1
0
Fork 0
Commit graph

527 commits

Author SHA1 Message Date
Scott Giese
3626b6c0a9 Maintenance: FlightPlan
convert to size_t
2021-02-21 15:50:28 -06:00
James Turner
0cc465d39f Tweak Natural-earth loading to reduce DB contention
Reduce ‘SQLITE BUSY’ issues with multiple running copies of FG
2021-02-05 11:14:34 +00:00
Automatic Release Builder
499dc79584 NavCache: add duplicate rebuild protection for Unixes 2020-12-22 10:34:58 +00:00
James Turner
a092902483 If loading plain text route, check for an XML header
Avoids errors when we try to parse an XML file as plain text.

Sentry-Id: FLIGHTGEAR-J6
2020-12-17 10:51:07 +00:00
James Turner
7a3f8276a6 Add unit-tests for Route-manager commands
Fix some bugs in insert-waypt command discovered during the testing
(w00t, tests FTW)
2020-12-14 11:45:46 +00:00
James Turner
72f1b84837 Improve waypointFromString logic
- exclude POIs and some other types from the ident match, to avoid
  confusing results
- improve how the search vicinity is computed when inserting a leg,
  which is the common case for a route with a destination set. Use
  the midpoint of the leg ending at the insert position, as the optimal
  search vicinity
- move the waypointFromString code into route.cxx, since it is mostly
  independent of a FlightPlan instance
- extend the tests to cover the bug which flagged these issues

Ticket-Id: https://sourceforge.net/p/flightgear/codetickets/2372/
2020-12-13 17:55:32 +00:00
James Turner
e8c8495d7b NavCache: clean up the tables check query
This very first query was left un-finalized, which apparently blocks
another process from obtaining the exclusive lock needed to
COMMIT. Found using the trace funtions, phew.

Sentry-Id: FLIGHTGEAR-8C
Sentry-Id: FLIGHTGEAR-8F
2020-11-18 09:52:56 +00:00
James Turner
5273c03c6c NavCache: tolerate missing branch nodes
Since we don't insert nodes in read-only mode, we need to
tolerate their absence when loading.
2020-11-18 09:52:56 +00:00
Automatic Release Builder
1583ad3b03 FlightPlans: don’t report XML errors on load
Assume FlightPlan content might be user sourced; don’t worry about
reporting errors for it.
2020-11-12 09:35:26 +00:00
Automatic Release Builder
be50f64821 FlightPlan: don’t use exceptions for load errors
Avoid spamming Sentry when loading problem flight-plans.
2020-11-12 09:34:10 +00:00
James Turner
2683a27513 NavCache: make Transaction a no-op when read-only.
Avoid BEGINing a transaction when we’re in read-only mode, since this
is not wise.
2020-11-10 23:43:44 +00:00
James Turner
134e06af68 Avoid NavCache races on multiple copies rebuilding
If impatient users start multiple copies of FlightGear when a cache
rebuild is required, we can get into a mess. Use an additional
named mutex on Windows to avoid this situation, and block the secondary
copies until the primary instance has completed its cache rebuild.

Sentry-Id: FLIGHTGEAR-8D
Sentry-Id: FLIGHTGEAR-FY
2020-11-10 23:25:04 +00:00
James Turner
2d7e7b3df6 Waypoints: use bool returns for load errors
Avoid using exception throws for malformed input, makes error
reporting very noisy when user-generated flight plans are loaded.
2020-11-10 23:18:03 +00:00
James Turner
e028c52e42 NavCache: fix writes to a read-only DB
When in read-only mode, we were still allowing writes during the
octree building, which is bad. Add the missing check.

Sentry-Id: FLIGHTGEAR-8F
2020-10-29 21:13:47 +00:00
Automatic Release Builder
77d7476ba6 Fix route-path crashes on empty
This could be triggered by doing ‘view route’ in the launcher flight-
planning code, wit no route defined.
2020-10-29 16:24:41 +00:00
Automatic Release Builder
fc037de312 NavDataCache: don’t duplicate FG_ROOT/Scenery path
Check if the path is in the list, before adding it again.
2020-10-29 16:23:26 +00:00
Automatic Release Builder
59e87d0570 Switch a message to MANDATORY_INFO 2020-10-29 14:51:32 +00:00
James Turner
afb43edbdf Make EDDF procedures work better.
Allow mis-match of STAR and approach, which is necessary for EDDF
operations. When there’s a mismatch, just route the common section
and don’t try to build a STAR -> approach transition.

Fix by Tobias Dammers.
2020-10-25 20:31:53 +00:00
James Turner
4bbff581f3 Fix crash in flight-plan UI
Sentry-Id: FLIGHTGEAR-ES
2020-10-25 20:14:10 +00:00
Florent Rougon
b1b12a924f Simplify NavDataCache::NavDataCachePrivate::findDatFiles()
- Use a range-based for loop instead of an iterator.

- Don't bother checking if the previously-visited directory was the
  same: the potential saving that the previous code was hoping should be
  null because FGGlobals::append_fg_scenery() doesn't add a path to
  FGGlobals::fg_scenery if it's already there---this assumes that all
  paths added to FGGlobals::fg_scenery go through
  FGGlobals::append_fg_scenery(), of course.

This should have no effect on FG's behavior.
2020-09-30 11:09:44 +02:00
James Turner
4402d7b81d NavCache: only remove if the file exists.
Otherwise we block trying to re-create the cache, which is dumb. This
showed up as Sentry issue:
https://sentry.io/organizations/flightgear/issues/1875854826

Will back-port to 2020.2 once verified.
2020-09-04 12:29:45 +01:00
Scott Giese
0ce2fdf190 REVIEW: Memory Leak - PositionedOctree.cxx
1,728 bytes in 36 blocks are still reachable
1,280 bytes in 8 blocks are still reachable
480 bytes in 3 blocks are still reachable
2020-08-23 00:15:42 -05:00
Scott Giese
8e9e16bac6 REVIEW: Memory Leak - Branch::childAtIndex
9,152 bytes in 52 blocks are still reachable
2020-08-22 18:01:15 -05:00
James Turner
a297c89144 Flightplans: add clearAll, rename clear -> clearLegs
Improve naming of clear() methods, and expose the new ones to Nasal.
this will allow RouteManager dialog ‘clear’ button to clear all, soon.

See ticket:
https://sourceforge.net/p/flightgear/codetickets/2064/
2020-08-17 16:11:10 +01:00
James Turner
06a14b4b27 Fix a use-after-free ASan spotted. 2020-08-17 16:05:20 +01:00
James Turner
977bfb06bf Reset: ensure mobile tacan drop its AI references
Add a slightly ugly hook, to force mobile nav records to drop their
refs into the property tree on reset.
2020-08-12 16:37:52 +01:00
legoboyvdlp R
3255105190 Ensure that cruise parameters are set when cloning flightplan 2020-07-22 11:31:07 +01:00
James Turner
2f4b2153ef Fix test: find airways.
Fix handling of ‘both’ level option in several places, and amend
the test to make it explicit that we want Clacton VOR, not NDB.
2020-07-16 11:28:50 +01:00
James Turner
f97ea1f621 Fix route-path hiding of waypoints
Also hide the path, if the preceding waypoint is hidden. Fixes the
route-manager test JonathanR added.
2020-07-02 12:19:19 +01:00
Stuart Buchanan
c4942eaa69 AirportBuilder to generate airports from apt.dat 2020-07-01 20:57:15 +01:00
James Turner
4e12748b8c Fix some compiler warnings 2020-06-19 10:46:01 +01:00
James Turner
c1afb053a1 Fix flight-plan cloning with approach transitions 2020-06-10 17:58:31 +01:00
James Turner
03563a1601 Fix DME intercept handling in route/GPS
Both the route-path cade and the RNAV code were using some bad logic
to compute the intersection point. All fixed now, but requires a
new helper in Simgear.
2020-05-30 15:59:07 +01:00
James Turner
6bb1478067 GPS fly-by support
Remove the non-functional turn-anticipation code from the GPS, and
enable fly-by mode in (some of the) RNAV controllers: initially the
leg controller.

The new config property is gps/config/enable-fly-by, defaults to off
for compatibility.
2020-05-30 15:59:07 +01:00
James Turner
58d7454976 Explicit support for approach transitions
Make the handling of approach transitions consistent with SID and STAR
transitions, including the Nasal API. Add some tests to cover the
new features.
2020-05-27 21:44:59 +01:00
James Turner
55ce199535 Fixes for cloning FPs with procedures 2020-05-27 16:09:18 +01:00
James Turner
fffcd14362 Fix FlightPlan procedure transition handling
Overhaul how transitions are stored in FlightPlan XML, and how
they’re exposed to Nasal. Simplify the Nasal access by making
‘sid_trans’ and ‘star_trans’ writeable.

Extend the unit-tests a lot to cover this, both from C++ and also
from Nasal

As part of this, overhaul the ownership of FlightPlan delegate
factories, to make it safer (use ref-counting of the factories,
and allow the factory to customise delegate clean-up behaviour)
2020-05-21 22:32:28 +01:00
James Turner
5f5a9d2a5e Fix RNAVProcedureTests::testIndexOf
Make FlightPlan::findIndexOfWp use the correct ::matches() method,
and special case this for basic waypoints. This fixes the behaviour
when loaded routes / procedures store a navaid-waypoint as basic.
2020-05-19 22:24:07 +01:00
James Turner
8af20fbcab Fix for loading of FGFP plans.
Where the plan does not contain departure / arrival WPs, fire the
corresponding delegate methods after load, to run the selection logic.

Also, add a ‘loaded’ delegate callback, to give delegates a chance to
perform validation and fix-up after all flight-plan loads.
2020-05-19 22:01:42 +01:00
James Turner
134685b527 Fix FP delegate when loading GPXs
When load a GPX route, run normal departure/arrival airport callbacks
on the delegates. (They are blocked for ‘native’ XML routes to avoid
losing SIDs and STARs)

https://sourceforge.net/p/flightgear/codetickets/2227/
2020-05-03 08:30:18 +01:00
James Turner
6baa55b36b Waypoint hiding support (for the A320, etc)
To support FMS which delete ‘past’ waypoints, add some helpers:
 - waypoints can be flagged as hidden, this property is exposed to Nasal
 - currentWP() accepts an offset parameter, allowing FlightPlan indexing
   to treat the current WP as index 0
 - a ‘numRemainingWaypoints()’ function on flightPlan in Nasal, again
   to range how many waypoints are left, starting from the current WP

Add some basic tests of these new features to the route-manager tests.
2020-05-01 09:28:42 +01:00
James Turner
b85048db2d GPS/FLightPlan test updates
- fix reseting of the NavData each test
- improve the test-pilot’s GPS following (reduce turn range close
   to the desired course)
- test DME intercepts
2020-04-21 15:48:41 +01:00
James Turner
17774dbe4c FlightPlan: DISCON works with pointAlongRoute
Add some tests for the default GPS / route-manager handling of
DISCON in flight-plans. Especially check behaviour when sequencing
a DISCON, and resuming with a Direct-to after one. 

This requires a corresponding FGData update to the default GPS Nasal.
2020-04-19 16:44:49 +01:00
James Turner
1a925b152d FlightPlan: don’t assert on invalid waypoints
When there is no first anchor waypoint (eg a runway or navaid), avoid
asserts from the RoutePath code.
2020-04-19 12:27:50 +01:00
James Turner
4c0965e3c1 Some initial tests for route discontinuities.
Probably many more to add, but this is some basic testing at least of
inserting, querying and removing discontinuities.
2020-04-15 16:11:33 +01:00
Stuart Buchanan
9b20c08db3 Fix compiler warnings 2020-04-12 21:52:19 +01:00
gallaert
fb65f64342 Replace BOOST_FOREACH by C++ 11 for range loop. 2020-04-12 14:40:36 +01:00
James Turner
27410761f3 Lots of UTF8 path correctness fixes 2020-03-12 14:23:44 +00:00
James Turner
3f14d53286 FlightPlans: fix save/load of restrictions, deleting legs from Nasal
- Fix Leg ownership to avoid segfaults when deleting them via Nasal
- Modify FP save/load code so restrictions and holds are correctly
  saved and restored

Extend the tests to cover these cases.
2020-03-10 14:00:40 +00:00
Erik Hofman
2d2d5dbb76 Switch to C++11 threads, mutexes and lock_guards. Switching to C++11 condition_variables requires some more thought. 2020-01-22 13:45:28 +01:00