logpile/internal/parser/archive.go

package parser

import (
	"archive/tar"
	"archive/zip"
	"compress/gzip"
	"fmt"
	"io"
	"os"
	"path/filepath"
	"strings"
)

// ExtractedFile represents a file extracted from archive
type ExtractedFile struct {
	Path    string
	Content []byte
}

// ExtractArchive extracts tar.gz or zip archive and returns file contents
func ExtractArchive(archivePath string) ([]ExtractedFile, error) {
	ext := strings.ToLower(filepath.Ext(archivePath))

	switch ext {
	case ".gz", ".tgz":
		return extractTarGz(archivePath)
	case ".zip":
		return extractZip(archivePath)
	default:
		return nil, fmt.Errorf("unsupported archive format: %s", ext)
	}
}

// ExtractArchiveFromReader extracts archive from reader
func ExtractArchiveFromReader(r io.Reader, filename string) ([]ExtractedFile, error) {
	ext := strings.ToLower(filepath.Ext(filename))

	switch ext {
	case ".gz", ".tgz":
		return extractTarGzFromReader(r)
	default:
		return nil, fmt.Errorf("unsupported archive format: %s", ext)
	}
}

func extractTarGz(archivePath string) ([]ExtractedFile, error) {
	f, err := os.Open(archivePath)
	if err != nil {
		return nil, fmt.Errorf("open archive: %w", err)
	}
	defer f.Close()

	return extractTarGzFromReader(f)
}

func extractTarGzFromReader(r io.Reader) ([]ExtractedFile, error) {
	gzr, err := gzip.NewReader(r)
	if err != nil {
		return nil, fmt.Errorf("gzip reader: %w", err)
	}
	defer gzr.Close()

	tr := tar.NewReader(gzr)
	var files []ExtractedFile

	for {
		header, err := tr.Next()
		if err == io.EOF {
			break
		}
		if err != nil {
			return nil, fmt.Errorf("tar read: %w", err)
		}

		// Skip directories
		if header.Typeflag == tar.TypeDir {
			continue
		}

		// Skip large files (>10MB)
		if header.Size > 10*1024*1024 {
			continue
		}

		content, err := io.ReadAll(tr)
		if err != nil {
			return nil, fmt.Errorf("read file %s: %w", header.Name, err)
		}

		files = append(files, ExtractedFile{
			Path:    header.Name,
			Content: content,
		})
	}

	return files, nil
}

func extractZip(archivePath string) ([]ExtractedFile, error) {
	r, err := zip.OpenReader(archivePath)
	if err != nil {
		return nil, fmt.Errorf("open zip: %w", err)
	}
	defer r.Close()

	var files []ExtractedFile

	for _, f := range r.File {
		if f.FileInfo().IsDir() {
			continue
		}

		// Skip large files (>10MB)
		if f.FileInfo().Size() > 10*1024*1024 {
			continue
		}

		rc, err := f.Open()
		if err != nil {
			return nil, fmt.Errorf("open file %s: %w", f.Name, err)
		}

		content, err := io.ReadAll(rc)
		rc.Close()
		if err != nil {
			return nil, fmt.Errorf("read file %s: %w", f.Name, err)
		}

		files = append(files, ExtractedFile{
			Path:    f.Name,
			Content: content,
		})
	}

	return files, nil
}

// FindFileByPattern finds files matching pattern in extracted files
func FindFileByPattern(files []ExtractedFile, patterns ...string) []ExtractedFile {
	var result []ExtractedFile
	for _, f := range files {
		for _, pattern := range patterns {
			if strings.Contains(strings.ToLower(f.Path), strings.ToLower(pattern)) {
				result = append(result, f)
				break
			}
		}
	}
	return result
}

// FindFileByName finds file by exact name (case-insensitive)
func FindFileByName(files []ExtractedFile, name string) *ExtractedFile {
	for _, f := range files {
		if strings.EqualFold(filepath.Base(f.Path), name) {
			return &f
		}
	}
	return nil
}