fix https://github.com/pkgconf/pkgconf/issues/291
As defined in the C standard:
In all cases the argument is an int, the value of which shall
be representable as an unsigned char or shall equal the value
of the macro EOF. If the argument has any other value, the
behavior is undefined.
This is because they're designed to work with the int values returned
by getc or fgetc; they need extra work to handle a char value.
If EOF is -1 (as it almost always is), with 8-bit bytes, the allowed
inputs to the ctype(3) functions are:
{-1, 0, 1, 2, 3, ..., 255}.
However, on platforms where char is signed, such as x86 with the
usual ABI, code like
char *ptr = ...;
... isspace(*ptr) ...
may pass in values in the range:
{-128, -127, -126, ..., -2, -1, 0, 1, ..., 127}.
This has two problems:
1. Inputs in the set {-128, -127, -126, ..., -2} are forbidden.
2. The non-EOF byte 0xff is conflated with the value EOF = -1, so
even though the input is not forbidden, it may give the wrong
answer.
Casting char to unsigned int first before passing the result to
ctype(3) doesn't help: inputs like -128 are unchanged by this cast,
because (on a two's-complement machine with 32-bit int and unsigned
int), converting the signed char with integer value -128 to unsigned
int gives integer value 2^32 - 128 = 0xffffff80, which is out of
range, and which is converted in int back to -128, which is also out
of range.
It is necessary to cast char inputs to unsigned char first; you can
then cast to unsigned int if you like but there's no need because the
functions will always convert the argument to int by definition. So
the above fragment needs to be:
char *ptr = ...;
... isspace((unsigned char)*ptr) ...
This patch changes unsigned int casts to unsigned char casts, and
adds unsigned char casts where they are missing.
GCC has a lovely bug (which I will report as soon as I have an account),
which causes -Wmisleading-indentation to miss cases of misleading
indentation after a `;;`, since the macro adds `;`, and in call cases
the caller also adds `;`, we end up with a double macro and gcc fails to
warn.
cache functions are the hottest part of the pkgconf code when
profiled, by removing the linked list for lookups, we can turn
lookups into an O(k) operation
in almost all cases, we partially solve the dependency graph multiple times, which
just wastes resources. if we record the solution to a given dependency node, further
iterations can make use of the previous solution without having to solve it again.
this is safe because all provides entries (including virtuals) are knowable prior to
solving the dependency graph the first time.
a nice side effect of this is that all packages are preloaded when querying
information about them (--cflags and related commands).