[17.0][ADD] attachment_preview_office: preview Office attachments via LibreOffice conversion

attachment_preview_office/README.rst

@@ -0,0 +1,112 @@
+=========================
+Attachment Preview Office
+=========================
+
+.. 
+   !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
+   !! This file is generated by oca-gen-addon-readme !!
+   !! changes will be overwritten.                   !!
+   !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
+   !! source digest: sha256:0445bfa8daff7a73f252d9cdcbc27806a805f6398a32851691d0a5b07dd35454
+   !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
+
+.. |badge1| image:: https://img.shields.io/badge/maturity-Beta-yellow.png
+    :target: https://odoo-community.org/page/development-status
+    :alt: Beta
+.. |badge2| image:: https://img.shields.io/badge/licence-AGPL--3-blue.png
+    :target: http://www.gnu.org/licenses/agpl-3.0-standalone.html
+    :alt: License: AGPL-3
+.. |badge3| image:: https://img.shields.io/badge/github-OCA%2Fknowledge-lightgray.png?logo=github
+    :target: https://github.com/OCA/knowledge/tree/17.0/attachment_preview_office
+    :alt: OCA/knowledge
+.. |badge4| image:: https://img.shields.io/badge/weblate-Translate%20me-F47D42.png
+    :target: https://translation.odoo-community.org/projects/knowledge-17-0/knowledge-17-0-attachment_preview_office
+    :alt: Translate me on Weblate
+.. |badge5| image:: https://img.shields.io/badge/runboat-Try%20me-875A7B.png
+    :target: https://runboat.odoo-community.org/builds?repo=OCA/knowledge&target_branch=17.0
+    :alt: Try me on Runboat
+
+|badge1| |badge2| |badge3| |badge4| |badge5|
+
+Preview office documents (DOCX, XLSX, PPTX, macro-enabled DOCM, XLSM,
+PPTM, legacy DOC, XLS, PPT, and ODT, ODS, ODP, ODG) directly in the
+browser, including chatter attachments.
+
+The document is converted to PDF on the Odoo server using LibreOffice in
+headless mode and rendered with Odoo's native PDF.js viewer. Nothing
+leaves your server: no external viewer service (Microsoft Office Online,
+Google Docs) is involved, which makes this suitable for sensitive
+documents.
+
+Converted PDFs are cached on disk (keyed by document checksum), so each
+document is only converted once. The *Download* button still downloads
+the original file, not the converted PDF.
+
+**Table of contents**
+
+.. contents::
+   :local:
+
+Installation
+============
+
+LibreOffice must be installed on the Odoo server and the ``libreoffice``
+binary must be available in the PATH of the Odoo process:
+
+::
+
+   apt-get install libreoffice
+
+On Odoo.sh, LibreOffice is already available by default.
+
+If LibreOffice is not available, the module still installs, but the
+preview button is simply not shown for office files (the availability is
+exposed to the web client through the session info).
+
+Usage
+=====
+
+1. Attach an office file (e.g. an ``.xlsx``) to any record with a
+   chatter, or send it in a message.
+2. Click on the attachment thumbnail.
+3. The file opens in the standard Odoo file viewer, rendered as PDF.
+
+Bug Tracker
+===========
+
+Bugs are tracked on `GitHub Issues <https://github.com/OCA/knowledge/issues>`_.
+In case of trouble, please check there if your issue has already been reported.
+If you spotted it first, help us to smash it by providing a detailed and welcomed
+`feedback <https://github.com/OCA/knowledge/issues/new?body=module:%20attachment_preview_office%0Aversion:%2017.0%0A%0A**Steps%20to%20reproduce**%0A-%20...%0A%0A**Current%20behavior**%0A%0A**Expected%20behavior**>`_.
+
+Do not contact contributors directly about support or help with technical issues.
+
+Credits
+=======
+
+Authors
+-------
+
+* Jarsa
+
+Contributors
+------------
+
+- Jesús Alan Ramos Rodríguez <alan.ramos@jarsa.com>
+
+Maintainers
+-----------
+
+This module is maintained by the OCA.
+
+.. image:: https://odoo-community.org/logo.png
+   :alt: Odoo Community Association
+   :target: https://odoo-community.org
+
+OCA, or the Odoo Community Association, is a nonprofit organization whose
+mission is to support the collaborative development of Odoo features and
+promote its widespread use.
+
+This module is part of the `OCA/knowledge <https://github.com/OCA/knowledge/tree/17.0/attachment_preview_office>`_ project on GitHub.
+
+You are welcome to contribute. To learn how please visit https://odoo-community.org/page/Contribute.

attachment_preview_office/__init__.py

@@ -0,0 +1,2 @@
+from . import controllers
+from . import models

attachment_preview_office/__manifest__.py

@@ -0,0 +1,21 @@
+# Copyright 2026 Jarsa
+# License AGPL-3.0 or later (http://www.gnu.org/licenses/agpl).
+{
+    "name": "Attachment Preview Office",
+    "summary": "Preview Office attachments in the browser "
+    "by converting them to PDF with LibreOffice",
+    "version": "17.0.1.0.0",
+    "category": "Knowledge Management",
+    "website": "https://github.com/OCA/knowledge",
+    "author": "Jarsa, Odoo Community Association (OCA)",
+    "license": "AGPL-3",
+    "depends": ["mail"],
+    "assets": {
+        "web.assets_backend": [
+            "attachment_preview_office/static/src/attachment_model_patch.esm.js",
+            "attachment_preview_office/static/src/file_viewer_patch.esm.js",
+            "attachment_preview_office/static/src/file_viewer_patch.xml",
+        ],
+    },
+    "installable": True,
+}

attachment_preview_office/controllers/__init__.py

@@ -0,0 +1 @@
+from . import main

attachment_preview_office/controllers/main.py

@@ -0,0 +1,234 @@
+# Copyright 2026 Jarsa
+# Copyright 2026 Ledoweb
+# License AGPL-3.0 or later (http://www.gnu.org/licenses/agpl).
+"""LibreOffice-based conversion endpoint for office-document preview.
+
+Converts office documents (ODF *and* OOXML: ODT/ODS/ODP, DOCX/XLSX/PPTX,
+macro-enabled DOCM/XLSM/PPTM, legacy DOC/XLS/PPT, ODG) to PDF so they can be
+rendered by Odoo's native PDF.js viewer.
+LibreOffice headless must be installed; if it is absent the endpoint returns
+HTTP 503.
+"""
+
+import base64
+import hashlib
+import logging
+import os
+import subprocess
+import tempfile
+
+from odoo import http, tools
+from odoo.http import request
+
+_logger = logging.getLogger(__name__)
+
+# Office formats LibreOffice can convert to PDF (ODF + OOXML + legacy binary).
+OFFICE_EXTENSIONS = frozenset(
+    {
+        "docx",
+        "docm",
+        "xlsx",
+        "xlsm",
+        "pptx",
+        "pptm",
+        "doc",
+        "xls",
+        "ppt",
+        "odt",
+        "ods",
+        "odp",
+        "odg",
+    }
+)
+
+# Spreadsheets are exported with the SinglePageSheets option (one full-size
+# page per sheet) so columns are not cut by Calc's print pagination. The
+# option is seeded into the isolated LibreOffice profile registry instead of
+# being passed as CLI filter options: the JSON filter-options syntax is only
+# parsed by LibreOffice >= 7.4, while the registry works from 7.2.
+SPREADSHEET_EXTENSIONS = frozenset({"xlsx", "xlsm", "xls", "ods"})
+SINGLE_PAGE_SHEETS_XCU = """<?xml version="1.0" encoding="UTF-8"?>
+<oor:items xmlns:oor="http://openoffice.org/2001/registry"
+ xmlns:xs="http://www.w3.org/2001/XMLSchema"
+ xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
+ <item oor:path="/org.openoffice.Office.Common/Filter/PDF/Export">
+  <prop oor:name="SinglePageSheets" oor:op="fuse"><value>true</value></prop>
+ </item>
+</oor:items>
+"""
+# Bump to invalidate PDFs cached before a conversion-behavior change.
+CACHE_KEY_VERSION = "2"
+
+# Reject documents larger than this before spawning a (slow) conversion.
+MAX_CONTENT_BYTES = 50 * 1024 * 1024  # 50 MB
+# Per-conversion wall-clock budget (seconds).
+CONVERT_TIMEOUT = 60
+
+
+class AttachmentPreviewOfficeController(http.Controller):
+    @http.route(
+        "/attachment_preview_office/to_pdf",
+        type="http",
+        auth="user",
+        methods=["GET"],
+    )
+    def office_to_pdf(  # pylint: disable=redefined-builtin
+        self, model, field, id, filename="file", **kwargs
+    ):
+        """Convert a binary field's office document to PDF for preview.
+
+        Query params:
+            model    – Odoo model name (e.g. 'ir.attachment')
+            field    – binary field name (e.g. 'datas')
+            id       – record id (integer)
+            filename – original filename (used to derive the extension)
+        """
+        try:
+            record_id = int(id)
+        except (TypeError, ValueError):
+            return request.make_response("Bad request", status=400)
+
+        # Extension allow-list (on the claimed filename) before touching the DB.
+        ext = os.path.splitext(filename)[-1].lstrip(".").lower()
+        if ext not in OFFICE_EXTENSIONS:
+            return request.make_response(
+                "Extension not supported for conversion", status=415
+            )
+
+        record = request.env[model].browse(record_id).exists()
+        if not record:
+            return request.make_response("Not found", status=404)
+        # Record-level ACL: raises AccessError (-> 403) if the user can't read.
+        record.check_access_rights("read")
+        record.check_access_rule("read")
+
+        # The field must exist on the model AND be a binary field — never read
+        # an arbitrary attribute supplied in the query string.
+        model_field = record._fields.get(field)
+        if model_field is None or model_field.type != "binary":
+            return request.make_response("Invalid field", status=400)
+
+        raw = record[field]
+        if not raw:
+            return request.make_response("No content", status=404)
+
+        content = base64.b64decode(raw)
+        if len(content) > MAX_CONTENT_BYTES:
+            return request.make_response("Document too large to preview", status=413)
+
+        pdf_bytes = self._get_pdf_cached(content, ext)
+        if pdf_bytes is None:
+            return request.make_response(
+                "LibreOffice not available — cannot convert document", status=503
+            )
+
+        return request.make_response(
+            pdf_bytes,
+            headers=[
+                ("Content-Type", "application/pdf"),
+                (
+                    "Content-Disposition",
+                    f'inline; filename="{os.path.splitext(filename)[0]}.pdf"',
+                ),
+                ("Cache-Control", "private, max-age=3600"),
+            ],
+        )
+
+    # -- conversion + cache ----------------------------------------------------
+
+    def _get_pdf_cached(self, content, ext):
+        """Return converted PDF bytes, using a checksum-keyed disk cache.
+
+        The cache lives under the Odoo filestore so it is shared across workers
+        on the same host; identical documents are converted only once. All cache
+        I/O is best-effort: a miss or any filesystem error falls back silently to
+        a fresh conversion.
+        """
+        # The extension is part of the key: the export options depend on it.
+        key = f"{hashlib.sha256(content).hexdigest()}-{ext}-{CACHE_KEY_VERSION}"
+        cache_path = self._cache_path(key)
+        if cache_path and os.path.exists(cache_path):
+            try:
+                with open(cache_path, "rb") as fh:
+                    return fh.read()
+            except OSError:
+                _logger.warning(
+                    "attachment_preview_office: cache read failed", exc_info=True
+                )
+
+        pdf_bytes = self._libreoffice_to_pdf(content, ext)
+        if pdf_bytes and cache_path:
+            try:
+                # Atomic publish so a concurrent reader never sees a partial file.
+                tmp = f"{cache_path}.{os.getpid()}.tmp"
+                with open(tmp, "wb") as fh:
+                    fh.write(pdf_bytes)
+                os.replace(tmp, cache_path)
+            except OSError:
+                _logger.warning(
+                    "attachment_preview_office: cache write failed", exc_info=True
+                )
+        return pdf_bytes
+
+    @staticmethod
+    def _cache_path(key):
+        """Return the cache file path for ``key`` under the filestore, or None."""
+        try:
+            base = os.path.join(
+                tools.config.filestore(request.env.cr.dbname),
+                "attachment_preview_office_cache",
+            )
+            os.makedirs(base, exist_ok=True)
+            return os.path.join(base, f"{key}.pdf")
+        except Exception:  # pragma: no cover - never let caching break preview
+            return None
+
+    @staticmethod
+    def _libreoffice_to_pdf(content, ext):
+        """Run LibreOffice headless conversion. Returns PDF bytes or None.
+
+        Each invocation uses an isolated ``UserInstallation`` profile so that
+        concurrent conversions do not collide on the shared LibreOffice profile
+        lock — the classic "another instance is already running" failure that
+        otherwise makes this endpoint flaky under any real load.
+        """
+        try:
+            with tempfile.TemporaryDirectory() as tmpdir:
+                src = os.path.join(tmpdir, f"source.{ext}")
+                with open(src, "wb") as fh:
+                    fh.write(content)
+                profile = os.path.join(tmpdir, "louser")
+                if ext in SPREADSHEET_EXTENSIONS:
+                    user_dir = os.path.join(profile, "user")
+                    os.makedirs(user_dir, exist_ok=True)
+                    xcu = os.path.join(user_dir, "registrymodifications.xcu")
+                    with open(xcu, "w", encoding="utf-8") as fh:
+                        fh.write(SINGLE_PAGE_SHEETS_XCU)
+                result = subprocess.run(
+                    [
+                        "libreoffice",
+                        f"-env:UserInstallation=file://{profile}",
+                        "--headless",
+                        "--convert-to",
+                        "pdf",
+                        "--outdir",
+                        tmpdir,
+                        src,
+                    ],
+                    timeout=CONVERT_TIMEOUT,
+                    capture_output=True,
+                )
+                if result.returncode != 0:
+                    _logger.warning(
+                        "LibreOffice conversion failed (exit %s): %s",
+                        result.returncode,
+                        result.stderr.decode("utf-8", errors="replace")[-500:],
+                    )
+                    return None
+                pdf_path = os.path.join(tmpdir, "source.pdf")
+                if not os.path.exists(pdf_path):
+                    return None
+                with open(pdf_path, "rb") as fh:
+                    return fh.read()
+        except (FileNotFoundError, subprocess.TimeoutExpired, OSError):
+            return None