Python 找不到回车符（^M 或 ) 对 nano 和 vi 可见

Question

我目前正在编写一个 bash 脚本，该脚本将其执行的函数的输出作为日志文件返回。当在记事本中查看时，显示函数过程的文本占据了文件的大部分。理想情况下，我不想要所有这些批量，我只想显示进度文本的最后一个输出片段，显示该过程已成功执行。

output_log_18_01_2024.log - Notepad
+---------------------------------------------------------------------------+
 Reading flash ............... 0x00000000 (0%) 
 Reading flash ............... 0x00001000 (0%) 
 Reading flash ............... 0x00002000 (0%) 
 Reading flash ............... 0x00003000 (0%) 
 Reading flash ............... 0x00004000 (0%) 
 Reading flash ............... 0x00005000 (0%) 
 Reading flash ............... 0x00006000 (0%) 
 Reading flash ............... 0x00007000 (0%) 
 Reading flash ............... 0x00008000 (0%) 
 Reading flash ............... 0x00009000 (0%) 
 Reading flash ............... 0x0000A000 (0%) 
 Reading flash ............... 0x0000B000 (0%) 
 Reading flash ............... 0x0000C000 (0%) 
 Reading flash ............... 0x0000D000 (0%) 
 Reading flash ............... 0x0000E000 (0%) 
 Reading flash ............... 0x0000F000 (0%) 
 Reading flash ............... 0x00010000 (0%) 
 Reading flash ............... 0x00011000 (0%) 
 Reading flash ............... 0x00012000 (0%) 
 Reading flash ............... 0x00013000 (0%)
 ...

在命令行上的 nano 或 vi 中查看时，此过程文本显示为由回车符 ^M 分隔的一行。

GNU nano 5.6.1                           output_log_18_01_2024.log
+---------------------------------------------------------------------------+
Reading flash ............... 0x00000000 (0%) ^M Reading flash ............... 0x00001000 (0%) ^M Reading flash ............... 0x00002000 (0%) ^M Reading flash ............... 0x00003000 (0%) ^M Reading flash ............... 0x00004000 (0%) etc...

因此，我尝试使用 python 脚本来检测这些回车符并相应地处理它们。这是我尝试过的脚本：

import sys, re

f = open(sys.argv[1], "r")
data = f.read()

counter = 0
for char in data:
    print(char, end="")
    if char == r'\r':
        counter += 1
        
print(counter)
# always returns 0


rx = r'\r'
rr = re.search(rx, data)

print(rr)
# always returns None

f.close()

为什么python看不到这些回车符？是否只是将它们解释为新线而我无能为力？我愿意接受 python 或 bash 中的任何支持，基本上任何能让我删除大部分日志并保留重要文本的支持。

编辑：删除错误图像，替换为文本副本。

Answer 1

这是一个最终对我有用的答案。可能有一种更有效的方法来做到这一点。但这是脚本的一部分已经非常非常慢了。

# formatter.py

#!/usr/bin/python3

import sys

### Split at newlines.
testlist = repr(sys.argv[1]).split(r'\n')

for i in range(0, len(testlist)):

    line = testlist[i]
    if line[1] == " ":
        line = line.replace(" ","",1)
    ### If a return carriage is found in line
    if (line.find(r'\r') > -1):
        old_idx = 0
        curr_idx = line.find(r'\r', old_idx)
        ### Find last occurence of \r   
        while curr_idx != -1:
            old_idx = curr_idx
            curr_idx = line.find(r'\r', (old_idx+1))
        
        ### Return line from position of penultimate carriage return to end of line.
        print(line[old_idx+2:].replace(r'\r',""))
    ### If no carriage returns are found, return script as is.
    else:
        print(line)

# echoline.sh

#!/bin/bash

echoLine()
{
    outfile="test.txt"

    # get input
    set -f # stops * being expanded
    if test ! -t 0 # pipe
    then
        indata=$(</dev/stdin)
    elif test -n "$1"
    then
        indata="$@"
    fi

    indata="${indata//^M/'\r'}"
    IFS='$'; arrIN=($indata); unset IFS;

    for (( x=0; x<${#arrIN[@]}; x++))
    do
        outdata=$(python3 soanswer.py "${arrIN[x]}" 2>/dev/null || python3 soanswer.py "${arrIN[x]: -300}")
        outdata="${outdata//"'"/""} \n"
        echo -ne "$outdata" >> $outfile
        echo -ne "$outdata"
    done
}

echoLine -s $(cat -e output_log_18_01_2024_old.log)