command was trying to parse their options and failing. (Commands
that aren't just disabled but have no option string even in the allyesconfig
case need an OPTSTR of 0.)
This is probably only a partial fix, there's some design work to think through
here.
a patch from Elliott Hughes, who said:
[PATCH] Add support for libcrypto for MD5/SHA.
Orders of magnitude faster (for architectures where OpenSSL/BoringSSL
has optimized assembler).
Also adds sha224sum, sha256sum, sha384sum, and sha512sum for folks
building with libcrypto.
The fallback portable C implementations could easily be refactored
to be API-compatible, but I don't know whether they'd stay here or
move to lib/ so I've left that part alone for now.
In main.c: used adjusted string with placeholers for removed options (so flag
values don't move based on config, allowing FORCE_FLAGS to work).
In scripts/mkflags.c: test was wrong, a bare longopt never matches a single
letter command, but it was treated as always matching when comparing
allyesconfig vs current config to determine which options were disabled, so
there was a corner case that got the flag values wrong.
Keep the low 32 bits of FLAG_x constants as 32 bit numbers so that at least
on little endian platforms it's still normal 32 bit math outside of lib/args.c.
This means the flag space is no longer packed, but leaves gaps where the zeroes
go. (Actual flag bit positions are the same for all configs.) Since the
option parsing needs to know where the holes are, the OPTSTR values are
now generated as part of flags.h with ascii 1 values for the disabled values.
(So generated/oldflags.h went away.)
This also means that the option string argument for OLDTOY() went away, it now
uses the same arguments as the NEWTOY() it references.
Only the first __long__ option, without any __short__ option had a proper
flag value, rest all were defined to ZERO. Becaus the _flist_ was not moved
to the next in this case.
I am of the opinion that __-__ be converted to '_' for generating the FLAG_xxx macros and be used in the command.
This will enable the user to 'abc-def', but be handled like 'abc_def' in code.
Don't #undefine TT in the CLEANUP blocks of generated/flags.h, and #ifdef
around the other TT definition. That way you can put a union at the start
of your GLOBALS() with the arguments filled out by option parsing, and then
have multiple main() functions with different argumetns and different
FLAG_x macros, while sharing infrastructure that's not under lib.
This should actually generate FLAG_longopt 0 #defines for disabled bare
longopts (ala ls without --color). Put temporary executables under "generated"
(including instlist for install).
Rob Landley
Elliott Hughes
Isaac Dunham
Hyejin Kim
Ashwini Sharma