Cfrgtkky

I spent plenty of time searching for an existing question about my issue but I falied to find any solution for this case in specific. I apologize in advance if there is an existing thread/question solving it, and would appreciate if anyone could point me forward.

I have to search a pattern (e.g. "GENE") among numerous files in a data system. However, the data set is named as follows:

foo_1549474392_93.nwk

foo_1549474392_93.ort.final.nwk

foo_1549474392_93.ort.nwk

foo_1549474392_93.ort_reroot.nwk

These are the text files for analysis n# 93. There are 550 in total. Thing is, I have to find the pattern only in the files named like "foo_1549474392_93.nwk" (i.e.: anything + underscore + digit(s) + .nwk, with numbers going from 1 to 550, not 001 to 550). Everything else before the underline is irrelevant.

I have already tried this

grep "GENE" *'/d'.nwk 

and many variations using [0-9] and so on.

Thank you very much for the help!

To search any file whose name matches anything + underscore + digits + .nwk, with the digits going from 1 to 550 and excluding 001, 551, etc, try:

shopt -s nullglob; grep GENE  *_{1..550}.nwk

Because {1..550} expands to the numbers that you want (and only the numbers you want), the glob *_{1..550}.nwk will include only the files that you want. To assure that all the files in the expansion actually exist in the directory, we set nullglob with shopt -s nullglob.

Since you may not want the change in nullglob to affect other commands, it may be useful to use parens to put the command in a subshell.

(shopt -s nullglob; grep GENE  *_{1..550}.nwk)

The change in nullglob affects only the subshell (what's inside the parens) and not anything before or after.

Example

Let's create four files with GENE, two that match your filename criteria and two that don't:

$ echo GENE | tee bad_1224_01.nwk bad_3456_551.nwk good_23456_1.nwk good_763456_550.nwk

GENE

$ ls

bad_1224_01.nwk  bad_3456_551.nwk  good_23456_1.nwk  good_763456_550.nwk

Now, let's run our command:

$ (shopt -s nullglob; grep GENE  *_{1..550}.nwk)

good_23456_1.nwk:GENE

good_763456_550.nwk:GENE

The unwanted files are excluded and the good names are found.

For comparison, let's try:

$ grep "GENE" *[0-9].nwk 

bad_1224_01.nwk:GENE

bad_3456_551.nwk:GENE

good_23456_1.nwk:GENE

good_763456_550.nwk:GENE

This matches the unwanted files.