When an empty nkArgList `varargs[untyped]` is passed around it is now
reused for efficiency sake and to prevent the introduction of a spurious
element: before this commit we'd pass the caller a
nkArgList[nkHiddenStdConv[nkBracket]] node instead of just an empty
nkArgList.
Fixes#8706
* Added simple interpreting event parser to pegs module.
* Has side-effects problem.
* Macro solution works.
* First flat callback test works.
* Fixed namespace pollution.
* Added handler for pkChar.
* Replaced event parser test.
* Started extensive docs.
* 'callback' to 'handler' renaming part 1.
* Renaming 'callback' to 'handler' part2, completed comments.
* Fixed exported API pollution.
* Added more event handler hooks, fixed comments.
* Changed event parser addition entry.
* Fixed variable names and comments.
* Enhanced comment.
* Leave handlers are not called for an unsuccessful match.
* The three varieties of back-reference matches are processed in separate
of-clauses now.
* Improved hygiene and (almost) eliminated exports.
* Trying to fix CI test breakage by eliminating export.
* Trying to fix CI test breakage by eliminating exports.
* Re-activated leave handler code execution for unsuccessful matches.
* Eliminated the last export statement (with a funny smelling hack).
* Make sure leave handler code is executed for all unsuccessful matcher
cases.
* Replaced local unicode.`==` with export.
* Don't consider tyAnd/tyNot/tyOr/tyAnything as generic
`containsGenericType` was too shallow and didn't check all the branches.
The resulting half-processed nodes are often simplified by the constant
folding pass but when that's not possible we get a nasty error during
codegen.
Fixes#8693
* Move the blame onto the semFold pass
Slightly better evaluation of `is` forms.
This is a temporary fix that will be reworked in a follow up commit
that aims to eliminate the tfExplicit flag from the compiler. The
complete and proper fix was considered too risky for inclusion just
before our 0.19 release.
* Goodbye postInitProc
* Give preInitProc its own scope
Avoid any conflict between the variables introduced by preInitProc and
initProc since both are codegen'd in the same function body.
* Fix codegen for global var init in emulated TLS
Fixes#8535
* Add test for #8535
* Keep a bogus stack frame around
* Remove more dead code
