API has w/form samples-per-entry rather than size

Informing the user of a desired waveform size is not the right approach,
as the waveform is still essentially meaningless without knowledge of
the scale it uses.  As such, the public API now exposes the required
number of _samples per waveform entry_, thereby creating a clear-cut
scaling factor for waveform interpretation in the context of samples.

1 files changed, 4 insertions, 2 deletions

diff --git a/test/enginelibrary/performance_data_test.cpp b/test/enginelibrary/performance_data_test.cpp
index 9e2cb58..b1fb830 100644
--- a/test/enginelibrary/performance_data_test.cpp
+++ b/test/enginelibrary/performance_data_test.cpp
@@ -110,7 +110,8 @@ static void populate_track_1(djinterop::track& t)
 
     // High-resolution waveform data
     std::vector<djinterop::waveform_entry> waveform;
-    int64_t waveform_size = t.recommended_waveform_size();
+    int64_t samples_per_entry = t.required_waveform_samples_per_entry();
+    int64_t waveform_size = (t.sampling()->sample_count / samples_per_entry) + 1;
     waveform.reserve(waveform_size);
     for (int64_t i = 0; i < waveform_size; ++i)
     {
@@ -261,7 +262,8 @@ static void populate_track_2(djinterop::track& t)
         1, {"Loop 2", 2345600, 2345700, el::standard_pad_colors::pad_2});
 
     // High-resolution waveform data
-    int64_t waveform_size = t.recommended_waveform_size();
+    int64_t samples_per_entry = t.required_waveform_samples_per_entry();
+    int64_t waveform_size = (t.sampling()->sample_count / samples_per_entry) + 1;
     std::vector<djinterop::waveform_entry> waveform;
     waveform.reserve(waveform_size);
     for (int64_t i = 0; i < waveform_size; ++i)