Find files containing multiple strings - string

Question

I use a command to recursively find files containing a certain string1:
find . -type f -exec grep -H string1 {} \;
I need to find files containing multiple strings, so the command should return those containing all strings. Something like this:
find . -type f -exec grep -H string1 AND string2 {} \;
I couldn't find a way. The strings can be anywhere in the files. Even a solution for only two strings would be nice.

Answer 1

you can also try this;
find . -type f -exec grep -l 'string1' {} \; | xargs grep -l 'string2'
this shows file names that contain string1 and string2

Answer 2

You can chain your actions and use the exit status of the first one to only execute the second one if the first one was successful. (Omitting the operator between primaries defaults to -and/-a.)
find . -type f -exec grep -q 'string1' {} \; -exec grep -H 'string2' {} \;
The first grep command uses -q, "quiet", which returns a successful exit status if the string was found.
To collect all files containing string1 and then run the search for string2 with just a single invocation of grep, you could use -exec ... {} +:
find . -type f -exec grep -q 'string1' {} \; -exec grep 'string2' {} +

Answer 3

with GNU grep
grep -rlZ 'string1' | xargs -0 grep -l 'string2'
from man grep
-r, --recursive
Read all files under each directory, recursively, following symbolic
links only if they are on the command line. Note that if no file
operand is given, grep searches the working directory. This is
equivalent to the -d recurse option.
-Z, --null
Output a zero byte (the ASCII NUL character) instead of the character that normally follows a file
name. For example, grep -lZ outputs a zero byte after each file name instead of the usual newline.
This option makes the output unambiguous, even in the presence of file names containing unusual
characters like newlines. This option can be used with commands like find -print0, perl -0, sort -z,
and xargs -0 to process arbitrary file names, even those that contain newline characters.

Answer 4

Amazed that this old question lacks the obvious simple Awk solution:
find . -type f -exec awk '/string1/ && /string2/ { print; r=1 } END { exit 1-r }' {} \;
The trickery with the r variable is just to emulate the exit code from grep (zero means found, one means not; if you don't care, you can take that out).
For efficiency, maybe switch from -exec ... {} \; to -exec ... {} + though then you might want to refactor the Awk script a bit (either throw out the exit code, or change it so the exit code indicates something like "no files matched" vs "only some files matched" vs "all files matched"?)
The above code looks for files which contain both strings on the same line. The case of finding them on any lines is an easy change.
awk '/string1/ { s1=1 }
/string2/ { s2=1 }
s1 && s2 { print FILENAME; exit }
END { exit(1 - (s1 && s2)) }' file
This just prints the name of the file, and assumes that you have a single input file. For processing multiple files, refactor slightly, to reset the values of s1 and s2 when visiting a new file:
awk 'FNR == 1 { s1 = s2 = 0 }
/string1/ { s1 = 1 }
/string2/ { s2 = 1 }
s1 && s2 { r=1; print FILENAME; nextfile }
END { exit 1-r }' file1 file2 file3 ...
Some ancient Awk versions might not support nextfile, though it is now in POSIX.

Answer 5

Answer
As you can see from the other answers on this page, there are several command-line tools that can be used to perform conjunctive searching across files. A fast and flexible solution that has not yet been posted is to use ag:
ag -l string1 | xargs ag -l string2
Useful variations
For case-insensitive searching, use the -i option of ag:
ag -il string1 | xargs ag -il string2
For additional search terms, extend the pipeline:
ag -l string1 | xargs ag -l string2 | xargs ag -l string3 | xargs ag -l string4

Answer 6

grep -rlZ string1 | xargs -0 grep -l string2
If your patterns are fixed strings, we can speed up the command by adding -F to grep:
grep -rlZF string1 | xargs -0 grep -lF string2