[SOLVED] paste -n functionality

mikudo · 09-17-2018, 06:46 AM

I am trying to merge two temp files, one with 2 columns and one with 1 column so that I get a file with 3 columns.

However, some of the columns in the second file are missing because a command is timing out.

I found an answer that I could use paste -n but this option is not in man pages in my system.

What are my options? How else could I do this?

edit: to be clear, i want to print the row with one of the columns blank not erase the row

1 2 3
3 4 5
6 7
8 9 10
11 12 13 etc

l0f4r0 · 09-17-2018, 07:03 AM

paste seems to be a good solution.
What's the problem exactly with using:

Code:

paste file1 file2

?

Quote:

Originally Posted by mikudo

to be clear, i want to print the row with one of the columns blank not erase the row

paste does not erase the row should a column be blank...

PS: I'm not aware of a "n" option

l0f4r0 · 09-17-2018, 07:05 AM

mikudo · 09-17-2018, 07:06 AM

What I am getting though from paste f1 f2

is disjointing

1 2 3
4 5 6
7 8 10
11 12 13

etc, it is so very close to being correct. i read elsewhere that paste -n would give me the skipped column rather than the shift but this is not in all version of paste. im reading on column, join, table, and just looking for what will do it.

mikudo · 09-17-2018, 07:14 AM

Link to OP then?

l0f4r0 · 09-17-2018, 07:18 AM

Quote:

Originally Posted by mikudo

What I am getting though from paste f1 f2 is disjointing

Can you define "disjointing" please?
Sorry but I don't manage in understanding your situation with the examples provided. It's not quite easy to get it just by the expected result. Could you try one of the following:

provide a more meaningful example with extracted actual content (if not confidential)?
provide the first 5 lines of 2 bogus files to "paste"?

mikudo · 09-17-2018, 07:27 AM

Sure.

first column, number of counts from uniq -c

second column, string

third column, result of command run on string

file1.txt
5 124124
20 93945
14 19009
10 90413

file 2 (some executed commands return empty row because operation times out and no way around this, in this example C is missing)
A
B
D

I need as output:
5 124124 A
20 93945 B
14 19009 blank (or string indicating command failed)
10 90413 D

paste -n is suppose to give me the blank space but this isn't in all shells.

l0f4r0 · 09-17-2018, 07:39 AM

Okay, it's clearer now.
Actually, your main issue is that the timeout doesn't return an "empty row" as you write it but no row at all. So I cannot imagine how shell could know that a line is missing so it goes on with the next one.
Either there is some logic between lines from file2 so you can indicate to shell when an expected line is missing OR you should try to write an empty row (or whatever default text you wish) in case of timeout. This done, you could paste your files as desired.

hydrurga · 09-17-2018, 07:40 AM

Quote:

Originally Posted by mikudo

Sure.

first column, number of counts from uniq -c

second column, string

third column, result of command run on string

file1.txt
5 124124
20 93945
14 19009
10 90413

file 2 (some executed commands return empty row because operation times out and no way around this, in this example C is missing)
A
B
D

I need as output:
5 124124 A
20 93945 B
14 19009 blank (or string indicating command failed)
10 90413 D

paste -n is suppose to give me the blank space but this isn't in all shells.

How is any join technique going to know, in this example, whether the third row in file 2 is in fact the third row (with the fourth being missed out) or the fourth row (with the third being missed out)?

P.S. Can you please use CODE tags (see https://www.linuxquestions.org/quest...og.php?b=36313) for your output.

mikudo · 09-17-2018, 07:46 AM

This is helpful thank you. I will use code tags in the future.

I will have to, ugh, put in more logic lol to make sure that there is something in the row to match the other rows.

Gosh this simple script got complicated fast. I guess that is pretty normal that dealing with something like this will double the size and complexity of the script.

rtmistler · 09-17-2018, 07:54 AM

People are correct, the paste command will not intelligently determine what lines to skip from file2. You'll have to figure out a way to match the file2 lines with file1 if you wish to use paste. The command description states this:

Quote:

Write lines consisting of the sequentially corresponding lines from each file

The key term is "sequentially corresponding lines"

EDIT: Sounds like you already got that point from the earlier posts.

hydrurga · 09-17-2018, 07:55 AM

Quote:

Originally Posted by mikudo

This is helpful thank you. I will use code tags in the future.

I will have to, ugh, put in more logic lol to make sure that there is something in the row to match the other rows.

Gosh this simple script got complicated fast. I guess that is pretty normal that dealing with something like this will double the size and complexity of the script.

If file2 were to have a blank line (simply a LF) for each failed command rather than no line at all then that could work.