Mercurial > hg > svcore
annotate base/LogRange.h @ 1301:6681027ff2ff 3.0-integration
Interpolate always, when interpolation requested. Previous test failed to take into account varying bin height in log-scale modes, which meant interpolation might be necessary even when more bins than pixels.
author | Chris Cannam <cannam@all-day-breakfast.com> |
---|---|
date | Mon, 28 Nov 2016 14:40:45 +0000 |
parents | 31f01931b781 |
children | 667e369cfeab |
rev | line source |
---|---|
Chris@224 | 1 /* -*- c-basic-offset: 4 indent-tabs-mode: nil -*- vi:set ts=8 sts=4 sw=4: */ |
Chris@224 | 2 |
Chris@224 | 3 /* |
Chris@224 | 4 Sonic Visualiser |
Chris@224 | 5 An audio file viewer and annotation editor. |
Chris@224 | 6 Centre for Digital Music, Queen Mary, University of London. |
Chris@224 | 7 This file copyright 2006 Chris Cannam. |
Chris@224 | 8 |
Chris@224 | 9 This program is free software; you can redistribute it and/or |
Chris@224 | 10 modify it under the terms of the GNU General Public License as |
Chris@224 | 11 published by the Free Software Foundation; either version 2 of the |
Chris@224 | 12 License, or (at your option) any later version. See the file |
Chris@224 | 13 COPYING included with this distribution for more information. |
Chris@224 | 14 */ |
Chris@224 | 15 |
Chris@224 | 16 #ifndef _LOG_RANGE_H_ |
Chris@224 | 17 #define _LOG_RANGE_H_ |
Chris@224 | 18 |
Chris@478 | 19 #include <vector> |
Chris@687 | 20 #include "Debug.h" |
Chris@478 | 21 |
Chris@224 | 22 class LogRange |
Chris@224 | 23 { |
Chris@224 | 24 public: |
Chris@224 | 25 /** |
Chris@224 | 26 * Map a linear range onto a logarithmic range. min and max are |
Chris@224 | 27 * passed as the extents of the linear range and returned as the |
Chris@224 | 28 * extents of the logarithmic range. thresh is the minimum value |
Chris@224 | 29 * for the log range, to be used if the linear range spans zero. |
Chris@224 | 30 */ |
Chris@1044 | 31 static void mapRange(double &min, double &max, double thresh = -10); |
Chris@224 | 32 |
Chris@224 | 33 /** |
Chris@224 | 34 * Map a value onto a logarithmic range. This just means taking |
Chris@224 | 35 * the base-10 log of the absolute value, or using the threshold |
Chris@224 | 36 * value if the absolute value is zero. |
Chris@224 | 37 */ |
Chris@1044 | 38 static double map(double value, double thresh = -10); |
Chris@266 | 39 |
Chris@266 | 40 /** |
Chris@266 | 41 * Map a value from the logarithmic range back again. This just |
Chris@266 | 42 * means taking the value'th power of ten. |
Chris@266 | 43 */ |
Chris@1044 | 44 static double unmap(double value); |
Chris@478 | 45 |
Chris@478 | 46 /** |
Chris@478 | 47 * Estimate whether a set of values would be more properly shown |
Chris@478 | 48 * using a logarithmic than a linear scale. This is only ever |
Chris@478 | 49 * going to be a rough guess. |
Chris@478 | 50 */ |
Chris@1044 | 51 static bool useLogScale(std::vector<double> values); |
Chris@478 | 52 |
Chris@224 | 53 }; |
Chris@224 | 54 |
Chris@224 | 55 #endif |