Data managers moving into or syncing with Odoo hit the same wall every migration cycle: the native Import tool is stateless. You rebuild the column-to-field mapping by hand for every file, re-tick "Don't import" on the same junk columns, and re-explain to a colleague which sheet feeds res.partner versus product.template. When a load half-fails, Odoo aborts the batch with a single terse error instead of telling you which of the 4,000 rows was bad. There is no way to save a mapping, no way to transform values on the way in (trim, upper-case, map "Yes/No" to booleans, look up an external ID), and no way to run the same export on a schedule. For teams doing repeated onboarding, multi-entity consolidation, or nightly syncs, that native ceiling turns a five-minute job into an afternoon.

Advanced Import / Export Templating is a build-to-order Odoo module that turns those one-off mappings into first-class, reusable records. We add an ir.import.template model (and matching export template) that stores the target model, the column-to-field mapping, per-field transforms, and validation rules as saved configuration you name once and reuse forever. Mappings target any model in the registry — including your custom addons — and support related fields via parent_id/name dot-notation and external-ID resolution (__export__/XML IDs) so relational links resolve correctly instead of failing silently. Transforms are declarative (trim, case, date-format normalization, default-on-empty, regex replace, boolean/selection value maps) with a hook for a Python expression when a client's data is genuinely messy.

Technically, the module is a clean Odoo addon: a proper __manifest__.py, models built on models.Model with fields and @api.depends computes where derived state is needed, an OWL/XML wizard for the upload-preview-validate-commit flow, and ir.model.access.csv plus record rules so template ownership and execution are permission-scoped (import/export can be gated by group). Validation runs in a dry-run pass first: every row is checked against field types, required fields, selection domains, and your custom rules, and failures are collected into a downloadable error-row report (row number, column, value, reason) rendered via a QWeb report — so you fix the source file, not guess. Both CSV and XLSX are supported on read and write. Recurring jobs are wired through ir.cron / automated actions so a template can pull a file from a configured location (or push an export) on a schedule, and every run is logged. The same templates are callable over XML-RPC / JSON-RPC, so an external ETL or another system can trigger a validated load programmatically.

Because this is build-to-order, ECOSIRE builds the module against your exact Odoo version (17.0, 18.0, or 19.0), your edition (Community or Enterprise — we note where Enterprise-only fields or studio models affect mapping), and your real target models and file formats. We start with a short scoping call to confirm the models, the transforms your data actually needs, and where scheduled files land. Typical delivery is one working week from confirmed scope. You get installable source, a UAT pass on your staging database with a rollback plan, documentation, a training session for your data team, and a post-go-live support window — not a black-box zip.