Feature request: 'merge' (line 'cat' for rows, but with column interpolation)

[jeetsukumaran]

The final output would have the union of all the field names across all input files.
Fields without data (e.g., that column was missing from the original) could be left blank or a user-supplied value (e.g., "--nodata='NA'").

[BurntSushi]

Could you give a small example please? Just so I make sure I understand.

[jeetsukumaran]

data1.csv:

f1,f2,f3,f4
 0, 1, 2, 3
 4, 5, 6, 7

data2.csv:

f1,f4,f5,f6,f7
 a, b, c, d, e
 f, g, h, i, j

Output of xsv cat rows data1.csv data2.csv --nodata="NA":

f1,f2,f3,f4,f5,f6,f7
 0, 1, 2, 3,NA,NA,NA
 4, 5, 6, 7,NA,NA,NA
 a,NA,NA, b, c, d, e
 f,NA,NA, g, h, i, j
~~

[jeetsukumaran]

Maybe "missing data" or "missing fields" would be a more precise term. The flag could be "--interpolate-missing" or something like that.

[danielecook]

4 years late to the party but...

@jeetsukumaran I have written a utility that does this:

https://github.com/danielecook/tut

tut stack data1.csv data2.csv

It's called stack, and I use it to merge all kinds of output files with heterogeneous columns... I also added an option to output the filename with --add-filename (full path) or basenames --add-basename of the files being merged. This makes it super easy to merge together a collection of related files for analysis.

@BurntSushi - it would be really cool if you were able to add this as I'm sure the xsv implementation would be much faster.

[geekscrapy]

Echo'ing @danielecook 's last comment: It'd be amazing if this was incorporated as a subcommand 👍 Currently using csvstack - it works, but I think xsv would be faster!

[data-man]

Maybe tsv-utils can help (written in D).

[ad-si]

I was pretty surprised to find out that this is not the default behavior 😳. Addition of this would be highly appreciated 😊

[alexmarco]

Maybe tsv-utils can help (written in D).

HI, don't exists option in tsv-utils for stack and align input data.
+1 for this option in xsv

[BurntSushi]

@alexmarco Please don't post comments just to +1 a feature requests.