salfter/fluxengine
1
0
Fork 0
mirror of https://github.com/davidgiven/fluxengine.git synced 2025-10-31 11:17:01 -07:00
Code Issues Packages Projects Releases Wiki Activity

The flux pattern matcher will now preserve leading zeroes (although it can't

Browse Source 
match them). The IBM decoder is now properly converted, autodetecting FM and
MFM.
This commit is contained in:
David Given
2019-04-27 22:11:39 +02:00
parent bec46419d6
commit 4e3d4e31af
6 changed files with 86 additions and 211 deletions
Download Patch File Download Diff File Expand all files Collapse all files

56
lib/decoders/fluxmapreader.cc
View File

@@ -62,8 +62,14 @@ FluxPattern::FluxPattern(unsigned bits, uint64_t pattern):
assert(pattern != 0);
unsigned lowbit = ffsll(pattern)-1;
pattern <<= 64 - bits;
_highzeroes = 0;
while (!(pattern & TOPBIT))
{
pattern <<= 1;
_highzeroes++;
}
_length = 0;
while (pattern != TOPBIT)
@@ -87,34 +93,37 @@ FluxPattern::FluxPattern(unsigned bits, uint64_t pattern):
}
}
unsigned FluxPattern::matches(const unsigned* end, double& clock) const
bool FluxPattern::matches(const unsigned* end, FluxMatch& match) const
{
const unsigned* start = end - _intervals.size();
unsigned candidatelength = std::accumulate(start, end - _lowzero, 0);
if (!candidatelength)
return 0;
clock = (double)candidatelength / (double)_length;
return false;
match.clock = (double)candidatelength / (double)_length;
unsigned exactIntervals = _intervals.size() - _lowzero;
for (unsigned i=0; i<exactIntervals; i++)
{
double ii = clock * (double)_intervals[i];
double ii = match.clock * (double)_intervals[i];
double ci = (double)start[i];
double error = fabs((ii - ci) / clock);
double error = fabs((ii - ci) / match.clock);
if (error > clockDecodeThreshold)
return 0;
return false;
}
if (_lowzero)
{
double ii = clock * (double)_intervals[exactIntervals];
double ii = match.clock * (double)_intervals[exactIntervals];
double ci = (double)start[exactIntervals];
double error = (ii - ci) / clock;
double error = (ii - ci) / match.clock;
if (error > clockDecodeThreshold)
return 0;
return false;
}
return _intervals.size();
match.matcher = this;
match.intervals = _intervals.size();
match.zeroes = _highzeroes;
return true;
}
FluxMatchers::FluxMatchers(const std::initializer_list<const FluxMatcher*> matchers):
@@ -125,15 +134,14 @@ FluxMatchers::FluxMatchers(const std::initializer_list<const FluxMatcher*> match
_intervals = std::max(_intervals, matcher->intervals());
}
unsigned FluxMatchers::matches(const unsigned* intervals, double& clock) const
bool FluxMatchers::matches(const unsigned* intervals, FluxMatch& match) const
{
for (const auto* matcher : _matchers)
{
unsigned m = matcher->matches(intervals, clock);
if (m)
return m;
if (matcher->matches(intervals, match))
return true;
}
return 0;
return false;
}
void FluxmapReader::seek(nanoseconds_t ns)
@@ -166,12 +174,12 @@ nanoseconds_t FluxmapReader::seekToPattern(const FluxMatcher& pattern)
while (!eof())
{
double clock;
unsigned m = pattern.matches(&candidates[intervalCount+1], clock);
if (m)
FluxMatch match;
if (pattern.matches(&candidates[intervalCount+1], match))
{
seek(positions[intervalCount-m]);
return clock * NS_PER_TICK;
seek(positions[intervalCount-match.intervals]);
_pos.zeroes = match.zeroes;
return match.clock * NS_PER_TICK;
}
for (unsigned i=0; i<intervalCount; i++)
@@ -190,16 +198,16 @@ nanoseconds_t FluxmapReader::seekToPattern(const FluxMatcher& pattern)
void FluxmapReader::seekToIndexMark()
{
readNextMatchingOpcode(F_OP_INDEX);
_pendingZeroBits = 0;
_pos.zeroes = 0;
}
bool FluxmapReader::readRawBit(nanoseconds_t clockPeriod)
{
assert(clockPeriod != 0);
if (_pendingZeroBits)
if (_pos.zeroes)
{
_pendingZeroBits--;
_pos.zeroes--;
return false;
}
@@ -208,7 +216,7 @@ bool FluxmapReader::readRawBit(nanoseconds_t clockPeriod)
if (clocks < 1.0)
clocks = 1.0;
_pendingZeroBits = (int)round(clocks) - 1;
_pos.zeroes = (int)round(clocks) - 1;
return true;
}