grep string without anything surrounding it - count

Question

I have a directory with multiple files, let's call these 1.txt, 2.txt, etc. Each file consists of 3 columns: an ID, a lat, and a lon. Unfortunately, some of the ID's may also be numbered (e.g., 1346, 248, 67912, etc.). I am trying to count the number of instances a station ID occurs among all files (1.txt, 2.txt, etc.) based on a master file (masterfile.txt). So far I have:
while IFS='' read line || [[ -n "$line" ]]
do
cat * | grep -w -c $line >> counting_filename.txt
done < masterfile.txt
Which works great. However, if the lat and/or long contain the particular ID, this will also count. For example, if I am looking for the station ID of 4575, if there is a lat of '47.4575' or a lon of '-77.4575', these will also go towards the count. Therefore, there are two solutions I could think of that I can't figure out:
1) Just grep the first column of instances, or
2) grep while NOT including the leading '.'
For example:
1.txt
4575 39.4575 -77.51
5010 38.3498 -78.4575
LAMS 38.4444 -78.3126
2.txt
3124 39.1010 -79.4575
4575 39.4575 -77.5010
PAOQ 39.2222 -78.0032
If I ran the above command, I would get a count of 6 for 4575, 2 for 5010, 1 for LAMS, 1 for 3124, and 1 for PAOQ.
What is desired is: 2 for 4575, 1 for 5010, 1 for LAMS, 1 for 3124, and 1 for PAOQ.
Any thoughts?

Answer 1

You're using the wrong tools - a small, simple awk script will be far more robust, efficient, and portable than a mixture of shell loops, read, grep, etc.
It's not clear what masterfile.txt is for as from your example it looks like this is all you need:
$ awk '{cnt[$1]++} END{for (id in cnt) print id, cnt[id]}' 1.txt 2.txt
LAMS 1
PAOQ 1
3124 1
4575 2
5010 1
If you need masterfile.txt to list a specific set of IDs rather than just producing counts for all IDs as above then you can do that too:
$ cat masterfile.txt
4575
3124
PAOQ
BLAH
$ awk 'NR==FNR{ids[$1];next} $1 in ids{cnt[$1]++} END{for (id in cnt) print id, cnt[id]}' masterfile.txt 1.txt 2.txt
PAOQ 1
3124 1
4575 2
$ awk 'NR==FNR{ids[$1];next} $1 in ids{cnt[$1]++} END{for (id in ids) print id, cnt[id]+0}' masterfile.txt 1.txt 2.txt
BLAH 0
PAOQ 1
3124 1
4575 2
I added BLAH to show the different options you have of handling an ID from masterfile.txt that doesn't appear in your other files.