quick overview of assignments without collecting

[ferdinandyb]

Feature request (kooplex-edu):
I would need a quick overview of assignments in one folder, without having to collect and then reassign (reassignment removes them again). Preferably, if only one .ipynb is present in the students assignment folder, copy that to a single folder, with the students name prepended and also convert it to html for quick preview (jupyter nbconvert --execute --to html notebook.ipynb.)

[oroszl]

you can do that manually through the student's work dir

[jegesm]

If you can direct the output of the conversion outside of the student's working directory, then probably that is the simplest solution

…

On Mon, Oct 15, 2018 at 6:54 PM László Oroszlány ***@***.***> wrote: you can do that manually through the student's work dir — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#36 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABCtxes4LU0bCMG9Hd4_s-39sKFtUQKnks5ulL3AgaJpZM4Xcqkj> .

[ferdinandyb]

This script will do the trick until there's a better solution:

eg: python3 gatherassigments.py workdir/EN-L/ assignment1 -c

import argparse
import os
import glob
import shutil
import pathlib
import subprocess

import pandas as pd

def gatherassignments(course,assignment, destination):
    stcounter = 0
    nbcounter = 0
    for st in os.listdir(course):
        snb = False
        try:
            name = studentmap[st]
        except KeyError as e:
            # students will not be in the list if they stopped attending the course at some point
            continue
        for nb in glob.glob(os.path.join(course,st,assignment,"*.ipynb")):
            snb = True
            nbcounter += 1
            shutil.copy(nb,os.path.join(destination,name + "_" + os.path.basename(nb)))
        if snb:
            stcounter += 1
    print("Found {} notebooks from {} students.".format(nbcounter,stcounter))
    
def convert_notebooks(folder):
    for nb in glob.glob(os.path.join(folder,"*.ipynb")):
        subprocess.run(["jupyter", "nbconvert", "--execute", "--to", "html",nb],timeout = 60)

if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('course',type=str)
    parser.add_argument('assignment',type=str)
    parser.add_argument('-c','--convert', action='store_true')
    parser.add_argument('-s','--students', type=str,default="hallgatok1.xlsx")
    args = parser.parse_args()
    
    studentmap = {line[1].lower() : "_".join(line[0].split()) for line in pd.read_excel(args.students).values}
    destination = os.path.join("assignments",args.assignment)
    pathlib.Path(destination).mkdir(parents=True, exist_ok=True) 
    
    gatherassignments(args.course,args.assignment,destination)
    if args.convert:
        convert_notebooks(destination)

[oroszl]

This was my solution:

WORKDIR=$1
PACKAGE=$2
for diak in `ls $WORKDIR`; do
if [ ! -d $diak/$PACKAGE/ ]; then
  mkdir -p $diak/$PACKAGE/
fi
cp $WORKDIR/$diak/$PACKAGE/MEGOLDASOK/*.ipynb $diak/$PACKAGE/
jupyter-nbconvert --to html $diak/$PACKAGE/*.ipynb --output-dir $diak/$PACKAGE
done

of course this requires a predefined structure

[ferdinandyb]

I don't touch bash if I don't have to, and I wanted to prepend their names, but definitely shorter :)