Compare commits
5 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 641e24896b | |||
| 88137410ee | |||
| cd64435c8e | |||
| 9896bef7d9 | |||
| 7e5601d5b5 |
@@ -1,88 +0,0 @@
|
|||||||
package parser
|
|
||||||
|
|
||||||
import (
|
|
||||||
"strings"
|
|
||||||
|
|
||||||
"asciigoat.org/core/lexer"
|
|
||||||
)
|
|
||||||
|
|
||||||
type commaArrayParser struct {
|
|
||||||
TextParser
|
|
||||||
|
|
||||||
out []string
|
|
||||||
}
|
|
||||||
|
|
||||||
func (p *commaArrayParser) lexStart() (lexer.StateFn, error) {
|
|
||||||
for {
|
|
||||||
r, _, err := p.ReadRune()
|
|
||||||
switch {
|
|
||||||
case err != nil:
|
|
||||||
// EOF
|
|
||||||
return nil, err
|
|
||||||
case r == RuneQuotes:
|
|
||||||
// Quoted Value
|
|
||||||
return p.lexQuotedString, nil
|
|
||||||
case IsNewLine(r):
|
|
||||||
// new lines are acceptable when parsing a string for
|
|
||||||
// comma delimited arrays. but make sure we discard it
|
|
||||||
// complete
|
|
||||||
p.UnreadRune()
|
|
||||||
p.AcceptNewLine()
|
|
||||||
p.Discard()
|
|
||||||
case lexer.IsSpace(r):
|
|
||||||
// discard whitespace outside quotes
|
|
||||||
p.Discard()
|
|
||||||
default:
|
|
||||||
p.UnreadRune()
|
|
||||||
return p.lexWord, nil
|
|
||||||
}
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
func (p *commaArrayParser) lexWord() (lexer.StateFn, error) {
|
|
||||||
for {
|
|
||||||
r, _, err := p.ReadRune()
|
|
||||||
switch {
|
|
||||||
case err != nil:
|
|
||||||
// done. store what we got and move on
|
|
||||||
_, s := p.Emit()
|
|
||||||
p.out = append(p.out, s)
|
|
||||||
return nil, err
|
|
||||||
case r == ',':
|
|
||||||
// done
|
|
||||||
_, s := p.Emit()
|
|
||||||
// remove comma, trim and append to output
|
|
||||||
s = strings.TrimRightFunc(s[:len(s)-1], IsSpace)
|
|
||||||
p.out = append(p.out, s)
|
|
||||||
return p.lexStart, nil
|
|
||||||
}
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
func (p *commaArrayParser) lexQuotedString() (lexer.StateFn, error) {
|
|
||||||
s, err := lexQuotedString(&p.TextParser)
|
|
||||||
if err != nil {
|
|
||||||
return nil, err
|
|
||||||
}
|
|
||||||
|
|
||||||
p.Discard()
|
|
||||||
p.out = append(p.out, s)
|
|
||||||
return p.lexStart, nil
|
|
||||||
}
|
|
||||||
|
|
||||||
func (p *commaArrayParser) Run() ([]string, error) {
|
|
||||||
err := lexer.Run(p.lexStart)
|
|
||||||
|
|
||||||
return p.out, err
|
|
||||||
}
|
|
||||||
|
|
||||||
// SplitCommaArray splits comma separated strings, removing whitespace
|
|
||||||
// and respecting quoted literals.
|
|
||||||
func SplitCommaArray(s string) ([]string, error) {
|
|
||||||
if s != "" {
|
|
||||||
var p commaArrayParser
|
|
||||||
p.InitString(s)
|
|
||||||
return p.Run()
|
|
||||||
}
|
|
||||||
return nil, nil
|
|
||||||
}
|
|
||||||
@@ -1,8 +1,6 @@
|
|||||||
package parser
|
package parser
|
||||||
|
|
||||||
import (
|
import (
|
||||||
"io/fs"
|
|
||||||
|
|
||||||
"asciigoat.org/core/lexer"
|
"asciigoat.org/core/lexer"
|
||||||
)
|
)
|
||||||
|
|
||||||
@@ -27,36 +25,3 @@ func ErrPlusPosition(pos lexer.Position, e *lexer.Error) *lexer.Error {
|
|||||||
|
|
||||||
return NewError(pos, e.Content, e.Hint, e.Err)
|
return NewError(pos, e.Content, e.Hint, e.Err)
|
||||||
}
|
}
|
||||||
|
|
||||||
// NewErrIncompleteQuotedString returns a [lexer.Error]
|
|
||||||
// indicating the quoted string being parsed wasn't correctly
|
|
||||||
// terminated
|
|
||||||
func NewErrIncompleteQuotedString(p *TextParser) *lexer.Error {
|
|
||||||
return newErrIncomplete(p, "incomplete quoted string")
|
|
||||||
}
|
|
||||||
|
|
||||||
// NewErrIncompleteEscaped returns a [lexer.Error]
|
|
||||||
// indicating the text being parsed wasn't correctly
|
|
||||||
// terminated
|
|
||||||
func NewErrIncompleteEscaped(p *TextParser) *lexer.Error {
|
|
||||||
return newErrIncomplete(p, "incomplete escaped string")
|
|
||||||
}
|
|
||||||
|
|
||||||
func newErrIncomplete(p *TextParser, hint string) *lexer.Error {
|
|
||||||
pos, s := p.Emit()
|
|
||||||
pos.Add(GetPositionalLength(s))
|
|
||||||
|
|
||||||
return NewError(pos, s, hint, fs.ErrInvalid)
|
|
||||||
}
|
|
||||||
|
|
||||||
// NewErrInvalidEscapeSequence returns a [lexer.Error] indicating
|
|
||||||
// the specified sequence, at the end of the accepted buffer,
|
|
||||||
// is invalid
|
|
||||||
func NewErrInvalidEscapeSequence(p *TextParser, seq string) *lexer.Error {
|
|
||||||
pos, s := p.Position(), p.String()
|
|
||||||
|
|
||||||
s = s[:len(s)-len(seq)]
|
|
||||||
pos.Add(GetPositionalLength(s))
|
|
||||||
|
|
||||||
return NewError(pos, seq, "invalid escape character", fs.ErrInvalid)
|
|
||||||
}
|
|
||||||
|
|||||||
@@ -7,13 +7,11 @@ import (
|
|||||||
)
|
)
|
||||||
|
|
||||||
const (
|
const (
|
||||||
RuneComment = ';' // RuneComment is the standard INI comment character
|
RuneComment = ';' // RuneComment is the standard dosini comment character
|
||||||
RuneCommentExtra = '#' // RuneCommentExtra is UNIX shell's comment character
|
RuneCommentExtra = '#' // RuneCommentExtra is UNIX shell's comment character
|
||||||
RuneSectionStart = '[' // RuneSectionStart indicates the start of a section declaration
|
RuneSectionStart = '[' // RuneSectionStart indicates the start of a section declaration
|
||||||
RuneSectionEnd = ']' // RuneSectionEnd indicates the end of a section declaration
|
RuneSectionEnd = ']' // RuneSectionEnd indiciates the end of a section declaration
|
||||||
RuneFieldEqual = '=' // RuneFieldEqual separates field keys from their values
|
RuneFieldEqual = '=' // RuneFieldEqual separates field keys from their values
|
||||||
RuneQuotes = '"' // RuneQuotes indicates the start and end of a quoted value
|
|
||||||
RuneEscape = '\\' // RuneEscape indicates the next rune is escaped
|
|
||||||
)
|
)
|
||||||
|
|
||||||
var (
|
var (
|
||||||
@@ -29,8 +27,6 @@ var (
|
|||||||
RuneSectionStart,
|
RuneSectionStart,
|
||||||
RuneSectionEnd,
|
RuneSectionEnd,
|
||||||
RuneFieldEqual,
|
RuneFieldEqual,
|
||||||
RuneQuotes,
|
|
||||||
RuneEscape,
|
|
||||||
})
|
})
|
||||||
)
|
)
|
||||||
|
|
||||||
|
|||||||
@@ -1,135 +0,0 @@
|
|||||||
package parser
|
|
||||||
|
|
||||||
import (
|
|
||||||
"strings"
|
|
||||||
|
|
||||||
"asciigoat.org/core/lexer"
|
|
||||||
)
|
|
||||||
|
|
||||||
// AcceptQuotedString consumes a quoted string from the source
|
|
||||||
// and returns it unquoted and unescaped
|
|
||||||
func (p *TextParser) AcceptQuotedString() (string, bool, error) {
|
|
||||||
r, _, err := p.ReadRune()
|
|
||||||
switch {
|
|
||||||
case err != nil:
|
|
||||||
// nothing here
|
|
||||||
return "", false, err
|
|
||||||
case r != RuneQuotes:
|
|
||||||
// not for us
|
|
||||||
p.UnreadRune()
|
|
||||||
return "", false, nil
|
|
||||||
default:
|
|
||||||
// let's roll
|
|
||||||
s, err := lexQuotedString(p)
|
|
||||||
switch {
|
|
||||||
case err != nil:
|
|
||||||
// bad quoted string
|
|
||||||
return "", false, err
|
|
||||||
default:
|
|
||||||
// success
|
|
||||||
return s, true, nil
|
|
||||||
}
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
func lexQuotedString(p *TextParser) (string, *lexer.Error) {
|
|
||||||
s, ok, err := lexQuotedStringNoEscape(p)
|
|
||||||
switch {
|
|
||||||
case err != nil:
|
|
||||||
return "", err
|
|
||||||
case ok:
|
|
||||||
return s, nil
|
|
||||||
default:
|
|
||||||
// escape character detected
|
|
||||||
return lexQuotedStringEscaped(p)
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
func lexQuotedStringNoEscape(p *TextParser) (string, bool, *lexer.Error) {
|
|
||||||
for {
|
|
||||||
r, _, err := p.ReadRune()
|
|
||||||
switch {
|
|
||||||
case err != nil:
|
|
||||||
// incomplete
|
|
||||||
return "", false, NewErrIncompleteQuotedString(p)
|
|
||||||
case r == RuneQuotes:
|
|
||||||
// end, just remove the quotes
|
|
||||||
s := p.String()
|
|
||||||
l := len(s)
|
|
||||||
return s[1 : l-1], true, nil
|
|
||||||
case r == RuneEscape:
|
|
||||||
// things just got complicated...
|
|
||||||
p.UnreadRune()
|
|
||||||
return "", false, nil
|
|
||||||
case IsNewLine(r):
|
|
||||||
// new lines within quoted values are acceptable
|
|
||||||
p.UnreadRune()
|
|
||||||
p.AcceptNewLine()
|
|
||||||
default:
|
|
||||||
// continue
|
|
||||||
}
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
// Unquoted removes quotes the content and unescapes the content
|
|
||||||
func lexQuotedStringEscaped(p *TextParser) (string, *lexer.Error) {
|
|
||||||
var result strings.Builder
|
|
||||||
|
|
||||||
// append what was accepted before the escape character
|
|
||||||
_, _ = result.WriteString(p.String()[1:])
|
|
||||||
|
|
||||||
for {
|
|
||||||
r, _, err := p.ReadRune()
|
|
||||||
switch {
|
|
||||||
case err != nil:
|
|
||||||
// incomplete quoted
|
|
||||||
return "", NewErrIncompleteQuotedString(p)
|
|
||||||
case r == RuneQuotes:
|
|
||||||
// end
|
|
||||||
return result.String(), nil
|
|
||||||
case r == RuneEscape:
|
|
||||||
// escaped
|
|
||||||
r2, _, err := p.ReadRune()
|
|
||||||
switch {
|
|
||||||
case err != nil:
|
|
||||||
// incomplete escaped
|
|
||||||
return "", NewErrIncompleteEscaped(p)
|
|
||||||
case IsNewLine(r2):
|
|
||||||
// escaped new line, skip
|
|
||||||
p.UnreadRune()
|
|
||||||
p.AcceptNewLine()
|
|
||||||
default:
|
|
||||||
// TODO: check valid escape character and
|
|
||||||
// append to result
|
|
||||||
s := string([]rune{r, r2})
|
|
||||||
err := NewErrInvalidEscapeSequence(p, s)
|
|
||||||
return "", err
|
|
||||||
}
|
|
||||||
default:
|
|
||||||
// normal, append to result
|
|
||||||
_, _ = result.WriteRune(r)
|
|
||||||
}
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
|
||||||
// Unquoted removes quotes and unescapes the content
|
|
||||||
func Unquoted(s string) (string, error) {
|
|
||||||
var p TextParser
|
|
||||||
if s == "" {
|
|
||||||
return "", nil
|
|
||||||
}
|
|
||||||
|
|
||||||
p.InitString(s)
|
|
||||||
unquoted, ok, err := p.AcceptQuotedString()
|
|
||||||
switch {
|
|
||||||
case err != nil:
|
|
||||||
// bad string
|
|
||||||
return "", err
|
|
||||||
case ok:
|
|
||||||
// success
|
|
||||||
return unquoted, nil
|
|
||||||
default:
|
|
||||||
// not quoted
|
|
||||||
return s, nil
|
|
||||||
}
|
|
||||||
}
|
|
||||||
Reference in New Issue
Block a user