don't swallow non-ASCII control characters, since they're parts of UTF-8
sequences
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2378 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
since we can't dynamically set that element of the array to its UTF-8
equivalent when in UTF-8 mode, we have to use the ISO-8859-1 version and
pass every string in the credits through make_mbstring() to make sure
they're all UTF-8 (sigh)
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2373 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
sync with their edit window counterparts, and add a few cosmetic fixes
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2340 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
into an infinite loop if the terminal it's running in dies unexpectedly
due to a signal
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2339 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
check_operating_dir() to make it use boolean values (plus a few more
boolean-related tweaks of mine); also stop wrapping the bad_chr
parameter of parse_mbchar() in #ifdefs for cleanliness, as the
allow_tabcomp parameter of check_operating_dir() isn't wrapped in
#ifdefs, and don't set bad_chr to TRUE when we get a null byte in a
multibyte string, as it's not an invalid multibyte character
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2316 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
increase efficiency and support multibyte characters; also add a few
miscellaneous tweaks of mine
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2309 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
home" routines to use it (since it returns the number of bytes taken up
by spacing characters at the beginning of the line), and rearrange a few
initializations near asserts
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2307 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
size_t, store the number of multibyte characters instead of the number
of single-byte characters in totsize, and add a few formatting fixes
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2302 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
toggle and a "morespace" rcfile option; when these are used, the
normally-unused blank line below the titlebar will be treated as part of
the edit window
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2281 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
efficient to just add the return value of parse_char() directly instead
of having move_mbright() return it (except in the cursor movement
functions, which should be kept simple)
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2271 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
work with multibyte strings, and tweak the "smart home" routines to use
move_right() for consistency
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2269 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
one page of a line and their end on a different page of the same line
would paint past the end of the regex by tweaking DB's changes to add
multibyte string support to edit_add(); now calculate paintlen the same
way the original working code did, but keep accounting for multibyte
characters as DB'S new code did
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2253 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
characters and strings instead of wide characters and strings as much as
possible, and move multibyte/wide character-specific functions into
their own source file, chars.c
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2248 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
indicate when we run a normal shortcut's associated function with the
ran_func parameter, reset the statusbar cursor position when we do, and
don't call print_view_warning() when we try to run a function not
allowed in view mode; also apply the first of these changes to
do_input() so that both the edit window and statusbar routines are in
sync again
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2246 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
prone to off-by-one errors that lead to memory corruption and I can't
find anything obviously wrong with it; always allocate enough space for
len columns of text now; any extra space that isn't used will be removed
by the null_at() call at the end of display_string() anyway
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2245 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
characters, found by valgrind (and caused by a bad merge of part of DB's
code)
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2234 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
since we can't add half a two-column character to converted in
display_string(), and we need the other to keep the "$" from spilling
over to the next line if the line contains two-column characters in the
right places)
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2232 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
two-column characters at the beginnings and ends of long lines (i.e,
lines with $'s added), as the problems don't occur with ncurses 5.4, and
there's no easy way to fix the same type of problem with doing a regex
replace of "^" on a line beginning with a two-column character
git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@2231 35c25a1d-7b9e-4130-9fde-d3aeb78583b8
David Lawrence Ramsey