DateKeys-App/scripts/zip-go-read.go

//go:build ignore

// Prints src/lib/dkc/testing/zip-vectors.json: how archive/zip of the Go
// standard library reads the ZIP samples that the page writes (plan of
// format 3 in datekeys-ts, step 5), which scripts/zip-ts-samples.mjs wrote
// into a directory. For each archive it records the SHA-256 of its bytes,
// the number of its entries and, for each entry, a line with its name,
// whether bit 11 marks the name as UTF-8, its method, its CRC-32, its sizes,
// the time archive/zip reads from its extra fields, in UTC, and the SHA-256
// of its content, read with the CRC-32 checked. It keeps the lines of an
// archive of 20 entries at most, and the SHA-256 of all of them joined by
// newlines for any archive, so that the archive of 65535 entries, which is
// ZIP64 by its number of entries, weighs little. zipsink.test.ts writes each
// archive again, requires its SHA-256, and computes the lines it must give.
//
// It needs only the standard library:
//
//	node scripts/zip-ts-samples.mjs DIR > zip-samples.json
//	go run zip-go-read.go DIR zip-samples.json > zip-vectors.json
package main

import (
	"archive/zip"
	"bytes"
	"crypto/sha256"
	"encoding/hex"
	"encoding/json"
	"fmt"
	"io"
	"os"
	"path/filepath"
	"runtime"
	"strings"
	"time"
)

type sampleIn struct {
	Name   string `json:"name"`
	File   string `json:"file"`
	SHA256 string `json:"sha256"`
}

type sampleOut struct {
	Name        string   `json:"name"`
	SHA256      string   `json:"sha256"`
	Count       int      `json:"count"`
	LinesSHA256 string   `json:"lines_sha256"`
	Lines       []string `json:"lines,omitempty"`
	Error       string   `json:"error,omitempty"`
}

type output struct {
	Description      string      `json:"description"`
	Generator        string      `json:"generator"`
	SamplesGenerator string      `json:"samples_generator"`
	Go               string      `json:"go"`
	Samples          []sampleOut `json:"samples"`
}

func must[T any](v T, err error) T {
	if err != nil {
		panic(err)
	}
	return v
}

// read reads an archive with archive/zip, and gives one line for each entry.
func read(b []byte) ([]string, error) {
	r, err := zip.NewReader(bytes.NewReader(b), int64(len(b)))
	if err != nil {
		return nil, err
	}
	var lines []string
	for _, f := range r.File {
		rc, err := f.Open()
		if err != nil {
			return nil, fmt.Errorf("%s: %w", f.Name, err)
		}
		data, err := io.ReadAll(rc)
		rc.Close()
		if err != nil {
			return nil, fmt.Errorf("%s: %w", f.Name, err)
		}
		sum := sha256.Sum256(data)
		lines = append(lines, fmt.Sprintf("%s\t%t\t%d\t%08x\t%d\t%d\t%s\t%s",
			f.Name, f.Flags&0x800 != 0, f.Method, f.CRC32, f.UncompressedSize64, f.CompressedSize64,
			f.Modified.UTC().Format(time.RFC3339), hex.EncodeToString(sum[:])))
	}
	return lines, nil
}

func main() {
	dir := os.Args[1]
	var in struct {
		Generator string     `json:"generator"`
		Samples   []sampleIn `json:"samples"`
	}
	if err := json.Unmarshal(must(os.ReadFile(os.Args[2])), &in); err != nil {
		panic(err)
	}
	out := output{
		Description:      "How archive/zip of Go reads the ZIP samples of the page; see the header of scripts/zip-go-read.go.",
		Generator:        "scripts/zip-go-read.go",
		SamplesGenerator: in.Generator,
		Go:               runtime.Version(),
	}
	for _, s := range in.Samples {
		b := must(os.ReadFile(filepath.Join(dir, s.File)))
		sum := sha256.Sum256(b)
		so := sampleOut{Name: s.Name, SHA256: hex.EncodeToString(sum[:])}
		if so.SHA256 != s.SHA256 {
			panic(fmt.Sprintf("%s: the file is not the sample written", s.Name))
		}
		lines, err := read(b)
		if err != nil {
			so.Error = err.Error()
		}
		so.Count = len(lines)
		all := sha256.Sum256([]byte(strings.Join(lines, "\n")))
		so.LinesSHA256 = hex.EncodeToString(all[:])
		if len(lines) <= 20 {
			so.Lines = lines
		}
		out.Samples = append(out.Samples, so)
	}
	e := json.NewEncoder(os.Stdout)
	e.SetEscapeHTML(false)
	e.SetIndent("", " ")
	if err := e.Encode(out); err != nil {
		panic(err)
	}
}