Edit build.lua for programs losing their private assert.h, so they
depend on a list of .h files excluding assert.h.
Remove modules/src/assert; it would be a dependency of cpp.ansi but we
didn't build it, so cpp.ansi uses the libc assert.
I hope that libc <assert.h> can better report failed assertions. Some
old "assert.h" files didn't report the expression. Some reported a
literal "x", because traditional C expanded the macro parameter x in
"x", but ANSI C89 doesn't expand macro parameters in string literals.
@dram reported a build failure in FreeBSD at
https://github.com/davidgiven/ack/issues/1#issuecomment-273668299
Linux manual for getopt(3) says:
> If the first character of optstring is '-', then each nonoption
> argv-element is handled as if it were the argument of an option with
> character code 1....
>
> The use of '+' and '-' in optstring is a GNU extension.
GNU/Linux and OpenBSD handle '-' in this special way, but FreeBSD
seems not to. If '-' is not special, then em_ego can't find its input
file, so the build must fail. This commit stops using '-' in both
em_b and em_ego, but doesn't change mcg.
Also fix em_ego -O3 to not act like -O4.
This prevents an overflow reported by @hexcoder- in
https://github.com/davidgiven/ack/issues/56
lang/cem/cpp.ansi/LLlex.c used a plain 1 << ... and caused an overflow
on machines where sizeof(int) < sizeof(long). Using 1L << ... would
work for now but might fail later if arith became long long.
C doesn't specify whether negative integers use 2's complement or some
other format. Therefore, (arith) 1 << ... has an undefined value. It
should still work because the value is some integer where the sign bit
is set and all other bits are clear.
(unsigned arith) 1 << ... would also get the sign bit, but casting it
from unsigned back to signed would make the same undefined value.
(arith) -1 << ... would assume 2's complement.
Reported by me in https://github.com/davidgiven/ack/issues/60
This doesn't change DIV. Right now a DIV b does floor division and
a MOD b has the sign of b. This is the same as Lua, Python, Ruby,
Tcl; but is different from other Modula-2 implementations.
Traditional C compilers had long but not unsigned long. I now assume
that everyone can compile unsigned long. Remove macro UNSIGNED_ARITH
and act like it is always defined. The type `unsigned arith` works
because arith is a macro for long.
If feof(fp) or ferror(fp) was set, then our libc returned EOF for all
later reads without trying to read. Our libc now behaves like BSD
(and probably Illumos and musl) by checking only feof(fp). For
difference, glibc doesn't check feof(fp).
I described the difference between our libc and BSD libc in
https://sourceforge.net/p/tack/mailman/message/35430300/
@hexcoder- reported in https://github.com/davidgiven/ack/issues/57
that our getpw() has bugs.
I don't fix these bugs, because Illumos and Linux manual pages say
that getpw() is obsolete. The function can overflow its buffer, so it
is never safe to use. Our libc did not build getpw().
This malloc.h might get confused with the private malloc.h in our
libc. C programs should #include <stdlib.h> for malloc().
This tgmath.h has no useful content, and never worked because
complex.h is missing.
Touch build.lua (by deleting some whitespace) so the *.h globs see
the deletions.
These functions are in POSIX; hypot() is in C99. Also build cabs()
because it rides with hypot(), but don't declare cabs() in any header
file, because our compiler can't parse C99 "double complex" type.
Touch build.lua so it sees that .c files moved.
This commit slightly improves the formatting of the manuals. My
OpenBSD machine uses mandoc(1) to format manuals. I check the manuals
with `mandoc -T lint` and fix most of the warnings. I also make
other changes where mandoc didn't warn me.
roff(7) says, "Each sentence should terminate at the end of an input
line," but we often forgot this rule. I insert some newlines after
sentences that had ended mid-line.
roff(7) also says that blank lines "are only permitted within literal
contexts." I delete blank lines. This removes some extra blank lines
from mandoc's output. If I do want a blank line in the output, I call
".sp 1" to make it in man(7). If I want a blank line in the source,
but not the output, I put a plain dot "." so roff ignores it.
Hyphens used for command-line options, like \-a, should be escaped by
a backslash. I insert a few missing backslashes.
mandoc warns if the date in .TH doesn't look like a date. Our manuals
had a missing date or the RCS keyword "$Revision$". Git doesn't
expand RCS keywords. I put in today's date, 2017-01-18.
Some manuals used tab characters in filled mode. That doesn't work.
I use .nf to turn off filled mode, or I use .IP in man(7) to make the
indentation without a tab character.
ack(1) defined a macro .SB but never used it, so I delete the
definition. I also remove a call to the missing macro .RF.
mandoc warns about empty paragraphs. I deleted them. mandoc also
warned about these macro pairs in anm(1):
.SM
.B text
The .SM did nothing because the .B text is on a different line. I
changed each pair to .SB for small bold text.
I make a few other small changes.
compiler flag, which is used to set the name of the patch table. The compiler
now understands C preprocessor line directives. Extend the standard library
somewhat.
standard library, because they never worked and come from an achingly old
version of the Pascal specification. Fix the implementations of New() and
Dispose() to use the standard C memory allocator rather than rolling their own
(also in C). Write test!
This undoes part of bfeb736, and returns to using DBL_MAX_EXP and
DBL_MIN_EXP from float.h.
Add a dependency on math/localmath.h and other local header files so
libc is rebuilt when those headers change.
This restores the correct values of DBL_MAX, DBL_MIN_EXP, and related
constants. This fixes some range checks within libc, causing
atof("-36e90") and atof("1.44e-288") to return the correct values.
