The way these are used have little to do with them being scalars;
they are just tuples with names, lengths and IS_* constants. We
will probably reuse these in the future with other types.
Start >= 0 already tells us whether or not the loop has a loop
variable, no need to add extra flags to opcodes.
Also added a test for a case where FREE_ON_RETURN is relevant,
we didn't seem to have any coverage for this.
This fixes a regression from 16a9bc1ec2.
Together with the recent closure related changes this should allow
all usages of self etc, while previously (in PHP 5) some things like
self::class did not work.
* Don't throw an error if self/parent/static are used in a closure
(outside a class).
* Don't propagate self:: constants into closures
* Use runtime fetch for self::class in closures
Fixes bug #66811.
we basically added a mechanism to store the token stream during parsing
and exposed the entire parser stack on the tokenizer extension through
an opt in flag: token_get_all($src, TOKEN_PARSE).
this change allows easy future language enhancements regarding context
aware parsing & scanning without further maintance on the tokenizer
extension while solves known inconsistencies "parseless" tokenizer
extension has when it handles `__halt_compiler()` presence.
Also fix a bug with return types where "self" was rejected inside
a class, but not on a method.
Fallout: A couple of tests changed to more generic error messages.
Introduce helper macro FC(x) for CG(file_context).x.
end_compilation() now handled by file_context_end().
While at it, dropped zval wrapper for ticcks.
Renamed compiler_context to oparray_context. Introduced per-file
file_context. Moved import tables into the file_context.
context_stack no longer exists, instead keeping backups of contexts
on C stack. Same for file contexts.
TODO: Move more things out of CG into file_context. There should be
a number of other things that we should not try to reuse in nested
compilations.
