Joining Text Files

[David Robertson_1]

How do I join text files with NO newline between them.
I've tried using 'cat file1 file2 file3' but this adds newlines.

Dave Robertson

[Andreas Voss · Accepted solution]

Hi,

this might be not the best solution but it works:

echo $(cat file1)$(cat file2) > file

OR if not ksh/posix-sh

echo `cat file1``cat file2` > file

Regards

[Lasse Knudsen]

Hi there,

'cat' does not add newlines - my guess is that your NLs are already in the files.

[Ralph Grothe]

I have the sneaking suspicion that the presented solution might fail on utterly large files, based on what echo presumably can swallow.
If I were sure about the redundant n in the last line of each file I'd simply send it through the stream editor (awk, perl, or whatever), like this
$ sed '$d' file1 file2 ... fileN > fileCatted
If unsure about the trailing n better use a regex to filter.
TIMTOWTDI as they would say in Perl-lingo.
But my feeling may be totally errant because I'm relatively new to Unix.

[Ralph Grothe]

Ooops, weren't aware of the disappearing of backslashes in the forum.
Of course, substitute \n for n

[Ralph Grothe]

Ooops, weren't aware of the disappearing of backslashes in the forum.
Of course, substitute \n for n

[Ralph Grothe]

Ooops, weren't aware of the disappearing of backslashes in the forum.
Of course, substitute \n for n

[Ralph Grothe]

Ooops, weren't aware of the disappearing of backslashes in the forum.
Of course, substitute \n for n

[James R. Ferguson]

David:

See man pages for 'join'. Is this what you seek?

...JRF...

[Ralph Grothe]

I have to apologize for my silly postings.
First, the annoying repetitive last posting was due to a webserver failure.
Second, one should first try things before giving advice. ;-)
Third, I think I misunderstood the issue.
If it was for chopping the trailing \n of each line (what the echo from above is doing on my machine) then you could use this Perl oneliner (even the obsolete Perl 4 shipped with HP-UX should do the trick)

$ perl -ne 'chop;print' file1 file2 ... fileN

But of course, this is also easily done using sed or awk.