Development

1. Clone ilias and ilias-cli to ~/go/src/github.com/krakowski/
2. cd to ilias-cli
3. GOOS=linux go get
4. GOOS=linux GOARCH=amd64 go build -o ilias-linux

Workflow

Admin: Distribute Submissions

You should start with an empty directory for each exercise sheet because workspace.yml is overriden when fetching submissions for an exercise sheet.

1. Create a file .tutors.yml:

- id: hehei001
  hours: 10
- id: hehei002
  hours: 9

2. Create CORRECTION.tmpl:

student: {{.Student}}
points: [0,0,0,0]
corrected: false
correction: |

  <b>Rückfragen bitte unter Angabe der Korrekturnummer '{{.Student}}' und der Blattnummer an <a href="mailto:{{.Tutor}}@hhu.de?subject=RA-Abgabe {{.Student}} Blatt n">{{.Tutor}}@hhu.de</a></b>

  Maximal erreichbare Punkte: 10+5+5+5

3. In Ilias, deactivate displaying the profile picture in “Abgaben und Noten”, activate “Abgegeben am” (such that the first three columns are “Name”, “Benutzername”, “Abgegeben am”; otherwise you'll get runtime error: index out of range [1] with length 1).
4. Distribute submissions using the ref_id (not Object ID) of the Übungsobjekt and ass_id of the Übungseinheit:

./ilias-linux exercises distribute 1005951 22122

5. Run the download script:

# copies workspace.yml to korrektorkennung/BlattNN/ and runs  workspace init <korrektorkennung>  for each corrector; for submissions which are no single pdf file, sets default feedback and correction status
./download_submissions.bash BlattNN

6. Upload the generated folders to sciebo using webdav + merge with existing folders.

Corrector Workflow

1. Download the folder for the current exercise sheet, including the file workspace.yml
2. For each student, create a Korrektur.pdf using Xournal++ and put it next to Korrektur.yml.
3. In Korrektur.yml, change the points array to reflect the points per exercise on the sheet, and set corrected to true when finished with all exercises.
4. When finished, run the sanity check script and fix any errors detected:

# checks existance of Korrekur.pdf, corrected=true, and maximum number per exercise
./check.py

5. zip the folder containing the Korrektur.yml and all subdirectories to a file called korrektorkennung.zip. In the end, you should have one zip file containing the folder BlattNN, which contains the subfolders for each student, check.bash, and workspace.yml.
6. Upload that zip file to the write-only sciebo shared folder for BlattNN.
7. In case you have to update the zip file because of an error, just re-upload it again.

Admin: Upload Feedback for Students

1. Download all zip files from write-only share to a folder containing the workspace.yml from above.
2. Manually remove re-uploaded files (they have a numeric suffix).
3. Run upload script:

# extracts zip files, runs check.bash for every corrector first, then checks that we have not lost any submissions, then  workspace upload <korrektorkennung>  to upload the feedback files and points for each corrector
./upload_feedbacks.bash BlattNN

4. Get the total points per student and exercise sheet (and subexercise) so far:

./ilias-linux grades export 1005951