hcl/zcl/zclsyntax/token.go
Martin Atkins 4a939a2b46 zclsyntax: allow scanner to support multiple modes
A scanner "mode" decides which state it starts in, allowing us to start
in template mode for parsing top-level templates. However, currently the
only mode implemented is "normal" mode, which is the behavior we had
before.
2017-05-28 15:44:22 -07:00

140 lines
3.4 KiB
Go
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

package zclsyntax
import (
"fmt"
"github.com/apparentlymart/go-textseg/textseg"
"github.com/zclconf/go-zcl/zcl"
)
// Token represents a sequence of bytes from some zcl code that has been
// tagged with a type and its range within the source file.
type Token struct {
Type TokenType
Bytes []byte
Range zcl.Range
}
// TokenType is an enumeration used for the Type field on Token.
type TokenType rune
//go:generate stringer -type TokenType -output token_type_string.go
const (
// Single-character tokens are represented by their own character, for
// convenience in producing these within the scanner. However, the values
// are otherwise arbitrary and just intended to be mnemonic for humans
// who might see them in debug output.
TokenOBrace TokenType = '{'
TokenCBrace TokenType = '}'
TokenOBrack TokenType = '['
TokenCBrack TokenType = ']'
TokenOParen TokenType = '('
TokenCParen TokenType = ')'
TokenOQuote TokenType = '«'
TokenCQuote TokenType = '»'
TokenOHeredoc TokenType = 'H'
TokenCHeredoc TokenType = 'h'
TokenDot TokenType = '.'
TokenStar TokenType = '*'
TokenSlash TokenType = '/'
TokenPlus TokenType = '+'
TokenMinus TokenType = '-'
TokenEqual TokenType = '='
TokenNotEqual TokenType = '≠'
TokenLessThan TokenType = '<'
TokenLessThanEq TokenType = '≤'
TokenGreaterThan TokenType = '>'
TokenGreaterThanEq TokenType = '≥'
TokenAnd TokenType = '∧'
TokenOr TokenType = ''
TokenBang TokenType = '!'
TokenQuestion TokenType = '?'
TokenColon TokenType = ':'
TokenTemplateInterp TokenType = '∫'
TokenTemplateControl TokenType = 'λ'
TokenTemplateSeqEnd TokenType = '∎'
TokenStringLit TokenType = 'S'
TokenNumberLit TokenType = 'N'
TokenIdent TokenType = 'I'
TokenNewline TokenType = '\n'
TokenEOF TokenType = '␄'
// The rest are not used in the language but recognized by the scanner so
// we can generate good diagnostics in the parser when users try to write
// things that might work in other languages they are familiar with, or
// simply make incorrect assumptions about the zcl language.
TokenBitwiseAnd TokenType = '&'
TokenBitwiseOr TokenType = '|'
TokenBitwiseNot TokenType = '~'
TokenBitwiseXor TokenType = '^'
TokenStarStar TokenType = '➚'
TokenBacktick TokenType = '`'
TokenSemicolon TokenType = ';'
TokenTabs TokenType = '␉'
TokenInvalid TokenType = '<27>'
TokenBadUTF8 TokenType = '💩'
)
func (t TokenType) GoString() string {
return fmt.Sprintf("zclsyntax.%s", t.String())
}
type scanMode int
const (
scanNormal scanMode = iota
scanTemplate
)
type tokenAccum struct {
Filename string
Bytes []byte
Pos zcl.Pos
Tokens []Token
}
func (f *tokenAccum) emitToken(ty TokenType, startOfs, endOfs int) {
// Walk through our buffer to figure out how much we need to adjust
// the start pos to get our end pos.
start := f.Pos
start.Column += startOfs - f.Pos.Byte // Safe because only ASCII spaces can be in the offset
start.Byte = startOfs
end := start
end.Byte = endOfs
b := f.Bytes[startOfs:endOfs]
for len(b) > 0 {
advance, seq, _ := textseg.ScanGraphemeClusters(b, true)
if len(seq) == 1 && seq[0] == '\n' {
end.Line++
end.Column = 1
} else {
end.Column++
}
b = b[advance:]
}
f.Pos = end
f.Tokens = append(f.Tokens, Token{
Type: ty,
Bytes: f.Bytes[startOfs:endOfs],
Range: zcl.Range{
Filename: f.Filename,
Start: start,
End: end,
},
})
}