code: add course types & improve debuggability

lflex_celcat_survival/events.py

@@ -11,7 +11,7 @@ import re
 from . import fetch
 
 ROOM_RE = re.compile(r'^(?:FSI|F2SMH) / (.*)$')
-COURSE_TYPE_RE = re.compile(r'COURS|COURS/TD|TD|TP|CONTROLE CONTINU|CONTROLE PARTIEL')
+COURSE_TYPE_RE = re.compile(r'COURS|COURS/TD|TD|TP|CONTROLE CONTINU|CONTROLE PARTIEL|EXAMEN|Controle de Substitution|CONSULTATION DE COPIES')
 STUDENT_GROUP_RE = re.compile(r'K?IN[A-Z0-9]+')
 
 class CelcatEvents:
@@ -98,47 +98,52 @@ def parse_description(row):
     groups = []
     course_type = 'unset'
 
-    if len(fields) == 0:
-        raise ValueError(f'There should be at least 1 field, but fields are {fields}')
-    elif len(fields) == 1:
-        # probably not a course. examples: "CONGES\r\n" or "FERIE\r\n"
-        course_type = fields[0]
-    else:
-        # first fields should be the room, but this is not always set
-        while (m := ROOM_RE.match(fields[0])) is not None:
-            rooms.append(m[1])
+    try:
+        if len(fields) == 0:
+            raise ValueError(f'There should be at least 1 field, but fields are {fields}')
+        elif len(fields) == 1:
+            # probably not a course. examples: "CONGES\r\n" or "FERIE\r\n"
+            course_type = fields[0]
+        else:
+            # first fields should be the room, but this is not always set
+            while (m := ROOM_RE.match(fields[0])) is not None:
+                rooms.append(m[1])
+                fields = fields[1:]
+
+            # assume that the next field is the course name, and skip it
             fields = fields[1:]
 
-        # assume that the next field is the course name, and skip it
-        fields = fields[1:]
+            # skip notes at the end of the fields until they look like a course type
+            while COURSE_TYPE_RE.match(fields[-1]) is None:
+                fields = fields[:-1]
+                if len(fields) <= 0:
+                    break
 
-        # skip notes at the end of the fields until they look like a course type
-        while COURSE_TYPE_RE.match(fields[-1]) is None:
+            # last field is a course type
+            course_type = fields[-1]
             fields = fields[:-1]
-            if len(fields) <= 0:
-                break
 
-        # last field is a course type
-        course_type = fields[-1]
-        fields = fields[:-1]
+            # the last field may be a teacher, but this is optional
+            if STUDENT_GROUP_RE.match(fields[-1]) is None:
+                teacher = fields[-1]
+                fields = fields[:-1]
 
-        # the last field may be a teacher, but this is optional
-        if STUDENT_GROUP_RE.match(fields[-1]) is None:
-            teacher = fields[-1]
-            fields = fields[:-1]
+            # all remaining fields should be student groups
+            groups = []
+            while len(fields) > 0 and (m := STUDENT_GROUP_RE.match(fields[0])) is not None:
+                groups.append(m[0])
+                fields = fields[1:]
 
-        # all remaining fields should be student groups
-        groups = []
-        while len(fields) > 0 and (m := STUDENT_GROUP_RE.match(fields[0])) is not None:
-            groups.append(m[0])
-            fields = fields[1:]
+        if len(rooms) == 0:
+            rooms = ['unset']
+        if len(groups) == 0:
+            groups = ['unset']
 
-    if len(rooms) == 0:
-        rooms = ['unset']
-    if len(groups) == 0:
-        groups = ['unset']
+        return pd.Series([rooms, teacher, course_type, groups], index=['rooms_parsed', 'teacher_parsed', 'course_type_parsed', 'groups_parsed'])
+    except Exception as e:
+        print(f"Could not parse an event description. fields={preparse_fields}")
+        raise e
 
-    return pd.Series([rooms, teacher, course_type, groups], index=['rooms_parsed', 'teacher_parsed', 'course_type_parsed', 'groups_parsed'])
 
 def request_slots_by_mod_code(flat_slot_df, session):
     subject_codes = list(flat_slot_df['mod_code'].dropna().unique())