curses.buildlink2.mk. This was wrong because we _really_ do want to
express that we want _n_curses when we include the buildlink2.mk file.
We should have a better way to say that the NetBSD curses doesn't
quite work well enough. In fact, it's far better to depend on ncurses
by default, and exceptionally note when it's okay to use NetBSD curses
for specific packages. We will look into this again in the future.
Changes since 1.4.8
1. src/slarray.c: superfluous call to SLclass_add_math_op removed
(Michael Noble)
2. src/slang.c: foreach (NULL) using("next"){} foo (); caused _NARGS=1
in foo.
3. src/slarrfunc.c: Fix to prevent sum(NULL) from causing a core-dump.
4. src/slimport.c: import (module, "") made equivalent to
import(module,"Global"); This way, import(module, current_namespace())
will work when the current namespace is anonymous.
5. src/slospath.c: Several users have requested that I add the ability
to define a load path and use that path when loading interpreter
files. To this end, several new functions were added to the API:
char *SLpath_get_load_path (void);
int SLpath_set_load_path (char *path);
/* Get and Set the path to be searched for files */
int SLpath_get_path_delimiter (void);
SLpath_set_path_delimiter (int delimiter);
/* Get and set the character delimiter for search paths */
int SLang_load_file_verbose (int verbose);
/* if non-zero, display file loading messages */
New intrinsics include:
set_slang_load_path
get_slang_load_path
path_get_delimiter
These functions, nor the intrinsics have an effect on applications
that use SLang_load_file_hook or SLns_load_file_hook for loading
files. The change should be transparant to applications that use
the stock load file mechanism. The main difference is that if one
attempts to load a file with no extension, e.g., "foo", but the
file does not exist, then the interpreter will try to load the more
recent of "foo.sl" and "foo.slc".
See src/slsh.c for how the functions may be used.
6. slsh/slsh.c: Updated to use the new search path code outlined
above. Also, slsh is distributed with a collection of general
purpose slang functions, including jed's provide/require functions.
See slsh/README for more information.
7. doc/tm/cslang.tm: Modified the section describing the implemetation
of intrinsic functions in an effort to clarify the discussion.
8. src/slang.c: tiny memory leak resulting from peephole optimzations
added earlier found and fixed.
9. src/slarrmisc.c: new intrinsic: cumsum computes the cumulative sum
of an array via the new SLarray_map_array function.
The automatic truncation in gensolpkg doesn't work for packages which
have the same package name for the first 5-6 chars.
e.g. amanda-server and amanda-client would be named amanda and amanda.
Now, we add a SVR4_PKGNAME and use amacl for amanda-client and amase for
amanda-server.
All svr4 packages also have a vendor tag, so we have to reserve some chars
for this tag, which is normaly 3 or 4 chars. Thats why we can only use 6
or 5 chars for SVR4_PKGNAME. I used 5 for all the packages, to give the
vendor tag enough room.
All p5-* packages and a few other packages have now a SVR4_PKGNAME.
BUILDLINK_PREFIX.<pkgname>. This allows buildlink to find X11BASE packages
regardless of whether they were installed before or after xpkgwedge was
installed. Idea by Alistair Crooks <agc@pkgsrc.org>.
Fixed a bug that shows up on 64 bit BigEndian machines--- it
affected no others.
Fixed potential problem in pre-parsing binary strings.
Bug a fixed affecting only pure termcap-based systems. It has
been around a while, I am surprised that it took so long to be
discovered.
on usenet:
Version 1.4.3 is primarily a bug-fix release, and will probably be the
last release in the 1.x series. It is time to move on to v2.x.
In fact, v1.4.3 adds DESTDIR support, and extends the ABI. The change log
given in the announcement is too long and detailed to included here.
Please refer to <URL:news:slrn97ou3d.87i.davis@aluche.mit.edu>, available
at your nearest usenet archive, for details.
out of date - it was based on a.out OBJECT_FMT, and added entries in the
generated PLISTs to reflect the symlinks that ELF packages uses. It also
tried to be clever, and removed and recreated any symbolic links that were
created, which has resulted in some fun, especially with packages which
use dlopen(3) to load modules. Some recent changes to our ld.so to bring
it more into line with other Operating Systems also exposed some cracks.
+ Modify bsd.pkg.mk and its shared object handling, so that PLISTs now contain
the ELF symlinks.
+ Don't mess about with file system entries when handling shared objects in
bsd.pkg.mk, since it's likely that libtool and the BSD *.mk processing will
have got it right, and have a much better idea than we do.
+ Modify PLISTs to contain "ELF symlinks"
+ On a.out platforms, delete any "ELF symlinks" from the generated PLISTs
+ On ELF platforms, no extra processing needs to be done in bsd.pkg.mk
+ Modify print-PLIST target in bsd.pkg.mk to add dummy symlink entries on
a.out platforms
+ Update the documentation in Packages.txt
With many thanks to Thomas Klausner for keeping me honest with this.
slang.c: Under certain conditions, the continue statement was not
properly handled in do..while statements. src/test/loops.sl added
for testing.
slparse.c: avoid potential (rare?) infinite loop when slang error occurs
slsmg.c: When SLsmg_init_smg is called, mark the display as trashed.
It is now possible to add intrinsics to their own namespace via
new SLns_add* functions. Moreover, the import function now takes
an optional additional argument that specifies a namespace.
New namespace intrinsics: use_namespace, current_namespace
Changed inner-product algorithm to minimize the number of cache misses.
sldisply.c: Kanji specific patch
sldisply.c: Assume that Eterm and rxvt are xterm-like
sldostty.c: mouse support added
slsearch.c: avoid infinite loop if search string has no length.
slregexp.c: regexp \d+ was not working properly
keyhash.c: typos involving USER_BLOCK keywords corrected.
New intrinsic variable: _slang_doc_dir. This specifies the
installation location of the doc files.
Make sure it can compile with SLTT_HAS_NON_BCE_SUPPORT set to 0.
Add a new USE_LIBTOOL definition that uses the libtool package instead of
pkglibtool which is now considered outdated.
USE_PKGLIBTOOL is available for backwards compatibility with old packages
but is deprecated for new packages.
New intrinsic function: strreplace.
SLang_roll_stack and SLang_reverse_stack functions made public.
SLang_free_function added.
src/slrline.c: Keybindings for ESC O A, etc added.
src/slsmg.c: SLsmg_write_nstring: avoid many loops if an extremely
large value is passed (> 0x7FFFFFFF).
src/slregexp.c made thread safe
src/slsmg.c: Cursor was not always properly positioned when
after SLsmg_touch_lines called.
If terminal does not have erase to eol capability, then use spaces.
doc/tm/strops.sl: doc for strcat updated to reflect its ability to
concatenate N strings.
src/slsmg.c: bug involving SLsmg_set_screen_start fixed.
src/slparse.c: parser was failing to catch misplaced `}'.
Changes since 1.3.10
1. If a floating point exception occurs and the OS allows the library
to handle it without forcing a longjmp, then SL_FLOATING_EXCEPTION
will get generated instead of SL_INTRINSIC_ERROR. Note: Linux
provides no way to handle floating point exceptions without
forcing a longjmp. In my opinion, this is a flaw.
2. SLang_pop_double was returning the wrong value for short and
character types.
3. New intrinsic: is_struct_type(X) ==> non-zero is X is a struct.
4. typecast operation from user defined type to Struct_Type added.
5. slkeypad.c: DOS/Windows DELETE_KEY definition added (Doug Kaufman
<dkaufman@rahul.net>)
6. slposdir.c: Do not depend upon the existence of rmdir on VMS
systems.
7. slang.c: abs, sign, mul2, chs, sqr were not being treated as
function calls.
8. sldisply.c:SLtt_cls: If the terminal is a color terminal but
being used as a black and white terminal, then reset colors before
clearing.
9. path_sans_extname intrinsic added.
10. slimport.c: If module defines deinit_NAME, will be be called prior
to unloading the module. (Ulrich Dessauer <des@gmx.de>)
