Shell Scripting

[NeedHelp]

I need to do an IF statement and if there is a count of two do one thing and a count of one do another. My question is, is there a count function? My function basically looks like this:

cat filename | grep ..... | {print 1}

If there are two results in I need the 2nd one, otherwise the first one. Any ideas?

[NeedHelp]

Bump, can anyone help?

[ghostdog74]

Originally Posted by NeedHelp

I need to do an IF statement and if there is a count of two do one thing and a count of one do another. My question is, is there a count function? My function basically looks like this:

cat filename | grep ..... | {print 1}

If there are two results in I need the 2nd one, otherwise the first one. Any ideas?

if i understand you, you wanted to count a word occurence in a file?

awk -v RS="gggg" 'END{print (NR?NR-1:0)}' file #gggg is the search string

[Jordan]

You can also use grep for that or WC. Are you counting lines or characters?

grep -c // Line count
wc -l // Line Count

So just do a normal shell if statement on a variable that contains those values.

[ghostdog74]

Originally Posted by Jordan

You can also use grep for that or WC. Are you counting lines or characters?

grep -c // Line count
wc -l // Line Count

So just do a normal shell if statement on a variable that contains those values.

depending on OP's requirement, if he is counting words, grep -c would not be accurate if 2 or more words occur on a line. grep -c will count as 1, because it returns matching number of lines which the word is found.also wc -l counts number of lines, not words.

[Jordan]

grep -c will only count lines, hence the "// Line Count" comment.You can change the -l of wc to "wc -w" for word count.

[ghostdog74]

wc -w counts total words in a file. it will not work if OP only wants to count a certain word in a file.

[Jordan]

Sorry, I meant to type -m

-m, --chars print the character counts

[ghostdog74]

for example this sample file:

# more file
this is a test line
this is second line
# wc -m file
40 file
# wc -c file
40 file

but if OP wants to count how many times the word "line" occurs, both the above will not work, whether -m or -c. however:

# awk -v RS="line" 'END{print (NR?NR-1:0)}' file
2

[NeedHelp]

I needed the line count. I was able to accomplish it doing:

COUNT=`cat $1 | grep -c  rtn`

if [ "$COUNT" -eq 1 ]; then
.....