10 Commits

Author SHA1 Message Date
43392a4132 fix: normalize List.aspx?folderID= to fragment form for yt-dlp
Canvas stores folder links as List.aspx?folderID=X (query param).
yt-dlp's PanoptoList extractor requires List.aspx#folderID="X"
(fragment with quoted ID) to scope the download to that folder.
Without the fragment form it downloaded the entire Panopto instance
(1806 items instead of 3).

Also drop --no-playlist for list URLs since they are intentional
playlists, and use title/%(title)s.%(ext)s output template for them.
2026-05-16 19:27:36 +02:00
13063c6cc5 add video-only mode 2026-05-16 17:33:40 +02:00
joren
14a71e7dca Update Readme 2026-03-11 20:12:08 +01:00
051b67de51 Merge pull request 'impl-me' (#4) from impl-me into main
Reviewed-on: #4
2026-03-11 20:10:55 +01:00
joren
9691ecd7a5 Fix course files 2026-03-11 20:06:46 +01:00
joren
b8e6180b35 add numbering 2026-03-11 19:56:37 +01:00
joren
8591ae283e Add support to download all the users courses 2026-03-11 19:36:54 +01:00
11d9867155 Merge pull request 'Prevent dupes in fo mode' (#3) from files-only into main
Reviewed-on: #3
2026-03-11 17:52:41 +01:00
9ffe1283c2 Merge branch 'main' into files-only 2026-03-11 17:52:31 +01:00
05ed4dd4ed Prevent dupes in fo mode 2026-03-11 17:51:34 +01:00
5 changed files with 178 additions and 37 deletions

View File

@@ -21,24 +21,31 @@ go build -o canvasarchiver ./cmd/canvasarchiver
./canvasarchiver
```
Or
Or for files-only mode:
```bash
./canvasarchiver -fo
```
For files-only mode.
2. On first run, you'll be prompted to authenticate:
- Visit the provided OAuth URL
- Authorize the application
- Copy the authorization code back to the terminal
3. Enter your Course ID when prompted
3. Enter your Course ID when prompted (or use `-me` to download all enrolled courses)
4. The tool will download:
- Regular course files (to `Course Files/`)
- Module content (to `Modules/`)
- Panopto recordings (to `Recordings/`)
### Flags
| Flag | Description |
|------|-------------|
| `-fo` | Files only mode - download all files to a single directory without module structure |
| `-me` | Download all enrolled courses |
| `-n` | Prefix modules with order numbers `[1]`, `[2]`, etc. |
## Configuration

View File

@@ -14,6 +14,9 @@ import (
func main() {
filesOnly := flag.Bool("fo", false, "Files only mode - download all files to a single directory without module structure")
videosOnly := flag.Bool("vo", false, "Videos only mode - download only Panopto videos to a single directory")
me := flag.Bool("me", false, "Download all enrolled courses")
moduleNumbers := flag.Bool("n", false, "Prefix modules with order numbers [1], [2], etc.")
flag.Parse()
httpClient := &http.Client{}
@@ -25,11 +28,31 @@ func main() {
return
}
if *me {
canvasClient := canvas.NewClient(httpClient, accessToken, "", *filesOnly, *videosOnly, *moduleNumbers)
courses, err := canvasClient.GetEnrolledCourses()
if err != nil {
fmt.Printf("Error fetching courses: %v\n", err)
return
}
fmt.Printf("[+] Found %d enrolled courses\n", len(courses))
for _, course := range courses {
fmt.Printf(" -> Downloading: %s (ID: %d)\n", course.Name, course.ID)
downloadCourse(httpClient, accessToken, fmt.Sprintf("%d", course.ID), *filesOnly, *videosOnly, *moduleNumbers)
}
return
}
var courseID string
fmt.Print("Enter Course ID: ")
fmt.Scanln(&courseID)
canvasClient := canvas.NewClient(httpClient, accessToken, courseID, *filesOnly)
downloadCourse(httpClient, accessToken, courseID, *filesOnly, *videosOnly, *moduleNumbers)
}
func downloadCourse(httpClient *http.Client, accessToken, courseID string, filesOnly, videosOnly, moduleNumbers bool) {
canvasClient := canvas.NewClient(httpClient, accessToken, courseID, filesOnly, videosOnly, moduleNumbers)
if err := canvasClient.GetCourseInfo(); err != nil {
fmt.Printf("Error: %v\n", err)
@@ -44,7 +67,8 @@ func main() {
canvasClient.DownloadModules(courseRoot)
if !*filesOnly {
panopto.DownloadMainRecordings(httpClient, accessToken, courseID, courseRoot)
// Run recordings: always in -vo mode; skipped in -fo mode; normal otherwise.
if videosOnly || !filesOnly {
panopto.DownloadMainRecordings(httpClient, accessToken, courseID, courseRoot, videosOnly)
}
}

View File

@@ -22,14 +22,20 @@ type Client struct {
CourseID string
CourseName string
FilesOnly bool
VideosOnly bool
ModuleNumbers bool
downloadedFiles map[string]bool
}
func NewClient(httpClient *http.Client, accessToken, courseID string, filesOnly bool) *Client {
func NewClient(httpClient *http.Client, accessToken, courseID string, filesOnly, videosOnly, moduleNumbers bool) *Client {
return &Client{
HTTPClient: httpClient,
AccessToken: accessToken,
CourseID: courseID,
FilesOnly: filesOnly,
VideosOnly: videosOnly,
ModuleNumbers: moduleNumbers,
downloadedFiles: make(map[string]bool),
}
}
@@ -49,7 +55,25 @@ func (c *Client) GetCourseInfo() error {
return nil
}
func (c *Client) GetEnrolledCourses() ([]models.Course, error) {
req, _ := http.NewRequest("GET", fmt.Sprintf("%s/api/v1/courses?enrollment_state=active&per_page=100", config.BaseURL), nil)
req.Header.Set("Authorization", "Bearer "+c.AccessToken)
resp, err := c.HTTPClient.Do(req)
if err != nil {
return nil, err
}
defer resp.Body.Close()
var courses []models.Course
json.NewDecoder(resp.Body).Decode(&courses)
return courses, nil
}
func (c *Client) DownloadCourseFiles(root string) {
if c.VideosOnly {
fmt.Println("\n[*] Skipping regular course files (videos only mode)")
return
}
fmt.Println("\n[*] Fetching regular course files...")
fReq, _ := http.NewRequest("GET", fmt.Sprintf("%s/api/v1/courses/%s/folders?per_page=100", config.BaseURL, c.CourseID), nil)
@@ -82,12 +106,23 @@ func (c *Client) DownloadCourseFiles(root string) {
fileCount := 0
for _, file := range files {
if c.FilesOnly {
if c.downloadedFiles[file.DisplayName] {
continue
}
c.downloadedFiles[file.DisplayName] = true
}
rawFolderPath := folderMap[file.FolderID]
safeFolderPath := utils.SanitizePath(rawFolderPath)
subDir := root
if !c.FilesOnly {
if safeFolderPath != "" && strings.ToLower(safeFolderPath) != "course files" {
subDir = filepath.Join(root, "Course Files", safeFolderPath)
} else {
subDir = filepath.Join(root, "Course Files")
}
}
os.MkdirAll(subDir, 0o755)
path := filepath.Join(subDir, utils.Sanitize(file.DisplayName))
@@ -129,15 +164,25 @@ func (c *Client) DownloadModules(courseRoot string) {
json.NewDecoder(resp.Body).Decode(&modules)
resp.Body.Close()
for _, mod := range modules {
for i, mod := range modules {
modName := mod.Name
if c.ModuleNumbers {
modName = fmt.Sprintf("[%d] %s", i+1, mod.Name)
}
// In videos-only mode everything goes flat into courseRoot.
// In files-only mode everything goes flat into courseRoot.
// Otherwise use the structured Modules/<name> path.
modBaseDir := courseRoot
if !c.FilesOnly {
modBaseDir = filepath.Join(courseRoot, "Modules", utils.Sanitize(mod.Name))
if !c.FilesOnly && !c.VideosOnly {
modBaseDir = filepath.Join(courseRoot, "Modules", utils.Sanitize(modName))
}
os.MkdirAll(modBaseDir, 0o755)
if !c.FilesOnly {
fmt.Printf("\n[Module] %s\n", mod.Name)
if !c.FilesOnly && !c.VideosOnly {
fmt.Printf("\n[Module] %s\n", modName)
} else if c.VideosOnly {
fmt.Printf("\n[Module] %s (scanning for videos)\n", modName)
}
subHeaderStack := []string{}
@@ -145,15 +190,16 @@ func (c *Client) DownloadModules(courseRoot string) {
for _, item := range mod.Items {
// In videos-only mode always download to the flat courseRoot.
targetDir := modBaseDir
if len(subHeaderStack) > 0 && !c.FilesOnly {
if len(subHeaderStack) > 0 && !c.FilesOnly && !c.VideosOnly {
targetDir = filepath.Join(modBaseDir, filepath.Join(subHeaderStack...))
}
os.MkdirAll(targetDir, 0o755)
switch item.Type {
case "SubHeader":
if c.FilesOnly {
if c.FilesOnly || c.VideosOnly {
continue
}
currentIndent := item.Indent
@@ -173,6 +219,9 @@ func (c *Client) DownloadModules(courseRoot string) {
fmt.Printf("%s--- %s ---\n", indent, item.Title)
case "File":
if c.VideosOnly {
continue
}
c.downloadModuleFile(item, targetDir)
case "ExternalTool":
@@ -223,11 +272,18 @@ func (c *Client) downloadModuleFile(item models.ModuleItem, dir string) {
return
}
if c.FilesOnly {
if c.downloadedFiles[fileMeta.DisplayName] {
return
}
c.downloadedFiles[fileMeta.DisplayName] = true
}
ext := filepath.Ext(fileMeta.DisplayName)
origBase := strings.TrimSuffix(fileMeta.DisplayName, ext)
fileName := fileMeta.DisplayName
if !strings.EqualFold(origBase, item.Title) && item.Title != "" {
if !c.FilesOnly && !strings.EqualFold(origBase, item.Title) && item.Title != "" {
fileName = fmt.Sprintf("%s (%s)%s", origBase, item.Title, ext)
}

View File

@@ -10,6 +10,7 @@ type TokenResponse struct {
}
type Course struct {
ID int `json:"id"`
Name string `json:"name"`
}

View File

@@ -17,6 +17,27 @@ import (
"git.directme.in/Joren/CanvasArchiver/internal/utils"
)
// normalizePanoptoURL converts the query-param form that Canvas stores
// (List.aspx?folderID=X) to the fragment form that yt-dlp's PanoptoList
// extractor understands (List.aspx#folderID="X"). Without this yt-dlp
// ignores the folder filter and downloads the entire Panopto instance.
func normalizePanoptoURL(rawURL string) string {
parsed, err := url.Parse(rawURL)
if err != nil {
return rawURL
}
if strings.Contains(parsed.Path, "List.aspx") {
folderID := parsed.Query().Get("folderID")
if folderID != "" {
// Strip query, set fragment: List.aspx#folderID="<id>"
parsed.RawQuery = ""
parsed.Fragment = fmt.Sprintf(`folderID="%s"`, folderID)
return parsed.String()
}
}
return rawURL
}
func getYoutubeDLCommand() string {
exePath, err := os.Executable()
if err == nil {
@@ -192,13 +213,35 @@ func DownloadVideo(httpClient *http.Client, accessToken, courseID, modDir, input
fmt.Printf(" [*] Downloading video: %s\n", title)
ytCmd := getYoutubeDLCommand()
cmd := exec.Command(ytCmd,
// Normalize folder URLs so yt-dlp scopes to the right folder.
normalizedURL := normalizePanoptoURL(targetURL)
// Folder/list URLs are intentional playlists; don't pass --no-playlist.
isList := strings.Contains(normalizedURL, "List.aspx")
var outputTpl string
var args []string
if isList {
outputTpl = utils.Sanitize(title) + "/%(title)s.%(ext)s"
args = []string{
"--cookies", cookieFile,
"--referer", config.BaseURL + "/",
"-P", modDir,
"-o", outputTpl,
normalizedURL,
}
} else {
outputTpl = utils.Sanitize(title) + ".%(ext)s"
args = []string{
"--no-playlist",
"--cookies", cookieFile,
"--referer", config.BaseURL + "/",
"-P", modDir,
"-o", utils.Sanitize(title)+".%(ext)s",
targetURL)
"-o", outputTpl,
normalizedURL,
}
}
cmd := exec.Command(ytCmd, args...)
cmd.Stdout = os.Stdout
cmd.Stderr = os.Stderr
@@ -212,7 +255,7 @@ func DownloadVideo(httpClient *http.Client, accessToken, courseID, modDir, input
}
}
func DownloadMainRecordings(httpClient *http.Client, accessToken, courseID, root string) {
func DownloadMainRecordings(httpClient *http.Client, accessToken, courseID, root string, videosOnly bool) {
fmt.Println("\n[*] Checking for main Panopto recordings...")
jar, _ := cookiejar.New(nil)
@@ -300,8 +343,18 @@ func DownloadMainRecordings(httpClient *http.Client, accessToken, courseID, root
}
os.WriteFile(cookieFile, []byte(cData), 0o644)
recordingsDir := filepath.Join(root, "Recordings")
os.MkdirAll(recordingsDir, 0o755)
var downloadDir string
var outputTemplate string
if videosOnly {
// Flat: all videos go directly into course root
downloadDir = root
outputTemplate = "%(title)s.%(ext)s"
} else {
downloadDir = filepath.Join(root, "Recordings")
os.MkdirAll(downloadDir, 0o755)
outputTemplate = "%(playlist_title)s/%(title)s.%(ext)s"
}
fmt.Println("[*] Starting yt-dlp download for main recordings...")
@@ -310,8 +363,8 @@ func DownloadMainRecordings(httpClient *http.Client, accessToken, courseID, root
cmd := exec.Command(ytCmd,
"--cookies", cookieFile,
"--referer", config.BaseURL+"/",
"-P", recordingsDir,
"-o", "%(playlist_title)s/%(title)s.%(ext)s",
"-P", downloadDir,
"-o", outputTemplate,
decodedURL,
)
cmd.Stdout = os.Stdout