Module shell (1)
This commit is contained in:
parent
57f6071552
commit
898822d78f
3 changed files with 646 additions and 0 deletions
125
shell/scanner/scanner.go
Normal file
125
shell/scanner/scanner.go
Normal file
|
|
@ -0,0 +1,125 @@
|
|||
package scanner
|
||||
|
||||
import (
|
||||
"bufio"
|
||||
"errors"
|
||||
"io"
|
||||
"unicode/utf8"
|
||||
|
||||
"gitea.zaclys.com/bvaudour/gob/collection"
|
||||
)
|
||||
|
||||
// Tokenizer est une interface implémentant une fonction de splittage.
|
||||
type Tokenizer interface {
|
||||
Split(data []byte, atEOF bool) (advance int, token []byte, err error)
|
||||
}
|
||||
|
||||
type tk struct {
|
||||
quote bool
|
||||
escape bool
|
||||
spaces []rune
|
||||
}
|
||||
|
||||
func runeToBytes(r rune) []byte {
|
||||
l := utf8.RuneLen(r)
|
||||
b := make([]byte, l)
|
||||
utf8.EncodeRune(b, r)
|
||||
return b
|
||||
}
|
||||
|
||||
func (t tk) Split(data []byte, atEOF bool) (advance int, token []byte, err error) {
|
||||
q, e := collection.NewSet[rune](), collection.NewSet[rune]()
|
||||
s := collection.NewSet(t.spaces...)
|
||||
if t.quote {
|
||||
q.Add('\'', '"')
|
||||
}
|
||||
if t.escape {
|
||||
e.Add('\\')
|
||||
}
|
||||
|
||||
// Skip leading spaces.
|
||||
start := 0
|
||||
for width := 0; start < len(data); start += width {
|
||||
var r rune
|
||||
r, width = utf8.DecodeRune(data[start:])
|
||||
if !s.Contains(r) {
|
||||
break
|
||||
}
|
||||
}
|
||||
|
||||
var quote, esc rune
|
||||
|
||||
// Scan until space, marking end of word.
|
||||
for width, i := 0, start; i < len(data); i += width {
|
||||
var r rune
|
||||
r, width = utf8.DecodeRune(data[i:])
|
||||
if s.Contains(r) && !q.Contains(quote) && !e.Contains(esc) {
|
||||
return i + width, token, nil
|
||||
}
|
||||
if e.Contains(esc) {
|
||||
if q.Contains(quote) && !e.Contains(r) && r != quote {
|
||||
token = append(token, runeToBytes(esc)...)
|
||||
}
|
||||
token = append(token, runeToBytes(r)...)
|
||||
esc = 0
|
||||
} else if e.Contains(r) {
|
||||
esc = r
|
||||
} else if q.Contains(r) {
|
||||
if !q.Contains(quote) {
|
||||
quote = r
|
||||
} else if quote == r {
|
||||
quote = 0
|
||||
} else {
|
||||
token = append(token, runeToBytes(r)...)
|
||||
}
|
||||
} else {
|
||||
token = append(token, runeToBytes(r)...)
|
||||
}
|
||||
}
|
||||
// If we're at EOF, we have a final, non-empty, non-terminated word. Return it.
|
||||
if atEOF && len(data) > start {
|
||||
if e.Contains(esc) || q.Contains(quote) {
|
||||
return start, nil, errors.New("Incomplete token")
|
||||
}
|
||||
return len(data), token, nil
|
||||
}
|
||||
// Request more data.
|
||||
return start, nil, nil
|
||||
}
|
||||
|
||||
// NewTokenizer retourne un tokenizer d’arguments de ligne de commande.
|
||||
//
|
||||
// Le split s’effectue au niveau des caractères spaces fourni (tous les espaces
|
||||
// si aucun de fourni) sauf si ces caractères sont échappés
|
||||
// (si escape) ou entre guillemets (si quote)
|
||||
// Par exemple, prenons la chaîne suivante :
|
||||
// unmot une\ phrase "une deuxième\" phrase"
|
||||
//
|
||||
// Le résultat va être décomposé en 3 éléments :
|
||||
// - unmot
|
||||
// - une\ phrase
|
||||
// - "une deuxième\" phrase"
|
||||
func NewTokenizer(quote, escape bool, spaces ...rune) Tokenizer {
|
||||
if len(spaces) == 0 {
|
||||
spaces = []rune(" \t\n\v\f\r")
|
||||
}
|
||||
return tk{
|
||||
quote: quote,
|
||||
escape: escape,
|
||||
spaces: spaces,
|
||||
}
|
||||
}
|
||||
|
||||
// NewScanner retourne un nouveau scanner utilisant le tokenizer spécifié
|
||||
// pour la fonction de splitage (NewTokenizer(true, true) si aucun tokenizer fourni).
|
||||
func NewScanner(r io.Reader, t ...Tokenizer) *bufio.Scanner {
|
||||
var s bufio.SplitFunc
|
||||
if len(t) > 0 {
|
||||
s = t[0].Split
|
||||
} else {
|
||||
s = (NewTokenizer(true, true)).Split
|
||||
}
|
||||
sc := bufio.NewScanner(r)
|
||||
sc.Split(s)
|
||||
return sc
|
||||
}
|
||||
Loading…
Add table
Add a link
Reference in a new issue