UNIX for Dummies Questions & Answers

My file contains the following:

uat2000.aspclient.active=true
uat2001.aspclient.active=true
uat2002.aspclient.active=true
uat2003.aspclient.active=true
uat2004.aspclient.active=false
uat2005.aspclient.active=false
uat2006.aspclient.active=false
uat2007.aspclient.active=false
uat2008.aspclient.active=false
uat2009.aspclient.active=false

I am greping which ones = true with
# cat filename | grep active=true

Results:

uat2000.aspclient.active=true
uat2001.aspclient.active=true
uat2002.aspclient.active=true
uat2003.aspclient.active=true

QUESTION: I want the results to display only the first 7 characters, ie uat2000, uat2001, etc - HOW DO I DO THAT?

Thanks in advance!

grep is not the correct tool to do this, here are some ways you could do it.
Also in you file you can use the . as a dilimiter. It is probably better to do so
then to depend on having 'n' characters.

Code:

awk -F\. '/active=true/{print $1}'

Code:

sed -n '/active=true/s/^\([^.][^.]*\)[.].*/\1/p'

Code:

IFS=.
while read one two three; do
    [[ "$3" == "active=true" ]] && echo $one
done < inputfile

Code:

grep  "active=true"  file | cut -d"." -f1

Thank you for your responses. This forum is great! I am very new to the unix environment and just trying to figure things out..

So now that you have helped me with this problem, I realize that what I am really looking for is the solution to this - with the same result as above:

1. From the root directory (or within a script), search all directories below to find any file called Application.properties

2. When an Application.properties file is found, search it for "active=true"

result:
uat2000.aspclient.active=true
uat2001.aspclient.active=true
uat2002.aspclient.active=true
uat2003.aspclient.active=true

3. Count the lines and display only first 7 characters, ie uat2000, uat2001, etc :

uat2000
uat2001
uat2002
uat2003

Thanks A Million!

a variation on the 'awk' theme:

Code:

awk -F'[.=]' '$(NF-1)=="active" && $NF=="true" {print $1}' myFile.txt

reborg,
a slight modification to the shell alternative:

Code:

while IFS='.' read one two three; do
    [[ "$3" == "active=true" ]] && echo $one
done < inputfile

that way you don't have to save the IFS and reset it after the loop.

True, good point Vlad, it prevents fogetting to reset it, like I did.

..and you forgot your usual "On Solaris use nawk" for the awk alternative.