summaryrefslogtreecommitdiff
path: root/gpsd.c
Commit message (Collapse)AuthorAgeFilesLines
* Rename config.h so it can be installed as gpsd_config.h...Chris Kuethe2006-10-261-4/+5
| | | | | | ...for those who want to build their own apps linked against libgps and want the headers to work. Works on OpenBSD, tested by Jeff Francis on OS X and Linux
* Restore reporting time to millisecond resolution. Eric S. Raymond2006-10-251-2/+2
| | | | | I was wrong about this; it isn't the GPS system that has a centisecond limit, it's some vendor implementations of NMEA.
* Stelian Pop's fix for the J command handling.Eric S. Raymond2006-10-251-15/+8
|
* Allow for gpsd_report and gpsd_hexdump to be suppressed to save cpu cycles.Chris Kuethe2006-10-241-0/+2
| | | | | This renders logging completely useless, but that may be acceptable in some embedded deployments.
* Don't pretend we offer more accuracy than the GPS system supports.Eric S. Raymond2006-10-231-2/+2
|
* I don't like fixed width fields.Chris Kuethe2006-10-211-21/+21
| | | | | Older parsers don't seem to care one way or t'other... no objections on the lists both times I posted this.
* Introducing a new configuration option "--disable-reconfigure"...Chris Kuethe2006-10-021-0/+4
| | | | | | | | | | | | | | | | | | | | ...which is used to prevent gpsd from taking actions which would alter the configuration of your receiver. This could include, but might not be limited to: * port speed * communication protocol * output messages sent * output message rate * resetting the receiver * clearing stored almanac/ephemeris/position/clock calibration This is useful in cases where you have your receiver set up in a particular manner (logging raw pseudoranges, carrier phases and NAV data, for example) and don't mind sharing that data, so long as it's read-only. For now, this affects the speed/protocol switch commands in gpsd.c, but the device drivers will need to be patched as well.
* Undoing my | change of 28 Sep 2006.Chris Kuethe2006-09-281-18/+1
|
* Undoing my |-protection change of 28 Sep 2006.Chris Kuethe2006-09-281-8/+3
|
* Protect the | command by adding a -x command line option.Chris Kuethe2006-09-281-3/+8
| | | | | | | | | Experimental or debugging features which could damage hardware could can be wrapped in if (dangerous) { /* ... */ ; } so that they will only work if -x is set.
* Add the | command.Chris Kuethe2006-09-281-1/+18
| | | | | | | | | This command allows gpsd to pass through a raw command from the client to the server; the reverse of super-raw (R=2) mode. This could conceivably be used to reflash your receiver's firmware, modify the port settings, tune the navigation engine to produce flawed results or any other thing one could do with a direct connection. This command is primarily for debugging purposes.
* Prevent gpsd from spinning if the dgpsip stream goes away.Chris Kuethe2006-09-281-1/+5
| | | | Reported by Neal Probert on gpsd-dev
* fix the format strings of the timestamps in O and Y messagesChris Kuethe2006-09-231-2/+2
|
* Add subversion ID tags near the beginning of every file.Chris Kuethe2006-09-211-0/+1
| | | | First line where possible, second line for scripts called with #!
* The big dead-storage elimination: we get rid of the newdata member.Eric S. Raymond2006-09-091-1/+1
| | | | All regression tests pass.
* Another step in eliminating dead storage. All regression tests pass.Eric S. Raymond2006-09-091-1/+1
|
* Do mode checks relative to the user fixbuffer.Eric S. Raymond2006-09-081-31/+29
|
* A gallimaufry of small fixes:Eric S. Raymond2006-09-051-5/+9
| | | | | | | Add a sanity check to garmin.c. Fix another regression test screwed up by the GPGLL bug. Add some documentation comments to gpsd.c. splint cleanup in libgpsd_core.c.
* Stelian Pop's cleanups for the J command.Eric S. Raymond2006-09-031-2/+1
|
* Ooops...bad capitalization.Eric S. Raymond2006-09-021-1/+1
|
* Implement no-CASOC policy in a more explicit (and correct!) way.Eric S. Raymond2006-09-021-13/+7
|
* Finally, a version of the J-command patch that passes regression tests.Eric S. Raymond2006-09-011-26/+1
| | | | Some cleanup will follow.
* Refactor the way stashing oldfixes is handled. Eric S. Raymond2006-09-011-3/+9
| | | | All regression tests pass.
* Put the error-modeller call in a place where it won't cause a segfault.Eric S. Raymond2006-09-011-4/+1
|
* Yet another minor data-management bug.Eric S. Raymond2006-09-011-0/+4
| | | | | Error estimates for PGRME must be computed from newdata, not the merged fix.
* Nailed a data-management bug. Eric S. Raymond2006-09-011-3/+0
| | | | | | The newdata member was not being properly cleared, which screwed up the error modeler in a way that only showed up on NMEA devices (not single-packet-per-cycle devices like SiRFs).
* Design decision: Trie North data gets mapped into existing fields (track, etc)Eric S. Raymond2006-08-311-8/+0
| | | | | | | | ...rather than new ones like heading. This gets rid of the HEADING_FIX code contributed but not compiled in by default. Also, fix a place in the zodiac driver that sets a fix rather than newdata member.
* Add a bit more logging to ntpshm_init.Gary E. Miller2006-08-301-0/+2
|
* Hmmm...this change may belong in libgpsd_core.c.Eric S. Raymond2006-08-211-0/+3
|
* J patch withdrawn, all regression tests pass.Eric S. Raymond2006-08-211-5/+15
|
* String safety, courtesy of snprintf, strlcat and strlcpy.Chris Kuethe2006-08-191-53/+53
| | | | | | | GPSD is now free from the often-misused strcat, strcpy and sprintf. Future code should not use unbounded string functions. Glibc users, please verify that the integrated strlcat and strlcpy are correctly linked in.
* strchr returns char *, so test it against NULL (a pointer) not 0 (an integer)Chris Kuethe2006-08-181-1/+1
|
* Fix the J option support.Eric S. Raymond2006-08-181-4/+4
|
* Robin Johnson's fixes for udev support.Eric S. Raymond2006-08-161-2/+3
|
* Reduce gpsd's default footprint by about 10MB.Eric S. Raymond2006-08-161-2/+4
|
* Added buffer policy switch.Eric S. Raymond2006-08-141-48/+107
|
* Eliminate use of fuser(1) from gpsfake.Eric S. Raymond2006-07-291-1/+1
|
* Fix typo pointed out my Marshall Crocker.Eric S. Raymond2006-07-261-1/+1
|
* Merge Eivind Tagseth's patch for limiting memory utilization on SBCs.Eric S. Raymond2006-07-261-14/+53
|
* Avoid a nasty bug when reclaiming fds.Eric S. Raymond2006-07-251-1/+1
|
* The -d option is no longer necessary.Eric S. Raymond2006-07-231-18/+23
| | | | | DGPSIP and NTRIP URLS are recognized when they occur in arguments. Currently only a single such source is allowed.
* Third stage: make *all* changes of all_fds have apprpriate adjust calls nearby.Eric S. Raymond2006-07-211-20/+30
|
* Second stage: lower maximum fd, when possible, on closes.Eric S. Raymond2006-07-211-6/+17
|
* Implement Owen Wallace's idea to track maximum fd...Eric S. Raymond2006-07-211-3/+20
| | | | ...so select() can poll a smaller set.
* Add dgnss.c, which should have been checked in with the NTRIP support.Eric S. Raymond2006-06-071-0/+8
| | | | Also check in the heading patch under -DHEADING_FIX.
* Full merge of O patch with 14th field enabled. Protocol level bumped to 3.Eric S. Raymond2006-06-071-2/+0
| | | | This requires rebuilding the regression tests.
* Robin Darroch's O-mode patch merged...Eric S. Raymond2006-06-071-1/+9
| | | | | | ...but actual generation of the mode field is disabled. Regression tests fail in a strange way; checksums are off by one. Must grovel back through the versions and figure out why.
* splint cleanupEric S. Raymond2006-06-071-23/+19
|
* It doesn't make much sense to allow port speed changes...Chris Kuethe2006-05-271-0/+2
| | | | ...when we build with a fixed port speed.
* This is a better way to do the fixed port speed.Chris Kuethe2006-05-221-6/+6
|
View Lorry Controller Status