summaryrefslogtreecommitdiff
path: root/gpsd.c
Commit message (Collapse)AuthorAgeFilesLines
* Fix up some line terminations.Eric S. Raymond2009-08-311-1/+0
|
* Refactor to make each dump function responsible for its own \r\n.Eric S. Raymond2009-08-311-18/+9
|
* Back out a change that broke regression tests. All tests pass.Eric S. Raymond2009-08-301-2/+2
|
* Implement rate-switching in new protocol (not tested).Eric S. Raymond2009-08-301-1/+9
|
* On the other hand, watchers *should* get notified when the device pool changes, Eric S. Raymond2009-08-301-0/+6
| | | | e.g. both on device additions and device deactivations.
* In new protocol, don't report device information on each channel assignment. Eric S. Raymond2009-08-301-4/+0
| | | | | | | We did that in old protocol so a client could know it's assigned device, but in new protocol "assignments" don't exist; the right way to get the corresponding information is to look at path fields in responses.
* Clean up device-state dumping.Eric S. Raymond2009-08-301-4/+2
|
* Finally give ?TPV; and ?SKY; the axe. Eric S. Raymond2009-08-301-26/+1
| | | | | | | Dealing with all the race conditions implied by the fact that you don't necessarily have a device bound when these are called is just too nasty - it's a maintainability problem and would have been a support nightmare.
* Simplify and regularize command parsing.Eric S. Raymond2009-08-301-66/+83
| | | | Code now deals more gracefully with JSON parse errors.
* Improved error messages.Eric S. Raymond2009-08-301-7/+9
|
* Address Debian bug #542384...Eric S. Raymond2009-08-301-0/+8
| | | | ...by adding a facility to specify a fallback dialout group.
* Create a way not to step on defaults.Eric S. Raymond2009-08-301-3/+3
|
* Typo fix.Eric S. Raymond2009-08-291-1/+1
|
* Prevent some SEGVs on ill-formed comands.Eric S. Raymond2009-08-291-2/+2
|
* AIS dumper's signature has changed. Integrate AIS parsing on the library side.Eric S. Raymond2009-08-291-2/+2
|
* Fix minor bug introdued by previous commit.Eric S. Raymond2009-08-251-19/+9
|
* Re-enable older behavior that devices aren't sniffed until a client connects.Eric S. Raymond2009-08-251-2/+23
| | | | Because new-protocol operation doesn't need to know device classes in advance.
* Simplify the implementation of ?WATCH.Eric S. Raymond2009-08-251-29/+32
|
* Unify CONFIGDEV and DEVICE responses.Eric S. Raymond2009-08-251-6/+6
|
* Fix up parsing of ?DEVICES response.Eric S. Raymond2009-08-241-1/+1
|
* Leave devices in the pool when they're deactivated.Eric S. Raymond2009-08-241-1/+1
|
* Fix a small but deadly bug in the allocation machinery.Eric S. Raymond2009-08-231-8/+10
|
* Arrange for clients to see a VERSION response when they first connect.Eric S. Raymond2009-08-231-4/+10
|
* Parse WATCH responses in the client library. All regression tests pass.Eric S. Raymond2009-08-231-2/+2
|
* Merge two structures that didn't need to exist separately. Eric S. Raymond2009-08-231-7/+11
| | | | All regression tests pass.
* Major rearrangement of device members in struct gpsdata_t... Eric S. Raymond2009-08-231-31/+31
| | | | | ...documented in the protocol-transition white paper. All regression tests pass.
* Parsing of JSON VERSION response, with unit test.Eric S. Raymond2009-08-191-1/+1
|
* Ship device flags as integer rather than string list.Eric S. Raymond2009-08-191-7/+7
|
* Cleanups made possible by last refactoring.Eric S. Raymond2009-08-181-1/+1
|
* Make the two different kinds of DEVICE notation conformable.Eric S. Raymond2009-08-181-1/+1
|
* More steps towards ?DEVICES decoding - refactor code.Eric S. Raymond2009-08-181-50/+22
|
* Delimit a version report field properly.Eric S. Raymond2009-08-181-1/+1
|
* Halfway to parsing ?DEVICES.Eric S. Raymond2009-08-181-1/+1
| | | | Adds devicelist_time member to struct gps_data_t.
* Document an advantage of the new protocol.Eric S. Raymond2009-08-181-0/+9
|
* Tighten up the logic for closing devices.Eric S. Raymond2009-08-171-13/+17
|
* Ensure that new-style watchers see new devices as they are added.Eric S. Raymond2009-08-171-0/+6
|
* Add cycle-time reporting and setting to new protocol. Eric S. Raymond2009-08-171-13/+2
| | | | | | Steps towards making gpsctl work with new protocol. Incidentally, fixes a bug where min_cycle was not reported correctly in the C command.
* Add code for changing modes in devconfig.Eric S. Raymond2009-08-171-3/+4
|
* Skip unallocated devices when watching.Eric S. Raymond2009-08-171-1/+2
|
* Big simplification of GPSD_NG interface. Eric S. Raymond2009-08-171-188/+69
| | | | | CONFIGCHAN is gone, WATCH now listens to everything and sets as per-user policy bits things that used to be per-channel.
* Don't let error messages step on other progress indications.Eric S. Raymond2009-08-151-4/+4
|
* Suppress garbage write on channel assignment.Eric S. Raymond2009-08-141-7/+8
|
* Remove a superfluous write.Eric S. Raymond2009-08-141-3/+1
|
* Channel assignment failure isn't a crash landing when done from WATCH,Eric S. Raymond2009-08-121-30/+38
|
* Clean up NG command parsing. Only require one introducer per request line.Eric S. Raymond2009-08-121-39/+55
|
* Multiple GPSD-NG commands per line are now supported.Eric S. Raymond2009-08-121-108/+130
|
* A refactoring step. All regression tests pass.Eric S. Raymond2009-08-121-256/+260
|
* Code simplification.Eric S. Raymond2009-08-121-16/+9
|
* GPSD-NG commands without object arguments now require an ending ';' delimiter.Eric S. Raymond2009-08-111-25/+47
| | | | Preparation for allowing multiple commands per line.
* Paul Fox's fix for the F command.Eric S. Raymond2009-08-111-14/+15
|