Tasmota/lib/IRremoteESP8266-2.5.2.03/src/ir_Sharp.cpp
Theo Arends 0924dfcfb7 Update IRRemoteESP8266 library
Update IRRemoteESP8266 library from 2.2.1 to 2.5.2
2018-11-20 15:53:56 +01:00

268 lines
10 KiB
C++

// Copyright 2009 Ken Shirriff
// Copyright 2017 David Conran
#include <algorithm>
#include "IRrecv.h"
#include "IRsend.h"
#include "IRutils.h"
// SSSS H H AAA RRRR PPPP
// S H H A A R R P P
// SSS HHHHH AAAAA RRRR PPPP
// S H H A A R R P
// SSSS H H A A R R P
// Equipment it seems compatible with:
// * Sharp LC-52D62U
// * <Add models (devices & remotes) you've gotten it working with here>
//
// Constants
// period time = 1/38000Hz = 26.316 microseconds.
// Ref:
// GlobalCache's IR Control Tower data.
// http://www.sbprojects.com/knowledge/ir/sharp.php
const uint16_t kSharpTick = 26;
const uint16_t kSharpBitMarkTicks = 10;
const uint16_t kSharpBitMark = kSharpBitMarkTicks * kSharpTick;
const uint16_t kSharpOneSpaceTicks = 70;
const uint16_t kSharpOneSpace = kSharpOneSpaceTicks * kSharpTick;
const uint16_t kSharpZeroSpaceTicks = 30;
const uint16_t kSharpZeroSpace = kSharpZeroSpaceTicks * kSharpTick;
const uint16_t kSharpGapTicks = 1677;
const uint16_t kSharpGap = kSharpGapTicks * kSharpTick;
// Address(5) + Command(8) + Expansion(1) + Check(1)
const uint64_t kSharpToggleMask =
((uint64_t)1 << (kSharpBits - kSharpAddressBits)) - 1;
const uint64_t kSharpAddressMask = ((uint64_t)1 << kSharpAddressBits) - 1;
const uint64_t kSharpCommandMask = ((uint64_t)1 << kSharpCommandBits) - 1;
#if (SEND_SHARP || SEND_DENON)
// Send a (raw) Sharp message
//
// Args:
// data: Contents of the message to be sent.
// nbits: Nr. of bits of data to be sent. Typically kSharpBits.
// repeat: Nr. of additional times the message is to be sent.
//
// Status: BETA / Previously working fine.
//
// Notes:
// This procedure handles the inversion of bits required per protocol.
// The protocol spec says to send the LSB first, but legacy code & usage
// has us sending the MSB first. Grrrr. Normal invocation of encodeSharp()
// handles this for you, assuming you are using the correct/standard values.
// e.g. sendSharpRaw(encodeSharp(address, command));
//
// Ref:
// http://www.sbprojects.com/knowledge/ir/sharp.htm
// http://lirc.sourceforge.net/remotes/sharp/GA538WJSA
// http://www.mwftr.com/ucF08/LEC14%20PIC%20IR.pdf
// http://www.hifi-remote.com/johnsfine/DecodeIR.html#Sharp
void IRsend::sendSharpRaw(uint64_t data, uint16_t nbits, uint16_t repeat) {
for (uint16_t i = 0; i <= repeat; i++) {
// Protocol demands that the data be sent twice; once normally,
// then with all but the address bits inverted.
// Note: Previously this used to be performed 3 times (normal, inverted,
// normal), however all data points to that being incorrect.
for (uint8_t n = 0; n < 2; n++) {
sendGeneric(0, 0, // No Header
kSharpBitMark, kSharpOneSpace, kSharpBitMark, kSharpZeroSpace,
kSharpBitMark, kSharpGap, data, nbits, 38, true,
0, // Repeats are handled already.
33);
// Invert the data per protocol. This is always called twice, so it's
// retured to original upon exiting the inner loop.
data ^= kSharpToggleMask;
}
}
}
// Encode a (raw) Sharp message from it's components.
//
// Args:
// address: The value of the address to be sent.
// command: The value of the address to be sent. (8 bits)
// expansion: The value of the expansion bit to use. (0 or 1, typically 1)
// check: The value of the check bit to use. (0 or 1, typically 0)
// MSBfirst: Flag indicating MSB first or LSB first order. (Default: false)
// Returns:
// An uint32_t containing the raw Sharp message for sendSharpRaw().
//
// Status: BETA / Should work okay.
//
// Notes:
// Assumes the standard Sharp bit sizes.
// Historically sendSharp() sends address & command in
// MSB first order. This is actually incorrect. It should be sent in LSB
// order. The behaviour of sendSharp() hasn't been changed to maintain
// backward compatibility.
//
// Ref:
// http://www.sbprojects.com/knowledge/ir/sharp.htm
// http://lirc.sourceforge.net/remotes/sharp/GA538WJSA
// http://www.mwftr.com/ucF08/LEC14%20PIC%20IR.pdf
uint32_t IRsend::encodeSharp(uint16_t address, uint16_t command,
uint16_t expansion, uint16_t check,
bool MSBfirst) {
// Mask any unexpected bits.
address &= ((1 << kSharpAddressBits) - 1);
command &= ((1 << kSharpCommandBits) - 1);
expansion &= 1;
check &= 1;
if (!MSBfirst) { // Correct bit order if needed.
address = reverseBits(address, kSharpAddressBits);
command = reverseBits(command, kSharpCommandBits);
}
// Concatinate all the bits.
return (address << (kSharpCommandBits + 2)) | (command << 2) |
(expansion << 1) | check;
}
// Send a Sharp message
//
// Args:
// address: Address value to be sent.
// command: Command value to be sent.
// nbits: Nr. of bits of data to be sent. Typically kSharpBits.
// repeat: Nr. of additional times the message is to be sent.
//
// Status: DEPRICATED / Previously working fine.
//
// Notes:
// This procedure has a non-standard invocation style compared to similar
// sendProtocol() routines. This is due to legacy, compatibility, & historic
// reasons. Normally the calling syntax version is like sendSharpRaw().
// This procedure transmits the address & command in MSB first order, which is
// incorrect. This behaviour is left as-is to maintain backward
// compatibility with legacy code.
// In short, you should use sendSharpRaw(), encodeSharp(), and the correct
// values of address & command instead of using this, & the wrong values.
//
// Ref:
// http://www.sbprojects.com/knowledge/ir/sharp.htm
// http://lirc.sourceforge.net/remotes/sharp/GA538WJSA
// http://www.mwftr.com/ucF08/LEC14%20PIC%20IR.pdf
void IRsend::sendSharp(uint16_t address, uint16_t command, uint16_t nbits,
uint16_t repeat) {
sendSharpRaw(encodeSharp(address, command, 1, 0, true), nbits, repeat);
}
#endif // (SEND_SHARP || SEND_DENON)
#if (DECODE_SHARP || DECODE_DENON)
// Decode the supplied Sharp message.
//
// Args:
// results: Ptr to the data to decode and where to store the decode result.
// nbits: Nr. of data bits to expect. Typically kSharpBits.
// strict: Flag indicating if we should perform strict matching.
// expansion: Should we expect the expansion bit to be set. Default is true.
// Returns:
// boolean: True if it can decode it, false if it can't.
//
// Status: STABLE / Working fine.
//
// Note:
// This procedure returns a value suitable for use in sendSharpRaw().
// TODO(crankyoldgit): Need to ensure capture of the inverted message as it can
// be missed due to the interrupt timeout used to detect an end of message.
// Several compliance checks are disabled until that is resolved.
// Ref:
// http://www.sbprojects.com/knowledge/ir/sharp.php
// http://www.mwftr.com/ucF08/LEC14%20PIC%20IR.pdf
// http://www.hifi-remote.com/johnsfine/DecodeIR.html#Sharp
bool IRrecv::decodeSharp(decode_results *results, uint16_t nbits, bool strict,
bool expansion) {
if (results->rawlen < 2 * nbits + kFooter - 1)
return false; // Not enough entries to be a Sharp message.
// Compliance
if (strict) {
if (nbits != kSharpBits) return false; // Request is out of spec.
// DISABLED - See TODO
#ifdef UNIT_TEST
// An in spec message has the data sent normally, then inverted. So we
// expect twice as many entries than to just get the results.
if (results->rawlen < 2 * (2 * nbits + kFooter)) return false;
#endif
}
uint64_t data = 0;
uint16_t offset = kStartOffset;
// No header
// But try to auto-calibrate off the initial mark signal.
if (!matchMark(results->rawbuf[offset], kSharpBitMark, 35)) return false;
// Calculate how long the common tick time is based on the header mark.
uint32_t tick = results->rawbuf[offset] * kRawTick / kSharpBitMarkTicks;
// Data
for (uint16_t i = 0; i < nbits; i++, offset++) {
// Use a higher tolerance value for kSharpBitMark as it is quite small.
if (!matchMark(results->rawbuf[offset++], kSharpBitMarkTicks * tick, 35))
return false;
if (matchSpace(results->rawbuf[offset], kSharpOneSpaceTicks * tick))
data = (data << 1) | 1; // 1
else if (matchSpace(results->rawbuf[offset], kSharpZeroSpaceTicks * tick))
data <<= 1; // 0
else
return false;
}
// Footer
if (!match(results->rawbuf[offset++], kSharpBitMarkTicks * tick))
return false;
if (offset < results->rawlen &&
!matchAtLeast(results->rawbuf[offset], kSharpGapTicks * tick))
return false;
// Compliance
if (strict) {
// Check the state of the expansion bit is what we expect.
if ((data & 0b10) >> 1 != expansion) return false;
// The check bit should be cleared in a normal message.
if (data & 0b1) return false;
// DISABLED - See TODO
#ifdef UNIT_TEST
// Grab the second copy of the data (i.e. inverted)
// Header
// i.e. The inter-data/command repeat gap.
if (!matchSpace(results->rawbuf[offset++], kSharpGapTicks * tick))
return false;
// Data
uint64_t second_data = 0;
for (uint16_t i = 0; i < nbits; i++, offset++) {
// Use a higher tolerance value for kSharpBitMark as it is quite small.
if (!matchMark(results->rawbuf[offset++], kSharpBitMarkTicks * tick, 35))
return false;
if (matchSpace(results->rawbuf[offset], kSharpOneSpaceTicks * tick))
second_data = (second_data << 1) | 1; // 1
else if (matchSpace(results->rawbuf[offset], kSharpZeroSpaceTicks * tick))
second_data <<= 1; // 0
else
return false;
}
// Footer
if (!match(results->rawbuf[offset++], kSharpBitMarkTicks * tick))
return false;
if (offset < results->rawlen &&
!matchAtLeast(results->rawbuf[offset], kSharpGapTicks * tick))
return false;
// Check that second_data has been inverted correctly.
if (data != (second_data ^ kSharpToggleMask)) return false;
#endif // UNIT_TEST
}
// Success
results->decode_type = SHARP;
results->bits = nbits;
results->value = data;
// Address & command are actually transmitted in LSB first order.
results->address = reverseBits(data, nbits) & kSharpAddressMask;
results->command =
reverseBits((data >> 2) & kSharpCommandMask, kSharpCommandBits);
return true;
}
#endif // (DECODE_SHARP || DECODE_DENON)