gosh/brace.go

package main

// Brace expansion like in bash: a{b,c}d becomes abd acd, {1..5} becomes
// 1 2 3 4 5. It happens first, on the text of the command line, so that the
// resulting words go through the other expansions (~, variables, wildcards).

import (
	"errors"
	"fmt"
	"strconv"
	"strings"
)

// maxBraceWords limits the number of words a single word may expand to.
const maxBraceWords = 100000

// expandBraceWords applies brace expansion to every word of a command (the
// text of a pipeline). Quoted or escaped braces, ${...} and $(...) are left
// alone.
// A redirection target must expand to a single word.
func expandBraceWords(input string) (string, error) {
	var (
		out           strings.Builder
		afterRedirect bool // the next word is the target of a redirection
		commandStart  = true
	)

	for i := 0; i < len(input); {
		ch := input[i]
		switch {
		case ch == ' ' || ch == '\t' || ch == '\n':
			out.WriteByte(ch)
			i++
		case strings.IndexByte(";&|<>", ch) >= 0:
			if ch == '<' || ch == '>' {
				afterRedirect = true
			} else {
				commandStart = true
			}
			out.WriteByte(ch)
			i++
		case ch == '#':
			// A comment, until the end of the line
			out.WriteString(input[i:])
			return out.String(), nil
		default:
			end := wordEnd(input, i)
			if commandStart && !afterRedirect && isAssignment(input[i:end]) {
				// name=value before the command: no brace expansion, like
				// in bash
				out.WriteString(input[i:end])
				i = end
				continue
			}
			if !afterRedirect {
				commandStart = false
			}
			words, err := expandBraces(input[i:end])
			if err != nil {
				return "", err
			}
			if afterRedirect && len(words) > 1 {
				return "", fmt.Errorf("%s: ambiguous redirect", sanitize(input[i:end]))
			}
			afterRedirect = false
			out.WriteString(strings.Join(words, " "))
			i = end
		}
	}
	return out.String(), nil
}

// isAssignment reports whether a word starts with name= (unquoted).
func isAssignment(word string) bool {
	name, _, ok := strings.Cut(word, "=")
	return ok && isVarName(name)
}

// wordEnd returns the end of the word starting at i: the next unquoted blank
// or operator character.
func wordEnd(s string, i int) int {
	var quote byte
	for ; i < len(s); i++ {
		ch := s[i]
		switch {
		case quote == '\'':
			if ch == '\'' {
				quote = 0
			}
		case ch == '\\':
			i++
		case isSubstStart(s, i):
			if j := substEnd(s, i); j >= 0 {
				i = j
			}
		case ch == '$' && i+1 < len(s) && s[i+1] == '{':
			if j := paramEnd(s, i); j >= 0 {
				i = j
			}
		case quote == '"':
			if ch == '"' {
				quote = 0
			}
		case ch == '"' || ch == '\'':
			quote = ch
		case strings.IndexByte(" \t\n;&|<>", ch) >= 0:
			return i
		}
	}
	return len(s)
}

// braceScan calls fn for every unquoted, unescaped character of s that is
// not part of a ${...} reference, with its index and the brace depth.
// Scanning stops when fn returns false.
func braceScan(s string, fn func(i, depth int) bool) {
	var quote byte
	depth := 0
	for i := 0; i < len(s); i++ {
		ch := s[i]
		switch {
		case quote == '\'':
			if ch == '\'' {
				quote = 0
			}
			continue
		case ch == '\\':
			i++
			continue
		case isSubstStart(s, i):
			// The subshell does its own brace expansion
			if j := substEnd(s, i); j >= 0 {
				i = j
			}
			continue
		case ch == '$' && i+1 < len(s) && s[i+1] == '{':
			if j := paramEnd(s, i); j >= 0 {
				i = j
			}
			continue
		case quote == '"':
			if ch == '"' {
				quote = 0
			}
			continue
		case ch == '"' || ch == '\'':
			quote = ch
			continue
		}
		if ch == '}' {
			depth--
		}
		if !fn(i, depth) {
			return
		}
		if ch == '{' {
			depth++
		}
	}
}

// expandBraces expands the first valid brace expression of a word and then
// recursively the results: a{b,c}d -> abd acd, x{1..3} -> x1 x2 x3. A brace
// expression is valid if it contains an unquoted comma at its top level or
// is a sequence; other braces ({}, {a}, ${x}) are kept as they are.
func expandBraces(word string) ([]string, error) {
	// Find the candidates: unquoted { with their matching }
	var opens []int
	pairs := map[int]int{}
	braceScan(word, func(i, depth int) bool {
		switch word[i] {
		case '{':
			opens = append(opens, i)
		case '}':
			if len(opens) > 0 {
				pairs[opens[len(opens)-1]] = i
				opens = opens[:len(opens)-1]
			}
		}
		return true
	})

	for start := 0; start < len(word); start++ {
		end, ok := pairs[start]
		if !ok {
			continue
		}
		inner := word[start+1 : end]
		alternatives := splitTopLevelCommas(inner)
		if len(alternatives) < 2 {
			seq, ok, err := braceSequence(inner)
			if err != nil {
				return nil, err
			}
			if !ok {
				continue
			}
			alternatives = seq
		}

		prefix, suffix := word[:start], word[end+1:]
		var words []string
		for _, alt := range alternatives {
			expanded, err := expandBraces(prefix + alt + suffix)
			if err != nil {
				return nil, err
			}
			words = append(words, expanded...)
			if len(words) > maxBraceWords {
				return nil, errors.New("brace expansion: too many words")
			}
		}
		return words, nil
	}
	return []string{word}, nil
}

// splitTopLevelCommas splits s at its unquoted commas that are not inside
// nested braces.
func splitTopLevelCommas(s string) []string {
	var parts []string
	last := 0
	braceScan(s, func(i, depth int) bool {
		if s[i] == ',' && depth == 0 {
			parts = append(parts, s[last:i])
			last = i + 1
		}
		return true
	})
	return append(parts, s[last:])
}

// braceSequence expands a sequence expression: x..y or x..y..step with
// integers (zero-padded to the same width if one of them has a leading zero)
// or single letters. ok is false if s is no sequence.
func braceSequence(s string) (words []string, ok bool, err error) {
	parts := strings.Split(s, "..")
	if len(parts) != 2 && len(parts) != 3 {
		return nil, false, nil
	}
	step := 1
	if len(parts) == 3 {
		step, err = strconv.Atoi(parts[2])
		if err != nil {
			return nil, false, nil
		}
		if step < 0 {
			step = -step
		}
		if step == 0 {
			step = 1
		}
	}

	from, errFrom := strconv.Atoi(parts[0])
	to, errTo := strconv.Atoi(parts[1])
	isLetter := func(p string) bool {
		return len(p) == 1 && (p[0] >= 'a' && p[0] <= 'z' || p[0] >= 'A' && p[0] <= 'Z')
	}
	format := func(n int) string { return strconv.Itoa(n) }
	switch {
	case errFrom == nil && errTo == nil:
		// {01..10}: pad to the width of the longer number
		if hasLeadingZero(parts[0]) || hasLeadingZero(parts[1]) {
			width := max(len(parts[0]), len(parts[1]))
			format = func(n int) string {
				if n < 0 {
					return fmt.Sprintf("-%0*d", width-1, -n)
				}
				return fmt.Sprintf("%0*d", width, n)
			}
		}
	case isLetter(parts[0]) && isLetter(parts[1]):
		from, to = int(parts[0][0]), int(parts[1][0])
		format = func(n int) string { return string(rune(n)) }
	default:
		return nil, false, nil
	}

	count := (max(from, to)-min(from, to))/step + 1
	if count > maxBraceWords {
		return nil, false, errors.New("brace expansion: too many words")
	}
	if to < from {
		step = -step
	}
	for n := from; (step > 0 && n <= to) || (step < 0 && n >= to); n += step {
		words = append(words, format(n))
	}
	return words, true, nil
}

// hasLeadingZero reports whether a number is written with a leading zero
// (01, -05), which makes a sequence zero-padded.
func hasLeadingZero(s string) bool {
	s = strings.TrimPrefix(s, "-")
	return len(s) > 1 && s[0] == '0'
}