This directive anyway does not make test run twice with OI, and with NI
It only once run the test with specific settings (// LANGUAGE)
and ignores irrelevant (OI or NI tags)
This new kind of expression encompasses the nullability of the original
expression after null check (or equivalent `is Nothing?` check).
Unlike FirExpressionWithSmartcast, this expression won't be materialized
during conversion to backend IR. Also, Nothing? is discarded when
computing the intersection of possible types from smartcast info.
In that way, Nothing? is not used during resolution, while such
smartcast info is stored in it (and the expression kind itself).
To do so, inside the root cause of inapplicable candidate errors,
we will record expected/actual type of receiver, if any.
That will help identifying inapplicable calls on nullable receiver.
Update includes:
- Changing syntax of `OI/`NI` tags from `<!NI;TAG!>` to `<!TAG{NI}!>`
- Fix some incorrect directives
- Change order of diagnostics in some places
- Remove ignored diagnostics from FIR test data (previously `DIAGNOSTICS` didn't work)
- Update FIR dumps in some places and add `FIR_IDENTICAL` if needed
- Replace all JAVAC_SKIP with SKIP_JAVAC directive
Call argument for conventional `contains` after expanding `in` may come from a `when` subject during its branch analysis.
In this case data flow info from a previous when branch was not considered,
because data flow info for subject had been used instead of data flow before argument.
Use of the latter one for the conventional `contains` solves the issue.
The old FE uses `isExternal` property of value arguments to skip smartcast reporting on `when` subject,
if they come from branches. To prevent undesired smartcasts on `when` subject after branch analysis in the new FE,
`isExternal` arguments are skipped in diagnostic reporter and during recorded type update.
Also, the new FE interprets `isExternal` completely differently from the old FE.
In the old FE this property is used exclusively by `when` with subject.
In the new FE it is also used for parially resolved calls, lambda return arguments and receivers.
This may be preventing the use of data flow info before argument in the first place, but this assumption requires additional investigation.
^KT-36818 Fixed
Add some classes to hierarchy of `ESValue`:
- `ESReceiver` -- representation of receiver
of lambda function
- `ESDataFlow` and `ESDataFlowReceiver` -- little refactoring of entities
that holds information for DFA (description in comments in code).
Also add kdoc to classes of `ESValue` hierarchy
Consider following expression: 'call() is Foo'. Suppose that we know
something about the 'call()', e.g. 'returns(foo) -> <condition>'
Previously, we've tried to re-use knowledge about 'call()', constructing
some smart clause, like 'returns(true) -> foo is Foo && <condition>'.
The conceptual error here is that *we can't* argue that <condition>
holds. Imagine that 'call()' actually has unspecified 'returns(foo2) ->
<!condition>', and 'foo2 is Foo' also holds. Then we would get
'returns(true) -> foo2 is Foo && <condition>' <=> 'returns(true) ->
<condition>' for the whole call, which is not correct.
More concrete example would be something like:
'if (!x.isNullOrEmpty() is Boolean)'
^KT-27241 Fixed
It is safe to treat DefaultValueArgument as UNKNOWN_COMPUTATION, because
default arguments can't break smartcasts.
Possibly, they can add new ones, but it can be supported later.
^KT-25278 Fixed
The order of types enumeration has been changed recently:
previously it was collected at first from a child and then from
its parent, but now it's being collected in order of appearance
- Introduce new language feature 'ReadDeserializedContracts', which
allows to deserialize contracts from metadata.
- Introduce new language feature 'AllowContractsForCustomFunctions',
which allows reading contracts from sources.
- Use new features instead of combination 'CallsInPlaceEffect ||
ReturnsEffect'
- Rename 'CallsInPlaceEffect' -> 'UseCallsInPlaceEffect',
'ReturnsEffect' -> 'UseReturnsEffect'. As names suggest, they control
if it is allowed to use corresponding effect in analysis.
We have to introduce separate 'ReadDeserializedContracts' to enable
contracts only in some modules of the project, because libraries are
read with project-wide settings (see KT-20692).
This commits introduces testdata changes, where NI behaviour strictly
improved, after several previous fixes.
For some tests, just WITH_NEW_INFERENCE directive was added. It
indicates, that some of previous commits first introduced error in that
test, and then some other commit fixed it (netting no overall testdata
change). It is preferrably to keep those annotations until we will
migrate to NI completely, to prevent unexpected regressions.
- Make AbstractDiagnosticsTest dump function contracts
- Add diagnostics tests on parsing contracts
- Add diagnostics tests on smartcats in presence of functions with
contracts
- Add diagnostics tests on initialization and flow in presence of
in-place called lambdas
==========
Introduction of EffectSystem: 16/18
Mikhail Glukhikh