overridden in UnitScanners: apply brace patch if one exists for this offset return true if subsequent end of line handling should be suppressed.

the base of a number

A character buffer for literals

the last read character

Parse character literal if current character is followed by \', or follow with given op and return a symbol literal token

The offset one past the last read character

Convert current strVal to char value

To prevent doc comments attached to expressions from leaking out of scope onto the next documentable entity, they are discarded upon passing a right brace, bracket, or parenthesis.

Convert current strVal, base to double value.

Determines whether this scanner should emit identifier deprecation warnings, e.g. when seeing macro or then, which are planned to become keywords in future versions of Scala.

An error routine to call on bad unicode escapes \\uxxxx.

read next token, filling TokenData fields of Scanner.

Convert current strVal, base to float value.

read fractional part and exponent of floating point number if one is present.

copy current character into cbuf, interpreting any escape sequences, and advance to next character.

Read a number into strVal.

The base can be 8, 10 or 16, where base 8 flags a leading zero. For ints, base 8 is legal only for the case of exactly one zero.

overridden in UnitScanners

Can token start a statement?

Can token end a statement?

signal an error where the input ended in the middle of a token

Initialization method: read first char, then first token

Convert current strVal, base to long value. This is tricky because of max negative value.

Conversions in base 10 and 16 are supported. As a permanent migration path, attempts to write base 8 literals except 0 emit a verbose error.

Is last character a unicode escape \\uxxxx?

The start offset of the line before the current one

the offset of the character following the token preceding this one

The start offset of the current line

A new reader that takes off at the current character position

the name of an identifier

we need one token lookahead and one token history

Advance one character; reducing CR;LF pairs to just LF

Advance one character, leaving CR;LF pairs intact. This is for use in multi-line strings, so there are no "potential line ends" here.

Produce next token, filling TokenData fields of Scanner.

Get next token, and allow the otherwise deprecated ident name

the offset of the first character of the current token

overridden in UnitScanners

append Unicode character to "cbuf" buffer

a stack of tokens which indicates whether line-ends can be statement separators also used for keeping track of nesting levels. We keep track of the closing symbol of a region. This can be RPAREN if region starts with '(' RBRACKET if region starts with '[' RBRACE if region starts with '{' ARROW if region starts with 'case' STRINGLIT if region is a string interpolation expression starting with '${' (the STRINGLIT appears twice in succession on the stack iff the expression is a multiline string literal).

Advance beyond a case token without marking the CASE in sepRegions. This method should be called to skip beyond CASE tokens that are not part of matches, i.e. no ARROW is expected after them.

read next token and return last offset

the string value of a literal

generate an error at the current token offset

generate an error at the given offset

the next token