A new print script, since the old print.sh was outdated and not working

print.py

+#!/usr/bin/python3
+"""
+Usage: ./print.py /path/to/results
+
+This script will print all <ip>.pdf files in /path/to/results in alphabetical
+order* w.r.t the students' last names (as opposed to ip addresses).
+A new directory /path/to/results/pdf_alphabetic of alphabetized (symlink)
+pdfs is created, and can be safely deleted afterwards.
+*Approximate alphabetical order, see create_symlinks.__doc__.
+
+By default, the pdfs are printed double-sided, two pages per side,
+and each student's solution is stapled. This can be configured in
+the LPR_OPTIONS variable.
+
+The default printer is p-hg-j-54. This can be configured in the PRINTER
+variable.
+Tip: Use "lpstat -p" to see the list of available printers.
+
+The path/to/results should point to a directory containing the pdfs of
+the student's solution, created by latex_creator.py, as well as the
+candidates.csv for the <ip> -> <last name> mapping.
+The pdfs should be named <ip>.pdf and other files, such as <ip>-nup.pdf,
+out.pdf, out_rotated.pdf are ignored.
+
+
+To do:
+    * Add --printer to command line arguments.
+    * Automatically split the job into batches, including a small test batch.
+    * Await user input before starting next batch. "Print next batch? (Y/n)".
+    * Allow a list of printers ["p-hg-j-42","p-hg-j-54",...] to print in
+        parallel. Make sure each printer gets consecutive batches, so that
+        the piles of paper can be easily joined while retaining the order.
+"""
+
+# Remove "-o StapleLocation=UpperRight" to skip stapling.
+LPR_OPTIONS = "-o sides=two-sided-short-edge -o landscape -o fit-to-page -o media=A4 -o number-up=2 -o number-up-layout=lr -o StapleLocation=UpperRight"
+# ~ PRINTER = "p-hg-j-54"
+PRINTER = "p-hg-e-62-2"
+
+import sys
+import os
+import subprocess
+import csv
+from string import ascii_letters
+
+def format_status(string):
+    start = '\033[92m' #green
+    end = '\033[0m'
+    return start + string + end
+    
+def is_ip_pdf( fname ):
+    '''
+    Any <ip>.pdf will satisfy this condition, while out.pdf,
+    out_rotated.pdf, <ip>-nup.pdf, and non-pdf files are ignored.
+    Not actually checking if valid ip address.
+    '''
+    fname_split = fname.split('.')
+    return len(fname_split)>1 and fname_split[-1] == "pdf" \
+            and all([x.isdigit() for x in fname_split[:-1]])
+
+def rm_non_ascii( name ):
+    # To do: instead of removing, replace non-ascii letters by their ascii alternatives.
+    allowed_letters = set(ascii_letters)
+    return "".join([letter for letter in name if letter in allowed_letters])
+
+def create_symlinks( pdf_dir ):
+    '''
+    pdf_dir needs to contain candidates.csv
+    
+    Creates a subdirectory of pdf_dir called pdf_alphabetic.
+    The new directory contains symbolic links to all <ip>.pdf files in pdf_dir,
+    with names <lastname>.<ip>.pdf, so that they are in alphabetic order.
+    The following caveats can cause interruptions of the alphabetic ordering:
+    Caveat 1: The <lastname> is provided by the student and sometimes the
+    student will instead give their first name.
+    Caveat 2: Any non-ascii characters of <lastname> are removed.
+    '''
+    pdf_dir = os.path.abspath( pdf_dir )
+    file_list = os.listdir( pdf_dir )
+    symlink_dir = os.path.abspath( pdf_dir+"/pdf_alphabetic" )
+    if not os.path.isdir( symlink_dir ):
+        print(format_status("Alphabetizing into directory:\n\t") + symlink_dir)
+        os.mkdir( symlink_dir ) # directory for for symlinks
+    else:
+        print(format_status("Overwriting existing alphabetized directory:\n\t") + symlink_dir)
+    for fname in file_list:
+        if is_ip_pdf( fname ):
+            pdf_ip = fname[:-4]
+            with open( pdf_dir+"/candidates.csv" ) as csvfile:
+                # Searching for the student's name.
+                candidates = csv.reader( csvfile, delimiter=';' )
+                for student in candidates:
+                    # student = [ <ip>, <login>, <fullname> ]
+                    if student[0] == pdf_ip:
+                        student_name = student[2].split()[0].lower()
+                        student_name = rm_non_ascii( student_name )
+                        if not student_name:
+                            # if name contains no ascii characters (!!!)
+                            # use login instead.
+                            student_name = student[1]
+                        new_fname = student_name + '.' + fname
+                        symlink_path = symlink_dir + '/' + new_fname
+                        # print(fname + ' -> ' + new_fname)
+                        if os.path.exists(symlink_path):
+                            os.remove(symlink_path)
+                        os.symlink(pdf_dir+'/'+fname,symlink_path)
+    return symlink_dir
+
+def print_all( symlink_dir ):
+    lpr_command = "lpr -P {} {}".format(PRINTER,LPR_OPTIONS)
+    # Create a list of arguments instead of string.
+    lpr_command = lpr_command.split()
+    # The path needs to be treated separately, in case any folder has
+    # a space in its name.
+    pdf_files = [f for f in os.listdir(symlink_dir) if f[-4:]==".pdf"]
+    print(format_status("Printing %d documents using:\n\t"%len(pdf_files))+" ".join(lpr_command) + " <filename>")
+    for fname in pdf_files:
+        # To glob (*.pdf), we need shell=True in subprocess.run().
+        # To avoid this, just print all .pdf in separate commands.
+        if fname[-4:] == ".pdf":
+            print(".",end="")
+            subprocess.run(lpr_command+[symlink_dir+'/'+fname],check=True,shell=False)
+    print(format_status("\nPrint job started! Go check printer: ") + PRINTER)
+
+if __name__ == "__main__":
+    if len(sys.argv) > 1 and os.path.isdir(sys.argv[1]):
+        path_to_results = sys.argv[1]
+        path_to_symlinks = create_symlinks( path_to_results )
+        print_all( path_to_symlinks )
+    else:
+        print("Please provide a valid /path/to/results.")