repository
by Stephen R. van den Berg 17 Aug '08

17 Aug '08

A few things: - The git repository tracking CVS works according to spec now (tracks pikefarm and trails it in less than 4 minutes at anytime for 7.8 and 7.9). - There are two repositories: git clone git://pike-git.lysator.liu.se/pike git clone git://pike-git.lysator.liu.se/pikex The first repository contains only the official CVS branches. The second repository contains the official CVS branches *and* any experimental branches developers wish to advertise. As an example of one of the experimental branches, I created a branch called "testing" which is intended to be the last know stable snapshot of the development branch (7.9). - As of now I would like to invite developers to ask for push/commit access to the pikex repository, in order to test and play around with committing directly to centralise git branches. The branches in pikex will be partly/selectively protected, depending on who wants to commit. There basically are several levels of protection: - People which I don't have the public key from cannot push at all. - For those that are able to push, the following restrictions will (currently) be enforced by the repository: + The official branches 0.5 ... 7.8 nt-tools extra_tests are protected and cannot be pushed into (currently). + Some branches are restricted to mere additions (regular commits). + Some branches cannot be committed to by all developers. + Some branches have no restrictions (i.e. can be created/removed or can even be destructively altered). - If we ever move to git as the leading central repository, then the central repository will become pushable as well, but with severe restrictions (only regular commits, nothing destructive, no branch creation). - In order to start granting access to pikex, I need public keys of developers. After you've handed me the keys, pushing into the repository should become possible using a URL like: git-pike@pike-git.lysator.liu.se:pikex which actually is a shorthand for: ssh://git-pike@pike-git.lysator.liu.se/~/pikex -- Sincerely, Stephen R. van den Berg. You are currently aboard a fully automated plane. There is no pilot on board. Rest assured, you have nothing to worry about... worry about... worry about...

5 13

Postgres wierdness
by Marcus Agehall (Roxen IS) ＠ Pike (-) developers forum 17 Aug '08

17 Aug '08

I think we have an old issue in the (old) postgres module which may be serious enough to fix in 7.8 before the release. Most of my systems doesn't have postgresql installed, but the other day I installed 8.3 on my laptop. Today I checked out a fresh 7.8 and ran make. Everything ran smooth until it tried to compile the postgres module, where it failed. Looking at the error, it was pretty obvious that linking failed because of clashing symbols. Further investigation lead me to conclude that both pgresult.c and postgres.c seems to define pike_postgres_mutex if the postgres library isn't thread-safe. This is the first problem. Both source files cannot define this symbol, and since it should only be defined when the library isn't thread-safe, I believe on of them should simply be turned into an extern. Changing the definition in pgresult.c to a simple extern makes the module compile. Now, I went on to investigate why my postgres libs weren't thread-safe. I would assume most libraries *ARE* thread-safe on modern platforms, so I found it kind of strange. After lots of digging, I found that we look for the function PQsetnonblocking() to determine if the library is thread-safe. This seems somewhat stupid, so I would propose that we also introduce a configure test to properly check if the library is reentrant by actually calling PQisthreadsafe(), with a fallback to the current test, in case older libs doesn't support that function. The whole thing seems to be caused by my postgres libs beeing installed from ports and not correctly found by Pike, which causes some confusion. If we are to keep the old module around for a prolonged period, we may want to clean up it's configure tests in general. For now, I do believe that the first issue has to be fixed before the 7.8 release, since it actually prevented me from compiling Pike, while the latter can probably be left out for now.

2 2

Tested pgsql driver in production, drop-in replacement for postgres
by Stephen R. van den Berg 16 Aug '08

16 Aug '08

Small heads up for Postgres users: I've now succesfully compiled pike 7.8 explicitly --without-postgres yet all existing applications can access Postgres databases just fine (pgsql automatically takes over the postgres:// url if libpq is excluded from the build). I'm running this in a production environment now, no ill effects so far. Obviously, pike --features doesn't list Postgres anymore in this case, but that doesn't mean that it can't access Postgres databases anymore (is this confusing?). -- Sincerely, Stephen R. van den Berg. "The state you are in while your HEAD is detached is not recorded by any branch (which is natural --- you are not on any branch)." -- man git-checkout

2 2

Re: 7.8 alpha 2
by Henrik Grubbstr�m (Lysator) ＠ Pike (-) developers forum 15 Aug '08

15 Aug '08

Also note that the bug was introduced in Pike 7.7.3, so it has only been present in development versions of Pike.

2 1

7.8 blockers
by Peter Bortas ＠ Pike developers forum 15 Aug '08

15 Aug '08

7.8 blockers: Owner Issue [/] Fallback to poll from epoll (from the conference road map). Implemented, but further testing is needed, as well as updated documentation. [ ] Fix gc of Stdio.File (from the conference road map). [ ] mast/zino New Windows build environment. [ ] per/marcus Performance regression --with-assembler. (per and marcus weren't present so item assigned without consent... per says memprotect() is responsible as far as -x benchmark goes.) [ ] zino/mast Building external modules should work. [ ] nilsson Write ChangeLog. [/] Module dumping should succeed for all modules. Currently it succeeds for all modules but 3: [ ] Tools.Standalone.dump [ ] Local [ ] Locale Non-blockers: [ ] jhs Build (SVG) graph from the -x benchmark results Pikefarm delivers to track development. [ ] An API for safely creating temporary files. Possibly wrapping mkstemp/mkdtemp. If you think you could help with any of the items please do. Especially the items that have no one assigned to them yet.

9 81

Overriding type checks in type^H^H^H^Hsoft casts
by Martin Stjernholm, Roxen IS ＠ Pike developers forum 14 Aug '08

14 Aug '08

Consider this condensed case from real code: class A { protected int x; void init (int i) {x = i;} int get() {return x;} } class B { protected int x; void init (string s) {x = (int) s;} int get() {return x;} } A create_a() { B b = B(); b->init ("17"); return [object(A)] b; } In create_a() I try to cheat and return a B as if it were an A, by using a type cast. It doesn't work: /home/mast/foo.pike:19:Soft cast of { B = object(implements /home/mast/foo()->B) } to { A = object(implements /home/mast/foo()->A) } isn't a valid cast. (Side note: This error message doesn't explain why the cast is invalid. Presumably it's due to the prototype of init(), but unlike other type errors it doesn't go into detail.) Let's assume that in this case I know that the type difference of the init() prototype isn't relevant (e.g. it's a function only used in the object factory and therefore never called outside create_a). Shouldn't it be possible to use type casts to make the compiler shut up? I can do this workaround: A create_a() { B b = B(); b->init ("17"); object b2 = b; return [object(A)] b2; } But avoiding that phony extra variable isn't possible; this variant is still nailed: A create_a() { B b = B(); b->init ("17"); return [object(A)] [object] b; }

3 5

Re: 7.8 alpha 2
by Lance Dillon 14 Aug '08

14 Aug '08

I'm not sure it is only a 64-bit problem. I have this [root@hobbes ~]# cat /proc/version Linux version 2.6.25.14-108.fc9.x86_64 (mockbuild@) (gcc version 4.3.0 20080428 (Red Hat 4.3.0-8) (GCC) ) #1 SMP Mon Aug 4 13:46:35 EDT 2008 [root@hobbes ~]# [root@hobbes ~]# cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 107 model name : AMD Athlon(tm) 64 X2 Dual Core Processor 5000+ but mine didn't fail: [riffraff@hobbes Pike-v7.8.8]$ pike Pike v7.8 release 8 running Hilfe v3.5 (Incremental Pike Frontend) > Image.TIFF; (1) Result: _Image_TIFF > string s=Stdio.read_bytes("/usr/share/sane/xsane/doc/xsane-restore.jpg"); > object i=Image.JPEG.decode(s); > i; (2) Result: Image.Image( 26 x 26 /* 2.0Kb */) > Image.JPEG.encode(i); (3) Result: "\377\330\377\340\0\20JFIF\0\1\1\0\0\1\0\1\0\0\377\333\0C\0\b\6\6\a\6\5\b\a\a\a\t\t\b\n" ... ----- Original Message ---- From: "Mirar @ Pike developers forum" <10353(a)lyskom.lysator.liu.se> To: pike-devel(a)lists.lysator.liu.se Sent: Thursday, August 14, 2008 8:15:03 AM Subject: 7.8 alpha 2 That it's a 64-bit-only problem, I mean. Hmm. But... | > string s = Stdio.read_bytes("sugar.jpg"); | > object i = Image.JPEG.decode(s); | >> i; | (1) Result: Image.Image( 300 x 300 /* 263.7Kb */) | > Image.JPEG.encode(i); | *** glibc detected *** pike: munmap_chunk(): invalid pointer: 0x0000000000a9b5d0 | *** | ======= Backtrace: ========= | /lib/libc.so.6(cfree+0x1b6)[0x2ba515677d46] | /usr/lib/libjpeg.so.62[0x2ba516a52ffa] | /usr/lib/libjpeg.so.62(jpeg_abort+0x15)[0x2ba516a52505] | /usr/local/pike/7.7.133/lib/modules/_Image_JPEG.so[0x2ba51681b888] | pike(low_mega_apply+0x598)[0x434908] | ... [ http://www.mirar.org/sugar.jpg ]

2 2

7.8 alpha 2
by Mirar ＠ Pike developers forum 14 Aug '08

14 Aug '08

And still no test case. :p | iris% cat test_jpeg.pike | int main() | { | string s = Stdio.read_bytes("sugar.jpg"); | object i = Image.JPEG.decode(s); | string out = Image.JPEG.encode(i); | write("%O\n",out[..5]); | return 0; | } | iris% pike test_jpeg.pike | "\377\330\377\340\0\20"

1 0

7.8 blockers
by Martin Stjernholm, Roxen IS ＠ Pike developers forum 14 Aug '08

14 Aug '08

Fixed that some time ago. Forgot to update the list. Sorry.

1 0

Pike git tracking repository is up
by Stephen R. van den Berg 12 Aug '08

12 Aug '08

For Pike there is a CVS-tracking git repository now which checks pikefarm for updates every minute. - The repository trails pikefarm by a minute of 4 tops for 7.8 (and 7.9). - It is synced daily at 06:00 for all other branches. - It has been pruned of all unnecessary branches and contains the repository proper now. - It basically is the version previously at git://git.cuci.nl/pike (which will continue to exist). - This is by no means the "final" version, things are still being tweaked (not in this version), but it works fine for current developmentwork. Suggested way of getting it: git clone git://pike-git.lysator.liu.se/pike cd pike for b in 7.8 7.6 7.4 7.2 7.0 0.6 0.5 nt-tools extra_tests do git branch $b origin/$b done git branch Sync at any later point with upstream: git pull (or just "git fetch" which doesn't alter your tracking branches) -- Sincerely, Stephen R. van den Berg. Father's Day Special at the local clinic -- Vasectomy!

3 2

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Pike-devel