2018-03-01 |
Chris Cannam |
Convert some cerrs to SVCERRs. Apart from anything else, this makes MSVC2017 happy
|
2017-12-11 |
Chris Cannam |
Rework canonicalisation so as to avoid theoretical possibility of integer overflow
|
2017-12-07 |
Chris Cannam |
Fix fallthrough which bothered compiler
|
2017-10-09 |
Chris Cannam |
Fix reversed parameter names (fortunately the only call to this function currently uses the correct order)
|
2017-10-09 |
Chris Cannam |
Add overlooked fclose in error case
|
2017-08-31 |
Chris Cannam |
Merge
|
2017-08-31 |
Chris Cannam |
Avoid displaying negative zero
|
2017-07-20 |
Chris Cannam |
Adjust rounding; fixes tests on 32-bit Linux test box. Also update one test. Next: recheck everywhere else.
|
2017-07-10 |
Chris Cannam |
Merge from branch "scale-ticks"
|
2017-05-04 |
Chris Cannam |
SVDEBUG, though it turns out not to be very useful
scale-ticks
|
2017-05-04 |
Chris Cannam |
Rework log scale calculation, update to changed interface
scale-ticks
|
2017-05-04 |
Chris Cannam |
We don't need to return the tick instruction gubbins
scale-ticks
|
2017-05-04 |
Chris Cannam |
Another test, that needs some more work
scale-ticks
|
2017-05-04 |
Chris Cannam |
After seeing them in action, I changed my mind about the wisdom of these trailing .0s
scale-ticks
|
2017-05-04 |
Chris Cannam |
Add logarithmic ticks. This is getting complicated!
scale-ticks
|
2017-05-04 |
Chris Cannam |
Tests now passing
scale-ticks
|
2017-05-04 |
Chris Cannam |
Change loop to something that feels more correct, though it makes no difference to the tests here. More tests, one failing.
scale-ticks
|
2017-05-03 |
Chris Cannam |
More tests & fixes
scale-ticks
|
2017-05-03 |
Chris Cannam |
These tests now pass (in some cases by fixing the tests)
scale-ticks
|
2017-05-03 |
Chris Cannam |
More fixes, more tests, including some more that now fail
scale-ticks
|
2017-05-03 |
Chris Cannam |
These tests now pass, but there's still a lot to be done
scale-ticks
|
2017-05-03 |
Chris Cannam |
Scale tick labeller, and tests (some failing so far)
scale-ticks
|
2017-03-07 |
Chris Cannam |
Untabify!
|
2017-03-07 |
Chris Cannam |
Change where the 32-bit memory calculation adjustment is carried out -- more transparent here
|
2017-03-07 |
Chris Cannam |
More debug logging
|
2017-03-07 |
Chris Cannam |
Debug output
|
2017-03-06 |
Chris Cannam |
Fix some one-off memory leaks (including distracting ones in tests)
|
2017-03-06 |
Chris Cannam |
Some bits and bobs to do with handling memory pressure
|
2017-03-06 |
Chris Cannam |
More useful diagnostics for MIDI startup
|
2017-03-06 |
Chris Cannam |
Avoid compiler warning for dummy build
|
2017-03-06 |
Chris Cannam |
Support pre-5.4 Qt builds. Also, don't use SVDEBUG in here, because it depends on this class!
|
2017-03-05 |
Chris Cannam |
Update RtMidi build to v2.1.1
|
2017-03-05 |
Chris Cannam |
Debug output
|
2017-02-28 |
Chris Cannam |
Fix equality comparison that caused test case failure on 32-bit build
|
2017-02-28 |
Chris Cannam |
Add Range01 normalisation method to ColumnOp. This is the normalisation that is actually used in the Colour 3D Plot layer historically when column normalisation is enabled (not Max1 after all).
|
2017-02-28 |
Chris Cannam |
Remove commented-out code
|
2017-02-28 |
Chris Cannam |
LogRange tests
|
2017-02-27 |
Chris Cannam |
Write aggregate models to .sv file when saving document. They still are not reloaded, but I think this is far as I'm prepared to go for 3.0.
|
2017-02-27 |
Chris Cannam |
And say so
|
2017-02-27 |
Chris Cannam |
Invalidate an aggregate model when one of its components announces it's about to be deleted
|
2017-02-27 |
Chris Cannam |
Minor updates to demo script
|
2017-02-23 |
Chris Cannam |
merge
|
2017-02-10 |
Chris Cannam |
Update to match slightly changed Piper Vamp C++ API
|
2017-02-23 |
Chris Cannam |
Debug notes, tidying
|
2017-02-22 |
Chris Cannam |
Fix various "model deleted with no aboutToDelete notification" warnings (and one associated potential memory leak)
|
2017-02-22 |
Chris Cannam |
Header required
|
2017-02-21 |
Chris Cannam |
Merge
|
2017-02-21 |
Chris Cannam |
Fixes to usage of fdopen, avoiding double-close in particular
|