for file in $(find . -type f -iname '*.json'); do
sed -i 's/"verified":"10"/"verified":"11"/' $file;
done
this post was submitted on 26 Aug 2023
33 points (100.0% liked)
Linux
48074 readers
895 users here now
From Wikipedia, the free encyclopedia
Linux is a family of open source Unix-like operating systems based on the Linux kernel, an operating system kernel first released on September 17, 1991 by Linus Torvalds. Linux is typically packaged in a Linux distribution (or distro for short).
Distributions include the Linux kernel and supporting system software and libraries, many of which are provided by the GNU Project. Many Linux distributions use the word "Linux" in their name, but the Free Software Foundation uses the name GNU/Linux to emphasize the importance of GNU software, causing some controversy.
Rules
- Posts must be relevant to operating systems running the Linux kernel. GNU/Linux or otherwise.
- No misinformation
- No NSFW content
- No hate speech, bigotry, etc
Related Communities
Community icon by Alpár-Etele Méder, licensed under CC BY 3.0
founded 5 years ago
MODERATORS
Find can actually do the sed itself if you don't want to use a subshell and a shell loop.
find . -type f -iname '*.json' -exec sed -i 's/"verified":"10"/"verified":"11"/' '{}' ';'-print0 | xargs -0 sed -i to get a single sed process working across multiple files.
Add a -P 8 to xargs to get 8 parallel processes.
Today I learned that xargs supports parallelization natively! That's gonna make some of my scripts much simpler
Change the ';' to a '+' for even more efficiency (no need to fork+exec a sed process per file, sed can take multiple files)
+1
And
In the off chance the files are not under git or some other VCS, might be a good idea to add the -b option to backup
With sed or awk, without a sample is hard to give a proper example.
sed -i 's/old/new/g' *.json
That rhymes!
With sed or awk, without a sample
is hard to give, a proper example
Name checks out, this is the way
Our sedvior!
You could use jq, which will work no matter how the json is formatted.
Without trying it out, something like the following might work:
jq '.path.to.key.to.change |= 11' file.json > file.json.tmp && mv file.json.tmp file.json
I also agree sed and some regex is your best bet
I recommend formatting the regex with regex101.com, I'm down to help you if you post some examples
Additionally there is a cli tool, I think jq or something like that, for processing json on the command line
I have foundry too, let me see if I can find the files that need to be updated
Here's the GitHub link to one of the batches of files I'm working with.
This line ,,"compatibility":{"minimum":"9","verified":"10"}," needs to say" 11" in all the files
I have made a python script and ran it on a clone of your git repo to confirm it works, simply run it at the root directory of wherever the files are, it will walk through and find module.json and do the replace.
#!/usr/bin/env python3
import re
import os
import fileinput
pattern = re.compile(r'(?P\.+)\"compatibility\":{\"minimum\":\"(?P\\d+)\",\"verified\":\"(?P\\d+)\"},(?P\.+)')
def make11(match):
if match.groupdict().get('min', None) and match.groupdict().get('ver', None):
return f"{match.groupdict()['pre']}\"compatibility\":{{\"minimum\":\"11\",\"verified\":\"11\"}},{match.groupdict()['post']}"
for root, dirs, files in os.walk("."):
for file in files:
if file == "module.json":
for line in fileinput.input(f"{root}/{file}", inplace=True):
print(re.sub(pattern, make11, line))
edit: lemmy is fucking with the formatting and removing the fucking regex group names, which will bork it. I've tried fixing it, dm me if you want me to send a downloadable link to the script
Neat script; just a touch overkill IMO compared to just using sed and bash!
Changing the minimum may be undesirable - I think it's only the latter value that needs to go from 10 -> 11.
If using Python, why not just use JSON module? Simpler and easier maintain without all those regex.
Still +1, on sed if one is on Linux.
Holy shit that's awesome! Thanks
I also agree sed and some regex is your best bet
Nah, regexes are okay if you really have no other choice, but they're a bit of a hamfisted tool. For a json file, which is a neatly structured format, I would always try to do it with jq first.
sed or awk might get you there, but something like jq which is meant for json might be a bit more ideal depending on your use-case.
find /path -name *.json -exec sed -i 's/from/to/g' {} ; -print
This, unless you want to mess with jq
Yeah, jq doesn't edit files, right? You'd have to have temp files or something? jq is so good handling json, I wish there was a way of using it to edit files.
You really want to do it that way anyway.. process the files to a new set of files. That way when you screw it up going back is just deleting the new files, fixing and rerunning.
Output redirect back into the file
That'll give you an empty file 🙂
Typical rookie mistake :D
Oh yea, you’re right
Since I don't know the structure of your files, I can't help entirely, but I would use find/locate to get a list of file paths, then use a script to take that list and use sed for the replacement, like this:
#!/bin/bash
for i in ListOfFilePaths.txt
do
sed -i "s/oldtext/newtext/g" $i
done
Please copy the entire line for oldtext and newtext to avoid accidental replacements.
Also, I am very new to scripting, and this likely has multiple problems with it. I am just throwing out ideas.
Looked at your github. I would do this in a script:
#!/bin/bash
find /base/path/of/files -type f -name "module.json" > ListOfFilePaths.txt
for i in ListOfFilePaths.txt
do
sed -i "s/oldtext/newtext/g" $i
done
Once again, probably not the most efficient way to do it, but it might work.
Thanks I'll give that a try!
Not sure but you can always ask chatgpt. Probably using something like sed or awk
I would love to see what it spit out if you feed it exactly what OP asked.
This was the response from chatgpt when I coppied OP's exact post. It wasn't too far off:
Yes, there's a way to automate this process using a script. You can use a combination of the find command and sed to search and replace the version number in all your files. Here's a sample command you can use:
find /path/to/assets -type f -name "*.asset" -exec sed -i 's/verified version 10/verified version 11/g' {} +
Replace /path/to/assets with the actual path to your asset folders. This command will recursively search for .asset files and replace "verified version 10" with "verified version 11". Make sure to have a backup of your files before running this command, just in case.
Also, consider testing this on a smaller set of files first to ensure it works as expected before applying it to all 400+ files.
Cool. Thanks for posting that.