smartersite

Doing things (for now grading) with files from UC Davis smartsite

grading.py

Most of the action is in grading.py, for information run:

python grading.py -h 

To be useful, this program must be run in the root of a directory tree as unzipped from a bulk_download.zip file as provided by the 'Grade' link in a smartsite Assignment.

The program will attempt to open submissions (.doc or .pdf as of now) and prompt the user for comments through the command line.

You have an option to edit those comments (if $EDITOR is set in your shell).

It prompts for a grade (must be <= 5 as of now) and if that grade is provided, includes it in the comments and also the grade to an output file with the format of grades.csv.

In a given 'session' of grading, any students for which grades aren't provided are written into an output file (also with the format of grades.csv) but with no grades.

The script takes arguments that allow you to grade only students in specified section, so long as an appropriate csv file to look up section membership is provided. There are also R-based tools to construct such a csv from Excel files downloaded from smartsite (see 'R scripts' below).

a workflow

1. download a zip archive from smartsite into bulk_download.zip
2. unpack it, which creates a single directory eg. to Homework _ Lab 1
3. cd to that directory
4. run grading.py as below (arguments to grade section 2 only)

~/TAwd/Homework _ Lab 1$ python ../grading.py -lu ../currentlut.csv -s 2 grades.csv outgrades.csv 'hw test' jaime notgraded.csv
# do some grading
# at some point, be done grading and exit by pressing `q` or `Q` at the appropriate prompt

5. next time, run the same script but with notgraded.csv as the input,

~/TAwd/Homework _ Lab 1$ python ../grading.py -lu ../currentlut.csv -s 2 notgraded.csv outgrades2.csv 'hw test' jaime notgraded2.csv
# do some grading
# at some point, be done grading and exit by pressing `q` or `Q` at the appropriate prompt

6. assuming you graded the rest of the section's assignments in this session, you have grades spread across two files, outgrades.csv and outgrades2.csv, use Rscript stackcsv.R to combine, but first delete grades.csv,

rm grades.csv
Rscript stackcsv.R outgrades.csv outgrades2.csv grades.csv
rm outgrades* notgraded.csv #cleanup 

7. zip up the entire directory and use 'Upload all' from smartsite to upload it. It's OK that grades.csv doesn't contain a grade for every student sub directory.

R scripts

Two .R scripts are for processing excel files provided by smartsite into lookup tables for section membership.

Another .R script is for stacking csvs that are like grades.csv provided by smartsite.