-
時間: 2024.07.30 作者: Yuan
-
問題簡述
這是一個發生在很久之前的問題,但是值得被記錄和澄清一下!
在2022年1月份參與的專案中,其中有一個需求是利用 SHT40-AD1B-R2 晶片,讀取環境溫度。其實就是透過 i2c 與這個溫度感測器建立通訊,然後讀取溫溼度資訊。對於上位機的我而言,需要解決的一點就是將從感測器中讀回的資料換算為以攝氏度為單位的十進位制數值。當時費了一番腦筋,最後找到了一個"解決"的方法。
在2022年6月份,梳理 Python 知識點的過程中,我猛然間發現似乎不用那麼複雜,使用Python內建的函式就可以直接轉換…… 頓時感覺自己“愚蠢且努力”!
現在是2024年7月份,事情過去兩年了,忽然想起需要整理下這個內容。由於當時對“位元組字串”的認識不足,以及對python基礎不夠清楚,所以走了彎路。現在再次確認,發現兩年前的解決辦法實際上是有bug的,現在就來解一解這個兩年前的bug!
- byte 字串初見
"""
當時在看到讀回的資料為類似 `b'fX'`、`b'f4'`、`b'd\xac\xd6\xa5\x9f\x1e'` 這樣的結果時,一時間有點摸不著頭腦,
於是請教了客戶那邊的軟體工程師,客戶給我做了一個示例。
有了客戶的示例和指導後,我就奔著目標開幹了:
將遇到的\x開頭的2位16進位制字元記為16進位制,非\x開頭的16進位制的字元轉為 ascii 碼!
"""
# Yuan: I2C 讀取溫度返回的資訊, 有點奇怪: b'f4' 、b'fX' 、b'f`5\x95\xA6\xE1'
# Kim: 什麼型號的溫度感測元件
# Yuan: SHT40-AD1B-R2
# Kim: b'f4' 算出來是24.866483558 °C
# 我覺得應該是正常的
# b'f4' = 0x66, 0x34
# 0x6634 = 26164
# ((26164/65535)*175) - 45 = 24.866483558
# b'f'' = 0x66, 0x27
# ' 這個的ascii code應該是0x27
# Yuan: 是用 ascii code 轉換成16進位制的對嗎?
# Kim: Yes,因為那是byte array
# 除非看到\x
# 不然就都要轉ASCII
# Yuan: 好的,\x開頭的都是2位16進位制, 對嗎
# Kim: Yes
- “笨拙的”解決方案
#!/usr/bin/env python3
"""
思路:
將 “\x後跟2位字元” 部分直接記錄為 Hex 值,將非“\x後跟2位字元”的單個字元轉換為 ASCII 碼,
於是,我突發奇想從utf-8中找了幾個希臘字母(因為他們不在ascii表中,避免了重複),先將 “\x後跟2位字元” 這類已知的Hex子字串替換成希臘字母,
然後,再遍歷替換後的字串,遇到的非希臘字母,就將其從 ASCII 碼,轉換為對應的Hex形式。
最終,將兩類Hex字元彙總成純Hex字元的字串,再統一轉換為十進位制數值,再進行下一步溫度的計算。
"""
import re
temp_list = [
b'dW',
b'd^',
b'dB',
b'd>',
b'dP',
b'dD',
b'dS',
b'dQ',
b'dP',
b'dO',
b'dK',
b'',
b'dU',
b'df',
b'd_',
b'dd(\xa6\x99\x95',
b'd\xa4o\xa5\xc2,',
b'd\xb0\xe8\xa5\xddA',
b'd\xa3\xf8\xa5\xaa.',
b'd\xb0\xe8\xa5\xac\x88',
b'd\xb8Q\xa5\xb3\xe5',
b'd\xac\xd6\xa5\x8b\x99',
b'd\xaap\xa5\xa41',
b'd\xae\xb4\xa5\xa7b',
b'd\xa7<\xa5\x93c',
b'd\xa5^\xa5\x98\x89',
b'd\xac\xd6\xa5\x9f\x1e',
b'd\xb9`\xa5\x9d|',
b'd\xa7<\xa5\xa3\xa6',
b'd\xb1\xd9\xa5\x9a\xeb',
b'd\xb8Q\xa5\x94\xf4',
b'd\xb9`\xa5\x89\xfb',
b'd\xb8Q\xa5xH',
b'',
b'd\xb3\xbb\xa5\x91\x01',
b'd\xb0\xe8\xa5\x96\x96',
b'd\xba3\xa5\x7f\xdf',
b'd\xb2\x8a\xa5xH',
b'd\xaap\xa5s\xa2',
b'd\xabA\xa5s\xa2',
b'd\xba3\xa5\x81B',
b'd\xb8Q\xa5n\xad',
b'd\xb2\x8a\xa5yy',
b'd\xac\xd6\xa5h\x0b',
b'd\xb6N\xa5n\xad',
b'd\xabA\xa5wf',
b'd\xad\xe7\xa5l\xcf',
b'd\xbc\x95\xa5i:',
b'd\xb3\xbb\xa5];',
b'd\xb1\xd9\xa5eG',
b'd\xbb\x02\xa5s\xa2',
b'd\xac\xd6\xa5p\xf1',
b'',
b'd\xad\xe7\xa5\\\n',
b'\n',
b'd\n',
b'd\xbc\x95\xa5O\x1a',
b'd\xb1\xd9\xa5LI']
def get_temperature(_data):
if _data:
data1 = str(_data).lstrip("b").strip("'")
reg = r'\\x([0-9a-f]{2})'
val_ret = re.findall(reg, data1)
unichr = 'δλξπσω'
unidict = {}
data2 = str(data1)
if val_ret:
for i, ret in enumerate(val_ret):
unidict[unichr[i]] = '0x{}'.format(ret)
data2 = data2.replace('\\x{}'.format(ret), unichr[i])
hex_str = []
for da in data2:
if da in unidict.keys():
hex_str.append(unidict[da])
else:
hex_str.append(hex(ord(da)))
data_pool = hex_str
value1 = int(''.join(data_pool[:2]).replace('0x', ''), 16)
degc = round(((value1 / 65535) * 175) - 45, 2)
# prh = round(((value2 / 65535) * 125) - 6, 2)
# if prh > 100:
# prh = 100
# if prh < 0:
# prh = 0
# unit = chr(8451)
unit = "°C"
if degc > 40 or degc < 10:
result = 'degc: {} {} ({}, {})'.format(degc, unit, _data, val_ret)
else:
result = 'degc: {} {}'.format(degc, unit)
# print('DEG: {} {}'.format(degc, unit), ', PRH: {} {}'.format(prh, '%'))
# print("{} ==> {} ==> Hex: {} ==> DEG: {} {}".format(_data, val_ret, data_pool, degc, unit))
print(f"{str(_data).ljust(25)} ==> hex: {repr(val_ret).ljust(30)} ==> full_hex: {repr(data_pool).ljust(64)} ==> DEG: {degc} {unit}")
else:
result = '-- no data --'
return result
if __name__ == '__main__':
for i, temp in enumerate(temp_list):
# print('{}: {} {}'.format(str(i).zfill(2), get_temperature(temp), temp))
get_temperature(temp)
"""
輸出:
b'dW' ==> hex: [] ==> full_hex: ['0x64', '0x57'] ==> DEG: 23.59 °C
b'd^' ==> hex: [] ==> full_hex: ['0x64', '0x5e'] ==> DEG: 23.61 °C
b'dB' ==> hex: [] ==> full_hex: ['0x64', '0x42'] ==> DEG: 23.54 °C
b'd>' ==> hex: [] ==> full_hex: ['0x64', '0x3e'] ==> DEG: 23.53 °C
b'dP' ==> hex: [] ==> full_hex: ['0x64', '0x50'] ==> DEG: 23.57 °C
b'dD' ==> hex: [] ==> full_hex: ['0x64', '0x44'] ==> DEG: 23.54 °C
b'dS' ==> hex: [] ==> full_hex: ['0x64', '0x53'] ==> DEG: 23.58 °C
b'dQ' ==> hex: [] ==> full_hex: ['0x64', '0x51'] ==> DEG: 23.58 °C
b'dP' ==> hex: [] ==> full_hex: ['0x64', '0x50'] ==> DEG: 23.57 °C
b'dO' ==> hex: [] ==> full_hex: ['0x64', '0x4f'] ==> DEG: 23.57 °C
b'dK' ==> hex: [] ==> full_hex: ['0x64', '0x4b'] ==> DEG: 23.56 °C
b'dU' ==> hex: [] ==> full_hex: ['0x64', '0x55'] ==> DEG: 23.59 °C
b'df' ==> hex: [] ==> full_hex: ['0x64', '0x66'] ==> DEG: 23.63 °C
b'd_' ==> hex: [] ==> full_hex: ['0x64', '0x5f'] ==> DEG: 23.61 °C
b'dd(\xa6\x99\x95' ==> hex: ['a6', '99', '95'] ==> full_hex: ['0x64', '0x64', '0x28', '0xa6', '0x99', '0x95'] ==> DEG: 23.63 °C
b'd\xa4o\xa5\xc2,' ==> hex: ['a4', 'a5', 'c2'] ==> full_hex: ['0x64', '0xa4', '0x6f', '0xa5', '0xc2', '0x2c'] ==> DEG: 23.8 °C
b'd\xb0\xe8\xa5\xddA' ==> hex: ['b0', 'e8', 'a5', 'dd'] ==> full_hex: ['0x64', '0xb0', '0xe8', '0xa5', '0xdd', '0x41'] ==> DEG: 23.83 °C
b'd\xa3\xf8\xa5\xaa.' ==> hex: ['a3', 'f8', 'a5', 'aa'] ==> full_hex: ['0x64', '0xa3', '0xf8', '0xa5', '0xaa', '0x2e'] ==> DEG: 23.8 °C
b'd\xb0\xe8\xa5\xac\x88' ==> hex: ['b0', 'e8', 'a5', 'ac', '88'] ==> full_hex: ['0x64', '0xb0', '0xe8', '0xa5', '0xac', '0x88'] ==> DEG: 23.83 °C
b'd\xb8Q\xa5\xb3\xe5' ==> hex: ['b8', 'a5', 'b3', 'e5'] ==> full_hex: ['0x64', '0xb8', '0x51', '0xa5', '0xb3', '0xe5'] ==> DEG: 23.85 °C
b'd\xac\xd6\xa5\x8b\x99' ==> hex: ['ac', 'd6', 'a5', '8b', '99'] ==> full_hex: ['0x64', '0xac', '0xd6', '0xa5', '0x8b', '0x99'] ==> DEG: 23.82 °C
b'd\xaap\xa5\xa41' ==> hex: ['aa', 'a5', 'a4'] ==> full_hex: ['0x64', '0xaa', '0x70', '0xa5', '0xa4', '0x31'] ==> DEG: 23.81 °C
b'd\xae\xb4\xa5\xa7b' ==> hex: ['ae', 'b4', 'a5', 'a7'] ==> full_hex: ['0x64', '0xae', '0xb4', '0xa5', '0xa7', '0x62'] ==> DEG: 23.83 °C
b'd\xa7<\xa5\x93c' ==> hex: ['a7', 'a5', '93'] ==> full_hex: ['0x64', '0xa7', '0x3c', '0xa5', '0x93', '0x63'] ==> DEG: 23.81 °C
b'd\xa5^\xa5\x98\x89' ==> hex: ['a5', 'a5', '98', '89'] ==> full_hex: ['0x64', '0xa5', '0x5e', '0xa5', '0x98', '0x89'] ==> DEG: 23.8 °C
b'd\xac\xd6\xa5\x9f\x1e' ==> hex: ['ac', 'd6', 'a5', '9f', '1e'] ==> full_hex: ['0x64', '0xac', '0xd6', '0xa5', '0x9f', '0x1e'] ==> DEG: 23.82 °C
b'd\xb9`\xa5\x9d|' ==> hex: ['b9', 'a5', '9d'] ==> full_hex: ['0x64', '0xb9', '0x60', '0xa5', '0x9d', '0x7c'] ==> DEG: 23.85 °C
b'd\xa7<\xa5\xa3\xa6' ==> hex: ['a7', 'a5', 'a3', 'a6'] ==> full_hex: ['0x64', '0xa7', '0x3c', '0xa5', '0xa3', '0xa6'] ==> DEG: 23.81 °C
b'd\xb1\xd9\xa5\x9a\xeb' ==> hex: ['b1', 'd9', 'a5', '9a', 'eb'] ==> full_hex: ['0x64', '0xb1', '0xd9', '0xa5', '0x9a', '0xeb'] ==> DEG: 23.83 °C
b'd\xb8Q\xa5\x94\xf4' ==> hex: ['b8', 'a5', '94', 'f4'] ==> full_hex: ['0x64', '0xb8', '0x51', '0xa5', '0x94', '0xf4'] ==> DEG: 23.85 °C
b'd\xb9`\xa5\x89\xfb' ==> hex: ['b9', 'a5', '89', 'fb'] ==> full_hex: ['0x64', '0xb9', '0x60', '0xa5', '0x89', '0xfb'] ==> DEG: 23.85 °C
b'd\xb8Q\xa5xH' ==> hex: ['b8', 'a5'] ==> full_hex: ['0x64', '0xb8', '0x51', '0xa5', '0x78', '0x48'] ==> DEG: 23.85 °C
b'd\xb3\xbb\xa5\x91\x01' ==> hex: ['b3', 'bb', 'a5', '91', '01'] ==> full_hex: ['0x64', '0xb3', '0xbb', '0xa5', '0x91', '0x01'] ==> DEG: 23.84 °C
b'd\xb0\xe8\xa5\x96\x96' ==> hex: ['b0', 'e8', 'a5', '96', '96'] ==> full_hex: ['0x64', '0xb0', '0xe8', '0xa5', '0x96', '0x96'] ==> DEG: 23.83 °C
b'd\xba3\xa5\x7f\xdf' ==> hex: ['ba', 'a5', '7f', 'df'] ==> full_hex: ['0x64', '0xba', '0x33', '0xa5', '0x7f', '0xdf'] ==> DEG: 23.86 °C
b'd\xb2\x8a\xa5xH' ==> hex: ['b2', '8a', 'a5'] ==> full_hex: ['0x64', '0xb2', '0x8a', '0xa5', '0x78', '0x48'] ==> DEG: 23.84 °C
b'd\xaap\xa5s\xa2' ==> hex: ['aa', 'a5', 'a2'] ==> full_hex: ['0x64', '0xaa', '0x70', '0xa5', '0x73', '0xa2'] ==> DEG: 23.81 °C
b'd\xabA\xa5s\xa2' ==> hex: ['ab', 'a5', 'a2'] ==> full_hex: ['0x64', '0xab', '0x41', '0xa5', '0x73', '0xa2'] ==> DEG: 23.82 °C
b'd\xba3\xa5\x81B' ==> hex: ['ba', 'a5', '81'] ==> full_hex: ['0x64', '0xba', '0x33', '0xa5', '0x81', '0x42'] ==> DEG: 23.86 °C
b'd\xb8Q\xa5n\xad' ==> hex: ['b8', 'a5', 'ad'] ==> full_hex: ['0x64', '0xb8', '0x51', '0xa5', '0x6e', '0xad'] ==> DEG: 23.85 °C
b'd\xb2\x8a\xa5yy' ==> hex: ['b2', '8a', 'a5'] ==> full_hex: ['0x64', '0xb2', '0x8a', '0xa5', '0x79', '0x79'] ==> DEG: 23.84 °C
b'd\xac\xd6\xa5h\x0b' ==> hex: ['ac', 'd6', 'a5', '0b'] ==> full_hex: ['0x64', '0xac', '0xd6', '0xa5', '0x68', '0x0b'] ==> DEG: 23.82 °C
b'd\xb6N\xa5n\xad' ==> hex: ['b6', 'a5', 'ad'] ==> full_hex: ['0x64', '0xb6', '0x4e', '0xa5', '0x6e', '0xad'] ==> DEG: 23.85 °C
b'd\xabA\xa5wf' ==> hex: ['ab', 'a5'] ==> full_hex: ['0x64', '0xab', '0x41', '0xa5', '0x77', '0x66'] ==> DEG: 23.82 °C
b'd\xad\xe7\xa5l\xcf' ==> hex: ['ad', 'e7', 'a5', 'cf'] ==> full_hex: ['0x64', '0xad', '0xe7', '0xa5', '0x6c', '0xcf'] ==> DEG: 23.82 °C
b'd\xbc\x95\xa5i:' ==> hex: ['bc', '95', 'a5'] ==> full_hex: ['0x64', '0xbc', '0x95', '0xa5', '0x69', '0x3a'] ==> DEG: 23.86 °C
b'd\xb3\xbb\xa5];' ==> hex: ['b3', 'bb', 'a5'] ==> full_hex: ['0x64', '0xb3', '0xbb', '0xa5', '0x5d', '0x3b'] ==> DEG: 23.84 °C
b'd\xb1\xd9\xa5eG' ==> hex: ['b1', 'd9', 'a5'] ==> full_hex: ['0x64', '0xb1', '0xd9', '0xa5', '0x65', '0x47'] ==> DEG: 23.83 °C
b'd\xbb\x02\xa5s\xa2' ==> hex: ['bb', '02', 'a5', 'a2'] ==> full_hex: ['0x64', '0xbb', '0x02', '0xa5', '0x73', '0xa2'] ==> DEG: 23.86 °C
b'd\xac\xd6\xa5p\xf1' ==> hex: ['ac', 'd6', 'a5', 'f1'] ==> full_hex: ['0x64', '0xac', '0xd6', '0xa5', '0x70', '0xf1'] ==> DEG: 23.82 °C
b'd\xad\xe7\xa5\\\n' ==> hex: ['ad', 'e7', 'a5'] ==> full_hex: ['0x64', '0xad', '0xe7', '0xa5', '0x5c', '0x5c', '0x5c', '0x6e'] ==> DEG: 23.82 °C
b'\n' ==> hex: [] ==> full_hex: ['0x5c', '0x6e'] ==> DEG: 18.19 °C
b'd\n' ==> hex: [] ==> full_hex: ['0x64', '0x5c', '0x6e'] ==> DEG: 23.61 °C
b'd\xbc\x95\xa5O\x1a' ==> hex: ['bc', '95', 'a5', '1a'] ==> full_hex: ['0x64', '0xbc', '0x95', '0xa5', '0x4f', '0x1a'] ==> DEG: 23.86 °C
b'd\xb1\xd9\xa5LI' ==> hex: ['b1', 'd9', 'a5'] ==> full_hex: ['0x64', '0xb1', '0xd9', '0xa5', '0x4c', '0x49'] ==> DEG: 23.83 °C
"""
- 後知後覺的內建函式
# 原始位元組資料
byte_data = b'd\xad\xe7\xa5\\\n'
# 轉換為十六進位制字串
hex_data = byte_data.hex()
# 列印結果
print(hex_data)
"""
輸出:
'64ade7a55c0a'
"""
# 如果需要將十六進位制字串格式化成每個位元組兩位並用空格分隔的形式,可以使用以下程式碼
# 原始位元組資料
byte_data = b'd\xad\xe7\xa5\\\n'
# 轉換為十六進位制字串並分隔每個位元組
hex_data = byte_data.hex()
formatted_hex = ' '.join([hex_data[i:i+2] for i in range(0, len(hex_data), 2)])
# 列印結果
print(formatted_hex)
- 兩種方案執行結果比對
# 我的方案
def byte_to_hex(_byte_data):
# 去頭去尾,轉為字串
data1 = str(_byte_data).lstrip("b").strip("'")
# 匹配\x開頭的2位字元
reg = r'\\x([0-9a-f]{2})'
val_ret = re.findall(reg, data1)
# 用拉丁字母替代\x開頭的2位字元
unichr = 'δλξπσω'
unidict = {}
data2 = str(data1)
if val_ret:
for i, ret in enumerate(val_ret):
unidict[unichr[i]] = '0x{}'.format(ret)
data2 = data2.replace('\\x{}'.format(ret), unichr[i])
# 遍歷加入拉丁字母代替的新字串, 轉換為hex
hex_chr_list = []
for da in data2:
if da in unidict.keys():
hex_chr_list.append(unidict[da])
else:
hex_chr_list.append(hex(ord(da)))
hex_str = ''.join(hex_chr_list).replace('0x', '')
return hex_str
# 新的方案
_byte_data.hex()
- 最終的結論
舊的方法基本可以滿足需求; 使用 python 內建方法, 更簡單更準確,且有ascii碼錶作為依據!
# 新、舊方案轉換和計算的溫度結果,在大部分情況下基本是一致的;
# 在現有的資料中有3筆資料轉換存在差異。引起差異的地方是:對 \n 的轉換, 比如 \\\n 這樣的資料
# 經過查詢 Ascii 碼錶推測,科學的做法是:\n 需要被看成一個整體轉換為 ascii 碼的 hex 形式,而不是舊的方案中分別對 '\' 和 'n' 進行轉換
-- the end --