= Relational Design =

This page documents the relational design for BEST Skopje Hub, based on ERModel_v11. The design uses partial transformation from the ER model into a PostgreSQL relational schema. Strong entities are mapped directly to relations, weak entities include the primary keys of their owner entities, specializations are mapped with one relation for the supertype and one relation for each subtype, and M:N relationships are mapped as separate relations.

== Descriptive representation of the relational schema ==

{{{
students(
    student_index, first_name, last_name, faculty, field_of_studies,
    year_of_studies, birthday, email, phone_number,
    PK(student_index)
)

members(
    student_index,
    PK(student_index),
    FK(student_index) -> students(student_index)
)

observers(
    student_index,
    PK(student_index),
    FK(student_index) -> members(student_index)
)

young_members(
    student_index, mentor_index,
    PK(student_index),
    FK(student_index) -> members(student_index),
    FK(mentor_index) -> full_members(student_index)
)

full_members(
    student_index,
    PK(student_index),
    FK(student_index) -> members(student_index)
)

alumni(
    student_index,
    PK(student_index),
    FK(student_index) -> members(student_index)
)

membership_stages(
    ms_id, ms_name,
    PK(ms_id),
    UNIQUE(ms_name)
)

has_stage(
    student_index, ms_id, valid_from, valid_to,
    PK(student_index, ms_id, valid_from),
    FK(student_index) -> members(student_index),
    FK(ms_id) -> membership_stages(ms_id)
)

organizational_functions(
    of_id, of_title,
    PK(of_id),
    UNIQUE(of_title)
)

board(
    of_id,
    PK(of_id),
    FK(of_id) -> organizational_functions(of_id)
)

non_board(
    of_id,
    PK(of_id),
    FK(of_id) -> organizational_functions(of_id)
)

function_mandates(
    student_index, of_id, mandate_year,
    PK(student_index, of_id, mandate_year),
    FK(student_index) -> full_members(student_index),
    FK(of_id) -> organizational_functions(of_id)
)

activity_types(
    at_id, at_name, at_description,
    PK(at_id),
    UNIQUE(at_name)
)

locations(
    l_id, l_name,
    PK(l_id),
    UNIQUE(l_name)
)

activity_sessions(
    as_id, at_id, l_id, as_start_time,
    PK(as_id),
    FK(at_id) -> activity_types(at_id),
    FK(l_id) -> locations(l_id)
)

related_to(
    as_id, ee_id,
    PK(as_id),
    FK(as_id) -> activity_sessions(as_id),
    FK(ee_id) -> event_editions(ee_id)
)

event_types(
    et_id, et_name, et_description,
    PK(et_id),
    UNIQUE(et_name)
)

event_editions(
    ee_id, et_id, ee_title, ee_start_time, ee_end_time,
    PK(ee_id),
    FK(et_id) -> event_types(et_id)
)

event_edition_locations(
    ee_id, l_id,
    PK(ee_id, l_id),
    FK(ee_id) -> event_editions(ee_id),
    FK(l_id) -> locations(l_id)
)

event_session_types(
    est_id, est_name, est_description,
    PK(est_id),
    UNIQUE(est_name)
)

event_function_types(
    eft_id, eft_name,
    PK(eft_id),
    UNIQUE(eft_name)
)

event_sessions(
    ee_id, es_no, est_id, l_id, es_start_time, es_title, needed_no_volunteers,
    PK(ee_id, es_no),
    FK(ee_id) -> event_editions(ee_id),
    FK(est_id) -> event_session_types(est_id),
    FK(l_id) -> locations(l_id)
)

event_functions(
    ee_id, ef_no, eft_id,
    PK(ee_id, ef_no),
    FK(ee_id) -> event_editions(ee_id),
    FK(eft_id) -> event_function_types(eft_id)
)

applications(
    application_id, student_index, ee_id, ef_no,
    motivational_letter, application_type, application_status,
    PK(application_id),
    FK(student_index) -> students(student_index),
    FK(ee_id) -> event_editions(ee_id),
    FK(ee_id, ef_no) -> event_functions(ee_id, ef_no)
)

holds_event_function(
    student_index, ee_id, ef_no,
    PK(student_index, ee_id, ef_no),
    FK(student_index) -> members(student_index),
    FK(ee_id, ef_no) -> event_functions(ee_id, ef_no)
)

participation(
    student_index, as_id, rsvp_status, rsvp_at, attendance_status,
    PK(student_index, as_id),
    FK(student_index) -> students(student_index),
    FK(as_id) -> activity_sessions(as_id)
)

required_attendance(
    student_index, as_id,
    PK(student_index, as_id),
    FK(student_index) -> members(student_index),
    FK(as_id) -> activity_sessions(as_id)
)

announces_absence(
    student_index, as_id, reason, announced_at,
    PK(student_index, as_id),
    FK(student_index, as_id) -> required_attendance(student_index, as_id)
)

responsible_for(
    student_index, as_id,
    PK(student_index, as_id),
    FK(student_index) -> members(student_index),
    FK(as_id) -> activity_sessions(as_id)
)

volunteers(
    student_index, ee_id, es_no, volunteer_role,
    PK(student_index, ee_id, es_no),
    FK(student_index) -> members(student_index),
    FK(ee_id, es_no) -> event_sessions(ee_id, es_no)
)

companies(
    c_id, c_name, naframa_reference,
    PK(c_id)
)

cooperation(
    c_id, ee_id, cooperation_type,
    PK(c_id, ee_id),
    FK(c_id) -> companies(c_id),
    FK(ee_id) -> event_editions(ee_id)
)
}}}

In the physical PostgreSQL schema, `application_id` is implemented as an automatically generated identity value. It is still listed as the primary key of `applications`, but it does not have to be manually inserted in the data loading script.

The `applications` relation is used only when there is an actual application process. Public attendance for events such as Job Fair, and open registration-style attendance for events such as beBESTie, is not represented as an application in this phase. For Hackathon and BEST Course in Summer participation applications, the `motivational_letter` attribute represents the most important motivation/answer kept from an external application form.

The specialization of students into members and then into observers, young members, full members, and alumni is represented through subtype tables whose primary keys are also foreign keys. `young_members` also contains `mentor_index`, which represents the mentorship relationship directly as a foreign key to `full_members`.

The weak entities from ERModel_v11 are represented with composite primary keys where they still depend on an owner entity. `event_functions` uses `(ee_id, ef_no)`, and `event_sessions` uses `(ee_id, es_no)`. Other concrete entities such as `activity_sessions` and `event_editions` use their own numeric identifiers.

The Locations entity is represented with `locations`. The relationship between !EventEditions and Locations is represented with `event_edition_locations`.

The relationships `has_stage`, `participation`, `required_attendance`, `announces_absence`, `responsible_for`, `volunteers`, `cooperation`, `related_to`, `holds_function`, `holds_event_function`, and the event-edition location relationship are represented as relations because they either have M:N cardinality, their own attributes, or are clearer without nullable foreign-key columns. The relationship `related_to` is represented separately because only some activity sessions are connected to event editions. The relationship `applies_with` is represented with the foreign key `student_index` in `applications`, because each application belongs to one student. The N:1 relationships `for_function`, `for_participation`, `takes_place_at`, `held_at`, `is_of_function_type`, and `mentorship` are represented directly with foreign keys on the N-side relation.

== DDL script for creating the database schema and objects ==

The DDL script recreates the official `project` schema and all database objects. It first drops the existing `project` schema if it exists, then creates the schema, tables, primary keys, foreign keys, and checks.

[attachment:schema_creation.sql schema_creation.sql]

== DML script for filling tables with data ==

The DML script recreates realistic sample data for all relations. It clears the current contents of the tables and inserts connected sample data for students, members, membership stages, activity sessions, event-related activity links, event editions, event sessions, event function types, event function slots, actual event function holders, applications, companies, volunteering, mentorship, attendance/RSVP, required attendance, announced absences, and organizational functions.

[attachment:data_load.sql data_load.sql]

== Relational diagram ==

The relational diagram was exported from DBeaver after running `schema_creation.sql` in the assigned PostgreSQL database and opening the `project` schema diagram. The diagram uses crow-feet notation.

[[Image(relational_schema.png, width=100%)]]

== AI Use ==

AI was used during this phase as a consultation, drafting, and checking tool while preparing the relational design from the already completed ERModel_v11. The AI-assisted parts were reviewed against the existing ER model and project requirements before being included.

Full AI usage documentation: [wiki:RelationalDesignAIUsage RelationalDesignAIUsage].