TODO (high level)

TODO (low level)

tree: add more tests
lstree: add tests
pstree: add tests
printf: add more tests
sum,count,min,max,avg: --group-by=col1,col2
uniq: --count
show: fix column width calc when input is in utf
source-tools: add option to add and remove columns
show: ability to lock columns (make them always visible)
show: implement "search" and "help" in the ncurses backend
sql/rpn: data(colname, relativerownum)
sql: implement case, cast
csv-from-txt --format "year,-,month,-,day,hour,:,min,:,sec,,,ms, ,level, ,file, ,message"
netstat: support more protocols
netstat: translate interface number to interface name
netstat: figure out how to print inet_diag_info
sql: add support for tables
sql: implement "group by", "having"
cut: multiple input files
ps: threads?
ps: figure out what to do with these columns: wchan, tty, tpgid, exit_signal, *signals*, alarm, kstk*, flags
grep: search on all columns
grep: -e by default

arbitrary precision arithmetic (with libgmp?)
rewrite csv-lstree and csv-pstree in C so that command line switches work correctly (e.g. -s/-S anywhere)
importing from other tools?
loops and temporary variables in rpn?
one multicommand binary? (csv ls, csv grep, ...)
binary data?
journalctl?
netstat: add pid
(auto?) input decompression/output compression
expand array support? deal with hierarchical data somehow
- ps: supgid_names, cmdline, environ, cgroup, what about fds/fdinfo?
- group-members: merge it with groups/users?
tool for generating dot scripts? (or other graphviz tools)
tool for transposing data?
tool for splitting files?
csv-add-printf?
sqlite: loading files into tables without csv-merge