summaryrefslogtreecommitdiffhomepage
path: root/qfunctionconfigurator
AgeCommit message (Collapse)Author
2015-10-24qfc: guard against unlikely division by zeroStanislaw Halik
2015-10-20all: update copyright where appropriateStanislaw Halik
2015-10-18qfc: elide NaN valuesStanislaw Halik
2015-10-08qfc: save retval's absolute valueStanislaw Halik
Closes #249
2015-10-07qfc: move var to outer scopeStanislaw Halik
2015-10-07qfc: no need to load lazily, it's received immediatelyStanislaw Halik
2015-10-07qfc: try harder to fill all memoized bucketsStanislaw Halik
2015-10-07qfc: revert to floats as in rc15Stanislaw Halik
Issue: #248 Issue: #231
2015-10-06qfc: increase precisionStanislaw Halik
Go from uint16_t to uint32_t. If it still doesn't help we have to back to using floats. Reported-by: @nanospork Issue: #231
2015-10-02nix warnings on osx clang buildStanislaw Halik
2015-09-05qfc: more gray colors when graph is disabledStanislaw Halik
Sponsored-by: TrackHat
2015-09-05qfc, mapping: disallow editing and change color when checkbox disabledStanislaw Halik
Sponsored-by: TrackHat
2015-09-05qfc: don't allow editing when disabledStanislaw Halik
Sponsored-by: TrackHat
2015-08-28qfc: also draw line from function start while movingStanislaw Halik
2015-08-22props changed onlyStanislaw Halik
Some source files had executable bit on. Remove it.
2015-08-11qfc: type mismatchStanislaw Halik
2015-08-11qfc: add commentStanislaw Halik
2015-07-27qfc: move division lastStanislaw Halik
2015-07-26fix OSX buildStanislaw Halik
2015-07-26qfc: qreal isn't float, use autoStanislaw Halik
2015-07-26qfc: use doubles for intermediate valuesStanislaw Halik
We could lose precision when multiplying by integral_max with floats. For doubles, we can count on up to 2^56 - 1 to be expressible somewhat exactly.
2015-07-26qfc: reload spline data lazilyStanislaw Halik
2015-07-26qfc: add missing mutexStanislaw Halik
2015-07-26qfc: tweaksStanislaw Halik
2015-07-26qfc: disallow placing new points if too close to existing onesStanislaw Halik
2015-07-26qfc: last value is always positive or zeroStanislaw Halik
2015-07-26qfc: use integers as storage, not floatsStanislaw Halik
Float mantissa has 23 bits. 9 bits get wasted per value. Instead, use uint16_t that doesn't lose much precision, but saves half the space.
2015-07-26qfc: we need positive retval hereStanislaw Halik
2015-07-26qfc: interpolate the right way for negative valuesStanislaw Halik
2015-07-26qfc: sensible step when drawingStanislaw Halik
Higher step value allows more precision when spline doesn't match the control points.
2015-07-26qfc: lift computation out of loopStanislaw Halik
2015-07-26qfc: abstract away internally-used data typeStanislaw Halik
Also, change value_type to more sensible value
2015-07-26qfc: reduce artifacts on high-frequency functionsStanislaw Halik
2015-07-25move portability classes to compat libraryStanislaw Halik
2015-07-24qfc: allow max range as floating pointStanislaw Halik
affects accela
2015-07-19fix osx, app bundle now buildsStanislaw Halik
2015-07-08retab and reformat onlyStanislaw Halik
2015-07-07qfc: only prepend leading (0,0) if there's more than 1 pointStanislaw Halik
2015-07-07qfc: use float intermediates, and consistentlyStanislaw Halik
2015-07-07qfc: use same precision no matter the max x valueStanislaw Halik
2015-07-07qfc: don't cancel moving if points come too closeStanislaw Halik
Instead, block moving but allow the user to move the other way.
2015-07-07qfc: don't cancel moving if points come too closeStanislaw Halik
Instead, block moving but allow the user to move the other way.
2015-07-07includes relative to project top directoryStanislaw Halik
2015-07-01qfc: always draw line to last pointStanislaw Halik
2015-07-01qfc: adapt step size to boundStanislaw Halik
2015-04-02qfc: simplifyStanislaw Halik
2015-03-20qfc: use floats consistentlyStanislaw Halik
2015-03-20qfc: remove some nonsenseStanislaw Halik
The code didn't do anything sensible.
2015-03-20qfc: start drawing control point lines from correct placeStanislaw Halik
2015-02-15tabsDaMichel