grep patterns from file1 into column of file2

Question

I have two files:

$ cat File1
A
B
C

$ cat File2
A aaa B
D bbb A
B aaa h

I would like to search patterns from File1 into File2, a sort of what grep -f File1 File2 would do, but searching the patterns reported in File1 only in $1 of File2

Sample output:

$cat File3
A aaa B
B aaa h

Answer 1

With awk:

awk 'NR==FNR{a[$0]=NR; next} a[$1]' f1.txt f2.txt

• NR==FNR{a[$0]=NR; next}: for first file (f1.txt) we are putting the record as key to an assiciative array with the corresponding record number as the value

• a[$1]: for second file (f2.txt), the record is only printed if the first field is a key of array a

Example:

% cat f1.txt                                       
A
B
C

% cat f2.txt                                       
A aaa B
D bbb A
B aaa h

% awk 'NR==FNR{a[$0]=NR; next} a[$1]' f1.txt f2.txt
A aaa B
B aaa h

Answer 2

Using join command:

join <(sort file1) <(sort file2)

If the files are sorted.

join file1 file2

Answer 3

With bash or any shell that understands process substitution:

$ grep -f <( awk '{ printf("^%s[[:blank:]]\n", $0) }' File1 ) File2
A aaa B
B aaa h

The idea here is to create the correct patterns for grep -f File1 to work directly on File2 by transforming each line in File from something to the regular expression ^something[[:blank:]] (prefix it with a circumflex and suffix it with [[:blank:]]).

The circumflex anchors the pattern to the start of the line and [[:blank:]] forces a match against a space or a tab character.

GNU grep may also read patterns from standard input:

$ awk '{ printf("^%s[[:blank:]]\n", $0) }' File1 | grep -f - File2
A aaa B
B aaa h

The awk command may be replaced by an equivalent sed command (if you prefer sed over awk):

$ sed -e 's/^/^/' -e 's/$/[[:blank:]]/' File1 | grep -f - File2