Adds support for checking ?Foo type hints in VerifyParamType.
The parameter must have type Foo or null. Failing to pass the hint is
reported as a warning instead of a recoverable error for now for
migration reasons---we'll want to convert it to be the same as normal
type hints later.
A poor man's formatter since I didn't like any of the other ones I found. The original C++ source sometimes put newlines and sometimes not.
Codemods:
codemod '([;{}])([^\n])' '\1\n\2'
codemod -m '\s*<\?php\s+' '<?php\n\n'
codemod '\t' ' '
I hand-fixed all the failing tests
A bunch of tests were failing because the behavior of getTypehintText()
changed. Instead of changing how getTypehintText() behaves, let's just
introduce a new reflection API, and then tests and application code can
be changed to use the new API later.
ReflectoinParamenter was fetching both type and type hint from the same field. That created an incompatible change with existing code and broke the mocking framework. Separated the type and type hint field to contain the possible different values.
@override-unit-failures
D774978 was checked in with a broken test. New tests go in test/slow
now. The test was broken because @drussi apparently didn't rebuild the
parser after his last diff update (changing 's' to 't').
This is intended so reflection can be used (via getTypehintText and getReturnTypehintText) to regenerate code the user annotated with types. Essentially using reflection to intrispect code in order to generate type safe
(hack safe) code. That is particularly important for the tools that do dependency injection. The runtime should be oblivious to the change as the rich type annotation is currently only stored for the sake of reflection. For
functions the values are in the shared portion which is cold and should also take care of traits.
Jordan DeLong