USAGE : cdx paste [options...] [files...]
paste combines the first lines of each file, then the second lines, and so on. If you want to combine columns based on a matching key, you want join
| short | long | description |
|---|---|---|
| -e | –end=[Exact,Early,Late] | What to do mif files have different lengths. Default Exact. See below. |
| -l | –last | In Late mode, repeat last line of shorteer files. |
| -d | –default=ScopedValue | In Late mode, use these value as the defaults, rather than the empty string. |
| -D | –dups=[Fail,Allow,Numeric] | What to do if duplicate column names are found. Default is Fail. See below. |
| -r | –rename=Old.New,Old.New,… | If the name Old is repeated, replace it with New. Eah pair is used only once, and all must be used |
| -R | –rename-sloppy | It is not an error is some renames not used. |
If aaa.txt contains
CDX one two
aaa bbb
ccc ddd
and bbb.txt contains
CDX three four
111 222
333 444
The cdx paste aaa.txt bbb.txt would produce
CDX one two three four
aaa bbb 111 222
ccc ddd 333 444
The output file has a CDX header if and only if every input file has a CDX header.
By default, all files must be the same length (have the same number of rows) or it is an error.
With --end Early, output stops when any file reaches its end.
With --end Late, output continues until all files have reched their end. Zero-length column values are used for shorter files, unless --last or --default are used.
When combining files this way, it is easy to find yourself with multi columns having the same name, which is malformed. By default, paste will check for this and fail if it occurrs. If you use --dups Allow then messages will be written to stderr, but you will be allowed to create a malformed file. With --dups Numeric if the column name foo appears more than once, then the second time it will be called foo1, the third time foo2 and so on. If you specify --rename foo,bar then the second time foo appears, it will be changed to bar.