[PATCH] color: use ui.formatted() to test TTYness, not sys.stdout.isatty()

[Matt Mackall]

On Wed, 2011-05-04 at 10:16 -0700, Danek Duvall wrote:
> Christian Ebert wrote:
> 
> > I tried -r, but it "feels" unreliable.
> 
> It's probably just fine when paging hg output, as it won't be doing
> anything odd that will throw off less's idea of where the position is.  But
> I agree; I'd much rather use -R.
> 
> > > The problem is that its heuristic for determining which sequences are for
> > > changing color are thrown off by the nsterm-16color and screen-256color
> > > terminal descriptions, where sgr0 is set to \E[m^O.  Less assumes by
> > > default that color sequences end in the character "m", which happens to be
> > > the case for xterm-256color, which presumably borrowed that from ansi
> > > (although the full escape sequence isn't the same).
> > > 
> > > You're supposed to be able to override this by setting LESSANSIENDCHARS
> > > appropriately, but I can't figure out how to make it work.  Setting it to
> > > "m^O" (where that's a literal control-O) doesn't seem to do anything, and
> > > putting simply "^O" there makes the color go away, too.
> > 
> > I haven't looked into how python curses works, but - as a shot in
> > the dark - this propagates nicely into less -R:
> > 
> > printf '\e[0m\e[34m%s\e[0m\e[0m\n%s\n' "blue" "no color"
> > 
> > So, in my cluelessness I wonder where the ^O comes from or why
> > it's there ...
> 
> Like I said, it comes from the terminfo entry for your terminal.  See the
> output of "infocmp $TERM", and it's right there, in sgr0.  It has nothing
> to do with python or (n)curses; it's a fundamental problem with the way
> that less tries to figure out which control sequences are related to color
> and which aren't.  You'll see exactly the same thing with any other program
> that emits terminfo-based color whose output you pipe through less.
> Unfortunately, I don't know of any program that does that -- colorized ls
> uses ansi color sequences, and all other programs I use that use color do
> it with terminfo, but none of them are usefully paged.

Grumble. Yes, the less docs say that -R only understands ANSI codes.

So, we can:

1) only support ANSI
2) force ANSI mode when using the pager
3) disable color when using the pager


Relatedly, less is generally not configured with -R out of the box,
which means you either need to turn color off or turn -R on to get a
working setup. I'd rather we had to turn color support on in pager and
documented it, eg:

 To get color support in pager, set:

 [pager]
 color = true
 pager = less -R  # handle ANSI color codes

-- 
Mathematics is the supreme nostalgia of our time.