ConfigKit/internal/engine/glob.go

package engine

import (
	"errors"
	"io/fs"
	"path"
	"slices"
	"sort"
	"strings"
)

const globReadBatch = 128

// selectGlobImportCandidates expands path components while retaining the
// lexical spelling used for filtering and later source admission.
func selectGlobImportCandidates(access *filesystemAccess, basePath string, spec importDeclarationSpec, policy resolvedImportPolicy, limits importLimits, accounting *importAccounting) (importSelection, []Diagnostic) {
	site := spec.fieldSites["glob"]
	segments := spec.globSegments
	if len(segments) == 0 {
		var valid bool
		segments, valid = validGlobPattern(spec.glob)
		if !valid {
			return importSelection{}, []Diagnostic{*importSelectionDiagnostic(site, CodeImportInvalid, "import glob pattern is invalid")}
		}
	}
	if accounting == nil {
		accounting = &importAccounting{}
	}
	rootPath := normalizedImportCandidatePath(access, basePath)
	candidates := []selectedImportCandidate{{path: rootPath, accessPath: basePath, site: site}}
	for index, segment := range segments {
		last := index == len(segments)-1
		literal, patterned := globLiteralSegment(segment)
		// nextByPath is the set union of candidates reached from each matching
		// parent at this segment; final exclusions below are set difference.
		nextByPath := make(map[string]selectedImportCandidate)
		for _, parent := range candidates {
			if !patterned {
				childPath := access.join(parent.path, literal)
				childAccessPath := access.join(parent.accessPath, literal)
				candidate, missing, diagnostics := explicitGlobCandidate(access, basePath, childPath, childAccessPath, site, last)
				if len(diagnostics) != 0 {
					return importSelection{}, diagnostics
				}
				if !missing {
					nextByPath[candidate.path] = candidate
				}
				continue
			}

			entries, missing, diagnostics := readGlobEntries(access, parent.accessPath, site.location, limits, accounting)
			if len(diagnostics) != 0 {
				return importSelection{}, diagnostics
			}
			if missing {
				continue
			}
			for _, entry := range entries {
				matched, err := path.Match(segment, entry.Name())
				if err != nil {
					return importSelection{}, []Diagnostic{*importSelectionDiagnostic(site, CodeImportInvalid, "import glob pattern is invalid")}
				}
				if !matched || !policy.hidden && strings.HasPrefix(entry.Name(), ".") {
					continue
				}
				accessPath := access.join(parent.accessPath, entry.Name())
				lexicalPath := access.join(parent.path, entry.Name())
				if !access.contains(accessPath) || !access.contains(lexicalPath) {
					return importSelection{}, []Diagnostic{*importSelectionDiagnostic(site, CodeImportOutsideRoot, "glob traversal is outside the configuration root")}
				}
				target, err := access.importEntryTarget(accessPath, entry, policy.follow)
				if err != nil {
					return importSelection{}, []Diagnostic{globCandidateDiagnostic(err, site.location, "glob match could not be inspected")}
				}
				if target.symlink && !policy.follow {
					continue
				}
				if !last && !target.dir {
					continue
				}
				physicalPath := target.identity
				// Native source identity and relative imports use the namespace
				// spelling; resolved link identities belong only to discovery.
				if access.native != nil {
					physicalPath = accessPath
				}
				if physicalPath == "" {
					physicalPath, err = access.canonical(accessPath)
					if err != nil {
						return importSelection{}, []Diagnostic{globFilesystemDiagnostic(err, "glob match could not be canonicalized", site.location)}
					}
				}
				candidate := selectedImportCandidate{
					path: normalizedImportCandidatePath(access, lexicalPath), accessPath: physicalPath,
					label: relativeImportLabel(access, basePath, lexicalPath), site: site,
				}
				nextByPath[candidate.path] = candidate
			}
		}
		candidates = candidatesFromMap(nextByPath)
		if len(candidates) == 0 {
			break
		}
	}

	selected := make([]selectedImportCandidate, 0, len(candidates))
	for _, candidate := range candidates {
		missing, _, diagnostic := validateNamedImportTarget(access, candidate, false, false)
		if diagnostic != nil {
			return importSelection{}, []Diagnostic{*diagnostic}
		}
		if missing || excludedBasename(path.Base(candidate.path), policy.exclude) {
			continue
		}
		selected = append(selected, candidate)
	}
	if policy.required && len(selected) == 0 {
		return importSelection{}, []Diagnostic{*importSelectionDiagnostic(site, CodeImportMissing, "required import glob selected no files")}
	}
	// Source allowance is charged by source admission, after this filtered set
	// reaches the traversal; that preserves repeated-source diagnostics first.
	selection := importSelection{candidates: selected}
	if len(selected) != 0 {
		selection.outcome = importSelectionHasCandidates
	}
	return selection, nil
}

func explicitGlobCandidate(access *filesystemAccess, basePath, candidatePath, accessPath string, site importFieldSite, last bool) (selectedImportCandidate, bool, []Diagnostic) {
	if !access.contains(candidatePath) || !access.contains(accessPath) {
		return selectedImportCandidate{}, false, []Diagnostic{*importSelectionDiagnostic(site, CodeImportOutsideRoot, "glob traversal is outside the configuration root")}
	}
	target, err := access.importEntryTarget(accessPath, nil, true)
	if err != nil {
		if errors.Is(err, fs.ErrNotExist) {
			return selectedImportCandidate{}, true, nil
		}
		return selectedImportCandidate{}, false, []Diagnostic{globCandidateDiagnostic(err, site.location, "glob match could not be inspected")}
	}
	canonical, err := access.canonical(accessPath)
	if err != nil {
		if errors.Is(err, fs.ErrNotExist) {
			return selectedImportCandidate{}, true, nil
		}
		return selectedImportCandidate{}, false, []Diagnostic{globFilesystemDiagnostic(err, "glob match could not be canonicalized", site.location)}
	}
	if !access.contains(canonical) {
		return selectedImportCandidate{}, false, []Diagnostic{*importSelectionDiagnostic(site, CodeImportOutsideRoot, "glob traversal resolves outside the configuration root")}
	}
	if !last {
		if !target.dir {
			return selectedImportCandidate{}, true, nil
		}
	}
	return selectedImportCandidate{
		path: normalizedImportCandidatePath(access, candidatePath), accessPath: canonical,
		label: relativeImportLabel(access, basePath, candidatePath), site: site,
	}, false, nil
}

func readGlobEntries(access *filesystemAccess, directory string, location Location, limits importLimits, accounting *importAccounting) ([]fs.DirEntry, bool, []Diagnostic) {
	target, err := access.importEntryTarget(directory, nil, true)
	if err != nil {
		if errors.Is(err, fs.ErrNotExist) {
			return nil, true, nil
		}
		return nil, false, []Diagnostic{globCandidateDiagnostic(err, location, "glob directory could not be inspected")}
	}
	canonical, err := access.canonical(directory)
	if err != nil {
		if errors.Is(err, fs.ErrNotExist) {
			return nil, true, nil
		}
		return nil, false, []Diagnostic{globFilesystemDiagnostic(err, "glob directory could not be canonicalized", location)}
	}
	if !access.contains(canonical) {
		return nil, false, []Diagnostic{importDiagnostic(CodeImportOutsideRoot, "glob directory resolves outside the configuration root", &location)}
	}
	if _, err := access.relative(canonical); err != nil {
		return nil, false, []Diagnostic{importDiagnostic(CodeImportOutsideRoot, "glob directory could not be addressed inside the configuration root", &location)}
	}
	readPath := target.identity
	if readPath == "" {
		readPath = canonical
	}
	entries := make([]fs.DirEntry, 0)
	err = visitBoundedImportDirectory(access, readPath, limits.entries, accounting, func(entry fs.DirEntry) error {
		entries = append(entries, entry)
		return nil
	})
	if errors.Is(err, errImportDirectoryEntryLimit) {
		return nil, false, []Diagnostic{importDiagnostic(CodeImportLimitExceeded, "glob traversal exceeds the directory-entry visit limit", &location)}
	}
	if missingDirectoryOpen(err) {
		return nil, true, nil
	}
	if err != nil {
		return nil, false, []Diagnostic{globFilesystemDiagnostic(err, "glob directory could not be traversed", location)}
	}
	sort.Slice(entries, func(left, right int) bool { return entries[left].Name() < entries[right].Name() })
	return entries, false, nil
}

func globCandidateDiagnostic(err error, location Location, message string) Diagnostic {
	if errors.Is(err, errImportDirectoryLinkCycle) {
		return importDiagnostic(CodeImportCycle, "directory traversal cycle detected", &location)
	}
	if errors.Is(err, errImportPathOutsideRoot) {
		return importDiagnostic(CodeImportOutsideRoot, "glob traversal resolves outside the configuration root", &location)
	}
	return globFilesystemDiagnostic(err, message, location)
}

func candidatesFromMap(candidates map[string]selectedImportCandidate) []selectedImportCandidate {
	paths := make([]string, 0, len(candidates))
	for candidatePath := range candidates {
		paths = append(paths, candidatePath)
	}
	slices.Sort(paths)
	result := make([]selectedImportCandidate, 0, len(paths))
	for _, candidatePath := range paths {
		result = append(result, candidates[candidatePath])
	}
	return result
}

func validGlobPattern(pattern string) ([]string, bool) {
	if !validImportPath(pattern) {
		return nil, false
	}
	segments := strings.Split(pattern, "/")
	for _, segment := range segments {
		if segment == "" || hasRecursiveStars(segment) {
			return nil, false
		}
		if _, err := path.Match(segment, ""); err != nil {
			return nil, false
		}
	}
	return segments, true
}

func hasRecursiveStars(segment string) bool {
	inClass := false
	for index := 0; index < len(segment); index++ {
		switch segment[index] {
		case '\\':
			index++
		case '[':
			inClass = true
		case ']':
			inClass = false
		case '*':
			if !inClass && index+1 < len(segment) && segment[index+1] == '*' {
				return true
			}
		}
	}
	return false
}

func globLiteralSegment(segment string) (string, bool) {
	var literal strings.Builder
	for index := 0; index < len(segment); index++ {
		switch segment[index] {
		case '*', '?', '[':
			return "", true
		case '\\':
			index++
			if index < len(segment) {
				literal.WriteByte(segment[index])
			}
		default:
			literal.WriteByte(segment[index])
		}
	}
	return literal.String(), false
}

func globFilesystemDiagnostic(err error, message string, location Location) Diagnostic {
	code := CodeImportRead
	if errors.Is(err, fs.ErrPermission) {
		code = CodeImportPermission
	}
	return importDiagnostic(code, message, &location)
}