ems.subjectems.subject represents an individual subject (course unit) that makes up one or more studies. It is the most widely referenced node in the curriculum hierarchy: teaching assignments, timetabling, planning, grading and attendance all key off subject_id. Every subject is automatically backed by a product.product (used to bill it through enrolments/sale.order), kept in sync by the model’s create()/write() overrides.
Module file: models/curriculum/subject.py
| Field | Type | Required | Stored | Description |
|---|---|---|---|---|
code |
Char |
Yes | Yes | Official code — unique per study, not globally (see below) |
acronym |
Char |
Yes | Yes | Short code |
name |
Char |
Yes | Yes | Full descriptive name |
ects |
Integer |
No | Yes | ECTS credits |
is_tutorship |
Boolean |
No | Yes | Marks a subject as a tutoring slot |
internal_hours |
Integer |
No | Yes | Hours taught at the centre |
external_hours |
Integer |
No | Yes | Hours in a work placement / external setting |
total_hours |
Integer (computed) |
— | No | internal_hours + external_hours |
product_id |
Many2one → product.product |
No | Yes | Auto-managed billing product for this subject |
outcome_ids |
One2many → ems.outcome |
No | No | Learning outcomes (inverse of outcome.subject_id) |
content_ids |
One2many → ems.content |
No | No | Content items (inverse of content.subject_id) |
study_ids |
Many2many → ems.study |
No | Yes | Studies this subject belongs to |
notes |
Text |
No | Yes | Free-form administrative notes |
display_name |
Char (computed) |
— | No | Format: Acronym: Name |
_rec_names_search = ['name', 'acronym'] — the name-search box matches on both fields, not just display_name.
graph TD
ST["ems.study (e.g. DAM)"]
S["ems.subject (e.g. Programming)"]
O["ems.outcome"]
C["ems.content"]
CR["ems.criteria"]
ST -->|study_ids| S
S -->|outcome_ids| O
S -->|content_ids| C
O -->|criteria_ids| CR
total_hours Computationflowchart LR
A["internal_hours"] --> C["total_hours = internal_hours + external_hours"]
B["external_hours"] --> C
The official code assigned to a professional module by the education department is not
guaranteed unique across every curriculum it’s ever used in — the same code can mean two
genuinely different subjects (different learning outcomes, different internal/external hours)
when each belongs to a different study (e.g. MP 3003 in a CFGB vs. the same code in a PFI —
see ems.subject_3003_sa and ems.subject_3003_pfi in data/cat/ems.subject.csv). A plain
unique(code) SQL constraint (the model’s behaviour before 2026-09-06) would force a fake
suffix onto one of them, which would then no longer match the real code used by any external
import or grade file, or by a convalidation request against the official curriculum.
_check_code_unique_per_study (an @api.constrains('code', 'study_ids') method, replacing the
old _sql_constraints = [('unique_code', ...)]) enforces a narrower rule instead: a duplicate
code is only a real conflict when the two subjects could actually be confused for one another —
either one of them has no study at all (nothing to disambiguate by, so treated as a genuine
duplicate — this preserves the original behaviour for a subject not yet assigned to a study),
or they share at least one study.
flowchart TD
A([Two subjects share the same code]) --> B{Does either have no study_ids?}
B -- Yes --> C[Conflict - raise 'duplicated code!']
B -- No --> D{Do their study_ids overlap?}
D -- Yes --> C
D -- No --> E[Allowed - different curricula, same official code]
Declared on study_ids, not just code, so re-validation also fires when a study is added to
or removed from a subject via the study’s own “Subjects” tab
(views/community/study/form.xml) — that tab writes the exact same many2many relation from its
other side, but Odoo does not automatically re-run ems.subject’s own constrains for that write
direction (confirmed empirically 2026-09-06). ems.study therefore declares its own mirroring
_check_subject_codes_unique_per_study (@api.constrains('subject_ids'),
models/curriculum/study.py), which delegates to ems.subject’s method on the changed
subjects rather than duplicating the logic.
Skipped entirely during data-file loading (self.env.context.get("install_mode"), the same
pattern already used by ems.planning.check_ponderation): each ems.subject.csv row sets its
own study_ids, but two subjects sharing a code can sit in different files, so the check would
see a transient state while they load. It never skips a real interactive edit through the UI.
Where the study link comes from in the data files: the study_ids:id column of
data/cat/ems.subject.csv (official subjects, re-applied on every upgrade) and of
data/custom/{eso,btx,ccff}/ems.subject.csv (the centre’s own, set once when the subject is
created and never touched again). No study file lists subjects, so a subject the centre creates
or links from the app keeps its studies across upgrades - see
data_loading.md.
product.product Sync (create() / write())Every subject needs a matching product.product so it can be sold/invoiced through an enrolment (sale.order). This is fully automatic — there is no manual “create product” step for an administrator.
flowchart TD
A([Create ems.subject]) --> B[Build product_vals from name/code]
B --> C{ems_category_academic exists?}
C -- Yes --> D[Set categ_id]
C -- No --> E[Skip categ_id]
D --> F[Create product.product]
E --> F
F --> G[ems_is_tutoria = code starts with T1_/T2_]
G --> H[subject.product_id = new product]
flowchart TD
A([Write ems.subject]) --> B{product_id missing?}
B -- Yes --> C[Self-heal: create a new product.product<br/>and link it]
B -- No --> D{name or code in vals?}
D -- Yes --> E[Push name/default_code/ems_is_tutoria<br/>to the linked product]
D -- No --> F[Nothing to sync]
flowchart TD
A([Admin user]) --> B[Community → Config → Curriculum → Subjects]
B --> C[Click New]
C --> D[Fill code + acronym + name]
D --> E[Save]
E --> F{ORM / DB validation}
F -- missing required field --> G[Error: required field constraint]
F -- duplicate code sharing a study --> H["Error: _check_code_unique_per_study"]
F -- valid --> I[(INSERT INTO ems_subject)]
I --> J[Auto-create linked product.product]
J --> K([Record created])
flowchart TD
A([Admin/Teacher/Secretary]) --> B[Community → Config → Curriculum → Subjects]
B --> C[List view: sorted by code ASC]
C --> D[Click record]
D --> E[Form view: main data · Studies / Learning Outcome / Content / Notes tabs]
flowchart TD
A([Admin user]) --> B[Open subject record]
B --> C[Edit fields]
C --> D[Save]
D --> E{Validation}
E -- required field cleared --> F[Error: required field constraint]
E -- valid --> G[(UPDATE ems_subject SET ...)]
G --> H[Sync linked product.product if name/code changed]
flowchart TD
A([Admin user]) --> B[Select subject in list]
B --> C[Action ⚙ → Delete]
C --> D[Confirm dialog]
D --> E{Linked records with a hard FK<br/>e.g. teaching assignments, grade sessions}
E -- Yes --> F[DB integrity error: cannot delete]
E -- No --> G[(DELETE FROM ems_subject WHERE id = ...)]
G --> H([Record removed; linked product.product is NOT deleted — ondelete='set null'])
Defined in security/ir.model.access.csv (lines 111–113).
| Role | Create | Read | Write | Delete | Group XML ID |
|---|---|---|---|---|---|
| Administrator | ✓ | ✓ | ✓ | ✓ | ems.group_academic_admin |
| Teacher | — | ✓ | — | — | ems.group_teacher |
| Secretary | — | ✓ | — | — | ems.group_secretary |
No record-level rules exist for this model. No portal access — subjects aren’t shown on the family/student portal directly.
ems.subject is one of the most referenced models in EMS. Selected consumers by area:
| Area | Model(s) | Field | Description |
|---|---|---|---|
| Curriculum | ems.study |
subject_ids |
Studies group their subjects |
| Curriculum | ems.outcome, ems.content |
subject_id |
Learning outcomes / content scoped to a subject |
| Contacts | ems.group |
subject_ids |
Group’s taught subjects |
| Contacts | ems.enrollment |
subject_id |
Enrolment line item |
| Employees | ems.teaching, ems.tracking |
subject_id |
Teaching assignments and follow-up |
| Employees | resource.calendar (working schedule) |
subject_id |
Weekly schedule slot |
| Attendance | ems.attendance_template, ems.attendance_session_header/_line |
subject_id |
Scheduling and roll-call |
| Planning | ems.planning |
subject_id (required) |
Planning scoped to a subject |
| Grades | ems.grade_session, ems.grade_subject_line, ems.grade_outcome_line, ems.student.year_record, ems.em_grading_wizard |
subject_id |
Grading flows |
| Communications | ems.limesurvey_recipient |
subject_id |
Survey targeting by subject |
| View | File | Notes |
|---|---|---|
| List | views/community/subject/list.xml |
Sorted by code; shows study_ids as tags |
| Form | views/community/subject/form.xml |
Main data group; Studies / Learning Outcome / Content / Notes tabs |
| Search | views/community/subject/search.xml |
— |
| Action + Menu | views/community/subject/menu.xml |
action_subject_tree, under Community → Configuration → Curriculum (sequence 3) |
| File | Purpose |
|---|---|
data/cat/ems.subject.csv |
Production catalog: Catalan VET/Baccalaureate subjects |
data/custom/btx/ems.subject.csv, data/custom/ccff/ems.subject.csv, data/custom/eso/ems.subject.csv |
Centre-specific subject catalogs |
Every optative class gathers students from all the groups of a course (1A, 1B…), so each one is an ems.group (reinforcement type, living data managed from the UI). A student keeps their main group and is also enrolled in (optative group, subject); the attendance roster and the grade session come from that enrollment. Which subject depends on the case:
| Case | Subject | Example |
|---|---|---|
| 1r-3r, the centre’s own optatives | One generic subject per course: 175_2022_OPT1/OPT2/OPT3, Optativa (1r)/(2n)/(3r), shared by every optative of that course |
Mediació (2n ESO) -> OPT2 |
| 1r-3r, optatives Esfera lists under their own name | Their own subject | Francès (1r ESO) -> FRA1; Robòtica I and II (2n ESO) -> ROB2 |
| 4t (all curricular) | Their own subject; a subject taught in two classes is one subject with two groups | Física i Química I and II (4t ESO) -> FIQ4 |
So the centre’s own offer changes every year with groups alone, without touching the repository. The generic subjects’ codes must never start with OPT: the Esfera grade importer (ems.grade_import_wizard) treats such a subject as “the student’s only optative in the study”. Since the subject’s name does not say which optative it was, the academic history copies the group’s name onto each subject line (ems.student.year_record.subject.group_name, see year_record.md).