This commit is contained in:
ron 2021-06-26 06:17:14 +02:00
parent ff02aa8a09
commit c817edc7a1

408
main.go
View File

@ -19,122 +19,118 @@ import (
"time"
"github.com/gocolly/colly"
"github.com/ipfs/go-cid"
ipfs "github.com/ipfs/go-ipfs-api"
"github.com/ipfs/ipfs-cluster/api"
cluster "github.com/ipfs/ipfs-cluster/api/rest/client"
"github.com/multiformats/go-multiaddr"
"github.com/schollz/progressbar/v3"
)
type Cache struct {
MostRecent string
Pages int
CID string
}
var cacheFile = flag.String("cache", "~/.taldl/cache.json", "cache file")
var cacheFile = "~/.taldl.json"
var outputDir = flag.String("output", "~/TAL", "output directory")
var tmpDir = flag.String("tmpdir", "~/.taldl", "tmp directory")
var formats = flag.String("formats", "zip,epub,pdf,a4.pdf,lt.pdf", "formats to download")
var progress = flag.Bool("progress", true, "show progress bar")
var verbose = flag.Bool("verbose", true, "verbose output")
var workers = flag.Int("workers", 5, "amount of workers")
var update = flag.Bool("update", false, "update all entries")
var formats = flag.String("formats", "zip,epub,pdf,a4.pdf,lt.pdf", "formats to download.")
var fullUpdate = flag.Bool("full", false, "check everything for modifications")
var verbose = flag.Bool("verbose", false, "verbose")
var workers = flag.Int("workers", 1, "amount of workers")
var ipfsAPI = flag.String("ipfs-api", "localhost:5001", "ipfs api")
var ipfsEnabled = flag.Bool("ipfs", false, "pin to ipfs")
var ipfsClusterEnabled = flag.Bool("ipfs-cluster", false, "pin to ipfs-cluster")
var ipfsClusterAPI = flag.String("ipfs-cluster-api", "/ip4/127.0.0.1/tcp/9094", "ipfs-cluster api")
var hrefs = []string{}
var hrefsMutex sync.Mutex
var mostRecent string
var done bool
var cache Cache
var useFormat map[string]bool
func main() {
var wg sync.WaitGroup
var cache Cache
var last int
var bar *progressbar.ProgressBar
var lastPage int
var hrefs = []string{}
var hrefsMutex sync.Mutex
var done bool
flag.Parse()
*cacheFile = fixpath(*cacheFile)
cacheData, _ := ioutil.ReadFile(*cacheFile)
if *verbose {
fmt.Printf("writing to %s\r\n", *outputDir)
}
useFormat = make(map[string]bool)
fmts := strings.Split(*formats, ",")
for _, f := range fmts {
useFormat[f] = true
}
cacheFile = fixPath(cacheFile)
cacheData, _ := ioutil.ReadFile(cacheFile)
_ = json.Unmarshal(cacheData, &cache)
*outputDir = fixpath(*outputDir)
*tmpDir = fixpath(*tmpDir)
*outputDir = fixPath(*outputDir)
os.MkdirAll(*outputDir, os.ModePerm)
os.MkdirAll(*tmpDir, os.ModePerm)
c := colly.NewCollector()
c.OnHTML("ul.pagination li:nth-last-child(2)", func(e *colly.HTMLElement) {
if last == 0 {
last, _ = strconv.Atoi(strings.TrimSpace(e.Text))
}
wg.Done()
lastPage, _ = strconv.Atoi(strings.TrimSpace(e.Text))
})
c.OnHTML("div.amw-listing-item:nth-child(1) a[href]", func(e *colly.HTMLElement) {
href := e.Attr("href")
mostRecent = href
})
c.OnHTML("div.amw-listing-item a[href]", func(e *colly.HTMLElement) {
if done {
return
}
href := e.Attr("href")
hrefsMutex.Lock()
if len(hrefs) == 0 {
mostRecent = href
}
if cache.MostRecent == href && *fullUpdate == false {
if cache.MostRecent == href {
done = true
} else {
}
if !done || *update {
hrefs = append(hrefs, href)
}
hrefsMutex.Unlock()
})
bar := progressbar.Default(1)
wg.Add(1)
c.Visit("https://theanarchistlibrary.org/latest")
wg.Wait()
url := "https://theanarchistlibrary.org/latest/1"
c.Visit(url)
newPages := 0
newPages := 1
if *fullUpdate {
newPages = last - 1
if *update {
newPages = lastPage
} else {
newPages = last - cache.Pages - 1
newPages = lastPage - cache.Pages
}
bar.ChangeMax(1 + newPages)
bar.Add(1)
for i := 0; i < newPages; i++ {
if done {
bar.Add(newPages - i)
break
if newPages > 0 {
if *verbose {
fmt.Fprintf(os.Stderr, "Checking latest entries... \r\n")
}
wg.Add(1)
for {
err := c.Visit(fmt.Sprintf("https://theanarchistlibrary.org/latest/%d", i+2))
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s", err)
} else {
break
}
if *progress {
bar = progressbar.Default(int64(newPages))
bar.Add(1)
}
bar.Add(1)
}
wg.Wait()
cache.Pages = last
scanJobs := make(chan int, newPages)
scanResults := make(chan []string, newPages)
for w := 1; w <= *workers; w++ {
go scanner(w, scanJobs, scanResults)
}
for a := 2; a <= newPages; a++ {
scanJobs <- a
}
close(scanJobs)
for a := 2; a <= newPages; a++ {
result := <-scanResults
hrefsMutex.Lock()
hrefs = append(hrefs, result...)
hrefsMutex.Unlock()
if *progress {
bar.Add(1)
}
}
close(scanResults)
cache.Pages = lastPage
cache.MostRecent = mostRecent
numJobs := len(hrefs)
@ -144,8 +140,14 @@ func main() {
return
}
bar.Reset()
bar.ChangeMax(numJobs)
if *verbose {
fmt.Fprintf(os.Stderr, "Checking %d entries for updates...\r\n", numJobs)
}
if *progress {
bar = progressbar.Default(int64(numJobs))
bar.ChangeMax(numJobs)
}
checkJobs := make(chan string, numJobs)
checkResults := make(chan string, numJobs)
@ -161,94 +163,59 @@ func main() {
}
close(checkJobs)
for w := 1; w <= *workers; w++ {
go downloader(w, downloadJobs, downloadResults)
}
for a := 1; a <= numJobs; a++ {
r := <-checkResults
bar.Add(1)
if *progress {
bar.Add(1)
}
if r != "" {
downloadCount++
downloadJobs <- r
downloadCount++
}
}
close(checkResults)
close(downloadJobs)
bar.Finish()
if downloadCount == 0 {
save(cache)
return
}
bar.Reset()
bar.ChangeMax(downloadCount)
if *verbose {
fmt.Fprintf(os.Stderr, "Downloading %d entries...\r\n", downloadCount)
}
if *progress {
bar = progressbar.Default(int64(downloadCount))
}
for w := 1; w <= *workers; w++ {
go downloader(w, downloadJobs, downloadResults)
}
buffer := ""
for a := 1; a <= downloadCount; a++ {
r := <-downloadResults
bar.Add(1)
if *progress {
bar.Add(1)
}
if r != "" {
buffer += r
}
}
close(downloadResults)
bar.Finish()
save(cache)
if buffer != "" {
fmt.Fprintln(os.Stderr, buffer)
}
if *ipfsEnabled {
sh := ipfs.NewShell(*ipfsAPI)
newCidStr, err := sh.AddDir(*outputDir)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s", err)
os.Exit(1)
if *verbose {
fmt.Fprintln(os.Stderr, buffer)
}
if cache.CID != "" && cache.CID != newCidStr {
sh.Unpin(cache.CID)
}
if *ipfsClusterEnabled {
ma, err := multiaddr.NewMultiaddr(*ipfsClusterAPI)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s", err)
os.Exit(1)
}
sh, err := cluster.NewDefaultClient(&cluster.Config{APIAddr: ma})
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s", err)
os.Exit(1)
}
newCid, err := cid.Parse(newCidStr)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s", err)
os.Exit(1)
}
sh.Pin(context.TODO(), newCid, api.PinOptions{})
if cache.CID != "" {
oldCid, err := cid.Parse(cache.CID)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s", err)
os.Exit(1)
}
sh.Unpin(context.TODO(), oldCid)
}
}
cache.CID = newCidStr
save(cache)
fmt.Fprintln(os.Stderr, newCidStr)
}
}
func fixpath(path string) string {
func fixPath(path string) string {
path = filepath.FromSlash(path)
if (path)[0] == '~' {
user, err := user.Current()
@ -266,7 +233,7 @@ func filenameForUrl(url string) string {
return path.Base(url)
}
func head(url string, path string, timeout time.Duration) (modified bool, err error) {
func check(url string, path string, timeout time.Duration) (modified bool, err error) {
request, err := http.NewRequest("HEAD", url, nil)
if err != nil {
return
@ -312,7 +279,7 @@ func head(url string, path string, timeout time.Duration) (modified bool, err er
return true, nil
}
if file.ModTime().Before(lm) {
if lm.After(file.ModTime()) {
return true, nil
}
@ -336,6 +303,10 @@ func get(url string, timeout time.Duration) (content []byte, err error) {
}
defer response.Body.Close()
if response.StatusCode == 404 {
return
}
if response.StatusCode != 200 {
return nil, fmt.Errorf("%s: %s", url, response.Status)
}
@ -353,13 +324,16 @@ func fileExists(path string) bool {
func download(url string, path string) (err error) {
for {
data, err := get(url, time.Second*10)
data, err := get(url, time.Second*60)
if err != nil {
return err
fmt.Fprintf(os.Stderr, "error: %s: %s\n", url, err)
time.Sleep(time.Second)
continue
}
err = ioutil.WriteFile(path, data, 0644)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s\n", err)
return err
}
break
@ -418,72 +392,18 @@ func unzip(src string, dest string) ([]string, error) {
}
func downloader(id int, jobs <-chan string, results chan<- string) {
for href := range jobs {
result := ""
dir := filenameForUrl(href)
dest := filepath.Join(*outputDir, dir)
os.MkdirAll(dest, 0700)
result := ""
ext := "zip"
url := href + "." + ext
name := filenameForUrl(url)
path := filepath.Join(*tmpDir, name)
err := download(url, path)
if err != nil {
result += err.Error() + "\r\n"
}
_, err = unzip(path, *outputDir)
if err != nil {
result += err.Error() + "\r\n"
}
os.Remove(path)
ext = "epub"
if strings.Contains(*formats, ext) {
url = href + "." + ext
name = filenameForUrl(url)
path = filepath.Join(dest, name)
err = download(url, path)
if err != nil {
result += err.Error() + "\r\n"
}
}
ext = "pdf"
if strings.Contains(*formats, ext) {
url = href + "." + ext
name = filenameForUrl(url)
path = filepath.Join(dest, name)
err = download(url, path)
if err != nil {
result += err.Error() + "\r\n"
}
}
ext = "a4.pdf"
if strings.Contains(*formats, ext) {
url = href + "." + ext
name = filenameForUrl(url)
path = filepath.Join(dest, name)
err = download(url, path)
if err != nil {
result += err.Error() + "\r\n"
}
}
ext = "lt.pdf"
if strings.Contains(*formats, ext) {
url = href + "." + ext
name = filenameForUrl(url)
path = filepath.Join(dest, name)
err = download(url, path)
if err != nil {
result += err.Error() + "\r\n"
}
}
downloadFormat(href, "zip", dest)
downloadFormat(href, "epub", dest)
downloadFormat(href, "pdf", dest)
downloadFormat(href, "a4.pdf", dest)
downloadFormat(href, "lt.pdf", dest)
if *verbose {
results <- result
@ -493,19 +413,63 @@ func downloader(id int, jobs <-chan string, results chan<- string) {
}
}
func scanner(id int, jobs <-chan int, results chan<- []string) {
var result []string
c := colly.NewCollector()
c.OnHTML("div.amw-listing-item a[href]", func(e *colly.HTMLElement) {
href := e.Attr("href")
if cache.MostRecent == href {
done = true
}
if !done {
result = append(result, href)
}
})
c.AllowURLRevisit = true
for i := range jobs {
url := fmt.Sprintf("https://theanarchistlibrary.org/latest/%d", i)
for {
result = []string{}
err := c.Visit(url)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s\n", err)
time.Sleep(time.Second)
} else {
break
}
}
results <- result
}
}
func checker(id int, jobs <-chan string, results chan<- string) {
var modified bool
var err error
for href := range jobs {
ext := "muse"
url := href + "." + ext
dir := filenameForUrl(href)
name := filenameForUrl(url)
path := filepath.Join(*outputDir, dir, name)
modified, err := head(url, path, time.Second*10)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s", err)
results <- ""
if !fileExists(path) {
results <- href
continue
}
for {
modified, err = check(url, path, time.Second*30)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s\n", err)
time.Sleep(time.Second)
continue
}
break
}
if modified {
results <- href
} else {
@ -516,5 +480,43 @@ func checker(id int, jobs <-chan string, results chan<- string) {
func save(cache Cache) {
cacheData, _ := json.Marshal(&cache)
_ = ioutil.WriteFile(*cacheFile, cacheData, 0644)
_ = ioutil.WriteFile(cacheFile, cacheData, 0644)
}
func downloadFormat(href string, ext string, dest string) error {
if !useFormat[ext] {
return nil
}
url := href + "." + ext
name := filenameForUrl(url)
if ext == "zip" {
tmpDir, err := ioutil.TempDir(os.TempDir(), "taldl")
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s\n", err)
return err
}
path := filepath.Join(tmpDir, name)
err = download(url, path)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s\n", err)
return err
}
_, err = unzip(path, *outputDir)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s\n", err)
return err
}
os.RemoveAll(tmpDir)
} else {
path := filepath.Join(dest, name)
err := download(url, path)
if err != nil {
fmt.Fprintf(os.Stderr, "error: %s\n", err)
return err
}
}
return nil
}