Is tree either a non-volatile type, or a path that does not include any of:

a reference to a mutable variable/fielda reference to a by-name parametera member selection on a volatile type (Spec: 3.6 Volatile Types)?

Such a tree is a suitable target for type selection.

Translates an Assign(_, _) node to NamedArg(_, _) if the lhs is a simple ident. Otherwise returns unchanged.

Does this CaseDef catch Throwable?

Returns a wrapper that knows how to destructure and analyze applications.

//------------------------ => effectivePatternArity(args)
case Extractor(a)          => 1
case Extractor(a, b)       => 2
case Extractor((a, b))     => 2
case Extractor(a @ (b, c)) => 2

The first constructor definitions in stats

The arguments to the first constructor in stats.

Does list of trees start with a definition of a class or module with given name (ignoring imports)

Is tree's type volatile? (Ignored if its symbol has the @uncheckedStable annotation.)

Is tpt a by-name parameter type of the form => T?

Is this pattern node a catch-all or type-test pattern?

Is tree a declaration or type definition?

Is this pattern node a catch-all (wildcard or variable) pattern?

Is tree an expression which can be inlined without affecting program semantics?

Note that this is not called "isExprPure" since purity (lack of side-effects) is not the litmus test. References to modules and lazy vals are side-effecting, both because side-effecting code may be executed and because the first reference takes a different code path than all to follow; but they are safe to inline because the expression result from evaluating them is always the same.

Is this case guarded?

Is tree legal as a member definition of an interface?

Is tree a path, defined as follows? (Spec: 3.1 Paths)

- The empty path ε (which cannot be written explicitly in user programs). - C.this, where C references a class. - p.x where p is a path and x is a stable member of p. - C.super.x or C.super[M].x where C references a class and x references a stable member of the super class or designated parent class M of C.

NOTE: Trees with errors are (mostly) excluded.

Path ::= StableId | [id ‘.’] this

Is tree a pure (i.e. non-side-effecting) definition?

As if the name of the method didn't give it away, this logic is designed around issuing helpful warnings and minimizing spurious ones. That means don't reuse it for important matters like inlining decisions.

Is tpt a vararg type of the form T* ?

Is tree a self constructor call this(...)? I.e. a call to a constructor of the same object?

Is tree a self or super constructor call?

Is this pattern node a sequence-valued pattern?

Is tree a stable identifier, a path which ends in an identifier?

StableId ::= id | Path ‘.’ id | [id ’.’] ‘super’ [‘[’ id ‘]’] ‘.’ id

Is tree admissible as a stable identifier pattern (8.1.5 Stable Identifier Patterns)?

We disregard volatility, as it's irrelevant in patterns (scala/bug#6815)

Assuming sym is a member of tree, is it a "stable member"?

Stable members are packages or members introduced by object definitions or by value definitions of non-volatile types (§3.6).

Is this tree a Star(_) after removing bindings?

Is tree a super constructor call?

a Match(Typed(_, tpt), _) must be translated into a switch if isSwitchAnnotation(tpt.tpe)

Is this CaseDef synthetically generated, e.g. by MatchTranslation.translateTry?

Is this pattern node a synthetic catch-all case, added during PartialFunction synthesis before we know whether the user provided cases are exhaustive.

Is tree a variable pattern?

Does this tree represent an irrefutable pattern match in the position for { <tree> <- expr } based only on information at the parser phase? To qualify, there may be no subtree that will be interpreted as a Stable Identifier Pattern, nor any type tests, even on TupleN. See scala/bug#6968.

For instance:

(foo @ (bar @ _)) = 0

is a not a variable pattern; if only binds names.

The following are not variable patterns.

`bar`
Bar
(a, b)
_: T

If the pattern is a simple identifier, it is always a variable pattern. For example, the following introduce new bindings:

for { X <- xs } yield X
for { `backquoted` <- xs } yield `backquoted`

Note that this differs from a case clause:

object X
scrut match {
   case X =>  // case _ if scrut == X
}

Background: https://groups.google.com/d/msg/scala-internals/qwa_XOw_7Ks/IktkeTBYqg0J

Is tree a mutable variable, or the getter of a mutable field?

Is the argument a wildcard argument of the form _ or x @ _?

Is this argument node of the form <expr> : _* ?

Does this argument list end with an argument of the form <expr> : _* ?

Is the argument a wildcard star type of the form _*?

can this type be a type pattern

Is symbol potentially a getter of a variable?

The value definitions marked PRESUPER in this statement sequence

Strips layers of .asInstanceOf[T] / _.$asInstanceOf[T]() from an expression

Named arguments can transform a constructor call into a block, e.g. <init>(b = foo, a = bar) is transformed to { val x$1 = foo val x$2 = bar <init>(x$2, x$1) }

If this tree has type parameters, those. Otherwise Nil.

The underlying pattern ignoring any bindings

Destructures applications into important subparts described in Applied class, namely into: core, targs and argss (in the specified order).

Trees which are not applications are also accepted. Their callee and core will be equal to the input, while targs and argss will be Nil.

The provided extractors don't expose all the API of the Applied class. For advanced use, call dissectApplied explicitly and use its methods instead of pattern matching.

Locates the synthetic Apply node corresponding to an extractor's call to unapply (unwrapping nested Applies) and returns the fun part of that Apply.