Code kata to print a text table from input data.
The table.in file contains empty-line separated groups. The first such group
are the column headers. Subsequent groups are the column data.
The table.out file contains the desired program output.
Write a program table that satisfies this shell command:
$ ./table < table.in | diff - table.out
This command should generate no output, meaning the output of the program is
identical to table.out.
Can you separate the input, processing, and output phases of the program, so that there are clean boundaries between them?
Do something reasonable for not-nice inputs:
- Empty input.
- Extra blank lines in arbitrary places.
- Missing elements from rows.
- Extra elements in rows.
Things you can do:
- Print appropriate messages to stderr.
- Try to render the table as best you can.