If I need to do regex replacement 27000 times, how to shorten it?

[tatahuft]

If I need to do regex replacement 27000 times with different pattern search and replacement, how to shorten it?

import re
from _Arm64HexDatabase import *

handle_file1 = open("example1", "rb")
hex_read = handle_file1.read()
handle_file1.close()

hfile3 = open("result1.txt", "a")

hex_read = re.sub(hex_suba64w0w0w0, hex_mova64w0c0, hex_read)
...
hex_read = re.sub(hex_suba64w31w31w31, hex_mova64w31c0, hex_read)

hex_read = re.sub(hex_suba64x0x0x0, hex_mova64x0c0, hex_read)
...
hex_read = re.sub(hex_suba64x31x31x31, hex_mova64x31c0, hex_read)

hex_read = re.sub(hex_suba64w0w0c0, hex_mova64w0w0, hex_read)
...
hex_read = re.sub(hex_suba64w31w31c0, hex_mova64w31w31, hex_read)

hex_read = re.sub(hex_suba64x0x0c0, hex_mova64x0x0, hex_read)
...
hex_read = re.sub(hex_suba64x31x31c0, hex_mova64x31x31, hex_read)

hex_read = re.sub(hex_suba64w0w0wzr, hex_mova64w0w0, hex_read)
...
hex_read = re.sub(hex_suba64w31w31wzr, hex_mova64w31w31, hex_read)

hex_read = re.sub(hex_suba64x0x0xzr, hex_mova64x0x0, hex_read)
...
hex_read = re.sub(hex_suba64x31x31xzr, hex_mova64x31x31, hex_read)

hex_read = re.sub(hex_adda64w0w0c0, hex_mova64w0w0, hex_read)
...
hex_read = re.sub(hex_adda64w31w31c0, hex_mova64w31w31, hex_read)

hex_read = re.sub(hex_adda64x0x0c0, hex_mova64x0x0, hex_read)
...
hex_read = re.sub(hex_adda64x31x31c0, hex_mova64x31x31, hex_read)

hex_read = re.sub(hex_adda64w0w0wzr, hex_mova64w0w0, hex_read)
...
hex_read = re.sub(hex_adda64w31w31wzr, hex_mova64w31w31, hex_read)

hex_read = re.sub(hex_adda64x0x0xzr, hex_mova64x0x0, hex_read)
...
hex_read = re.sub(hex_adda64x31x31xzr, hex_mova64x31x31, hex_read)

The re.sub need to be done about 27000 times with different first and second argument

I wrote it 27000 times (27000 lines) in python

How to shorten it?

[Gribouillis]

We don't know what the patterns and replacements are, so it is almost impossible to help.

[paul18fr]

with few informations and nothing to test, i would propose the following snippet:

import re
from _Arm64HexDatabase import *


inputVar = [
            "hex_suba64w0w0w0", 
            "hex_suba64w31w31w31", 
            "hex_suba64x0x0x0", 
            "hex_suba64x31x31x31", 
            ...
            ]

outputVar = [
             "hex_mova64w0c0", 
             "hex_mova64w31c0", 
             "hex_mova64x0c0", 
             "hex_mova64x31c0", 
             ...
             ]
# note: the 2 lists might be written once (and for all) in a file and loaded
#       the snippet assumes the pattern is exactly this one (otherwise if .. then ... must be used)

i = 0
with open ("example1", "rb") as handle_file1, open("result1.txt", "a") as hfile3:
    inputLine = handle_file1.readline()

    while (inputLine != ''):
        outputLine = re.sub(inputVar[i], outputVar[i], inputLine)
        hfile3.write(f"{outputLine}\n")
        i += 1
        inputLine = handle_file1.readline()

[ndc85430]

Also, could you maybe tell us what overall problem you're trying to solve? It just seems a bit much that you're having to do 27000 regex replacements in the first place. If you tell us what you were trying to do that ended you up here, it might help us to provide a more appropriate solution.

[deanhystad]

Looking through your threads I’m left wondering if you are solving a very unusual problem, or if it is your solution that is unusual and the problem can be solved more simply. Please describe the problem.