airflow-pytest-plugin 0.2.1__tar.gz → 0.3.1__tar.gz

{airflow_pytest_plugin-0.2.1 → airflow_pytest_plugin-0.3.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: airflow-pytest-plugin
-Version: 0.2.1
+Version: 0.3.1
 Summary: View airflow-pytest-operator results in the Airflow 3 web UI.
 Project-URL: Homepage, https://github.com/IKrysanov/airflow-pytest-plugin
 Project-URL: Repository, https://github.com/IKrysanov/airflow-pytest-plugin
@@ -70,7 +70,9 @@ The operator runs a `pytest` suite as an Airflow task and parses the JUnit
 report into a structured result. This plugin archives each of those reports —
 keyed by `dag_id / run_id / task_id / try` — and serves a small web UI to browse
 them: pass/fail counts per run, durations, and the per-test breakdown (with
-failure messages) for any run.
+failure messages) for any run. On top of that it adds **cross-run analytics** —
+flaky-test detection, per-test history, run-to-run comparison, a duration
+histogram, and a searchable catalogue of unique tests.
 
 It has two halves that share one on-disk layout:
 
@@ -96,19 +98,34 @@ It has two halves that share one on-disk layout:
 ## Screenshots
 
 **Overview** — the run list with the historical chart (per-status legend
-toggles, run numbers, and a carousel beyond 30 runs), KPI cards, and Airflow-
-matched colours and font:
+toggles, run numbers, a carousel beyond 30 runs, an optional **pass-rate trend
+line** with a success-threshold gridline, and **tick runs in the list to filter
+the chart to just their trend**) beside a **flaky-tests panel** (with its own
+search and a quarantined-only toggle), KPI cards (including a clickable **unique
+tests** count), and Airflow-matched colours and font:
 
 ![Pytest Reports — overview](https://raw.githubusercontent.com/IKrysanov/airflow-pytest-plugin/main/docs/screenshots/overview.png)
 
-**A single run** — a clickable success donut (pass-rate in the centre; click a
-slice to filter by status), the counts, and every test's captured output on
-expand:
+**A single run** — a clickable success donut (pass-rate over the test count;
+click a slice to filter by status), a **test-duration histogram** (10-second
+buckets, scrollable), case search / group-by-module, and every test's captured
+output on expand:
 
 ![Pytest Reports — a single run](https://raw.githubusercontent.com/IKrysanov/airflow-pytest-plugin/main/docs/screenshots/detail.png)
 
-**Failed tests** — clicking the *Failures* KPI lists every failed/errored case
-across the visible runs (paginated); a row jumps to that run:
+**Flaky tests & comparison** — from a run, *Flaky tests* lists the tests that
+both pass and fail across recent runs, each with a recent-outcome strip, a
+**flakiness score**, a **trend** arrow, and a **quarantine** badge, over a
+configurable analysis window; *Compare to previous* diffs it against the prior
+run; expanding a case offers its full **history**:
+
+![Pytest Reports — flaky tests](https://raw.githubusercontent.com/IKrysanov/airflow-pytest-plugin/main/docs/screenshots/flaky.png)
+
+**Unique tests & failures** — the *Unique tests* KPI opens the searchable,
+paginated catalogue of every distinct test; the *Failures* KPI lists every
+failed/errored case across the visible runs — a row jumps to that run:
+
+![Pytest Reports — unique tests](https://raw.githubusercontent.com/IKrysanov/airflow-pytest-plugin/main/docs/screenshots/unique.png)
 
 ![Pytest Reports — failed tests](https://raw.githubusercontent.com/IKrysanov/airflow-pytest-plugin/main/docs/screenshots/failures.png)
 
@@ -214,10 +231,15 @@ runtime. Endpoints (relative to the mount):
 | `GET /api/reports?dag_id=&run_id=` | summaries, newest first |
 | `GET /api/reports/{report_id}` | one report with per-case rows |
 | `GET /api/failures?dag_id=&run_id=&task_id=` | failed/errored cases across the visible runs |
+| `GET /api/compare?base=&head=` | per-test diff between two runs (newly failed / fixed / …) |
+| `GET /api/flaky?dag_id=&task_id=&window=` | flaky tests with score, trend, and a quarantine flag |
+| `GET /api/test-history?dag_id=&task_id=&node_id=&limit=` | one test's outcome per run |
+| `GET /api/unique-tests?dag_id=&task_id=&run_id=&full=` | distinct test count (+ list when `full`) |
 | `DELETE /api/reports/{report_id}` | delete a report (RBAC-gated) |
 | `GET /api/reports/{report_id}/allure.zip` | raw Allure results as a zip (if any) |
-| `GET /api/health` | `{"status": "ok"}` |
-| `GET /api/docs` | OpenAPI docs |
+| `GET /api/health` | liveness + readiness: `status`, `ready`, `reports_root`(+`_exists`), `auth`, `secure_xml` |
+| `GET /api/version` | `{"name": ..., "version": ...}` from package metadata |
+| `GET /api/docs` | OpenAPI docs (Swagger UI) |
 
 The reads (`GET`) and the delete are gated by Airflow RBAC — see below.
 
@@ -274,6 +296,61 @@ Download them from a report's detail view, or `GET
 | `AIRFLOW_PYTEST_REPORTS_ROOT` (env) | — | report root (highest precedence) |
 | `[pytest_reports] reports_root` (cfg) | — | report root |
 | built-in default | `/opt/airflow/pytest-reports` | fallback |
+| `AIRFLOW_PYTEST_PLUGIN_ENABLE` (env) | `True` | reader on/off — see below |
+| `AIRFLOW_PYTEST_SCAN_CACHE_TTL` (env) | `2.0` | seconds a directory scan is reused (`0` disables) |
+| `AIRFLOW_PYTEST_RETENTION_MAX_AGE_DAYS` (env/cfg) | — | delete runs older than N days |
+| `AIRFLOW_PYTEST_RETENTION_MAX_RUNS` (env/cfg) | — | keep at most N newest runs per dag·task |
+| `AIRFLOW_PYTEST_RETENTION_MAX_TOTAL_MB` (env/cfg) | — | total report-tree budget in MB |
+| `AIRFLOW_PYTEST_FLAKY_WINDOW` (env/cfg) | `30` | default recent runs the flaky detector scans |
+| `AIRFLOW_PYTEST_FLAKY_QUARANTINE_SCORE` (env/cfg) | `0.5` | flakiness score (0–1) that flags a test for quarantine |
+| `AIRFLOW_PYTEST_FLAKY_MIN_SCORE` (env/cfg) | `0.1` | flakiness score (0–1) below which a test is not counted as flaky |
+| `AIRFLOW_PYTEST_SUCCESS_THRESHOLD` (env/cfg) | `0.85` | pass-rate (0–1) over executed tests at/above which a run counts as successful (*Passing runs*); `1.0` = strict, zero failures/errors |
+
+**Enable / disable the reader.** Set `AIRFLOW_PYTEST_PLUGIN_ENABLE` to a falsey
+value (`0`, `false`, `no`, `off`) to stop the plugin registering its UI and API
+with Airflow; any other value, or leaving it unset, keeps it on (the default).
+This is a kill switch for the reader only — the producer-side
+`ArchivingResultParser` still archives reports regardless. It is read once at
+plugin discovery, so toggling it takes effect on the next API-server restart.
+
+```bash
+export AIRFLOW_PYTEST_PLUGIN_ENABLE=false   # hide the Pytest Reports UI + API
+```
+
+**Scan cache.** Loading the home page hits several summary-driven endpoints (the run
+list, the flaky panel, the unique-tests count), and the filter box queries as you
+type. To avoid walking the report tree once per call, the filesystem source reuses a
+single scan for `AIRFLOW_PYTEST_SCAN_CACHE_TTL` seconds (default `2.0`; deletes
+invalidate it immediately). New runs therefore appear within a couple of seconds (or
+on **Refresh**); set it to `0` for no caching, or higher on a very large tree.
+
+## Retention (auto-cleanup)
+
+Reports accumulate forever unless you prune them. Set any of the
+`AIRFLOW_PYTEST_RETENTION_*` knobs above (all opt-in; unset = keep everything) and
+schedule `prune_reports` from a maintenance DAG:
+
+```python
+from airflow import DAG
+from airflow.providers.standard.operators.python import PythonOperator
+from airflow_pytest_plugin import prune_reports
+
+with DAG("pytest_reports_retention", schedule="@daily", catchup=False, ...):
+    PythonOperator(task_id="prune", python_callable=prune_reports)
+```
+
+Each knob is a dimension and they combine as a union — a run is deleted if **any**
+applies:
+
+- **age** — older than `…_MAX_AGE_DAYS`;
+- **count** — beyond the newest `…_MAX_RUNS` of its dag·task;
+- **size** — oldest-first until the tree fits `…_MAX_TOTAL_MB`.
+
+The **newest run of each dag·task is always kept**, so a task's latest result never
+disappears. `prune_reports(dry_run=True)` reports what *would* go without deleting
+(its `RetentionResult` carries `deleted`, `freed_bytes`, `scanned`). Cleanup is
+scheduler-driven — the plugin never deletes on its own. For a custom policy, build a
+`RetentionPolicy(...)` and pass it (`prune_reports(policy)`).
 
 ## Architecture (SOLID)
 
@@ -285,6 +362,7 @@ Mirrors the operator's layering — each piece has one reason to change:
 | `producer.ArchivingResultParser` | write JUnit XML + `meta.json` (extends the operator's parser) |
 | `sources.ReportSource` / `FileSystemReportSource` | read/index reports behind an interface (Dependency Inversion) |
 | `web.create_app` | map HTTP onto a `ReportSource` — knows nothing about the filesystem |
+| `retention` | pure `select_expired` decision + a `prune` orchestrator over any `ReportSource` |
 | `plugin.PytestReportsPlugin` | register the app with Airflow |
 | `compat` | the only module that imports Airflow; version differences resolved once |
 | `models` | JSON-serializable view types; the web layer never sees operator types |

{airflow_pytest_plugin-0.2.1 → airflow_pytest_plugin-0.3.1}/README.md

@@ -26,7 +26,9 @@ The operator runs a `pytest` suite as an Airflow task and parses the JUnit
 report into a structured result. This plugin archives each of those reports —
 keyed by `dag_id / run_id / task_id / try` — and serves a small web UI to browse
 them: pass/fail counts per run, durations, and the per-test breakdown (with
-failure messages) for any run.
+failure messages) for any run. On top of that it adds **cross-run analytics** —
+flaky-test detection, per-test history, run-to-run comparison, a duration
+histogram, and a searchable catalogue of unique tests.
 
 It has two halves that share one on-disk layout:
 
@@ -52,19 +54,34 @@ It has two halves that share one on-disk layout:
 ## Screenshots
 
 **Overview** — the run list with the historical chart (per-status legend
-toggles, run numbers, and a carousel beyond 30 runs), KPI cards, and Airflow-
-matched colours and font:
+toggles, run numbers, a carousel beyond 30 runs, an optional **pass-rate trend
+line** with a success-threshold gridline, and **tick runs in the list to filter
+the chart to just their trend**) beside a **flaky-tests panel** (with its own
+search and a quarantined-only toggle), KPI cards (including a clickable **unique
+tests** count), and Airflow-matched colours and font:
 
 ![Pytest Reports — overview](https://raw.githubusercontent.com/IKrysanov/airflow-pytest-plugin/main/docs/screenshots/overview.png)
 
-**A single run** — a clickable success donut (pass-rate in the centre; click a
-slice to filter by status), the counts, and every test's captured output on
-expand:
+**A single run** — a clickable success donut (pass-rate over the test count;
+click a slice to filter by status), a **test-duration histogram** (10-second
+buckets, scrollable), case search / group-by-module, and every test's captured
+output on expand:
 
 ![Pytest Reports — a single run](https://raw.githubusercontent.com/IKrysanov/airflow-pytest-plugin/main/docs/screenshots/detail.png)
 
-**Failed tests** — clicking the *Failures* KPI lists every failed/errored case
-across the visible runs (paginated); a row jumps to that run:
+**Flaky tests & comparison** — from a run, *Flaky tests* lists the tests that
+both pass and fail across recent runs, each with a recent-outcome strip, a
+**flakiness score**, a **trend** arrow, and a **quarantine** badge, over a
+configurable analysis window; *Compare to previous* diffs it against the prior
+run; expanding a case offers its full **history**:
+
+![Pytest Reports — flaky tests](https://raw.githubusercontent.com/IKrysanov/airflow-pytest-plugin/main/docs/screenshots/flaky.png)
+
+**Unique tests & failures** — the *Unique tests* KPI opens the searchable,
+paginated catalogue of every distinct test; the *Failures* KPI lists every
+failed/errored case across the visible runs — a row jumps to that run:
+
+![Pytest Reports — unique tests](https://raw.githubusercontent.com/IKrysanov/airflow-pytest-plugin/main/docs/screenshots/unique.png)
 
 ![Pytest Reports — failed tests](https://raw.githubusercontent.com/IKrysanov/airflow-pytest-plugin/main/docs/screenshots/failures.png)
 
@@ -170,10 +187,15 @@ runtime. Endpoints (relative to the mount):
 | `GET /api/reports?dag_id=&run_id=` | summaries, newest first |
 | `GET /api/reports/{report_id}` | one report with per-case rows |
 | `GET /api/failures?dag_id=&run_id=&task_id=` | failed/errored cases across the visible runs |
+| `GET /api/compare?base=&head=` | per-test diff between two runs (newly failed / fixed / …) |
+| `GET /api/flaky?dag_id=&task_id=&window=` | flaky tests with score, trend, and a quarantine flag |
+| `GET /api/test-history?dag_id=&task_id=&node_id=&limit=` | one test's outcome per run |
+| `GET /api/unique-tests?dag_id=&task_id=&run_id=&full=` | distinct test count (+ list when `full`) |
 | `DELETE /api/reports/{report_id}` | delete a report (RBAC-gated) |
 | `GET /api/reports/{report_id}/allure.zip` | raw Allure results as a zip (if any) |
-| `GET /api/health` | `{"status": "ok"}` |
-| `GET /api/docs` | OpenAPI docs |
+| `GET /api/health` | liveness + readiness: `status`, `ready`, `reports_root`(+`_exists`), `auth`, `secure_xml` |
+| `GET /api/version` | `{"name": ..., "version": ...}` from package metadata |
+| `GET /api/docs` | OpenAPI docs (Swagger UI) |
 
 The reads (`GET`) and the delete are gated by Airflow RBAC — see below.
 
@@ -230,6 +252,61 @@ Download them from a report's detail view, or `GET
 | `AIRFLOW_PYTEST_REPORTS_ROOT` (env) | — | report root (highest precedence) |
 | `[pytest_reports] reports_root` (cfg) | — | report root |
 | built-in default | `/opt/airflow/pytest-reports` | fallback |
+| `AIRFLOW_PYTEST_PLUGIN_ENABLE` (env) | `True` | reader on/off — see below |
+| `AIRFLOW_PYTEST_SCAN_CACHE_TTL` (env) | `2.0` | seconds a directory scan is reused (`0` disables) |
+| `AIRFLOW_PYTEST_RETENTION_MAX_AGE_DAYS` (env/cfg) | — | delete runs older than N days |
+| `AIRFLOW_PYTEST_RETENTION_MAX_RUNS` (env/cfg) | — | keep at most N newest runs per dag·task |
+| `AIRFLOW_PYTEST_RETENTION_MAX_TOTAL_MB` (env/cfg) | — | total report-tree budget in MB |
+| `AIRFLOW_PYTEST_FLAKY_WINDOW` (env/cfg) | `30` | default recent runs the flaky detector scans |
+| `AIRFLOW_PYTEST_FLAKY_QUARANTINE_SCORE` (env/cfg) | `0.5` | flakiness score (0–1) that flags a test for quarantine |
+| `AIRFLOW_PYTEST_FLAKY_MIN_SCORE` (env/cfg) | `0.1` | flakiness score (0–1) below which a test is not counted as flaky |
+| `AIRFLOW_PYTEST_SUCCESS_THRESHOLD` (env/cfg) | `0.85` | pass-rate (0–1) over executed tests at/above which a run counts as successful (*Passing runs*); `1.0` = strict, zero failures/errors |
+
+**Enable / disable the reader.** Set `AIRFLOW_PYTEST_PLUGIN_ENABLE` to a falsey
+value (`0`, `false`, `no`, `off`) to stop the plugin registering its UI and API
+with Airflow; any other value, or leaving it unset, keeps it on (the default).
+This is a kill switch for the reader only — the producer-side
+`ArchivingResultParser` still archives reports regardless. It is read once at
+plugin discovery, so toggling it takes effect on the next API-server restart.
+
+```bash
+export AIRFLOW_PYTEST_PLUGIN_ENABLE=false   # hide the Pytest Reports UI + API
+```
+
+**Scan cache.** Loading the home page hits several summary-driven endpoints (the run
+list, the flaky panel, the unique-tests count), and the filter box queries as you
+type. To avoid walking the report tree once per call, the filesystem source reuses a
+single scan for `AIRFLOW_PYTEST_SCAN_CACHE_TTL` seconds (default `2.0`; deletes
+invalidate it immediately). New runs therefore appear within a couple of seconds (or
+on **Refresh**); set it to `0` for no caching, or higher on a very large tree.
+
+## Retention (auto-cleanup)
+
+Reports accumulate forever unless you prune them. Set any of the
+`AIRFLOW_PYTEST_RETENTION_*` knobs above (all opt-in; unset = keep everything) and
+schedule `prune_reports` from a maintenance DAG:
+
+```python
+from airflow import DAG
+from airflow.providers.standard.operators.python import PythonOperator
+from airflow_pytest_plugin import prune_reports
+
+with DAG("pytest_reports_retention", schedule="@daily", catchup=False, ...):
+    PythonOperator(task_id="prune", python_callable=prune_reports)
+```
+
+Each knob is a dimension and they combine as a union — a run is deleted if **any**
+applies:
+
+- **age** — older than `…_MAX_AGE_DAYS`;
+- **count** — beyond the newest `…_MAX_RUNS` of its dag·task;
+- **size** — oldest-first until the tree fits `…_MAX_TOTAL_MB`.
+
+The **newest run of each dag·task is always kept**, so a task's latest result never
+disappears. `prune_reports(dry_run=True)` reports what *would* go without deleting
+(its `RetentionResult` carries `deleted`, `freed_bytes`, `scanned`). Cleanup is
+scheduler-driven — the plugin never deletes on its own. For a custom policy, build a
+`RetentionPolicy(...)` and pass it (`prune_reports(policy)`).
 
 ## Architecture (SOLID)
 
@@ -241,6 +318,7 @@ Mirrors the operator's layering — each piece has one reason to change:
 | `producer.ArchivingResultParser` | write JUnit XML + `meta.json` (extends the operator's parser) |
 | `sources.ReportSource` / `FileSystemReportSource` | read/index reports behind an interface (Dependency Inversion) |
 | `web.create_app` | map HTTP onto a `ReportSource` — knows nothing about the filesystem |
+| `retention` | pure `select_expired` decision + a `prune` orchestrator over any `ReportSource` |
 | `plugin.PytestReportsPlugin` | register the app with Airflow |
 | `compat` | the only module that imports Airflow; version differences resolved once |
 | `models` | JSON-serializable view types; the web layer never sees operator types |

{airflow_pytest_plugin-0.2.1 → airflow_pytest_plugin-0.3.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "airflow-pytest-plugin"
-version = "0.2.1"
+version = "0.3.1"
 description = "View airflow-pytest-operator results in the Airflow 3 web UI."
 readme = "README.md"
 requires-python = ">=3.10"

{airflow_pytest_plugin-0.2.1 → airflow_pytest_plugin-0.3.1}/src/airflow_pytest_plugin/__init__.py

@@ -22,6 +22,7 @@ from .config import get_reports_root
 from .layout import ReportLayout
 from .models import CaseView, ReportDetail, ReportRef, ReportSummary
 from .producer import ArchivingResultParser
+from .retention import RetentionPolicy, RetentionResult, prune_reports
 from .sources import FileSystemReportSource, ReportSource
 from .version import __version__ as __version__
 
@@ -38,6 +39,9 @@ __all__ = [
     "ReportDetail",
     "CaseView",
     "ReportLayout",
+    "RetentionPolicy",
+    "RetentionResult",
+    "prune_reports",
     "get_reports_root",
     "create_app",
     "__version__",

airflow_pytest_plugin-0.3.1/src/airflow_pytest_plugin/config.py

@@ -0,0 +1,177 @@
+# Copyright 2026 the airflow-pytest-plugin contributors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Resolve the report root (env, then Airflow config, then default)."""
+
+from __future__ import annotations
+
+import os
+
+from .compat import get_conf_value
+
+ENV_VAR = "AIRFLOW_PYTEST_REPORTS_ROOT"
+CONF_SECTION = "pytest_reports"
+CONF_KEY = "reports_root"
+DEFAULT_ROOT = "/opt/airflow/pytest-reports"
+
+#: Toggles whether the reader plugin (UI + API) registers with Airflow. Default on.
+ENABLE_ENV_VAR = "AIRFLOW_PYTEST_PLUGIN_ENABLE"
+_FALSEY = frozenset({"0", "false", "no", "off", "n", "f"})
+
+
+def is_plugin_enabled() -> bool:
+    """Whether the reader plugin should register with Airflow.
+
+    Reads ``AIRFLOW_PYTEST_PLUGIN_ENABLE`` -- ``True`` (the default when unset/empty)
+    registers the UI + API; a falsey value (``0``/``false``/``no``/``off``) disables
+    it. Only gates the reader; the producer-side parser is unaffected.
+    """
+    raw = os.environ.get(ENABLE_ENV_VAR)
+    if raw is None or not raw.strip():
+        return True
+    return raw.strip().lower() not in _FALSEY
+
+
+#: How long (seconds) the filesystem source may reuse a directory scan. A short
+#: window lets the several summary-driven endpoints on one page load (list + flaky +
+#: unique-tests, plus filter typing) share one tree walk instead of rescanning each.
+SCAN_TTL_ENV_VAR = "AIRFLOW_PYTEST_SCAN_CACHE_TTL"
+DEFAULT_SCAN_TTL = 2.0
+
+
+def get_scan_cache_ttl() -> float:
+    """Resolve the directory-scan cache TTL in seconds (``0`` disables caching).
+
+    Reads ``AIRFLOW_PYTEST_SCAN_CACHE_TTL``; falls back to ``DEFAULT_SCAN_TTL``.
+    A malformed or negative value falls back to the default.
+    """
+    raw = os.environ.get(SCAN_TTL_ENV_VAR)
+    if raw is None or not raw.strip():
+        return DEFAULT_SCAN_TTL
+    try:
+        ttl = float(raw.strip())
+    except ValueError:
+        return DEFAULT_SCAN_TTL
+    return ttl if ttl >= 0 else DEFAULT_SCAN_TTL
+
+
+#: Retention knobs (all opt-in; unset = keep everything). Each reads its env var,
+#: then the matching ``[pytest_reports]`` cfg key. Non-positive/invalid -> unset.
+RETENTION_MAX_AGE_DAYS_ENV = "AIRFLOW_PYTEST_RETENTION_MAX_AGE_DAYS"
+RETENTION_MAX_RUNS_ENV = "AIRFLOW_PYTEST_RETENTION_MAX_RUNS"
+RETENTION_MAX_TOTAL_MB_ENV = "AIRFLOW_PYTEST_RETENTION_MAX_TOTAL_MB"
+
+
+def _positive_int_setting(env_var: str, conf_key: str) -> int | None:
+    raw = os.environ.get(env_var)
+    if raw is None or not raw.strip():
+        raw = get_conf_value(CONF_SECTION, conf_key)
+    if raw is None or not str(raw).strip():
+        return None
+    try:
+        value = int(str(raw).strip())
+    except ValueError:
+        return None
+    return value if value > 0 else None
+
+
+def get_retention_max_age_days() -> int | None:
+    """Delete runs older than this many days (``None`` = no age limit)."""
+    return _positive_int_setting(RETENTION_MAX_AGE_DAYS_ENV, "retention_max_age_days")
+
+
+def get_retention_max_runs() -> int | None:
+    """Keep at most this many newest runs per dag·task (``None`` = unlimited)."""
+    return _positive_int_setting(RETENTION_MAX_RUNS_ENV, "retention_max_runs")
+
+
+def get_retention_max_total_mb() -> int | None:
+    """Total report-tree budget in MB (``None`` = unlimited)."""
+    return _positive_int_setting(RETENTION_MAX_TOTAL_MB_ENV, "retention_max_total_mb")
+
+
+#: Flaky-detector defaults (overridable per request via the ``window`` query param).
+FLAKY_WINDOW_ENV = "AIRFLOW_PYTEST_FLAKY_WINDOW"
+DEFAULT_FLAKY_WINDOW = 30
+FLAKY_QUARANTINE_SCORE_ENV = "AIRFLOW_PYTEST_FLAKY_QUARANTINE_SCORE"
+DEFAULT_FLAKY_QUARANTINE_SCORE = 0.5
+#: Flakiness-score floor: below it a test is too steady to count as flaky, so a lone
+#: blip in a long history (a near-zero flip rate) drops off the list.
+FLAKY_MIN_SCORE_ENV = "AIRFLOW_PYTEST_FLAKY_MIN_SCORE"
+DEFAULT_FLAKY_MIN_SCORE = 0.1
+
+
+def get_flaky_window() -> int:
+    """Default number of recent runs the flaky detector looks at per dag·task."""
+    return (
+        _positive_int_setting(FLAKY_WINDOW_ENV, "flaky_window") or DEFAULT_FLAKY_WINDOW
+    )
+
+
+def _unit_float_setting(env_var: str, conf_key: str, default: float) -> float:
+    """A 0–1 float from env, then cfg; default on missing/invalid/out-of-range."""
+    raw = os.environ.get(env_var)
+    if raw is None or not raw.strip():
+        raw = get_conf_value(CONF_SECTION, conf_key)
+    if raw is None or not str(raw).strip():
+        return default
+    try:
+        value = float(str(raw).strip())
+    except ValueError:
+        return default
+    return value if 0.0 <= value <= 1.0 else default
+
+
+def get_flaky_quarantine_score() -> float:
+    """Flakiness score (0–1) at/above which a test is flagged for quarantine."""
+    return _unit_float_setting(
+        FLAKY_QUARANTINE_SCORE_ENV,
+        "flaky_quarantine_score",
+        DEFAULT_FLAKY_QUARANTINE_SCORE,
+    )
+
+
+def get_flaky_min_score() -> float:
+    """Flakiness score (0–1) below which a test is NOT counted as flaky."""
+    return _unit_float_setting(
+        FLAKY_MIN_SCORE_ENV, "flaky_min_score", DEFAULT_FLAKY_MIN_SCORE
+    )
+
+
+#: Pass-rate (0–1) at/above which a run counts as successful ("Passing runs"). The
+#: rate is over executed tests, so a run can carry a few failures and still pass.
+#: At 1.0 a run is successful only with zero failures/errors (the strict default of
+#: older versions). 0.85 is a common "good enough" bar (ISTQB defines no fixed number).
+SUCCESS_THRESHOLD_ENV = "AIRFLOW_PYTEST_SUCCESS_THRESHOLD"
+DEFAULT_SUCCESS_THRESHOLD = 0.85
+
+
+def get_success_threshold() -> float:
+    """Pass-rate (0–1) at/above which a run counts as successful."""
+    return _unit_float_setting(
+        SUCCESS_THRESHOLD_ENV, "success_threshold", DEFAULT_SUCCESS_THRESHOLD
+    )
+
+
+def get_reports_root() -> str:
+    """Resolve the report root directory (absolute path)."""
+    env = os.environ.get(ENV_VAR)
+    if env and env.strip():
+        return os.path.abspath(env.strip())
+
+    conf = get_conf_value(CONF_SECTION, CONF_KEY)
+    if conf and conf.strip():
+        return os.path.abspath(conf.strip())
+
+    return os.path.abspath(DEFAULT_ROOT)

{airflow_pytest_plugin-0.2.1 → airflow_pytest_plugin-0.3.1}/src/airflow_pytest_plugin/models.py

@@ -117,6 +117,19 @@ class ReportSummary:
         }
 
 
+def run_succeeds(passed: int, failed: int, errors: int, threshold: float) -> bool:
+    """Whether a run counts as successful at a given pass-rate ``threshold`` (0–1).
+
+    The rate is over *executed* tests -- ``passed / (passed + failed + errors)`` --
+    so skipped tests neither help nor hurt; a run with nothing executed is a pass.
+    At ``threshold == 1.0`` this is exactly "no failures or errors".
+    """
+    executed = passed + failed + errors
+    if executed <= 0:
+        return True
+    return passed / executed >= threshold
+
+
 @dataclass(frozen=True)
 class CaseView:
     """One test case, flattened for the detail table."""

{airflow_pytest_plugin-0.2.1 → airflow_pytest_plugin-0.3.1}/src/airflow_pytest_plugin/plugin.py

@@ -20,6 +20,7 @@ import logging
 from typing import TYPE_CHECKING, Any
 
 from .compat import get_airflow_plugin_base
+from .config import ENABLE_ENV_VAR, is_plugin_enabled
 
 _log = logging.getLogger(__name__)
 
@@ -73,9 +74,16 @@ else:
         _Base = object
 
 
+#: Kill switch: when AIRFLOW_PYTEST_PLUGIN_ENABLE is falsey the reader registers
+#: nothing (no app, no nav link), so the UI/API are unavailable.
+_ENABLED = is_plugin_enabled()
+if not _ENABLED:
+    _log.info("Pytest Reports reader disabled via %s; not registering.", ENABLE_ENV_VAR)
+
+
 class PytestReportsPlugin(_Base):
-    """Exposes the Pytest Reports UI to Airflow."""
+    """Exposes the Pytest Reports UI to Airflow (unless disabled via the env var)."""
 
     name = "pytest_reports"
-    fastapi_apps = _build_fastapi_apps()
-    external_views = _build_external_views()
+    fastapi_apps = _build_fastapi_apps() if _ENABLED else []
+    external_views = _build_external_views() if _ENABLED else []

{airflow_pytest_plugin-0.2.1 → airflow_pytest_plugin-0.3.1}/src/airflow_pytest_plugin/producer/archiving_parser.py

@@ -163,6 +163,9 @@ class ArchivingResultParser(JUnitResultParser):  # type: ignore[misc]
             "report_file": REPORT_FILENAME,
             "allure": self._archive_allure(out_dir, ref),
             "summary": result.to_xcom(),
+            # Compact per-test outcomes [node_id, outcome, duration] so cross-run
+            # views (compare/flaky/history) need not re-parse junit.xml.
+            "tests": _test_rows(result),
         }
         # Atomic write: a reader scanning concurrently never sees a half file.
         tmp = os.path.join(out_dir, f".{META_FILENAME}.{uuid.uuid4().hex}.tmp")
@@ -229,6 +232,18 @@ def _first_int(*values: Any, default: int) -> int:
     return default
 
 
+def _test_rows(result: TestRunResult) -> list[list[Any]]:
+    """Compact ``[node_id, outcome, duration]`` rows for cross-run views."""
+    rows: list[list[Any]] = []
+    for c in getattr(result, "cases", ()) or ():
+        try:
+            dur = round(float(getattr(c, "time", 0.0) or 0.0), 3)
+        except (TypeError, ValueError):
+            dur = 0.0
+        rows.append([c.node_id, c.outcome, dur])
+    return rows
+
+
 def _logical_date(context: dict[str, Any] | None) -> str | None:
     if not context:
         return None