Tags: #bash #find #filter #execute #grep #uniq #sort #search
#!/usr/local/bin/bash
DIR="$HOME/code/project-foo/"
pushd "$DIR"
# the \; is necessary when using -exec flag
find . -name 'README*' ! -path '*node_modules*' -exec wc -l {} \;
popd
# the `-type f` is not needed as I'm using the `-name` flag to anchor to Python files
# but I kept it in for the sake of me needing to remember the flag and how to use it.
#
# some files use "double" quotes while others use 'single' quotes
# to catch both variations we have to make sure the pattern is wrapped in double quotes
# as this allows us to escape the double quotes used in the pattern itself
# if your pattern was wrapped in single quotes, then the shell will ignore any \ escaping of a single quote within your pattern
time find . -type f -name '*.py' -exec egrep -o "\(r?f?['\"](\/[^'\"]+)" {} \; | sort | uniq -c | sort
# find all yaml files inside of a 'src' directory but ignore 'settings.yaml'
# TODO: double check we didn't mean -name instead of -path
$ find -E ~/Code/fastly/api-documentation -iregex '.+/src/.+\.yaml$' ! -path '*settings.yaml' -print
# now find all API endpoints defined in those files
# e.g. I want to find things like '/customer/{customer_id}/users:'
$ find -E ~/Code/fastly/api-documentation -iregex '.+/src/.+\.yaml$' -print -exec grep -E '\s/.+:$' {} \;
# now normalise the api by replacing the interpolation syntax {} with ...
# we do this by using gsed's extended regex -E and then starting by finding { and keep going until something that's not an } and then ending match with }
$ find -E ~/Code/fastly/api-documentation -iregex '.+/src/.+\.yaml$' -exec grep -E '\s/.+:$' {} \; | gsed -E 's/\{[^}]+\}/.../g'
# find all files named pipfile or requirement
$ find -E . -iregex '.+(pipfile|requirements).+'
# print all matching files that include bf_tornado (or bf-tornado) + print the bf_tornado value itself
# we do the latter by searching lines that match the bf_tornado pattern and then use -B ("before") to specify that the output should include one line before
$ find -E . -iregex '.+(pipfile|requirements).+' -print -exec egrep 'bf[_-]tornado' {} \; | egrep -B 1 '^bf\-'
# find any files that contain the given string ("nginx").
# then we use the grep command's -l, --files-with-matches flag to ensure only file names are printed.
# we use the ! operator (same as -not flag) to exclude matches.
# we also have to be careful with files that have spaces in their name, so we use '* *' to try and capture those.
# we also want to avoid dotfiles such as .git
# the output will be paths like './webapp_waf/nginx.conf.j2' so we use the cut command to extract the service name (e.g. webapp_waf)
$ time find . -type f ! -path '*.git/*' ! -path '*.venv*' ! -path '* *' ! -name '*.md' ! -name '*.json' ! -name '*.yml' | xargs grep -l -i nginx | sort | uniq | sort | cut -d / -f 2 | uniq | sort