goparsify/parser.go
2017-08-10 22:06:08 +10:00

259 lines
6.8 KiB
Go

package goparsify
import (
"errors"
"fmt"
"regexp"
"strings"
"unicode/utf8"
)
// Result is the output of a parser. Usually only one of its fields will be set and should be though of
// more as a union type. having it avoids interface{} littered all through the parsing code and makes
// the it easy to do the two most common operations, getting a token and finding a child.
type Result struct {
Token string
Child []Result
Result interface{}
}
// Parser is the workhorse of parsify. A parser takes a State and returns a result, consuming some
// of the State in the process.
// Given state is shared there are a few rules that should be followed:
// - A parser that errors must set state.Error
// - A parser that errors must not change state.Pos
// - A parser that consumed some input should advance state.Pos
type Parser func(*State) Result
// VoidParser is a special type of parser that never returns anything but can still consume input
type VoidParser func(*State)
// Parserish types are any type that can be turned into a Parser by Parsify
// These currently include *Parser and string literals.
//
// This makes recursive grammars cleaner and allows string literals to be used directly in most contexts.
// eg, matching balanced paren:
// ```go
// var group Parser
// group = Seq("(", Maybe(&group), ")")
// ```
// vs
// ```go
// var group ParserPtr{}
// group.P = Seq(Exact("("), Maybe(group.Parse), Exact(")"))
// ```
type Parserish interface{}
// Parsify takes a Parserish and makes a Parser out of it. It should be called by
// any Parser that accepts a Parser as an argument. It should never be called during
// instead call it during parser creation so there is no runtime cost.
//
// See Parserish for details.
func Parsify(p Parserish) Parser {
switch p := p.(type) {
case func(*State) Result:
return p
case Parser:
return p
case *Parser:
// Todo: Maybe capture this stack and on nil show it? Is there a good error library to do this?
return func(ptr *State) Result {
return (*p)(ptr)
}
case string:
return Exact(p)
default:
panic(fmt.Errorf("cant turn a `%T` into a parser", p))
}
}
// ParsifyAll calls Parsify on all parsers
func ParsifyAll(parsers ...Parserish) []Parser {
ret := make([]Parser, len(parsers))
for i, parser := range parsers {
ret[i] = Parsify(parser)
}
return ret
}
// Run applies some input to a parser and returns the result, failing if the input isnt fully consumed.
// It is a convenience method for the most common way to invoke a parser.
func Run(parser Parserish, input string, ws ...VoidParser) (result interface{}, err error) {
p := Parsify(parser)
ps := NewState(input)
if len(ws) > 0 {
ps.WS = ws[0]
}
ret := p(ps)
ps.AutoWS()
if ps.Error.expected != "" {
return ret.Result, ps.Error
}
if ps.Get() != "" {
return ret.Result, errors.New("left unparsed: " + ps.Get())
}
return ret.Result, nil
}
// WS will consume whitespace, it should only be needed when AutoWS is turned off
func WS() Parser {
return NewParser("AutoWS", func(ps *State) Result {
ps.WS(ps)
return Result{}
})
}
// Cut prevents backtracking beyond this point. Usually used after keywords when you
// are sure this is the correct path. Improves performance and error reporting.
func Cut(ps *State) Result {
ps.Cut = ps.Pos
return Result{}
}
// Regex returns a match if the regex successfully matches
func Regex(pattern string) Parser {
re := regexp.MustCompile("^" + pattern)
return NewParser(pattern, func(ps *State) Result {
ps.AutoWS()
if match := re.FindString(ps.Get()); match != "" {
ps.Advance(len(match))
return Result{Token: match}
}
ps.ErrorHere(pattern)
return Result{}
})
}
// Exact will fully match the exact string supplied, or error. The match will be stored in .Token
func Exact(match string) Parser {
if len(match) == 1 {
matchByte := match[0]
return NewParser(match, func(ps *State) Result {
ps.AutoWS()
if ps.Pos >= len(ps.Input) || ps.Input[ps.Pos] != matchByte {
ps.ErrorHere(match)
return Result{}
}
ps.Advance(1)
return Result{Token: match}
})
}
return NewParser(match, func(ps *State) Result {
ps.AutoWS()
if !strings.HasPrefix(ps.Get(), match) {
ps.ErrorHere(match)
return Result{}
}
ps.Advance(len(match))
return Result{Token: match}
})
}
func parseRepetition(defaultMin, defaultMax int, repetition ...int) (min int, max int) {
min = defaultMin
max = defaultMax
switch len(repetition) {
case 0:
case 1:
min = repetition[0]
case 2:
min = repetition[0]
max = repetition[1]
default:
panic(fmt.Errorf("Dont know what %d repetion args mean", len(repetition)))
}
return min, max
}
// parseMatcher turns a string in the format a-f01234A-F into:
// - an alphabet of matches string(01234)
// - a set of ranges [][]rune{{'a', 'f'}, {'A', 'F'}}
func parseMatcher(matcher string) (alphabet string, ranges [][]rune) {
runes := []rune(matcher)
for i := 0; i < len(runes); i++ {
if i+2 < len(runes) && runes[i+1] == '-' && runes[i] != '\\' {
start := runes[i]
end := runes[i+2]
if start <= end {
ranges = append(ranges, []rune{start, end})
} else {
ranges = append(ranges, []rune{end, start})
}
} else if i+1 < len(runes) && runes[i] == '\\' {
alphabet += string(runes[i+1])
} else {
alphabet += string(runes[i])
}
}
return alphabet, ranges
}
// Chars is the swiss army knife of character matches. It can match:
// - ranges: Chars("a-z") will match one or more lowercase letter
// - alphabets: Chars("abcd") will match one or more of the letters abcd in any order
// - min and max: Chars("a-z0-9", 4, 6) will match 4-6 lowercase alphanumeric characters
// the above can be combined in any order
func Chars(matcher string, repetition ...int) Parser {
return NewParser("["+matcher+"]", charsImpl(matcher, false, repetition...))
}
// NotChars accepts the full range of input from Chars, but it will stop when any
// character matches.
func NotChars(matcher string, repetition ...int) Parser {
return NewParser("!["+matcher+"]", charsImpl(matcher, true, repetition...))
}
func charsImpl(matcher string, stopOn bool, repetition ...int) Parser {
min, max := parseRepetition(1, -1, repetition...)
alphabet, ranges := parseMatcher(matcher)
return func(ps *State) Result {
ps.AutoWS()
matched := 0
for ps.Pos+matched < len(ps.Input) {
if max != -1 && matched >= max {
break
}
r, w := utf8.DecodeRuneInString(ps.Input[ps.Pos+matched:])
anyMatched := strings.ContainsRune(alphabet, r)
if !anyMatched {
for _, rng := range ranges {
if r >= rng[0] && r <= rng[1] {
anyMatched = true
}
}
}
if anyMatched == stopOn {
break
}
matched += w
}
if matched < min {
ps.ErrorHere(matcher)
return Result{}
}
result := ps.Input[ps.Pos : ps.Pos+matched]
ps.Advance(matched)
return Result{Token: result}
}
}