[COMMIT] Be better about searching for chars typed via XIM and x-compose.el, isearch

[Aidan Kehoe]

Good evening!

Myself and Mats accidentally went off-list with this discussion. In summary
of what we said there:

-- Mats didn’t see any immediate benefit with my patch. He saw the problems
he described earlier in isearch in Gnus Summary buffers and in isearches in
dired, and a distinct set of problems in isearches in Gnus Article buffers.

-- I managed to reproduce his dired and Gnus summary problem when I switched
to a German layout (rather than my modified Irish layout), where ä and ö
have keys to themselves). This can be avoided with a nil value for
try-alternate-layout-for-commands, and I now think that only language
environments with non-Latin scripts should have a non-nil default for this
variable.

-- Mats still has the distinct set of problems in isearches in Gnus Article
buffers, which I have yet to reproduce. (Am renewing my news.individual.net
subscription to that end.)

Mats is curious about the need for try-alternate-layouts-for-commands, so
here’s an attempt to explain what it does; for people in Russia or Greece,
the keyboard by default doesn’t generate Roman-alphabet characters. So it’s
actually comparatively a lot of work to type control f or control b. Not
*so* much of an issue for GUI programs, but much more of an issue for
XEmacs. If try-alternate-layouts-for-commands is non-nil, and command lookup
is about to fail (that is, a message like “C-ь C-ф not defined” would
normally be shown), XEmacs pretends for a second that the keyboard has a US
layout, works out that C-ь C-ф would correspond to C-x C-f on that, and
calls #'find-file interactively. It comes into play in isearch because of
the rewriting of commands that #'isearch-pre-command-hook does in order to
add characters that would normally self-insert to the search string.

Best,

Aidan

 Ar an dara lá déag de mí Márta, scríobh Aidan Kehoe: 

 > [...]

-- 
“Apart from the nine-banded armadillo, man is the only natural host of
Mycobacterium leprae, although it can be grown in the footpads of mice.”
  -- Kumar & Clark, Clinical Medicine, summarising improbable leprosy research