htwkalender/services/data-manager/service/professor/professorService.go

package professor

import (
	"fmt"
	"htwkalender/data-manager/model"
	"log/slog"
	"strings"

	"github.com/pocketbase/dbx"
	"github.com/pocketbase/pocketbase"
)

type ProfessorService struct {
	app *pocketbase.PocketBase
}

func NewProfessorService(app *pocketbase.PocketBase) *ProfessorService {
	return &ProfessorService{app: app}
}

func (s *ProfessorService) GetModulesForProfessor(email string) ([]model.ModuleDTO, error) {
	// Extract name from email
	// Format: firstname.lastname@htwk-leipzig.de
	parts := strings.Split(email, "@")
	if len(parts) != 2 {
		return nil, fmt.Errorf("invalid email format")
	}

	nameParts := strings.Split(parts[0], ".")
	if len(nameParts) < 2 {
		slog.Warn("Email does not contain dot separator", "email", email)
		return []model.ModuleDTO{}, nil
	}

	// Extract first and last name
	firstName := nameParts[0]
	lastName := nameParts[len(nameParts)-1]

	// Capitalize first letter
	if len(firstName) > 0 {
		firstName = strings.ToUpper(firstName[:1]) + firstName[1:]
	}
	if len(lastName) > 0 {
		lastName = strings.ToUpper(lastName[:1]) + lastName[1:]
	}

	slog.Info("Searching for modules for professor", "firstName", firstName, "lastName", lastName, "email", email)

	// First, get all distinct modules with their professors
	type EventProf struct {
		Name      string `db:"Name" json:"name"`
		EventType string `db:"EventType" json:"eventType"`
		Prof      string `db:"Prof" json:"prof"`
		Course    string `db:"course" json:"course"`
		Semester  string `db:"semester" json:"semester"`
		UUID      string `db:"uuid" json:"uuid"`
	}

	var allEvents []EventProf
	err := s.app.DB().
		Select("Name", "EventType", "Prof", "course", "semester", "uuid").
		From("events").
		Where(dbx.NewExp("Prof != ''")).
		GroupBy("Name", "course", "Prof").
		Distinct(true).
		All(&allEvents)

	if err != nil {
		return nil, err
	}

	// Filter events by matching professor name and calculate confidence scores
	var modules []model.ModuleDTO
	seenModules := make(map[string]bool) // key: Name+Course to avoid duplicates

	for _, event := range allEvents {
		score := calculateConfidenceScore(event.Prof, firstName, lastName)
		if score > 0 { // Include all modules with any match
			key := event.Name + "|" + event.Course
			if !seenModules[key] {
				modules = append(modules, model.ModuleDTO{
					Name:            event.Name,
					EventType:       event.EventType,
					Prof:            event.Prof,
					Course:          event.Course,
					Semester:        event.Semester,
					UUID:            event.UUID,
					ConfidenceScore: score,
				})
				seenModules[key] = true
			}
		}
	}

	slog.Info("Found modules for professor", "count", len(modules), "lastName", lastName)
	return modules, nil
}

// calculateConfidenceScore returns a score from 0.0 to 1.0 indicating how confident we are
// that this professor string matches the given first and last name
// 1.0 = perfect match (both first and last name exact)
// 0.7-0.9 = good match (last name exact, first name fuzzy or present)
// 0.4-0.6 = possible match (last name fuzzy or partial)
// 0.1-0.3 = weak match (last name substring)
// 0.0 = no match
func calculateConfidenceScore(profString, firstName, lastName string) float64 {
	// Normalize the professor string: remove common titles and split into words
	profString = strings.ToLower(profString)

	// Remove common titles
	titles := []string{"prof.", "dr.", "arch.", "ing.", "dipl.", "m.sc.", "b.sc.", "ph.d."}
	for _, title := range titles {
		profString = strings.ReplaceAll(profString, title, "")
	}

	// Split by spaces, hyphens, and other separators
	words := strings.FieldsFunc(profString, func(r rune) bool {
		return r == ' ' || r == '-' || r == ',' || r == '.'
	})

	// Normalize firstName and lastName
	firstNameLower := strings.ToLower(firstName)
	lastNameLower := strings.ToLower(lastName)

	lastNameExact := false
	lastNameFuzzy := false
	lastNameSubstring := false
	firstNameExact := false
	firstNameFuzzy := false

	for _, word := range words {
		word = strings.TrimSpace(word)
		if word == "" {
			continue
		}

		// Check last name
		if word == lastNameLower {
			lastNameExact = true
		} else if levenshteinDistance(word, lastNameLower) <= 1 && len(lastNameLower) > 3 {
			lastNameFuzzy = true
		} else if strings.Contains(word, lastNameLower) || strings.Contains(lastNameLower, word) {
			lastNameSubstring = true
		}

		// Check first name
		if word == firstNameLower {
			firstNameExact = true
		} else if levenshteinDistance(word, firstNameLower) <= 1 && len(firstNameLower) > 3 {
			firstNameFuzzy = true
		}
	}

	// Calculate confidence score based on matches
	score := 0.0

	if lastNameExact {
		if firstNameExact {
			score = 1.0 // Perfect match
		} else if firstNameFuzzy {
			score = 0.9 // Excellent match
		} else {
			score = 0.8 // Good match (last name exact, no first name match)
		}
	} else if lastNameFuzzy {
		if firstNameExact || firstNameFuzzy {
			score = 0.6 // Decent match (fuzzy last name but first name matches)
		} else {
			score = 0.5 // Medium match (fuzzy last name, no first name)
		}
	} else if lastNameSubstring {
		score = 0.2 // Weak match (substring only)
	}

	return score
}

// matchesProfessor is deprecated, use calculateConfidenceScore instead
func matchesProfessor(profString, firstName, lastName string) bool {
	// Normalize the professor string: remove common titles and split into words
	profString = strings.ToLower(profString)

	// Remove common titles
	titles := []string{"prof.", "dr.", "arch.", "ing.", "dipl.", "m.sc.", "b.sc.", "ph.d."}
	for _, title := range titles {
		profString = strings.ReplaceAll(profString, title, "")
	}

	// Split by spaces, hyphens, and other separators
	words := strings.FieldsFunc(profString, func(r rune) bool {
		return r == ' ' || r == '-' || r == ',' || r == '.'
	})

	// Normalize firstName and lastName
	firstNameLower := strings.ToLower(firstName)
	lastNameLower := strings.ToLower(lastName)

	lastNameFound := false
	firstNameFound := false

	for _, word := range words {
		word = strings.TrimSpace(word)
		if word == "" {
			continue
		}

		// Exact match for last name
		if word == lastNameLower {
			lastNameFound = true
		}

		// Exact match for first name (optional, but increases confidence)
		if word == firstNameLower {
			firstNameFound = true
		}

		// Also check Levenshtein distance for typos
		if !lastNameFound && levenshteinDistance(word, lastNameLower) <= 1 {
			lastNameFound = true
		}
		if !firstNameFound && levenshteinDistance(word, firstNameLower) <= 1 {
			firstNameFound = true
		}
	}

	// Match if last name is found (first name is optional for additional confidence)
	// We require at least the last name to match
	return lastNameFound
}

// levenshteinDistance calculates the Levenshtein distance between two strings
func levenshteinDistance(s1, s2 string) int {
	if len(s1) == 0 {
		return len(s2)
	}
	if len(s2) == 0 {
		return len(s1)
	}

	// Create a 2D array for dynamic programming
	d := make([][]int, len(s1)+1)
	for i := range d {
		d[i] = make([]int, len(s2)+1)
	}

	// Initialize first column and row
	for i := 0; i <= len(s1); i++ {
		d[i][0] = i
	}
	for j := 0; j <= len(s2); j++ {
		d[0][j] = j
	}

	// Fill the matrix
	for i := 1; i <= len(s1); i++ {
		for j := 1; j <= len(s2); j++ {
			cost := 0
			if s1[i-1] != s2[j-1] {
				cost = 1
			}

			d[i][j] = min(
				d[i-1][j]+1,      // deletion
				d[i][j-1]+1,      // insertion
				d[i-1][j-1]+cost, // substitution
			)
		}
	}

	return d[len(s1)][len(s2)]
}

func min(a, b, c int) int {
	if a < b {
		if a < c {
			return a
		}
		return c
	}
	if b < c {
		return b
	}
	return c
}