Fix release blockers and deployment build

scripts/benchmark.py

@@ -4,6 +4,7 @@ from __future__ import annotations
 import argparse
 import hashlib
 import json
+import math
 import os
 import platform
 import tracemalloc
@@ -58,7 +59,7 @@ def _percentile(values: list[float], p: float) -> float:
     if not values:
         return 0.0
     ordered = sorted(values)
-    index = round((len(ordered) - 1) * p / 100)
+    index = max(0, math.ceil(len(ordered) * p / 100) - 1)
     return ordered[index]
 
 
@@ -98,7 +99,7 @@ class BenchmarkArtifact:
 
 def _load_text(path: Path) -> str:
     _is_supported_file_path(path)
-    return path.read_text(encoding="utf-8")
+    return path.read_text(encoding="utf-8-sig")
 
 
 def load_benchmark_dataset(path: Path) -> dict[str, Any]:
@@ -297,8 +298,6 @@ def _create_seed_jobs(
         )
         job = JobPosting.objects.create(
             employer=employer,
-            requirements=payload["job"]["requirements"],
-            benefits=payload["job"]["benefits"],
             description_html_sanitized=payload["job"]["description_text"],
             extraction_confidence=Decimal("0.95"),
             analysis_features={},

scripts/feedback_learning_report.py

@@ -11,7 +11,7 @@ import django
 os.environ.setdefault("DJANGO_SETTINGS_MODULE", "config.settings")
 django.setup()
 
-from apps.profiles.models import SearchProfile
+from apps.profiles.models import SearchProfile  # noqa: E402
 
 
 def _load_profile(profile_id: int | None, username: str | None) -> SearchProfile:
@@ -35,7 +35,9 @@ def _collect_feedback_churn(profile: SearchProfile) -> dict:
     false_negatives = []
 
     for feedback in all_feedback:
-        learning = feedback.metadata.get("learning") if isinstance(feedback.metadata, dict) else None
+        learning = (
+            feedback.metadata.get("learning") if isinstance(feedback.metadata, dict) else None
+        )
         if not isinstance(learning, dict):
             continue
         reason_code = learning.get("reason_code", "")
@@ -112,7 +114,10 @@ def main(argv: list[str] | None = None) -> int:
             ]
             + [f" - {item['feature']}: {item['delta']:+.3f}" for item in report["weight_churn"]]
             + [f"\nMogelijke non-learn false negatives (max {args.top}):"]
-            + [f" - {item['job_id']} ({item['reason_code']})" for item in feedback_report["queued_non_learning_signals"][: args.top]]
+            + [
+                f" - {item['job_id']} ({item['reason_code']})"
+                for item in feedback_report["queued_non_learning_signals"][: args.top]
+            ]
         )
         + "\n",
         encoding="utf-8",

scripts/release_smoke.py

@@ -109,24 +109,27 @@ def _run_fixture_import(*, fixture: str, url: str, source_name: str, source_type
 
 
 def _assert_idempotent_import(url: str) -> tuple[JobPosting, int]:
-    aliases = list(
-        JobSourceAlias.objects.filter(url=url).select_related("job").order_by("pk")
-    )
+    aliases = list(JobSourceAlias.objects.filter(url=url).select_related("job").order_by("pk"))
     if not aliases:
         raise RuntimeError("Release-smoke: fixtureimport leverde geen alias op.")
     job_ids = {alias.job_id for alias in aliases if alias.job_id}
     if len(job_ids) != 1:
         raise RuntimeError(
-            f"Release-smoke: fixtureimport leidde niet tot één aliascluster (aantal jobids={len(job_ids)})."
+            "Release-smoke: fixtureimport leidde niet tot één aliascluster "
+            f"(aantal jobids={len(job_ids)})."
         )
     return aliases[-1].job, len(aliases)
 
 
-def _build_report(*, label: str, alias_count: int, job: JobPosting, profile: SearchProfile) -> dict[str, object]:
+def _build_report(
+    *, label: str, alias_count: int, job: JobPosting, profile: SearchProfile
+) -> dict[str, object]:
     return {
         "label": label,
         "alias_count": alias_count,
-        "source_job_count": JobPosting.objects.filter(source_aliases__url=job.canonical_url).count(),
+        "source_job_count": JobPosting.objects.filter(
+            source_aliases__url=job.canonical_url
+        ).count(),
         "job_total_count": JobPosting.objects.count(),
         "feedback_count": Feedback.objects.filter(user=profile.user, job=job).count(),
         "application_exists": JobPosting.objects.filter(id=job.id).exists(),