degreelink

Category-Level and Group-Level Constraints

Overview

The system supports hybrid constraint scope, allowing constraints to apply either to an entire requirement category or to specific groups within that category.

Constraint Scoping

Constraint Key Structure

Constraints are identified by a 5-tuple: (program_id, category, semester, year, group_name)

Scope Logic:

• If group_name exists → group-level constraint (applies only to that specific group)
• If group_name is None → category-level constraint (applies to all groups in category)

# Determine constraint scope based on group_name presence
constraint_key = (program.id, category, semester, year, group_name)

Constraint Storage

JSON-Based Parameters

Constraints use JSON fields for flexible parameter storage:

Constraint types supported:

1. credits - Min/max credit requirements
   • Params: {"credits_min": X, "credits_max": Y}
2. courses - Min/max course count requirements
   • Params: {"courses_min": X, "courses_max": Y}
3. min_courses_at_level - Level-based requirements
   • Params: {"level": 3000, "courses": 2}
4. min_tag_courses / max_tag_credits - Tag-based requirements
   • Params: {"tag": "true", "courses": 2} or {"tag": "lab", "credits": 7}
   • Scope: {"tag_field": "has_lab"} or {"tag_field": "course_type"}

Scope Filter Structure

{
  "group_name": "Biology Lab",  // Optional: for group-level constraints
  "subject_codes": ["BIOS", "CHEM"],  // Optional: limit to specific subjects
  "tag_field": "has_lab"  // Optional: specify which tag field to check
}

Constraint Evaluation

Scope Filtering Logic

During constraint evaluation, the system filters courses based on the scope filter:

Group name filtering (for group-level constraints):

if 'group_name' in scope:
    pc_group_name = getattr(pc, 'group_name', None)
    if pc_group_name != scope['group_name']:
        match = False

Subject codes list filtering:

if 'subject_codes' in scope:
    if course.subject_code not in scope['subject_codes']:
        match = False

PlanCourse Group Tracking

The PlanCourse model includes group name tracking for constraint evaluation:

requirement_group = db.relationship('RequirementGroup', foreign_keys=[requirement_group_id])

@property
def group_name(self):
    """Get the group name from the related RequirementGroup."""
    if self.requirement_group:
        return self.requirement_group.group_name
    return None

Examples

Category-Level Constraint

CSV Row:

category,semester,year,type,group_name,...,constraint_type,description,min_credits,...
Core Major Requirements,1,1,grouped,,,credits,Minimum 15 credits,15,...

• First row of category with constraint_type but NO group_name
• Creates constraint with: scope_filter = {}
• Applies to ALL courses in “Core Major Requirements” category across all groups

Group-Level Constraint

CSV Row:

category,semester,year,type,group_name,...,constraint_type,description,min_courses,...
Core Major Requirements,1,1,grouped,Biology Lab,min_tag_courses,At least 2 lab courses,,,2,...

• First row of “Biology Lab” group with constraint_type AND group_name
• Creates constraint with: scope_filter = {"group_name": "Biology Lab"}
• Applies ONLY to courses assigned to “Biology Lab” group

Use Cases

Biology B.S. Program Example

Category: Core Major Requirements (40-45 credits) Groups within category:

1. Major Required Courses (specific BIOS courses)
2. Biology Electives (choose from list)
3. Biology Lab (must have lab component)
4. Biology 2000 (2000-level requirement)
5. Biology Research/Apprentice (special types)
6. Biology 4000 (4000-level requirement)

Constraints:

• Category-level: “Minimum 40 credits in Core Major Requirements”
  • Applies to ALL groups combined
• Group-level: “Biology Lab group must have at least 2 courses with has_lab:true”
  • Applies ONLY to courses in “Biology Lab” group
• Group-level: “Biology 4000 group must have at least 3 courses at 4000 level”
  • Applies ONLY to courses in “Biology 4000” group

Benefits

1. Flexibility: Can apply constraints at both category and group levels
2. Granular Control: Different groups can have different requirements
3. Backwards Compatible: Category-level constraints work exactly as before
4. Clear Semantics: group_name in scope_filter clearly indicates group-scoped constraint
5. Extensible: Easy to add more scope filters (subject codes, level ranges, etc.)

Database Schema

Uses existing scope_filter JSON column in requirement_constraints table:

• params (TEXT/JSON): Constraint-specific parameters
• scope_filter (TEXT/JSON): Optional filtering criteria including group_name
• constraint_type (VARCHAR): Type of constraint (‘credits’, ‘courses’, ‘min_courses_at_level’, etc.)
• description (TEXT): Human-readable description

This site is open source. Improve this page.