13
0
Commit Graph

53 Commits

Author SHA1 Message Date
259499fc5f require use of BBT_Argument as both parameter and return type from most methods (libs edition) 2023-02-15 16:02:56 -07:00
4beca98e28 temporal: fix various incorrect uses of ::samples() in audio time arithmetic
It is a wonder that this has not caused more visible issues since 7.0
2022-12-07 11:13:49 -07:00
bdacfb8724 Fix invalid use of Doxygen "@param" command
This is never for inline references to parameters, only for starting parameter
documentation blocks.  The "@p" command is for this, although unfortunately
Doxygen doesn't actually do anything with it and it's just an alias for code
text.
2022-10-30 20:44:28 -04:00
14da117bc8 Add explicit round/floor integer multiply/divide
This fixes various rounding issues. Notably superclock to sample
conversion must always round down when playing forward.

`::process (start, end, speed = 1)` uses exclusive end.
Processing begins at `start` and end ends just before `end`.
Next cycle will begin with the current end.

One example where this failed:
 - New session at 48kHz
 - Change tempo to 130 BPM
 - Enable snap to 1/8 note
 - Snap playhead to 1|3|0
 - Enable Metronome
 - Play

`assert (superclock_to_samples ((*i).sclock(), sample_rate()) < end);`

end = 177231 samples == superclock 1042118280
A grid point is found at superclock 1042116920 (that is < 1042118280).
However converting it back to samples rounded it to sample 177231 == end,
while actual location is 1360 super-clock ticks before end.

The metronome click has to be started this cycle, since the same
position will not be found at the beginning of the next cycle, with
start = 177232.

Similarly a samplecnt_t t, converted to music-time and back must not be
later than the given sample.

```
timepos_t tsc (t);
assert (timepos_t::from_ticks (tsc.ticks ()).samples () <= t);
```

IOW. When playing forward, all super-clock time between 1|1|0 and 1|1|1
should round down to 1|1|0. "We have not yet reached the first tick".
2022-10-22 02:10:05 +02:00
23010a910d
Be liberal about what to accept
This addressee a bug where ardour 6 was able to write negative
duration `length="-1"` `length-beats="-3.3650500597559585e-05"`

Ideally timecnt_t::string_to should check for invalid,
negative, duration. But this also catches a more generic case.

```
exception at str.substr (1)
#3 Temporal::timepos_t::string_to (this=0x7fffffff7bb0, str="") at libs/temporal/timeline.cc:904
#4 Temporal::timecnt_t::string_to (this=0x7fffffff7ba0, str="-2") at libs/temporal/timeline.cc:294
#5 PBD::string_to<Temporal::timecnt_t> (str="-2") at libs/ardour/ardour/types_convert.h:131
```
2022-10-16 22:30:35 +02:00
2c1f36201c
Fix timecnt_t modulo operation when mixing time-domains 2022-06-30 01:07:32 +02:00
cca3e6e061 temporal: fix bug introduced in 48f4f9bf9c
timecnt_t and timepos_t constructors with the initial argument as an int64_t
are assumed to be using samples. We need to use the explicit factory methods
instead.
2022-05-28 12:59:35 -06:00
48f4f9bf9c temporal: remove some arithmetic operators and rewrite others to use muldiv() 2022-05-27 12:47:44 -06:00
60e5b84d78 temporal: alternative solution to overflow in timeline operator*()
This uses boost::multiprecision::int512_t when multiplying and dividing by the numerator
and denominator of a ratio_t. 128 bits would be sufficient but for some reason, the boost
docs show the 512 bit variant being very slightly faster.

This is a better solution than using a double, which although it will prevent overflow
has fairly limited resolution.
2022-05-24 21:46:10 -06:00
878393e68b temporal: fix construction of timepos_t and timecnt_t with max_sample{pos,cnt}
max_samplepos and max_samplecnt and both INT64_MAX which is (a) too large to fit into a signed 62 bit
integer and (b) definitely too large to be represented in a signed 62 bit superclock value.

Move the constructors that use samplepos_t into the .cc file, and treat these two values as special
cases that mean "as large/late/huge/long as possible".
2021-11-22 10:35:52 -07:00
d71b3100d8 temporal: handle negative positional values that somehow appear in older sessions 2021-11-22 10:28:13 -07:00
1532ebb38f temporal: truncate ::to_string() method for timeline types to ::str() for easier use in debugger 2021-11-17 12:16:38 -07:00
8c632c8446 temporal: allow negative timecnt_t to be used when constructing a timepos_t
Negative positions are legal and should be handled by higher level logic
2021-11-14 15:22:53 -07:00
0acc1a3fec temporal: use correct implementation of timepos_t::operator+ (timecnt_t) (and +=)
If time domains differ, it is necessary to first convert the argument duration into a duration
at the position of "this", in the correct time domain. Then we recursively call the operator
again, but this time we will use the fast path that just adds two timepos_t values.
2021-10-04 17:45:02 -06:00
ed3d374f47
Fix --no-nls compilation 1/2 -- #8802
pbd/i18n.h MUST NEVER be included from header files and always be
the last include. This is because `_` is declared other headers
notably boost and some apple headers.

leading to issues like
../libs/pbd/gettext.h:58:27: error: expected unqualified-id before ‘const’
   58 | # define gettext(Msgid) ((const char *) (Msgid))
2021-09-27 16:28:44 +02:00
9c2e4ec6d2 libtemporal: fix timepos_t::set_time_domain()
For legacy reasons, this was going via a sample intermediate representation
2021-08-13 12:51:36 -06:00
e7cdca1ea2 remove some debug output 2021-08-13 12:51:36 -06:00
8bada4b5e1 remove public ::bbt_at (INTEGER) methods, because of superclock_t/samplepos_t confusion
libtemporal still needs ::bbt_at (superclock_t) fairly often so retain it as a private method, but public
interfaces take either Beats or timepos_t
2021-08-13 12:51:35 -06:00
092a454648 fix timepos_t::string_to (wrong argument to samples_to_superclock() 2021-08-13 12:51:35 -06:00
445ad2e383 rearrange if/else logic to avoid NOTREACHED condition 2021-08-13 12:51:35 -06:00
daad7b74ae libtemporal: rename TempoMap::full_duration_at() to ::convert_duration() and reorder parameters 2021-08-13 12:51:34 -06:00
c0c3fd07c9 libtemporal: add timecnt_t::end() and timecnt_t::set_time_domain() 2021-08-13 12:51:34 -06:00
2049cc44b2 libtemporal: tweak: don't both with local tempo map ptr 2021-08-13 12:51:34 -06:00
433a9ebe48 libtemporal: catch attempts to convert max audio time values into beats, early 2021-08-13 12:51:34 -06:00
4e6c06bc95 libtemporal: add maybe-not-required operator>> for timeline types 2021-08-13 12:51:34 -06:00
ce82efafcf libtemporal: slightly improve output from conversion statistics 2021-08-13 12:51:34 -06:00
25375ee6f4 add timepos_t::set_time_domain() to potentially change time domains 2021-08-13 12:51:33 -06:00
016aada4df libtemporal: fix/improve/correct parsing of timecnt_t from string 2021-08-13 12:51:32 -06:00
d165c492af temporal: fix/simplify/improve parsing of time values from strings 2021-08-13 12:51:32 -06:00
209908f700 libtemporal: improve/fix implementation of timepos_t::operator+=(timepos_t) 2021-08-13 12:51:32 -06:00
9438295607 libtemporal: fix implementation of timecnt_t::operator+ (timecnt_t) 2021-08-13 12:51:32 -06:00
80986b9445 libtemporal: add (commented) debug output for timepos_t::distance() methods 2021-08-13 12:51:31 -06:00
6fe4394c54 fix parameter name in comment 2021-08-13 12:51:31 -06:00
df0ac42f54 Temporal: cleanup for explicit Beats::operator bool() 2021-08-13 12:51:31 -06:00
43d84620e0 Temporal: clean up API to remove ambiguity about ::quarter*_at_*
superclock_t and samplepos_t have the same underlying C++ type, so methods that accept one or the
other as an argument need to be named to make it clear which type they accept. We do not need
the superclock_t variant publically, but it turns out to be useful within TempoMap.
2021-08-13 12:51:31 -06:00
c3d325b56a Temporal: clean up API for timepos_t to remove unneeded duplicates and remove ambiguity 2021-08-13 12:51:31 -06:00
5887ee524e Temporal: further improvements to a timepos_t::distance() method 2021-08-13 12:51:31 -06:00
7c56b9b2fc Temporal: fix insanely incorrect implementation of timepos_t::distance() 2021-08-13 12:51:30 -06:00
def35cf043 mark BBT_Offset (double) constructor explicit to avoid implicit conversion in timeline expressions.
Also clean up the mess this had caused.
2021-08-13 12:51:30 -06:00
f0008898af add operator<<() for timecnt_t() 2021-08-13 12:51:30 -06:00
b876baa757 Temporal: various changes to arithmetic/logic methods to avoid assert() on domain mismatches where none is really needed 2021-08-13 12:51:30 -06:00
cdab3a5e8f fix a bunch of missing symbols related to temporal 2021-08-13 12:51:29 -06:00
8fb70885ec changes needed to get audio_clock.cc to work 2021-08-13 12:51:29 -06:00
bca8e2e8cc libtemporal: changes derived from converting libardour 2021-08-13 12:51:28 -06:00
f06e2dd6d1 Temporal: remove constructors accepting scalar values from timepos_t/timecnt_t and force use of factory methods
This allows us to differentiate between superclock_t and samplepos_t (and related types) which are all typedef'ed to the
same underlying primitive C++ type. Without this, it would be impossible for the compiler or someone reading the code
to know whether a scalar passed to a constructor for a timeline type is in units of samples or superclocks
2021-08-13 12:51:28 -06:00
c75fc64306 Temporal: various changes to int62_t and timeline types based on code review 2021-08-13 12:51:28 -06:00
b5bf518a0a Temporal: fixes, changes, redesigns, extensions to be able to compile ARDOUR::Region with timepos_t/timecnt_t members 2021-08-13 12:51:28 -06:00
56795a97a3 changes required to get ardour to start up (missing vars, incorrect values) 2021-08-13 12:51:28 -06:00
1f0d9159a7 Temporal: use, don't fetch tempo map pointer 2021-08-13 12:51:28 -06:00
8ec81c2308 Temporal: correctly construct return timepos_t for * and / 2021-08-13 12:51:28 -06:00