mirror of
https://github.com/esphome/esphome.git
synced 2026-02-13 13:07:41 -07:00
862 lines
29 KiB
C++
862 lines
29 KiB
C++
#include "esphome/core/helpers.h"
|
|
|
|
#include "esphome/core/defines.h"
|
|
#include "esphome/core/hal.h"
|
|
#include "esphome/core/log.h"
|
|
#include "esphome/core/progmem.h"
|
|
#include "esphome/core/string_ref.h"
|
|
|
|
#include <strings.h>
|
|
#include <algorithm>
|
|
#include <cctype>
|
|
#include <cmath>
|
|
#include <cstdarg>
|
|
#include <cstdio>
|
|
#include <cstring>
|
|
|
|
#ifdef USE_ESP32
|
|
#include "rom/crc.h"
|
|
#endif
|
|
|
|
namespace esphome {
|
|
|
|
static const char *const TAG = "helpers";
|
|
|
|
static const uint16_t CRC16_A001_LE_LUT_L[] = {0x0000, 0xc0c1, 0xc181, 0x0140, 0xc301, 0x03c0, 0x0280, 0xc241,
|
|
0xc601, 0x06c0, 0x0780, 0xc741, 0x0500, 0xc5c1, 0xc481, 0x0440};
|
|
static const uint16_t CRC16_A001_LE_LUT_H[] = {0x0000, 0xcc01, 0xd801, 0x1400, 0xf001, 0x3c00, 0x2800, 0xe401,
|
|
0xa001, 0x6c00, 0x7800, 0xb401, 0x5000, 0x9c01, 0x8801, 0x4400};
|
|
|
|
#ifndef USE_ESP32
|
|
static const uint16_t CRC16_8408_LE_LUT_L[] = {0x0000, 0x1189, 0x2312, 0x329b, 0x4624, 0x57ad, 0x6536, 0x74bf,
|
|
0x8c48, 0x9dc1, 0xaf5a, 0xbed3, 0xca6c, 0xdbe5, 0xe97e, 0xf8f7};
|
|
static const uint16_t CRC16_8408_LE_LUT_H[] = {0x0000, 0x1081, 0x2102, 0x3183, 0x4204, 0x5285, 0x6306, 0x7387,
|
|
0x8408, 0x9489, 0xa50a, 0xb58b, 0xc60c, 0xd68d, 0xe70e, 0xf78f};
|
|
#endif
|
|
|
|
#if !defined(USE_ESP32) || defined(USE_ESP32_VARIANT_ESP32S2)
|
|
static const uint16_t CRC16_1021_BE_LUT_L[] = {0x0000, 0x1021, 0x2042, 0x3063, 0x4084, 0x50a5, 0x60c6, 0x70e7,
|
|
0x8108, 0x9129, 0xa14a, 0xb16b, 0xc18c, 0xd1ad, 0xe1ce, 0xf1ef};
|
|
static const uint16_t CRC16_1021_BE_LUT_H[] = {0x0000, 0x1231, 0x2462, 0x3653, 0x48c4, 0x5af5, 0x6ca6, 0x7e97,
|
|
0x9188, 0x83b9, 0xb5ea, 0xa7db, 0xd94c, 0xcb7d, 0xfd2e, 0xef1f};
|
|
#endif
|
|
|
|
// Mathematics
|
|
|
|
uint8_t crc8(const uint8_t *data, uint8_t len, uint8_t crc, uint8_t poly, bool msb_first) {
|
|
while ((len--) != 0u) {
|
|
uint8_t inbyte = *data++;
|
|
if (msb_first) {
|
|
// MSB first processing (for polynomials like 0x31, 0x07)
|
|
crc ^= inbyte;
|
|
for (uint8_t i = 8; i != 0u; i--) {
|
|
if (crc & 0x80) {
|
|
crc = (crc << 1) ^ poly;
|
|
} else {
|
|
crc <<= 1;
|
|
}
|
|
}
|
|
} else {
|
|
// LSB first processing (default for Dallas/Maxim 0x8C)
|
|
for (uint8_t i = 8; i != 0u; i--) {
|
|
bool mix = (crc ^ inbyte) & 0x01;
|
|
crc >>= 1;
|
|
if (mix)
|
|
crc ^= poly;
|
|
inbyte >>= 1;
|
|
}
|
|
}
|
|
}
|
|
return crc;
|
|
}
|
|
|
|
uint16_t crc16(const uint8_t *data, uint16_t len, uint16_t crc, uint16_t reverse_poly, bool refin, bool refout) {
|
|
#ifdef USE_ESP32
|
|
if (reverse_poly == 0x8408) {
|
|
crc = crc16_le(refin ? crc : (crc ^ 0xffff), data, len);
|
|
return refout ? crc : (crc ^ 0xffff);
|
|
}
|
|
#endif
|
|
if (refin) {
|
|
crc ^= 0xffff;
|
|
}
|
|
#ifndef USE_ESP32
|
|
if (reverse_poly == 0x8408) {
|
|
while (len--) {
|
|
uint8_t combo = crc ^ (uint8_t) *data++;
|
|
crc = (crc >> 8) ^ CRC16_8408_LE_LUT_L[combo & 0x0F] ^ CRC16_8408_LE_LUT_H[combo >> 4];
|
|
}
|
|
} else
|
|
#endif
|
|
{
|
|
if (reverse_poly == 0xa001) {
|
|
while (len--) {
|
|
uint8_t combo = crc ^ (uint8_t) *data++;
|
|
crc = (crc >> 8) ^ CRC16_A001_LE_LUT_L[combo & 0x0F] ^ CRC16_A001_LE_LUT_H[combo >> 4];
|
|
}
|
|
} else {
|
|
while (len--) {
|
|
crc ^= *data++;
|
|
for (uint8_t i = 0; i < 8; i++) {
|
|
if (crc & 0x0001) {
|
|
crc = (crc >> 1) ^ reverse_poly;
|
|
} else {
|
|
crc >>= 1;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
return refout ? (crc ^ 0xffff) : crc;
|
|
}
|
|
|
|
uint16_t crc16be(const uint8_t *data, uint16_t len, uint16_t crc, uint16_t poly, bool refin, bool refout) {
|
|
#if defined(USE_ESP32) && !defined(USE_ESP32_VARIANT_ESP32S2)
|
|
if (poly == 0x1021) {
|
|
crc = crc16_be(refin ? crc : (crc ^ 0xffff), data, len);
|
|
return refout ? crc : (crc ^ 0xffff);
|
|
}
|
|
#endif
|
|
if (refin) {
|
|
crc ^= 0xffff;
|
|
}
|
|
#if !defined(USE_ESP32) || defined(USE_ESP32_VARIANT_ESP32S2)
|
|
if (poly == 0x1021) {
|
|
while (len--) {
|
|
uint8_t combo = (crc >> 8) ^ *data++;
|
|
crc = (crc << 8) ^ CRC16_1021_BE_LUT_L[combo & 0x0F] ^ CRC16_1021_BE_LUT_H[combo >> 4];
|
|
}
|
|
} else {
|
|
#endif
|
|
while (len--) {
|
|
crc ^= (((uint16_t) *data++) << 8);
|
|
for (uint8_t i = 0; i < 8; i++) {
|
|
if (crc & 0x8000) {
|
|
crc = (crc << 1) ^ poly;
|
|
} else {
|
|
crc <<= 1;
|
|
}
|
|
}
|
|
}
|
|
#if !defined(USE_ESP32) || defined(USE_ESP32_VARIANT_ESP32S2)
|
|
}
|
|
#endif
|
|
return refout ? (crc ^ 0xffff) : crc;
|
|
}
|
|
|
|
// FNV-1 hash - deprecated, use fnv1a_hash() for new code
|
|
uint32_t fnv1_hash(const char *str) {
|
|
uint32_t hash = FNV1_OFFSET_BASIS;
|
|
if (str) {
|
|
while (*str) {
|
|
hash *= FNV1_PRIME;
|
|
hash ^= *str++;
|
|
}
|
|
}
|
|
return hash;
|
|
}
|
|
|
|
float random_float() { return static_cast<float>(random_uint32()) / static_cast<float>(UINT32_MAX); }
|
|
|
|
// Strings
|
|
|
|
bool str_equals_case_insensitive(const std::string &a, const std::string &b) {
|
|
return strcasecmp(a.c_str(), b.c_str()) == 0;
|
|
}
|
|
bool str_equals_case_insensitive(StringRef a, StringRef b) {
|
|
return a.size() == b.size() && strncasecmp(a.c_str(), b.c_str(), a.size()) == 0;
|
|
}
|
|
#if __cplusplus >= 202002L
|
|
bool str_startswith(const std::string &str, const std::string &start) { return str.starts_with(start); }
|
|
bool str_endswith(const std::string &str, const std::string &end) { return str.ends_with(end); }
|
|
#else
|
|
bool str_startswith(const std::string &str, const std::string &start) { return str.rfind(start, 0) == 0; }
|
|
bool str_endswith(const std::string &str, const std::string &end) {
|
|
return str.rfind(end) == (str.size() - end.size());
|
|
}
|
|
#endif
|
|
|
|
bool str_endswith_ignore_case(const char *str, size_t str_len, const char *suffix, size_t suffix_len) {
|
|
if (suffix_len > str_len)
|
|
return false;
|
|
return strncasecmp(str + str_len - suffix_len, suffix, suffix_len) == 0;
|
|
}
|
|
|
|
std::string str_truncate(const std::string &str, size_t length) {
|
|
return str.length() > length ? str.substr(0, length) : str;
|
|
}
|
|
std::string str_until(const char *str, char ch) {
|
|
const char *pos = strchr(str, ch);
|
|
return pos == nullptr ? std::string(str) : std::string(str, pos - str);
|
|
}
|
|
std::string str_until(const std::string &str, char ch) { return str.substr(0, str.find(ch)); }
|
|
// wrapper around std::transform to run safely on functions from the ctype.h header
|
|
// see https://en.cppreference.com/w/cpp/string/byte/toupper#Notes
|
|
template<int (*fn)(int)> std::string str_ctype_transform(const std::string &str) {
|
|
std::string result;
|
|
result.resize(str.length());
|
|
std::transform(str.begin(), str.end(), result.begin(), [](unsigned char ch) { return fn(ch); });
|
|
return result;
|
|
}
|
|
std::string str_lower_case(const std::string &str) { return str_ctype_transform<std::tolower>(str); }
|
|
std::string str_upper_case(const std::string &str) { return str_ctype_transform<std::toupper>(str); }
|
|
std::string str_snake_case(const std::string &str) {
|
|
std::string result = str;
|
|
for (char &c : result) {
|
|
c = to_snake_case_char(c);
|
|
}
|
|
return result;
|
|
}
|
|
char *str_sanitize_to(char *buffer, size_t buffer_size, const char *str) {
|
|
if (buffer_size == 0) {
|
|
return buffer;
|
|
}
|
|
size_t i = 0;
|
|
while (*str && i < buffer_size - 1) {
|
|
buffer[i++] = to_sanitized_char(*str++);
|
|
}
|
|
buffer[i] = '\0';
|
|
return buffer;
|
|
}
|
|
|
|
std::string str_sanitize(const std::string &str) {
|
|
std::string result;
|
|
result.resize(str.size());
|
|
str_sanitize_to(&result[0], str.size() + 1, str.c_str());
|
|
return result;
|
|
}
|
|
std::string str_snprintf(const char *fmt, size_t len, ...) {
|
|
std::string str;
|
|
va_list args;
|
|
|
|
str.resize(len);
|
|
va_start(args, len);
|
|
size_t out_length = vsnprintf(&str[0], len + 1, fmt, args);
|
|
va_end(args);
|
|
|
|
if (out_length < len)
|
|
str.resize(out_length);
|
|
|
|
return str;
|
|
}
|
|
std::string str_sprintf(const char *fmt, ...) {
|
|
std::string str;
|
|
va_list args;
|
|
|
|
va_start(args, fmt);
|
|
size_t length = vsnprintf(nullptr, 0, fmt, args);
|
|
va_end(args);
|
|
|
|
str.resize(length);
|
|
va_start(args, fmt);
|
|
vsnprintf(&str[0], length + 1, fmt, args);
|
|
va_end(args);
|
|
|
|
return str;
|
|
}
|
|
|
|
// Maximum size for name with suffix: 120 (max friendly name) + 1 (separator) + 6 (MAC suffix) + 1 (null term)
|
|
static constexpr size_t MAX_NAME_WITH_SUFFIX_SIZE = 128;
|
|
|
|
size_t make_name_with_suffix_to(char *buffer, size_t buffer_size, const char *name, size_t name_len, char sep,
|
|
const char *suffix_ptr, size_t suffix_len) {
|
|
size_t total_len = name_len + 1 + suffix_len;
|
|
|
|
// Silently truncate if needed: prioritize keeping the full suffix
|
|
if (total_len >= buffer_size) {
|
|
// NOTE: This calculation could underflow if suffix_len >= buffer_size - 2,
|
|
// but this is safe because this helper is only called with small suffixes:
|
|
// MAC suffixes (6-12 bytes), ".local" (5 bytes), etc.
|
|
name_len = buffer_size - suffix_len - 2; // -2 for separator and null terminator
|
|
total_len = name_len + 1 + suffix_len;
|
|
}
|
|
|
|
memcpy(buffer, name, name_len);
|
|
buffer[name_len] = sep;
|
|
memcpy(buffer + name_len + 1, suffix_ptr, suffix_len);
|
|
buffer[total_len] = '\0';
|
|
return total_len;
|
|
}
|
|
|
|
std::string make_name_with_suffix(const char *name, size_t name_len, char sep, const char *suffix_ptr,
|
|
size_t suffix_len) {
|
|
char buffer[MAX_NAME_WITH_SUFFIX_SIZE];
|
|
size_t len = make_name_with_suffix_to(buffer, sizeof(buffer), name, name_len, sep, suffix_ptr, suffix_len);
|
|
return std::string(buffer, len);
|
|
}
|
|
|
|
std::string make_name_with_suffix(const std::string &name, char sep, const char *suffix_ptr, size_t suffix_len) {
|
|
return make_name_with_suffix(name.c_str(), name.size(), sep, suffix_ptr, suffix_len);
|
|
}
|
|
|
|
// Parsing & formatting
|
|
|
|
size_t parse_hex(const char *str, size_t length, uint8_t *data, size_t count) {
|
|
size_t chars = std::min(length, 2 * count);
|
|
for (size_t i = 2 * count - chars; i < 2 * count; i++, str++) {
|
|
uint8_t val = parse_hex_char(*str);
|
|
if (val > 15)
|
|
return 0;
|
|
data[i >> 1] = (i & 1) ? data[i >> 1] | val : val << 4;
|
|
}
|
|
return chars;
|
|
}
|
|
|
|
std::string format_mac_address_pretty(const uint8_t *mac) {
|
|
char buf[18];
|
|
format_mac_addr_upper(mac, buf);
|
|
return std::string(buf);
|
|
}
|
|
|
|
// Internal helper for hex formatting - base is 'a' for lowercase or 'A' for uppercase
|
|
static char *format_hex_internal(char *buffer, size_t buffer_size, const uint8_t *data, size_t length, char separator,
|
|
char base) {
|
|
if (length == 0) {
|
|
buffer[0] = '\0';
|
|
return buffer;
|
|
}
|
|
// With separator: total length is 3*length (2*length hex chars, (length-1) separators, 1 null terminator)
|
|
// Without separator: total length is 2*length + 1 (2*length hex chars, 1 null terminator)
|
|
uint8_t stride = separator ? 3 : 2;
|
|
size_t max_bytes = separator ? (buffer_size / stride) : ((buffer_size - 1) / stride);
|
|
if (max_bytes == 0) {
|
|
buffer[0] = '\0';
|
|
return buffer;
|
|
}
|
|
if (length > max_bytes) {
|
|
length = max_bytes;
|
|
}
|
|
for (size_t i = 0; i < length; i++) {
|
|
size_t pos = i * stride;
|
|
buffer[pos] = format_hex_char(data[i] >> 4, base);
|
|
buffer[pos + 1] = format_hex_char(data[i] & 0x0F, base);
|
|
if (separator && i < length - 1) {
|
|
buffer[pos + 2] = separator;
|
|
}
|
|
}
|
|
buffer[length * stride - (separator ? 1 : 0)] = '\0';
|
|
return buffer;
|
|
}
|
|
|
|
char *format_hex_to(char *buffer, size_t buffer_size, const uint8_t *data, size_t length) {
|
|
return format_hex_internal(buffer, buffer_size, data, length, 0, 'a');
|
|
}
|
|
|
|
std::string format_hex(const uint8_t *data, size_t length) {
|
|
std::string ret;
|
|
ret.resize(length * 2);
|
|
format_hex_to(&ret[0], length * 2 + 1, data, length);
|
|
return ret;
|
|
}
|
|
std::string format_hex(const std::vector<uint8_t> &data) { return format_hex(data.data(), data.size()); }
|
|
|
|
char *format_hex_pretty_to(char *buffer, size_t buffer_size, const uint8_t *data, size_t length, char separator) {
|
|
return format_hex_internal(buffer, buffer_size, data, length, separator, 'A');
|
|
}
|
|
|
|
char *format_hex_pretty_to(char *buffer, size_t buffer_size, const uint16_t *data, size_t length, char separator) {
|
|
if (length == 0 || buffer_size == 0) {
|
|
if (buffer_size > 0)
|
|
buffer[0] = '\0';
|
|
return buffer;
|
|
}
|
|
// With separator: each uint16_t needs 5 chars (4 hex + 1 sep), except last has no separator
|
|
// Without separator: each uint16_t needs 4 chars, plus null terminator
|
|
uint8_t stride = separator ? 5 : 4;
|
|
size_t max_values = separator ? (buffer_size / stride) : ((buffer_size - 1) / stride);
|
|
if (max_values == 0) {
|
|
buffer[0] = '\0';
|
|
return buffer;
|
|
}
|
|
if (length > max_values) {
|
|
length = max_values;
|
|
}
|
|
for (size_t i = 0; i < length; i++) {
|
|
size_t pos = i * stride;
|
|
buffer[pos] = format_hex_pretty_char((data[i] & 0xF000) >> 12);
|
|
buffer[pos + 1] = format_hex_pretty_char((data[i] & 0x0F00) >> 8);
|
|
buffer[pos + 2] = format_hex_pretty_char((data[i] & 0x00F0) >> 4);
|
|
buffer[pos + 3] = format_hex_pretty_char(data[i] & 0x000F);
|
|
if (separator && i < length - 1) {
|
|
buffer[pos + 4] = separator;
|
|
}
|
|
}
|
|
buffer[length * stride - (separator ? 1 : 0)] = '\0';
|
|
return buffer;
|
|
}
|
|
|
|
// Shared implementation for uint8_t and string hex formatting
|
|
static std::string format_hex_pretty_uint8(const uint8_t *data, size_t length, char separator, bool show_length) {
|
|
if (data == nullptr || length == 0)
|
|
return "";
|
|
std::string ret;
|
|
size_t hex_len = separator ? (length * 3 - 1) : (length * 2);
|
|
ret.resize(hex_len);
|
|
format_hex_pretty_to(&ret[0], hex_len + 1, data, length, separator);
|
|
if (show_length && length > 4)
|
|
return ret + " (" + std::to_string(length) + ")";
|
|
return ret;
|
|
}
|
|
|
|
std::string format_hex_pretty(const uint8_t *data, size_t length, char separator, bool show_length) {
|
|
return format_hex_pretty_uint8(data, length, separator, show_length);
|
|
}
|
|
std::string format_hex_pretty(const std::vector<uint8_t> &data, char separator, bool show_length) {
|
|
return format_hex_pretty(data.data(), data.size(), separator, show_length);
|
|
}
|
|
|
|
std::string format_hex_pretty(const uint16_t *data, size_t length, char separator, bool show_length) {
|
|
if (data == nullptr || length == 0)
|
|
return "";
|
|
std::string ret;
|
|
size_t hex_len = separator ? (length * 5 - 1) : (length * 4);
|
|
ret.resize(hex_len);
|
|
format_hex_pretty_to(&ret[0], hex_len + 1, data, length, separator);
|
|
if (show_length && length > 4)
|
|
return ret + " (" + std::to_string(length) + ")";
|
|
return ret;
|
|
}
|
|
std::string format_hex_pretty(const std::vector<uint16_t> &data, char separator, bool show_length) {
|
|
return format_hex_pretty(data.data(), data.size(), separator, show_length);
|
|
}
|
|
std::string format_hex_pretty(const std::string &data, char separator, bool show_length) {
|
|
return format_hex_pretty_uint8(reinterpret_cast<const uint8_t *>(data.data()), data.length(), separator, show_length);
|
|
}
|
|
|
|
char *format_bin_to(char *buffer, size_t buffer_size, const uint8_t *data, size_t length) {
|
|
if (buffer_size == 0) {
|
|
return buffer;
|
|
}
|
|
// Calculate max bytes we can format: each byte needs 8 chars
|
|
size_t max_bytes = (buffer_size - 1) / 8;
|
|
if (max_bytes == 0 || length == 0) {
|
|
buffer[0] = '\0';
|
|
return buffer;
|
|
}
|
|
size_t bytes_to_format = std::min(length, max_bytes);
|
|
|
|
for (size_t byte_idx = 0; byte_idx < bytes_to_format; byte_idx++) {
|
|
for (size_t bit_idx = 0; bit_idx < 8; bit_idx++) {
|
|
buffer[byte_idx * 8 + bit_idx] = ((data[byte_idx] >> (7 - bit_idx)) & 1) + '0';
|
|
}
|
|
}
|
|
buffer[bytes_to_format * 8] = '\0';
|
|
return buffer;
|
|
}
|
|
|
|
std::string format_bin(const uint8_t *data, size_t length) {
|
|
std::string result;
|
|
result.resize(length * 8);
|
|
format_bin_to(&result[0], length * 8 + 1, data, length);
|
|
return result;
|
|
}
|
|
|
|
ParseOnOffState parse_on_off(const char *str, const char *on, const char *off) {
|
|
if (on == nullptr && ESPHOME_strcasecmp_P(str, ESPHOME_PSTR("on")) == 0)
|
|
return PARSE_ON;
|
|
if (on != nullptr && strcasecmp(str, on) == 0)
|
|
return PARSE_ON;
|
|
if (off == nullptr && ESPHOME_strcasecmp_P(str, ESPHOME_PSTR("off")) == 0)
|
|
return PARSE_OFF;
|
|
if (off != nullptr && strcasecmp(str, off) == 0)
|
|
return PARSE_OFF;
|
|
if (ESPHOME_strcasecmp_P(str, ESPHOME_PSTR("toggle")) == 0)
|
|
return PARSE_TOGGLE;
|
|
|
|
return PARSE_NONE;
|
|
}
|
|
|
|
static inline void normalize_accuracy_decimals(float &value, int8_t &accuracy_decimals) {
|
|
if (accuracy_decimals < 0) {
|
|
auto multiplier = powf(10.0f, accuracy_decimals);
|
|
value = roundf(value * multiplier) / multiplier;
|
|
accuracy_decimals = 0;
|
|
}
|
|
}
|
|
|
|
std::string value_accuracy_to_string(float value, int8_t accuracy_decimals) {
|
|
char buf[VALUE_ACCURACY_MAX_LEN];
|
|
value_accuracy_to_buf(buf, value, accuracy_decimals);
|
|
return std::string(buf);
|
|
}
|
|
|
|
size_t value_accuracy_to_buf(std::span<char, VALUE_ACCURACY_MAX_LEN> buf, float value, int8_t accuracy_decimals) {
|
|
normalize_accuracy_decimals(value, accuracy_decimals);
|
|
// snprintf returns chars that would be written (excluding null), or negative on error
|
|
int len = snprintf(buf.data(), buf.size(), "%.*f", accuracy_decimals, value);
|
|
if (len < 0)
|
|
return 0; // encoding error
|
|
// On truncation, snprintf returns would-be length; actual written is buf.size() - 1
|
|
return static_cast<size_t>(len) >= buf.size() ? buf.size() - 1 : static_cast<size_t>(len);
|
|
}
|
|
|
|
size_t value_accuracy_with_uom_to_buf(std::span<char, VALUE_ACCURACY_MAX_LEN> buf, float value,
|
|
int8_t accuracy_decimals, StringRef unit_of_measurement) {
|
|
if (unit_of_measurement.empty()) {
|
|
return value_accuracy_to_buf(buf, value, accuracy_decimals);
|
|
}
|
|
normalize_accuracy_decimals(value, accuracy_decimals);
|
|
// snprintf returns chars that would be written (excluding null), or negative on error
|
|
int len = snprintf(buf.data(), buf.size(), "%.*f %s", accuracy_decimals, value, unit_of_measurement.c_str());
|
|
if (len < 0)
|
|
return 0; // encoding error
|
|
// On truncation, snprintf returns would-be length; actual written is buf.size() - 1
|
|
return static_cast<size_t>(len) >= buf.size() ? buf.size() - 1 : static_cast<size_t>(len);
|
|
}
|
|
|
|
int8_t step_to_accuracy_decimals(float step) {
|
|
// use printf %g to find number of digits based on temperature step
|
|
char buf[32];
|
|
snprintf(buf, sizeof buf, "%.5g", step);
|
|
|
|
std::string str{buf};
|
|
size_t dot_pos = str.find('.');
|
|
if (dot_pos == std::string::npos)
|
|
return 0;
|
|
|
|
return str.length() - dot_pos - 1;
|
|
}
|
|
|
|
// Use C-style string constant to store in ROM instead of RAM (saves 24 bytes)
|
|
static constexpr const char *BASE64_CHARS = "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
|
|
"abcdefghijklmnopqrstuvwxyz"
|
|
"0123456789+/";
|
|
|
|
// Helper function to find the index of a base64/base64url character in the lookup table.
|
|
// Returns the character's position (0-63) if found, or 0 if not found.
|
|
// Supports both standard base64 (+/) and base64url (-_) alphabets.
|
|
// NOTE: This returns 0 for both 'A' (valid base64 char at index 0) and invalid characters.
|
|
// This is safe because is_base64() is ALWAYS checked before calling this function,
|
|
// preventing invalid characters from ever reaching here. The base64_decode function
|
|
// stops processing at the first invalid character due to the is_base64() check in its
|
|
// while loop condition, making this edge case harmless in practice.
|
|
static inline uint8_t base64_find_char(char c) {
|
|
// Handle base64url variants: '-' maps to '+' (index 62), '_' maps to '/' (index 63)
|
|
if (c == '-')
|
|
return 62;
|
|
if (c == '_')
|
|
return 63;
|
|
const char *pos = strchr(BASE64_CHARS, c);
|
|
return pos ? (pos - BASE64_CHARS) : 0;
|
|
}
|
|
|
|
// Check if character is valid base64 or base64url
|
|
static inline bool is_base64(char c) { return (isalnum(c) || (c == '+') || (c == '/') || (c == '-') || (c == '_')); }
|
|
|
|
std::string base64_encode(const std::vector<uint8_t> &buf) { return base64_encode(buf.data(), buf.size()); }
|
|
|
|
std::string base64_encode(const uint8_t *buf, size_t buf_len) {
|
|
std::string ret;
|
|
int i = 0;
|
|
int j = 0;
|
|
char char_array_3[3];
|
|
char char_array_4[4];
|
|
|
|
while (buf_len--) {
|
|
char_array_3[i++] = *(buf++);
|
|
if (i == 3) {
|
|
char_array_4[0] = (char_array_3[0] & 0xfc) >> 2;
|
|
char_array_4[1] = ((char_array_3[0] & 0x03) << 4) + ((char_array_3[1] & 0xf0) >> 4);
|
|
char_array_4[2] = ((char_array_3[1] & 0x0f) << 2) + ((char_array_3[2] & 0xc0) >> 6);
|
|
char_array_4[3] = char_array_3[2] & 0x3f;
|
|
|
|
for (i = 0; (i < 4); i++)
|
|
ret += BASE64_CHARS[static_cast<uint8_t>(char_array_4[i])];
|
|
i = 0;
|
|
}
|
|
}
|
|
|
|
if (i) {
|
|
for (j = i; j < 3; j++)
|
|
char_array_3[j] = '\0';
|
|
|
|
char_array_4[0] = (char_array_3[0] & 0xfc) >> 2;
|
|
char_array_4[1] = ((char_array_3[0] & 0x03) << 4) + ((char_array_3[1] & 0xf0) >> 4);
|
|
char_array_4[2] = ((char_array_3[1] & 0x0f) << 2) + ((char_array_3[2] & 0xc0) >> 6);
|
|
char_array_4[3] = char_array_3[2] & 0x3f;
|
|
|
|
for (j = 0; (j < i + 1); j++)
|
|
ret += BASE64_CHARS[static_cast<uint8_t>(char_array_4[j])];
|
|
|
|
while ((i++ < 3))
|
|
ret += '=';
|
|
}
|
|
|
|
return ret;
|
|
}
|
|
|
|
size_t base64_decode(const std::string &encoded_string, uint8_t *buf, size_t buf_len) {
|
|
return base64_decode(reinterpret_cast<const uint8_t *>(encoded_string.data()), encoded_string.size(), buf, buf_len);
|
|
}
|
|
|
|
size_t base64_decode(const uint8_t *encoded_data, size_t encoded_len, uint8_t *buf, size_t buf_len) {
|
|
size_t in_len = encoded_len;
|
|
int i = 0;
|
|
int j = 0;
|
|
size_t in = 0;
|
|
size_t out = 0;
|
|
uint8_t char_array_4[4], char_array_3[3];
|
|
bool truncated = false;
|
|
|
|
// SAFETY: The loop condition checks is_base64() before processing each character.
|
|
// This ensures base64_find_char() is only called on valid base64 characters,
|
|
// preventing the edge case where invalid chars would return 0 (same as 'A').
|
|
while (in_len-- && (encoded_data[in] != '=') && is_base64(encoded_data[in])) {
|
|
char_array_4[i++] = encoded_data[in];
|
|
in++;
|
|
if (i == 4) {
|
|
for (i = 0; i < 4; i++)
|
|
char_array_4[i] = base64_find_char(char_array_4[i]);
|
|
|
|
char_array_3[0] = (char_array_4[0] << 2) + ((char_array_4[1] & 0x30) >> 4);
|
|
char_array_3[1] = ((char_array_4[1] & 0xf) << 4) + ((char_array_4[2] & 0x3c) >> 2);
|
|
char_array_3[2] = ((char_array_4[2] & 0x3) << 6) + char_array_4[3];
|
|
|
|
for (i = 0; i < 3; i++) {
|
|
if (out < buf_len) {
|
|
buf[out++] = char_array_3[i];
|
|
} else {
|
|
truncated = true;
|
|
}
|
|
}
|
|
i = 0;
|
|
}
|
|
}
|
|
|
|
if (i) {
|
|
for (j = i; j < 4; j++)
|
|
char_array_4[j] = 0;
|
|
|
|
for (j = 0; j < 4; j++)
|
|
char_array_4[j] = base64_find_char(char_array_4[j]);
|
|
|
|
char_array_3[0] = (char_array_4[0] << 2) + ((char_array_4[1] & 0x30) >> 4);
|
|
char_array_3[1] = ((char_array_4[1] & 0xf) << 4) + ((char_array_4[2] & 0x3c) >> 2);
|
|
char_array_3[2] = ((char_array_4[2] & 0x3) << 6) + char_array_4[3];
|
|
|
|
for (j = 0; j < i - 1; j++) {
|
|
if (out < buf_len) {
|
|
buf[out++] = char_array_3[j];
|
|
} else {
|
|
truncated = true;
|
|
}
|
|
}
|
|
}
|
|
|
|
if (truncated) {
|
|
ESP_LOGW(TAG, "Base64 decode: buffer too small, truncating");
|
|
}
|
|
|
|
return out;
|
|
}
|
|
|
|
std::vector<uint8_t> base64_decode(const std::string &encoded_string) {
|
|
// Calculate maximum decoded size: every 4 base64 chars = 3 bytes
|
|
size_t max_len = ((encoded_string.size() + 3) / 4) * 3;
|
|
std::vector<uint8_t> ret(max_len);
|
|
size_t actual_len = base64_decode(encoded_string, ret.data(), max_len);
|
|
ret.resize(actual_len);
|
|
return ret;
|
|
}
|
|
|
|
/// Decode base64/base64url string directly into vector of little-endian int32 values
|
|
/// @param base64 Base64 or base64url encoded string (both +/ and -_ accepted)
|
|
/// @param out Output vector (cleared and filled with decoded int32 values)
|
|
/// @return true if successful, false if decode failed or invalid size
|
|
bool base64_decode_int32_vector(const std::string &base64, std::vector<int32_t> &out) {
|
|
// Decode in chunks to minimize stack usage
|
|
constexpr size_t chunk_bytes = 48; // 12 int32 values
|
|
constexpr size_t chunk_chars = 64; // 48 * 4/3 = 64 chars
|
|
uint8_t chunk[chunk_bytes];
|
|
|
|
out.clear();
|
|
|
|
const uint8_t *input = reinterpret_cast<const uint8_t *>(base64.data());
|
|
size_t remaining = base64.size();
|
|
size_t pos = 0;
|
|
|
|
while (remaining > 0) {
|
|
size_t chars_to_decode = std::min(remaining, chunk_chars);
|
|
size_t decoded_len = base64_decode(input + pos, chars_to_decode, chunk, chunk_bytes);
|
|
|
|
if (decoded_len == 0)
|
|
return false;
|
|
|
|
// Parse little-endian int32 values
|
|
for (size_t i = 0; i + 3 < decoded_len; i += 4) {
|
|
int32_t timing = static_cast<int32_t>(encode_uint32(chunk[i + 3], chunk[i + 2], chunk[i + 1], chunk[i]));
|
|
out.push_back(timing);
|
|
}
|
|
|
|
// Check for incomplete int32 in last chunk
|
|
if (remaining <= chunk_chars && (decoded_len % 4) != 0)
|
|
return false;
|
|
|
|
pos += chars_to_decode;
|
|
remaining -= chars_to_decode;
|
|
}
|
|
|
|
return !out.empty();
|
|
}
|
|
|
|
// Colors
|
|
|
|
float gamma_correct(float value, float gamma) {
|
|
if (value <= 0.0f)
|
|
return 0.0f;
|
|
if (gamma <= 0.0f)
|
|
return value;
|
|
|
|
return powf(value, gamma);
|
|
}
|
|
float gamma_uncorrect(float value, float gamma) {
|
|
if (value <= 0.0f)
|
|
return 0.0f;
|
|
if (gamma <= 0.0f)
|
|
return value;
|
|
|
|
return powf(value, 1 / gamma);
|
|
}
|
|
|
|
void rgb_to_hsv(float red, float green, float blue, int &hue, float &saturation, float &value) {
|
|
float max_color_value = std::max(std::max(red, green), blue);
|
|
float min_color_value = std::min(std::min(red, green), blue);
|
|
float delta = max_color_value - min_color_value;
|
|
|
|
if (delta == 0) {
|
|
hue = 0;
|
|
} else if (max_color_value == red) {
|
|
hue = int(fmod(((60 * ((green - blue) / delta)) + 360), 360));
|
|
} else if (max_color_value == green) {
|
|
hue = int(fmod(((60 * ((blue - red) / delta)) + 120), 360));
|
|
} else if (max_color_value == blue) {
|
|
hue = int(fmod(((60 * ((red - green) / delta)) + 240), 360));
|
|
}
|
|
|
|
if (max_color_value == 0) {
|
|
saturation = 0;
|
|
} else {
|
|
saturation = delta / max_color_value;
|
|
}
|
|
|
|
value = max_color_value;
|
|
}
|
|
void hsv_to_rgb(int hue, float saturation, float value, float &red, float &green, float &blue) {
|
|
float chroma = value * saturation;
|
|
float hue_prime = fmod(hue / 60.0, 6);
|
|
float intermediate = chroma * (1 - fabs(fmod(hue_prime, 2) - 1));
|
|
float delta = value - chroma;
|
|
|
|
if (0 <= hue_prime && hue_prime < 1) {
|
|
red = chroma;
|
|
green = intermediate;
|
|
blue = 0;
|
|
} else if (1 <= hue_prime && hue_prime < 2) {
|
|
red = intermediate;
|
|
green = chroma;
|
|
blue = 0;
|
|
} else if (2 <= hue_prime && hue_prime < 3) {
|
|
red = 0;
|
|
green = chroma;
|
|
blue = intermediate;
|
|
} else if (3 <= hue_prime && hue_prime < 4) {
|
|
red = 0;
|
|
green = intermediate;
|
|
blue = chroma;
|
|
} else if (4 <= hue_prime && hue_prime < 5) {
|
|
red = intermediate;
|
|
green = 0;
|
|
blue = chroma;
|
|
} else if (5 <= hue_prime && hue_prime < 6) {
|
|
red = chroma;
|
|
green = 0;
|
|
blue = intermediate;
|
|
} else {
|
|
red = 0;
|
|
green = 0;
|
|
blue = 0;
|
|
}
|
|
|
|
red += delta;
|
|
green += delta;
|
|
blue += delta;
|
|
}
|
|
|
|
uint8_t HighFrequencyLoopRequester::num_requests = 0; // NOLINT(cppcoreguidelines-avoid-non-const-global-variables)
|
|
void HighFrequencyLoopRequester::start() {
|
|
if (this->started_)
|
|
return;
|
|
num_requests++;
|
|
this->started_ = true;
|
|
}
|
|
void HighFrequencyLoopRequester::stop() {
|
|
if (!this->started_)
|
|
return;
|
|
num_requests--;
|
|
this->started_ = false;
|
|
}
|
|
bool HighFrequencyLoopRequester::is_high_frequency() { return num_requests > 0; }
|
|
|
|
std::string get_mac_address() {
|
|
uint8_t mac[6];
|
|
get_mac_address_raw(mac);
|
|
char buf[13];
|
|
format_mac_addr_lower_no_sep(mac, buf);
|
|
return std::string(buf);
|
|
}
|
|
|
|
std::string get_mac_address_pretty() {
|
|
char buf[MAC_ADDRESS_PRETTY_BUFFER_SIZE];
|
|
return std::string(get_mac_address_pretty_into_buffer(buf));
|
|
}
|
|
|
|
void get_mac_address_into_buffer(std::span<char, MAC_ADDRESS_BUFFER_SIZE> buf) {
|
|
uint8_t mac[6];
|
|
get_mac_address_raw(mac);
|
|
format_mac_addr_lower_no_sep(mac, buf.data());
|
|
}
|
|
|
|
const char *get_mac_address_pretty_into_buffer(std::span<char, MAC_ADDRESS_PRETTY_BUFFER_SIZE> buf) {
|
|
uint8_t mac[6];
|
|
get_mac_address_raw(mac);
|
|
format_mac_addr_upper(mac, buf.data());
|
|
return buf.data();
|
|
}
|
|
|
|
#ifndef USE_ESP32
|
|
bool has_custom_mac_address() { return false; }
|
|
#endif
|
|
|
|
bool mac_address_is_valid(const uint8_t *mac) {
|
|
bool is_all_zeros = true;
|
|
bool is_all_ones = true;
|
|
|
|
for (uint8_t i = 0; i < 6; i++) {
|
|
if (mac[i] != 0) {
|
|
is_all_zeros = false;
|
|
}
|
|
if (mac[i] != 0xFF) {
|
|
is_all_ones = false;
|
|
}
|
|
}
|
|
return !(is_all_zeros || is_all_ones);
|
|
}
|
|
|
|
void IRAM_ATTR HOT delay_microseconds_safe(uint32_t us) {
|
|
// avoids CPU locks that could trigger WDT or affect WiFi/BT stability
|
|
uint32_t start = micros();
|
|
|
|
const uint32_t lag = 5000; // microseconds, specifies the maximum time for a CPU busy-loop.
|
|
// it must be larger than the worst-case duration of a delay(1) call (hardware tasks)
|
|
// 5ms is conservative, it could be reduced when exact BT/WiFi stack delays are known
|
|
if (us > lag) {
|
|
delay((us - lag) / 1000UL); // note: in disabled-interrupt contexts delay() won't actually sleep
|
|
while (micros() - start < us - lag)
|
|
delay(1); // in those cases, this loop allows to yield for BT/WiFi stack tasks
|
|
}
|
|
while (micros() - start < us) // fine delay the remaining usecs
|
|
;
|
|
}
|
|
|
|
} // namespace esphome
|