This patch adds two improvements to view.pike. The first one is to give an error message if the path to a new engine does not even point to a file, instead of crashing. The second one tweaks the heur
First one question: there are lots of `static' modifiers across the script and many of them seem to be out of place (like the first: `static object engine'). As far as I know, `static' in Pike is the
Don't know whether it is intentional, but i do know how this happened. Patch is below. Well, you don't have to know Pike to use scripts written in it. That's a patch for `Makefile.am'. Once Dan runs
Yes the logic there had confused me and know I understand why. The if (has_value("1234...", testline[0..0]")) happens after sccanf has already stripped testline from the starting id, maybe I had move
When testing it I found some things had broken. This patch repairs those and fixes a few odd indentations. - bugfixes in view.pike /Gunnar Index: regression/view.pike == RCS file: /cvsroot/gnugo/gnug
For the record I've been using Pike 7.4.27 from Debian testing while developing view.pike. I cannot reproduce the crashes. I know for sure that the Pike developers consider crashes (of the core dump
Hi Arend, Hi Paul, thank you for your responses! some remarks: Am Fri, 9 Jan 2004 20:19:33 +0000 schrieb Paul Pogonyshev <address@hidden>: aquarius wrote: ... The Makefile in $(top_builddir)/interfac
This patch allows view.pike to be started with a list of several testcases: ./view.pike owl:1 owl:2 It will then allow to go forward and backward in this list with two (primitive) buttons. Also, the
- make test case label contain initial test name instead of `owl:1' - switch to move generation tab after loading new test case if appropriate Paul -- view.pike 10 Jan 2005 15:20:22 +0200 1.15 +++ vi
This fixes a bug when view.pike didn't set up the testcase correctly in some cases when starting a new engine or switching to a new testcase. Arend Index: regression/view.pike == RCS file: /cvsroot/g
The patch below works for me. Please check that it doesn't cause unwanted behaviour (i.e. warnings or errors) on your systems. If everything is OK, I'll commit it. Paul -- view.pike 10 Sep 2004 23:45
- view.pike: allow to directly switch to new testcase Very useful when having a 2nd engine started. Arend Index: regression/view.pike == RCS file: /cvsroot/gnugo/gnugo/regression/view.pike,v retrievi
- support for starting sgf viewer directly from view.pike This allows to automatically start an sgf viewer (specified in the "cgoban -edit %s" format) when doing reading analysis. The .sgf-file names
This patch makes `view.pike' not send lots of `loadsgf' commands. Paul Index: regression/view.pike == RCS file: /cvsroot/gnugo/gnugo/regression/view.pike,v retrieving revision 1.8 diff -u -p -r1.8 vi
Gunnar, while you are resurfaced on the list, maybe you have comments on this ;) I have modified view.pike to allow it to connect simultaneously to several engines. Of course the intended use is to c
On one computer, view.pike crashes, for example the following commands cause segmentation faults: ./view.pike games/strategy11.sgf:79 view.pike games/cgf2004/aya.sgf:114 This is with Pike v7.4 relea
It's a further development of the debugboard idea. It differs significantly though, primarily by using a graphical board and by being a GTP application. It's not finished but it has reached a state w
What is view.pike? It was added to the CVS in November by Gunnar. There is no documentation, no license, no discussion of this script in the mailing list archive. http://mail.gnu.org/archive/cgi-bin
Thanks, Gunnar! I installed libgtk1.2-dev, and had to blow away the previous build directory and build from scratch -- view.pike works like a charm now! Much appreciated!