How to use the text-processing tag effectively?

Question

Sometimes, a question about text processing is tagged with the utilities like awk or sed, when they are just soliciting any valid solutions. Should we attempt to clean-up such questions by replacing these specific tags with just text-processing?

For these questions, my justification is that using the utilities-specific tags are either too strict, or will be repetitive across questions if enforced. In other words, if the question doesn't care whether awk or sed gets the job done, why should the question be tagged as such? What if a perl solution comes in? Hence, I feel that the all-purpose text-processing tag should suffice.

Of course, if the question started as an awk question, and is seeking help specifically for that, then we can leave that tag in place.

Answer 1

There is already guidance on the tag text-processing in the tag wiki, which I think covers the issue quite well.

Unix systems tend to favor text files, often consisting of one record per line. Most Unix configuration files are text files. Unix systems come with many tools to manipulate such files. Most tools process the file in a stream: read a line, process it, emit the corresponding output; this makes it possible to chain scripts with pipes.

Use this tag when your question is about processing text files and you're not sure which tool to use. If your question is about a specific tool, use its tag.

I believe when this topic has come up in the past we've rationalized it like this.

• Try to get clarification from the OP if they strictly want an awk, sed, etc. solution or are open to any command. Both situations come up.
• We're generally very tolerant on unix.stackexchange.com vs. other SE's with respect to not flagging answers to questions that make use of alternative tools, when a question is tagged with something specific.
• Part of our interaction with OP's is to help get this clarification up front via comments, before providing a solution, so we should focus on doing this as the questions come. Often times, users new to Unix are unaware that they have choices. They will have some vague notions about there being tools like awk, sed, etc. but don't fully grasp their subtleties.

Cleaning up the site is always an appreciated service so I would encourage you to pull some searches together to get an idea of how many questions we're looking at here and then post another question on meta to help track the work. Let us know through comments/chat if you're unclear how to proceed and what tasks you need others to help with.

We'll generally post a search or list of questions and divide the work up amongst other members so that it isn't too burdensome for a single individual.