Problems found locating distfiles:
Package colorls: missing distfile ls.tar.gz
Package molden: missing distfile molden-4.6/molden4.6.tar.gz
Package softmaker-office-demo: missing distfile ofl06trial.tgz
Otherwise, existing SHA1 digests verified and found to be the same on
the machine holding the existing distfiles (morden). All existing
SHA1 digests retained for now as an audit trail.
* set LICENSE=gnu-gpl-v2
* add user-destdir support
* reset WRKSRC setting, use CONFIGURE_DIRS instead,
so that pkgsrc magick affect to all files
* regen patches, split out patch-ad from patch-ac
(one patch file to multiple files is not good).
* need termcap to build.
From hiramatu /at/ boreas _dot_ dti _dot_ ne _dot_ jp in PR#29890,
with a few modifications by me.
* ver 4.51 (Jan.16th,2004)
+ fixed keyboard-interrupt handling while reloading a file to
avoid segmentation fault. (pointed out by Masanori GOTO
<gotom@debian.or.jp>)
+ modified definitions of boolean_t, PAGE_SIZE, VDISCARD, and
autoconf for libcurses for AIX. (pointed out by SHIOTA
Shoichi <Shoichi.Shiota@lightwell.co.jp>)
+ omitted to specify owner and group to install command, and
omitted include langinfo.h when HAVE_LANGINFO_CODESET is not
defined, for Max OS X. (pointed out by Koki NAKASHIRO
<nakasiro@sdl.hitachi.co.jp>)
Change summary from the release note:
o Big5 to Unicode mapping didn't work at all (fixed offset of coding
system table)
(See http://lists.debian.or.jp/debian-devel/200311/msg00006.html)
o -D option now specifies default (fall-back) coding system, not default
EUC coding system. This means that -D coding system is used not only
for determination of EUC coding systems but also determination of other
coding systems.
o Added ISO-8859-10,11,13,14,15,16.
o Many bug fixes.
It compiles, but would not work due to unaligned access.
Solaris and Linux (and NetBSD/alpha) traps unaligned access exceptions
and fixes them up automatically but NetBSD/sparc64 does not.
This problem is already reported to the original author; I'd
wait for his fix. Until then use the 32bit binary instead.
http://www.ff.iij4u.or.jp/~nrt/lv/relnote.html
ver 4.49.4 (Aug.24th,2000)
fixed a bug of non-regular file loading (file.c).
added option -i to enable casefold search and +i to disable it.
added keybind 'w' to 'previous line'.
modified input routine to accept DEL as delete key. Thanks John.
modified console code to care putchar() as a macro. Thanks Kenji.
renamed some variables for clarity.
added JIS X 0213-2000 Plane 1,2 as independent charsets of JIS X 0208.
I heard that KS C 5601 has changed its name to KS X 1001.
Changes:
ver 4.49.3 (Nov.2nd,1999)
modified import.h to remove needless #ifndef.
added key bindings for PageUp/PageDown to move prev/next page, and
changed Left/Right bindings to moving prev/next half page.
added bzip2 support for *.bz2 files with bzcat(1).
modified ConfFilename() to refer environment variable MAN_PN and
show that instead of "(stdin)" when man(1) passes the variable to lv.
modified iso-8859-7/Unicode mapping table.
fixed a bug that standout sequences remaine when you search a
string before suspend/edit, and after that you immediately cancel
the search.
fixed a bug again, lv blocks stdin stream by itself, before the
first page cache is filled (once fixed in ver 4.2beta2).
ver 4.49.2 (Sep.14th,1999)
added run-time command '=' that acts equally as 'C-g'.
changed LV_FILES_MAX constant to LV_FILES_UNLIMITED for UNIX
version.
ver 4.49.1 (Jul.25th,1999)
Translation of utf-8 and utf-7 into iso-8859-* was mishandled. Since
iso-8859-* are ISO 2022 8bit system and output routine of that was
shared with iso-2022-jp inside lv, so Unicode was converted into JIS
Japanese codeset by mistake. Furthermore, some glyphs in iso-8859-2..9
that iso-8859-1 also contains were incorrectly mapped into iso-8859-1.
Then I changed following points:
Rebuilt iso-8859-* mapping table.
Separated iso-8859-* encoding routine from iso-2022-jp.
This bug report was posted by <elad@tsur1.dhis.org>.
pre ver 4.48.2
some bugfix.
