Shell Programming and Scripting

Code:

awk '/Start/{p=1;next}p && !int($0){exit}p' file

instead of

Code:

/^[0-9]/

try:

Code:

/[^0-9]/

Scrutinizer: Reading some more on awk, I thought the ^ inside the [ ] means that the script will look for a number at the beginning of the line, and the ^ outside the [ ] means it will exclude that. In any case, it unfortunately gave me the same results as before.

Franklin52: I think that worked! However, I realized that sometimes this output file may have more than one instance of such a sequence of numbers like this:

Code:

Start
1  10
2  20
 ...
 ...
<some text characters>

Start 
3 30
4 40
 ...
 ...
<some text characters>

Is there an easy way to tweak your command to let it find all such instances? The awk command without the extra code you put in did find it.

Thanks again to all of you for spending time to help me!

With a slightly modification of the code:

Code:

awk '/Start/{p=1;next}p && !int($0){p=0}p' file

Wonderful! Now I'm going to have to go read to figure out what you did!

I also realized something that might have made this easier. The list of numbers seems to always include one positive exponent. So it actually looks like:

Code:

Start
1  1.0E+05
2  2.0E+04
 ...
 ...
<some text characters>

Start 
3  3.0E+06
4  4.0E+02
 ...
 ...
<some text characters>

Knowing this, could we also use a simple search string to begin by searching for Start and then end when an exponent is no longer found? e.g.:

awk '/Start/,/^E+/' <filename>

I tried that but the tricky part seems to be forcing the search to look for E+ as exact text and not treat the + as a special character. I tried combinations of putting the E+ in quotes as well as with a backslash before the + but it didn't work.

Hi NickC it really is the other way around.

So it is because of the scientific notation that was not in your original sample that the other solutions did not work.

Well this should work as well then:

Code:

awk '/Start/ {flag=1;next} /^[a-zA-Z]/ {flag=0} flag {print}'  <filename>

or

Code:

awk '/Start/ {flag=1;next} /^[^0-9]/ {flag=0} flag {print}'  <filename>

or for a more cryptic one

Code:

awk '/^[^0-9]/{p=0};p;/Start/{p=1}' <filename>

Thanks Scrutinizer for the clarification!

I'm confused by your other code suggestions though. If I want awk to look for the next time that a scientific notation is NOT used, then how does /^[a-zA-Z]/ or /^[^0-9]/ work? Wouldn't they get confused as well whether it's looking for a word or the letter "E" in the scientific notation?

Could we not search for lines between /Start/ and the next line where E+ is no longer encountered? Or is that what your code does? I tried it but it showed all lines from Start until the end of the file.