add pl support (#66)

Author: gregbell26

AzureAD.py

@@ -31,18 +31,24 @@ def __init__(self, tenantId: str) -> None:
 
 
 
-    async def getCWIDFromEmail(self, username: str) -> str:
-        query = UsersRequestBuilder.UsersRequestBuilderGetQueryParameters(
-                select=["employeeId"],
-            )
+    async def getCWIDFromEmail(self, usernames: list[str]) -> list[tuple[str, str]]:
+        cwidMap = []
 
-        requestConfig = UsersRequestBuilder.UsersRequestBuilderGetRequestConfiguration(query_parameters=query)
-        userCwid = await self.client.users.by_user_id(username).get(requestConfig)
+        for i in range(0, len(usernames), 14):
+            query = UsersRequestBuilder.UsersRequestBuilderGetQueryParameters(
+                    select=["employeeId", "userPrincipalName"],
+                    filter=f"userPrincipalName in ['{'\',\''.join(usernames[i:i+14])}'] and accountEnabled eq true",
+                )
 
-        if userCwid is None or userCwid.employee_id is None:
-            return ""
+            requestConfig = UsersRequestBuilder.UsersRequestBuilderGetRequestConfiguration(query_parameters=query)
+            userCwids = await self.client.users.get(requestConfig)
+
+            if userCwids is None:
+                continue
+
+            cwidMap.extend([(val.user_principal_name, val.employee_id) for val in userCwids.value])
 
-        return userCwid.employee_id
+        return cwidMap
 
     async def getEmailFromCWID(self, cwid: str) -> str:
         query = UsersRequestBuilder.UsersRequestBuilderGetQueryParameters(

FileHelpers/csvLoaders.py

@@ -1,5 +1,8 @@
 # The biggest change that needs to be made is to remap mutlipasses to cwids
-
+import csv
+import re
+import sys
+from typing import List, Dict
 
 import pandas as pd
 from FileHelpers import fileHelper
@@ -12,6 +15,7 @@
 # Grace Period of 15 minutes
 GRADESCOPE_GRACE_PERIOD = 15
 
+csv.field_size_limit(sys.maxsize)
 
 def loadCSV(_filename: str, promptIfError: bool = False, directoriesToCheck: list[str] = None):
     """
@@ -115,6 +119,69 @@ def loadGradescope(_filename):
     print("Done.")
     return gradescopeDF
 
+def extractGroupFromPL(group: str):
+    r = re.compile(r"[\[\"\]]")
+
+    group = re.sub(r, "", group)
+
+    return group.split(",")
+
+def convertGroupSubmissionToIndividualSubmission(header: List[str], data: List[List[str]]):
+    GROUP_MEMBER_IDX = header.index("Usernames")
+    SUBMISSION_DATE_INDEX = header.index("Submission date")
+    QUESTION_POINTS_IDX = header.index("Question points")
+
+    normalizedSubmission = []
+
+    for line in data:
+        members = extractGroupFromPL(line[GROUP_MEMBER_IDX])
+        for member in members:
+            normalizedSubmission.append([member, line[SUBMISSION_DATE_INDEX], float(line[QUESTION_POINTS_IDX])])
+
+    return normalizedSubmission
+
+def parseLinePL(students: Dict[str, List[str]], line: List[str]):
+    USER_ID_IDX = 0
+    SUBMISSION_DATE_IDX = 1
+    POINTS_IDX = 2
+
+    if not line[USER_ID_IDX]:
+        # empty group
+        return
+
+    if line[USER_ID_IDX] in students.keys():
+        students[line[USER_ID_IDX]][SUBMISSION_DATE_IDX] = line[SUBMISSION_DATE_IDX]
+        students[line[USER_ID_IDX]][POINTS_IDX] += line[POINTS_IDX]
+        return
+
+    students[line[USER_ID_IDX]] = [line[USER_ID_IDX], line[SUBMISSION_DATE_IDX], line[POINTS_IDX]]
+
+def loadPrairieLearn(filename):
+    data = []
+    try:
+        with open(filename) as r:
+            reader = csv.reader(r, quotechar='"')
+            for line in reader:
+                data.append(line)
+    except FileNotFoundError:
+        return pd.DataFrame()
+
+    data = convertGroupSubmissionToIndividualSubmission(data[0], data[1:])
+
+    scores = {}
+
+    for line in data[1:]:
+        parseLinePL(scores, line)
+
+    plDF = pd.DataFrame({
+        'email': [value[0] for value in scores.values()],
+        'hours_late': [0 for _ in range(len(scores))],
+        'Total Score': [value[2] for value in scores.values()],
+        'Status': ['Graded' for _ in range(len(scores))],
+        'lateness_comment': ['' for _ in range(len(scores))],
+    })
+
+    return plDF
 
 def loadRunestone(_filename, assignment: str):
     """

Grade/gradesheets.py

@@ -11,12 +11,16 @@
 This file **non-destructively** creates gradesheets. The generated gradesheets still have to be 'selected' in Canvas
 in order to be scored and posted with everything else.
 """
+from typing import List
+
 import pandas as pd
 
 from Bartik.Bartik import Bartik
 from AzureAD import AzureAD
 
-async def convertBartikToGradesheet(_azure: AzureAD, _bartik: Bartik, _students: pd.DataFrame, _assignment: str, _maxPoints: float, _requiredProbems: int) -> pd.DataFrame:
+
+async def convertBartikToGradesheet(_azure: AzureAD, _bartik: Bartik, _students: pd.DataFrame, _assignment: str,
+                                    _maxPoints: float, _requiredProbems: int) -> pd.DataFrame:
     bartikGradesheet: pd.DataFrame = pd.DataFrame()
     bartikGradesheet['multipass'] = ""
     bartikGradesheet['Total Score'] = ""
@@ -30,12 +34,11 @@ async def convertBartikToGradesheet(_azure: AzureAD, _bartik: Bartik, _students:
         print(f"Now grading {row['name']} ({counter}/{len(_students)})...", end="")
 
         studentEmail: str = await _azure.getEmailFromCWID(row['sis_id'])
-        
+
         if studentEmail == "":
             print(f"Failed to map email for {row['name']}")
             continue
 
-
         missing: bool = False
         score: float = 0
 
@@ -46,16 +49,15 @@ async def convertBartikToGradesheet(_azure: AzureAD, _bartik: Bartik, _students:
             print(f"Missing")
 
         bartikGradesheet = pd.concat([bartikGradesheet, pd.DataFrame(
-                {
-                    'multipass': row['sis_id'],
-                    'Total Score': score,
-                    'lateness_comment': "",
-                }, index=[0]
-            )], ignore_index=True)
+            {
+                'multipass': row['sis_id'],
+                'Total Score': score,
+                'lateness_comment': "",
+            }, index=[0]
+        )], ignore_index=True)
 
         if not missing:
             print("Done")
-        
 
     _bartik.closeSession()
     return bartikGradesheet
@@ -97,7 +99,6 @@ def createGradesheetForPassFailAssignment(_passFailAssignment: pd.DataFrame, _st
                                           checkProofOfAttendance: bool = False,
                                           proofOfAttendanceColumn: (str, None) = None) \
         -> pd.DataFrame:
-
     if proofOfAttendanceColumn:
         proofOfAttendanceColumn = proofOfAttendanceColumn.replace(' ', '_')
 
@@ -141,4 +142,13 @@ def createGradesheetForPassFailAssignment(_passFailAssignment: pd.DataFrame, _st
     return _passFailAssignment
 
 
+async def finalizeGradesheet(azure: AzureAD, assignment: pd.DataFrame):
+    emails = assignment['email'].tolist()
+
+    emailsWithCwids = await azure.getCWIDFromEmail(emails)
+    assignment['multipass'] = ''
+
+    for i, row in assignment.iterrows():
+        assignment.loc[i, 'multipass'] = [email[1] for email in emailsWithCwids if email[0] == row['email']][0]
 
+    return assignment

UI/standardGrading.py

@@ -20,8 +20,8 @@ async def standardGrading(**kwargs):
         gradesheetsToGrade: dict[int, pd.DataFrame] = uiHelpers.setupGradescopeGrades(kwargs['canvas'])
     elif choice == 2:
         gradesheetsToGrade: dict[int, pd.DataFrame] = uiHelpers.setupRunestoneGrades(kwargs['canvas'])
-    elif choice == 3:
-        return NotImplementedError # TODO add PL support
+    else:
+        gradesheetsToGrade: dict[int, pd.DataFrame] = await uiHelpers.setupPLGrades(kwargs['canvas'], kwargs['azure'])
 
     specialCasesDF = uiHelpers.setupSpecialCases()
 

UI/uiHelpers.py

@@ -1,9 +1,12 @@
 """
 """
 import pandas as pd
-from FileHelpers.csvLoaders import loadGradescope, loadRunestone
+
+from AzureAD import AzureAD
+from FileHelpers.csvLoaders import loadGradescope, loadRunestone, loadPrairieLearn
 from FileHelpers.excelLoaders import loadSpecialCases, loadPassFailAssignment
 from Canvas import Canvas
+from Grade.gradesheets import finalizeGradesheet
 
 
 def getUserInput(allowedUserInput: str = None, allowedLowerRange: int = None, allowedUpperRange: int = None):
@@ -93,6 +96,23 @@ def setupGradescopeGrades(_canvas: Canvas) -> dict[int, pd.DataFrame]:
 
     return assignmentMap
 
+async def setupPLGrades(canvas: Canvas, azure: AzureAD) -> dict[int, pd.DataFrame]:
+    # the IDs will always be unique per course - using those over the common names
+    selectedAssignments: pd.DataFrame = canvas.getAssignmentsToGrade()
+    assignmentMap: dict[int, pd.DataFrame] = {}
+    if selectedAssignments is None:
+        return assignmentMap
+    for i, row in selectedAssignments.iterrows():
+        print(f"Enter path to pl grades for {row['common_name']}")
+        path = getUserInput(allowedUserInput="./path/to/pl/grades.csv")
+        plDf: pd.DataFrame = await finalizeGradesheet(azure, loadPrairieLearn(path))
+        if plDf.empty:
+            print(f"Failed to load file '{path}'")
+            # TODO handle this case more elegantly
+            return {}
+        assignmentMap[row['id']] = plDf
+
+    return assignmentMap
 
 def setupRunestoneGrades(_canvas: Canvas) -> dict[int, pd.DataFrame]:
     # the IDs will always be unique per course - using those over the common names