On 29Apr2016 03:06, Chris Angelico <ros...@gmail.com> wrote:
On Fri, Apr 29, 2016 at 2:33 AM, Steven D'Aprano <st...@pearwood.info> wrote:
I have an application written in Python which accepts -h or --help to show
help. I can:
(1) print the help text to stdout;
(2) run the help text through a pager;
(3) do something else?
Many command line tools simply output help to stdout (or stderr, if they're
evil),
I'm not sure stderr is particularly more evil than stdout, but whatever :)
I'm sure. If I _ask_ for the help information, it is not an error message.
Instead it is desired output. The logic is clear.
If I screw up and elicit an error and the usage message (which might be the
help text), that is an error message. But -h/--help? That is standard output,
even if it is the same usage text.
And on this topic, let me preemptively flame commands whose entire usage
message is "use -h/--help for usage". More evil.
What you could do is run the help text through a pager only if stdout
is a TTY. That allows it to be grepped conveniently, but also browsed
easily. But I wouldn't worry about that unless you have at least 3-5
pages of screed - for most programs, just dumping the text to a
standard stream is usually enough.
+1 to this. I'm -0 on conditionally offering a pager based on isatty; Marko
Rauhamaa has illustrated one downside, and personally I find _unexpectedly_
paged output to be jarring, requiring me to modeshift my brain.
So I'm for keeping it simple and just sending it to stdout.
Cheers,
Cameron Simpson <c...@zip.com.au>
--
https://mail.python.org/mailman/listinfo/python-list
