[Tfug] OOCalc opens OOWriter

[t takahashi]

n.b., fwiw (ymmv) (and you didn't ask):

csv is not a standard.  even with commas, there are 5 or 6 syntax
issues, such as quoting newlines, that are not only rarely specified,
but often swept under the carpet.  even 2 different versions of an ms
product can parse csv differently.

"dsv" is sometimes used for delimiter-separated values.

tabs are probably the most common in unix ad-hoc pipelines, since they
are understood by default by awk, aren't usually in pathnames,
sometimes line up nicely, and can be specified by programs that
require a single character delimiter and can't use a regexp.  that
doesn't mean they are a good idea, just that they are often useful.

to convert, use sed or awk.  csv2 is sometimes useful.  a few programs
can understand more than one format.  to sort and subset and group,
consider sqlite and uniq and sort.  afaik there is no converter to
rule them all and bind them.

there are many formats for tabular data that are human-readable or
not, and many have lots of programs.  these include stuff like stats
and db formats, netcdf (seems to be n-dimensional array actually) type
programs, xml type programs, xml-to-sexp (partial sanity, if it
works), and so on.  but no generic converter.

note that we have been talking about tuples.  tuples can also be
expressed as tranches or stanzas or rfc-822-style messages, a la
debian package file format and the like.  again, no generic converter
that i know of.

just as find(1) understands lots of different data formats and is
well-maintained, it would be convenient to have a tableconvert(1).

it's all a kludge.  99% of software is kludgey.

--
Webmaster: do you believe that people will (a) switch browsers to view
your "best viewed with" page or (b) go to your competitor?