If your file structure is always groups of three lines, then you could
create a tab delimited file like so:
<< 'eof' paste - - -
line 1
line 2
line 1
line 2
line 1
line 2
eof
If the file structure is not that consistent, awk works well:
<< 'eof' awk -v FS='\n' -v RS='' '{print $1 "\t" $2 }'
line 1
line 2
line 1
line 2
line 1
line 2
eof
Regards,
- Robert
On Thu, Jan 11, 2024 at 5:17 PM Robert Citek <[email protected]> wrote:
> Have a look at the IFS and RS variables in awk. IIRC, you can specify the
> RS as '\n\n' and the IFS as '\n'. So printing the first line in a record
> becomes ' { print $1 } ' once those variables are set.
>
> ChatGPT came up with a similar answer.
>
> https://chat.openai.com/share/8b993033-af63-4954-a715-2e5de5291232
>
>
> I’d try them , but I’m mobile at the moment.
>
> Regards,
> - Robert
>
>
> On Thu, Jan 11, 2024 at 8:29 AM Rich Shepard <[email protected]>
> wrote:
>
>> I have a file with two-line paragraphs and want to extract only the first
>> line of each paragraph to an output file. Sample input file:
>> line 1
>> line 2
>>
>> line 1
>> line 2
>>
>> line 1
>> line 2
>>
>> I thought the awk 'next' statement would do this but my attempts failed.
>> For example:
>> $ gawk '{ print $0, next, next }' testfile.txt > out.txt
>> gawk: cmd. line:1: { print $0, next, next }
>> gawk: cmd. line:1: ^ syntax error
>> gawk: cmd. line:1: { print $0, next, next }
>> gawk: cmd. line:1: ^ syntax error
>>
>> and
>>
>> $ gawk '{ print $0, \n, \n }' testfile.txt > out.txt
>> gawk: cmd. line:1: { print $0, \n, \n }
>> gawk: cmd. line:1: ^ backslash not last character on line
>> gawk: cmd. line:1: { print $0, \n, \n }
>> gawk: cmd. line:1: ^ syntax error
>>
>> I didn't find an example in my SED AWK book or my web searches. I'm sure
>> there must be a way to do this and I'd like to learn how.
>>
>> TIA,
>>
>> Rich
>>
>
