Trying to identify the difference between the two methods.
I recently had to find a the string abcABC in ~/path/. In that same directory were the subdirectories ~/path/files/ and ~/path/docs/ along with a series of .h, .cpp and .txt files. I only needed to look in the ~/path/ level so using:
grep -r -i abcABC ~/path/ produced findings in subdirectories which I didn't want.
Using the * wildcard to extend the path to be ~/path/*.*/ allowed me to use:
grep -i abcABC ~/path/*.* which found the subdirectories and files. However, I needed further refinement to only look at the .h and .cpp files for abcABC.
Did some reading in man grep and learned about --include=GLOB and altered my command to the following:
grep -i --include \*.h --include \*.cpp abcABC ~/path/*.*
This worked, but then I tried (for S&G's) the following:
grep -i abcABC ~/path/*.{h,cpp} and produced the same results I needed.
What I would like to know is what does { } do for grep? man grep puts {n,m} in repetition but I am not sure how doing this method equates to --include=GLOB.
EDIT: Last response by Jordan gives great detailed Wikipedia example. The duplicate (slightly) problem provides great use case.
