Hi, When GPS compiles a file and an error occurs, the error (with file/line number) is usually displayed in the 'Locations' window, which allows easy navigation to the error source line. Atm, this does not happen. With a simple test file which contains an error, GPS shows ... gprbuild -ws -c -f -u -P/usr/home/rod/Desktop/default.gpr hello_world.adb ada -c -g -O2 hello_world.adb [2015-08-14 18:59:31] process exited with status 4, elapsed time: 00.31s ... in the 'Messages' window, and nothing in the 'Locations' window. The test code was ... procedure hello_World is begin -- null; end hello_World; Regards.
I can't reproduce this. Using the provided example, On windows GPL 2015, Locations tab: V Builder results (1 item) V hello_world.adb (1 item) 5:1 statement expected On DragonFly using GPS 6.1.1 (20150118): V Builder results (1 item) V hello_world.adb (1 item) 5:1 statement expected Exactly the same!
The problem does *not* occur on dragonflyBSD. It does occur with freeBSD 10.2 and 11. Also, the gps log file (in ~/.gps) intermittently shows an exception (with callback trace) related to tasking, though this made no difference to the presence/absence of the problem. It occurs during startup, prior to the user doing the build command which produces the prob.
how new is your FreeBSD 11?
I installed from "FreeBSD-11.0-CURRENT-amd64-20151119-r291085-disc1.iso".
No change after 'pkg update' on fBSD11.
A commit references this bug: Author: marino Date: Sat Jun 11 18:41:55 UTC 2016 New revision: 416739 URL: https://svnweb.freebsd.org/changeset/ports/416739 Log: devel/gps: Upgrade to 2016 release (plus 2 dependencies) x11-toolkits/gtkada3: Upgrade version 3.8.3.2 => 3.14.2 devel/gnatcoll: Upgrade version 2015 => 2016 devel/gps: Upgrade version 6.1.1.0 => 2016 This upgrade appears to work flawlessly on DragonFly, but on FreeBSD the PR list below regarding navigation to code error is unfortunately still valid. PR: 202317 Changes: head/devel/gnatcoll/Makefile head/devel/gnatcoll/distinfo head/devel/gnatcoll/files/patch-Makefile head/devel/gnatcoll/files/patch-configure head/devel/gnatcoll/pkg-plist head/devel/gps/Makefile head/devel/gps/distinfo head/devel/gps/files/patch-Makefile.in head/devel/gps/files/patch-gps_Makefile head/devel/gps/files/patch-shared.gpr head/devel/gps/pkg-plist head/x11-toolkits/gtkada3/Makefile head/x11-toolkits/gtkada3/distinfo head/x11-toolkits/gtkada3/files/patch-Makefile.in head/x11-toolkits/gtkada3/files/patch-configure head/x11-toolkits/gtkada3/files/patch-docs_Makefile head/x11-toolkits/gtkada3/files/patch-docs_Makefile.in head/x11-toolkits/gtkada3/files/patch-namespace head/x11-toolkits/gtkada3/files/patch-src_opengl_gtkglarea.h head/x11-toolkits/gtkada3/pkg-plist
A commit references this bug: Author: marino Date: Fri Jul 8 22:28:05 UTC 2016 New revision: 418247 URL: https://svnweb.freebsd.org/changeset/ports/418247 Log: devel/gps: Fix compiler error highlight on FreeBSD On DragonFly, GPS has been properly highlighting code errors during compilation, but on FreeBSD nothing happens. Luckily Rod Kay has been persistent in diagnosing the cause and he finally isolated it to a too-short timeout interval for Expect calls. The timeout period was set to 1 millisecond, the absolute shortest interval possible. While Linux and DF could complete the function call during this time, FreeBSD requires at least 10 milliseconds to do the same. I'm setting the timeout to 100 milliseconds to ensure this functionality works on FreeBSD. Rod and I can't immediately see any negative impact to extending the timeout and we're casually pinging Adacore to understand the orignal 1-ms value. PR: 202317 Changes: head/devel/gps/Makefile head/devel/gps/files/patch-kernel_src_gps-kernel-timeout.adb
Good job, Charlie5!