summaryrefslogtreecommitdiff
path: root/libgps_core.c
Commit message (Collapse)AuthorAgeFilesLines
...
* Get back to a state where the shm export works.Eric S. Raymond2011-09-271-26/+0
|
* Converge the library APIs for different transports more closely.Eric S. Raymond2011-09-271-0/+26
|
* Add some debugging.Eric S. Raymond2011-09-161-1/+4
|
* Make the sock_export=no build.Eric S. Raymond2011-08-261-0/+59
| | | | | | | | | | The way I fixed this extends the library API so some functions which were previously undefined for shm transport are now defined. This doesn't change the binary API of the library in the normal (sock_export=yes) case at all, so I'm not bumping its version. No changes in the daemon. All regression tests pass (in the normal sock_export=yes build).
* More code slimming.Eric S. Raymond2011-06-111-1/+1
|
* Break the libgps test code our of libgps_core.c.Eric S. Raymond2011-04-031-104/+1
|
* Allow gps_errstr() to report shm-specific errors.Eric S. Raymond2011-03-291-2/+10
|
* Squeeze more code out of the minimal build.Eric S. Raymond2011-03-291-2/+9
|
* Scan-build and splint cleanup. All regression tests pass.Eric S. Raymond2011-03-281-0/+2
|
* Cast gpsdata->gps_fd to intptr_t as int might not be big enough.Bernd Zeimetz2011-03-281-2/+2
| | | | This also fixes building with g++ >= 4.5.
* The shm transport finallly works through gps_open().Eric S. Raymond2011-03-281-2/+1
|
* Part repair of the shm dispatch.Eric S. Raymond2011-03-271-1/+1
| | | | | Uh oh. One of my rests was broken. The gps_shm_* functions work, but dispatch to them through the public API is broken. This partway fixes it.
* Shared-memory export is now integrated and documented.Eric S. Raymond2011-03-271-58/+20
|
* Core client library now hands off to libgps_sock.c and libgps_shm.c.Eric S. Raymond2011-03-271-116/+47
| | | | Live test with cgps works.
* Begin splitting apart libgps_core.c into socket-export and generic functions.Eric S. Raymond2011-03-271-473/+86
|
* Clear up some confusion about *_SET vs.. *_IS defines.Eric S. Raymond2011-03-261-1/+2
| | | | | | | | | *_SET masks are library-side and for the set member of the fisx structure. *_IS masks are daemon-side and used only in the status mask returned by the packet analyzer methods, and code in the core library that uses same. Never the twain shall meet. All regression tests pass.
* remove unused variable (received) in gps_readJon Schlueter2011-03-221-2/+1
|
* Moinor correction to previous commit; buffer must be static.Eric S. Raymond2011-03-211-1/+1
|
* Banish compiler warnings in the Qt support.Eric S. Raymond2011-03-211-3/+5
| | | | All regression tests pass.
* Shut up a compiler warning on unused var if not defined USE_QTGary E. Miller2011-03-211-0/+2
|
* In the client API, take the data buffer out of the gpsdata_t structure.Eric S. Raymond2011-03-181-21/+29
| | | | | | | | | | | | | | | | | | tz has shown me that we had a point of brittleness where the size of the client's data buffer was exposed in gps_data_t. Solution: move it (and the data-waiting count) to the client's private-data member. Requires a new but trivial client-library entry point to get the buffer pointer when a client needs it. Had to be done before API Version 5 froze, or we'd have been stuck breaking a shipped API again. No implications for the daemon or other tools, as this buffer is never used there. (Except it gets rid of several K of never-used static buffer space in the daemon, which is a good thing.) All regression tests pass. Code splints clean. cgps (the only place the API fix matters in our code) live-tests correctly.
* Clear errno at an appropruaye place in gps_read().Eric S. Raymond2011-03-101-0/+2
|
* Give timestamps their own typedef. No logic changes.Eric S. Raymond2011-03-101-1/+1
| | | | All regression tests pass, splint and cppcheck pass.
* Make the Qt implementation honor timeout.Eric S. Raymond2011-03-091-1/+1
|
* Give gps_waiting() a timeout argument. Use it systematically in test clients.Eric S. Raymond2011-03-091-4/+7
| | | | cgps and gpxlogger are live-tested and work. All regression tests pass.
* Update documentation about gps_read() and gps_waiting().Eric S. Raymond2011-03-071-4/+0
|
* warning hunting const char * in libgps_core.cJon Schlueter2011-03-031-2/+2
|
* Tell the compiler that gos_errstr() ia an observer.Eric S. Raymond2011-03-011-1/+1
| | | | All regression tests pass. Patch due to Jon Schlueter.
* More name changes related to GST.Eric S. Raymond2011-02-011-1/+1
| | | | All regression tests pass.
* Change Noise sentence to use ISO8601 timestamp.Eric S. Raymond2011-02-011-2/+2
|
* Cleanup after the NOISE change. All regression tests pass.Eric S. Raymond2011-02-011-2/+3
|
* Process GPGST into noise statistics. All regression tests pass.Thomas Sprinkmeier2011-02-011-1/+1
| | | | Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
* Reorganize stream argument flags, and create one for timing data.Eric S. Raymond2011-01-261-0/+4
|
* Fix typos pointed out in Berlios tracker bug #17858.Eric S. Raymond2011-01-251-1/+1
| | | | All regression tests pass.
* The raw_hook mwember and set_raw_hook are gone; client API cleanup is done.Eric S. Raymond2011-01-251-22/+2
|
* More deheader testing.Eric S. Raymond2011-01-121-0/+1
|
* Remove subframesflag so they're omitted uncinditionally.Eric S. Raymond2011-01-101-5/+1
|
* WATCH_SUBFRAMES should not imply WATCH_JSON.Eric S. Raymond2011-01-041-1/+1
|
* Enable gpspipe to start subframe reporting.Eric S. Raymond2011-01-041-0/+4
|
* Quite down the JSON parser a bit.Eric S. Raymond2011-01-011-11/+13
|
* More header portability audting with a new version of deheader.Eric S. Raymond2010-12-221-0/+5
|
* Correct a previous commit changing <time.h> to <sys/time.h> eceywhere.Eric S. Raymond2010-12-211-1/+1
| | | | <sys/time.h> is needed for select(2).
* Use <time.h> a la SuS, rather than <sys/time.h>, for select(2) etc.Eric S. Raymond2010-12-211-1/+1
|
* Abolish gpsd_report.c. No more hardwired logging from the client libraries.Eric S. Raymond2010-12-161-3/+3
| | | | Only the daemon now uses this function.
* Refactor to turf all uses of gpsd_report() out of netlib.cEric S. Raymond2010-12-161-0/+4
| | | | | | A recent bug report makes it clear that all such uses have to be removed from the client libraries. Otherwise a GUI client might dummp log messages uselessly in the X session error log. This is a start.
* Taking SuS seriously some more.Eric S. Raymond2010-12-141-8/+4
|
* More header simplification. All regression tests pass.Eric S. Raymond2010-12-141-3/+3
|
* For simplification on SuS assumptions.Eric S. Raymond2010-12-131-3/+1
|
* More partial splint cleanup.Eric S. Raymond2010-12-081-1/+1
|
* Unused-header removals by deheader. All regression tests pass.Eric S. Raymond2010-12-011-3/+0
|