From cb93dbe05a6bdcc98275da8737307ebd749abc26 Mon Sep 17 00:00:00 2001
From: Colin Maudry <colin@maudry.com>
Date: Fri, 12 Jun 2026 13:58:50 +0200
Subject: [PATCH] feat: chargements de pages best-effort au boot (tableau,
 sources) (#78)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 src/figures.py           |  8 ++++--
 src/pages/tableau.py     | 15 +++++++---
 src/utils/__init__.py    | 16 +++++++++++
 tests/test_page_loads.py | 61 ++++++++++++++++++++++++++++++++++++++++
 4 files changed, 93 insertions(+), 7 deletions(-)
 create mode 100644 tests/test_page_loads.py

diff --git a/src/figures.py b/src/figures.py
index 4670723..966fd9b 100644
--- a/src/figures.py
+++ b/src/figures.py
@@ -1,6 +1,5 @@
 from datetime import datetime
 from typing import Literal
-from urllib.error import HTTPError, URLError
 
 import dash_bootstrap_components as dbc
 import dash_leaflet as dl
@@ -120,9 +119,12 @@ def get_barchart_sources(lff: pl.LazyFrame, type_date: str):
 
 def get_sources_tables(source_path) -> html.Div:
     try:
+        if not source_path:
+            raise ValueError("SOURCE_STATS_CSV_PATH non défini")
         dff = pl.read_csv(source_path)
-    except (URLError, HTTPError):
-        return html.Div("Erreur de connexion")
+    except Exception as e:
+        logger.warning(f"Sources de données indisponibles ({e})")
+        return html.Div("Sources de données momentanément indisponibles.")
     dff = dff.with_columns(
         (
             pl.lit('<a href = "')
diff --git a/src/pages/tableau.py b/src/pages/tableau.py
index 167b8ca..805dcb0 100644
--- a/src/pages/tableau.py
+++ b/src/pages/tableau.py
@@ -21,7 +21,7 @@ from dash import (
 
 from src.db import query_marches, schema
 from src.figures import DataTable, make_column_picker
-from src.utils import get_last_modified, logger
+from src.utils import get_data_update_timestamp, logger
 from src.utils.seo import META_CONTENT
 from src.utils.table import (
     COLUMNS,
@@ -33,9 +33,16 @@ from src.utils.table import (
 )
 from src.utils.tracking import track_search
 
-update_date_timestamp = get_last_modified(os.getenv("DATA_FILE_PARQUET_PATH", ""))
-update_date = datetime.fromtimestamp(update_date_timestamp).strftime("%d/%m/%Y")
-update_date_iso = datetime.fromtimestamp(update_date_timestamp).isoformat()
+update_date_timestamp = get_data_update_timestamp(
+    os.getenv("DATA_FILE_PARQUET_PATH", ""),
+    os.getenv("DUCKDB_PATH", "./decp.duckdb"),
+)
+if update_date_timestamp is not None:
+    update_date = datetime.fromtimestamp(update_date_timestamp).strftime("%d/%m/%Y")
+    update_date_iso = datetime.fromtimestamp(update_date_timestamp).isoformat()
+else:
+    update_date = "date inconnue"
+    update_date_iso = ""
 
 
 NAME = "Tableau"
diff --git a/src/utils/__init__.py b/src/utils/__init__.py
index 797209d..4a3a8b9 100644
--- a/src/utils/__init__.py
+++ b/src/utils/__init__.py
@@ -41,3 +41,19 @@ DOMAIN_NAME = (
     if os.getenv("DEVELOPMENT", "False").lower() == "true"
     else "decp.info"
 )
+
+
+def get_data_update_timestamp(
+    parquet_path: str, fallback_path: str | None = None
+) -> float | None:
+    """Date de MAJ des données, best-effort, sans jamais lever (usage au boot)."""
+    try:
+        return get_last_modified(parquet_path)
+    except Exception as e:
+        logger.warning(f"Date de mise à jour des données indisponible ({e})")
+    if fallback_path:
+        try:
+            return os.path.getmtime(fallback_path)
+        except OSError:
+            pass
+    return None
diff --git a/tests/test_page_loads.py b/tests/test_page_loads.py
new file mode 100644
index 0000000..039090a
--- /dev/null
+++ b/tests/test_page_loads.py
@@ -0,0 +1,61 @@
+import os
+
+
+def test_update_timestamp_falls_back_to_db_mtime(tmp_path, monkeypatch):
+    import src.utils as u
+    from src.utils import get_data_update_timestamp
+
+    def boom(*a, **k):
+        raise RuntimeError("net down")
+
+    monkeypatch.setattr(u, "get_last_modified", boom)
+    fb = tmp_path / "decp.duckdb"
+    fb.write_bytes(b"x")
+    assert get_data_update_timestamp("http://x", str(fb)) == os.path.getmtime(str(fb))
+
+
+def test_update_timestamp_none_when_all_fail(monkeypatch):
+    import src.utils as u
+    from src.utils import get_data_update_timestamp
+
+    def boom(*a, **k):
+        raise RuntimeError("net down")
+
+    monkeypatch.setattr(u, "get_last_modified", boom)
+    assert get_data_update_timestamp("http://x", None) is None
+
+
+def test_update_timestamp_nominal(monkeypatch):
+    import src.utils as u
+    from src.utils import get_data_update_timestamp
+
+    monkeypatch.setattr(u, "get_last_modified", lambda p: 123.0)
+    assert get_data_update_timestamp("http://x", None) == 123.0
+
+
+def test_sources_tables_none_path():
+    from src.figures import get_sources_tables
+
+    div = get_sources_tables(None)
+    assert "indisponible" in str(div.children).lower()
+
+
+def test_sources_tables_missing_file():
+    from src.figures import get_sources_tables
+
+    div = get_sources_tables("/does/not/exist.csv")
+    assert "indisponible" in str(div.children).lower()
+
+
+def test_sources_tables_valid_csv(tmp_path):
+    from dash import dash_table
+
+    from src.figures import get_sources_tables
+
+    csv = tmp_path / "s.csv"
+    csv.write_text(
+        "nom,organisation,nb_marchés,nb_acheteurs,code,url,unique\n"
+        "Source A,Org A,5,2,XA,http://a,1\n"
+    )
+    div = get_sources_tables(str(csv))
+    assert isinstance(div.children, dash_table.DataTable)