feat: TMS-40196 add at layers

.github/workflows/test.yml

@@ -91,8 +91,9 @@ jobs:
        run: |
          dotnet build --configuration Debug --property WarningLevel=0 api-validator-dotnet
          pip3 install -r python-examples/${{ matrix.project_name }}/requirements_ci.txt
-         pip3 install ./testit-python-commons
-         pip3 install ./${{ matrix.adapter_name }}
+         pip3 install -e ./testit-python-commons
+         pip3 install -e ./${{ matrix.adapter_name }}
+         python3 -c "from adapters_api.model.custom_attribute_api_result import CustomAttributeApiResult; assert 'code' in CustomAttributeApiResult.openapi_types"
 
      - name: Test
        run: |
@@ -129,8 +130,9 @@ jobs:
        run: |
          dotnet build --configuration Debug --property WarningLevel=0 api-validator-dotnet
          pip3 install -r python-examples/${{ matrix.project_name }}/requirements_ci.txt
-         pip3 install ./testit-python-commons
-         pip3 install ./${{ matrix.adapter_name }}
+         pip3 install -e ./testit-python-commons
+         pip3 install -e ./${{ matrix.adapter_name }}
+         python3 -c "from adapters_api.model.custom_attribute_api_result import CustomAttributeApiResult; assert 'code' in CustomAttributeApiResult.openapi_types"
 
 
      - name: Test import realtime false

docs/tz-autotest-layer.md

@@ -0,0 +1,228 @@
+# Technical assignment: autotest layer (test pyramid)
+
+## Context
+
+TMS supports a **test pyramid layer** on an **autotest** card. The layer can be assigned manually, by rules, or from automation runs.
+
+Adapters can set the layer when creating or updating an autotest during a test run. The layer source in API is **`Run`** (set by the adapter).
+
+## Business need
+
+As an automation engineer, I want to declare the pyramid level of each autotest **in test code** so that after a run the autotest card shows:
+
+- the layer name (e.g. `API`, `E2E`);
+- the source **Run** (adapter), distinct from Manual / Rule / Report.
+
+## Goal
+
+In every adapter that creates or updates autotests, support an optional **layer** metadata field:
+
+- read from test code only (annotation / mark / tag — per language);
+- send on **create** and **update** autotest API calls when present;
+- **omit** the field when the user did not specify a layer.
+
+## Scope
+
+Apply consistently to all Test IT adapters (any language / framework) and any entry point that upserts autotests via the standard adapters API.
+
+**In scope:** autotest `layer` on create/update.
+
+**Out of scope:**
+
+- layer on **test run** entity;
+- layer on **test result** as a separate API field (UI may show layer from the linked autotest);
+- configuration / env / CLI default layer for a whole run;
+- runtime API to set layer during test execution;
+- validating layer names against a fixed whitelist (adapters pass the string as-is).
+
+## API contract
+
+Use existing autotest endpoints and models:
+
+```json
+{
+  "layer": {
+    "name": "API",
+    "source": "Run"
+  }
+}
+```
+
+| Operation | Behaviour |
+|-----------|-----------|
+| **Create autotest** | Include `layer` only when the test declares a non-empty layer |
+| **Update autotest** | Always send `resetLayer: false`; include `layer` only when the test declares a non-empty layer |
+| **No layer in test** | Send `resetLayer: false`; do **not** send `layer` |
+
+Recommended layer names (constants in client libraries, not enforced): `E2E`, `UI`, `API`, `Contract`, `Integration`, `Component`, `Unit`.
+
+Any other non-empty string is valid.
+
+`LayerSource` values: `Manual`, `Report`, `Run`, `Rule`. Adapters always use **`Run`**.
+
+## Functional requirements
+
+### 1. Declaration in test code only
+
+The user sets layer on the **test method / scenario**, not via adapter config.
+
+| Stack | Suggested syntax |
+|-------|------------------|
+| Java (JUnit / TestNG) | `@Layer("API")` or `@Layer(TestLayers.API)` on test method |
+| Python (pytest) | `@pytest.mark.layer("api")` |
+| C# (NUnit / xUnit) | `[Layer("API")]` on test method |
+| JavaScript (Jest / Mocha) | `@layer('API')` or framework-specific decorator |
+| Cucumber (Gherkin) | `@Layer=API` on scenario |
+| JBehave | Meta: `@Layer API` on scenario |
+
+Parameter substitution (e.g. `{param}` in Java) should work if the framework already supports it for other annotations.
+
+### 2. Mapping flow
+
+```
+test code (annotation / tag)
+    → internal test result model (optional layer field)
+    → Converter / mapper
+    → AutoTestCreateApiModel.layer / AutoTestUpdateApiModel.layer
+    → TMS
+```
+
+### 3. Independence from other metadata
+
+- Layer ≠ labels / tags on autotest.
+- Layer ≠ test run tags.
+- Do not change existing labels/tags behaviour.
+
+### 4. Failed-test update path
+
+Some adapters send a minimal update when a test fails (copy existing autotest from TMS). Still apply `layer` from the test annotation when present.
+
+## Non-goals
+
+- Default layer for all tests in a run via config.
+- `Adapter.addLayer()` or similar dynamic API.
+- Resetting layer when annotation is absent (`resetLayer: true`).
+- OpenAPI / client regeneration in adapter repos (use existing generated `LayerApiModel`).
+
+## Acceptance criteria
+
+1. User adds layer in test code → autotest in TMS shows that layer with source **Run** after create/update.
+2. User omits layer → adapter does not send `layer` on create/update.
+3. Custom layer string is accepted without validation.
+4. Recommended constants documented; arbitrary strings work.
+5. Cross-language doc describes the same rules with per-language syntax examples.
+6. Unit tests: with layer / without layer / custom string / failed-update path.
+
+## Suggested implementation outline (language-agnostic)
+
+1. Add `@Layer` (or equivalent) in shared annotations package.
+2. Add optional `layer` field to internal `TestResult` (or equivalent).
+3. Extract layer from test method in each framework listener.
+4. In create/update mapper: if layer non-empty → `LayerApiModel { name, source: Run }`; on update always send `resetLayer: false`.
+5. BDD adapters: parse `@Layer=` / meta `Layer` like other scenario tags.
+6. Document in adapter README; add samples and tests.
+
+## Example (Java)
+
+```java
+import ru.testit.annotations.Layer;
+import ru.testit.models.TestLayers;
+
+@Layer(TestLayers.API)
+@Test
+void createUser() {
+    // ...
+}
+
+@Layer("my-custom-layer")
+@Test
+void customLayer() {
+    // ...
+}
+```
+
+## Example (pytest)
+
+```python
+import pytest
+from testit import layer, TestLayers
+
+@pytest.mark.layer("api")
+def test_create_user_with_mark():
+    ...
+
+@layer(TestLayers.API)
+def test_create_user_with_decorator():
+    ...
+
+@layer("my-custom-layer")
+def test_with_custom_layer():
+    ...
+```
+
+---
+
+## Java adapters status
+
+Implemented in `testit-java-commons` (shared by all Java adapters):
+
+- `@Layer` on test method (`ru.testit.annotations.Layer`)
+- `TestLayers` constants (`ru.testit.models.TestLayers`)
+- `TestResult.layer`, `Utils.extractLayer`
+- `Converter`: `layer` + `LayerSource.RUN` on create when set; on update always `resetLayer: false`, `layer` only when set
+- Framework listeners: JUnit 4/5, TestNG, Cucumber 4–7 (`@Layer=`), JBehave (meta `Layer`)
+
+## Python adapters status
+
+Implemented in `testit-python-commons` (shared by pytest / behave / nose / robot):
+
+- Decorator: `@layer("API")` or `@layer(TestLayers.API)` (`testit.layer`, `testit.TestLayers`)
+- `TestResult.set_layer` / `get_layer`
+- `Converter.layer_to_api_model`: `LayerApiModel(name, source=Run)` on create when set; on update always `resetLayer: false`, `layer` only when set
+- No config/env/CLI layer; no layer on test run or test result
+
+### Syntax by adapter
+
+| Adapter | How to declare layer in test code |
+|---------|-----------------------------------|
+| pytest | `@pytest.mark.layer("API")` or `@layer(TestLayers.API)` from `testit` |
+| behave | `@Layer=API` on scenario/feature |
+| nose (nose2) | `@layer("API")` from `testit` on test function |
+| robotframework | `[Tags] testit.layer:API` |
+
+### Recommended constants
+
+```python
+from testit import TestLayers
+
+TestLayers.E2E
+TestLayers.UI
+TestLayers.API
+TestLayers.CONTRACT
+TestLayers.INTEGRATION
+TestLayers.COMPONENT
+TestLayers.UNIT
+```
+
+Any other non-empty string is accepted without validation.
+
+### API mapping rules
+
+| Scenario | Adapter behaviour |
+|----------|-------------------|
+| Layer set in test | create: send `layer: { name, source: Run }` |
+| Layer set in test | update: send `layer` + `resetLayer: false` |
+| Layer not set | send `resetLayer: false`; do not send `layer` |
+
+### Cross-language implementation guide
+
+When porting to another language, keep the same contract:
+
+1. Read layer **only from test code** (annotation / mark / tag).
+2. Store it in the internal test result model as an optional string.
+3. On autotest **create**, include `layer` only when non-empty.
+4. On autotest **update**, always send `resetLayer: false`; include `layer` only when non-empty.
+5. Always set `source` to `Run`.
+6. Do not validate layer name against a whitelist; document recommended constants only.
+7. Do not add config/env/CLI defaults for run-level layer.
+8. Do not send layer on test run or test result entities.

testit-adapter-behave/setup.py

@@ -1,6 +1,6 @@
 from setuptools import find_packages, setup
 
-VERSION = "5.1.6"
+VERSION = "5.2.0"
 
 setup(
     name='testit-adapter-behave',

testit-adapter-behave/src/testit_adapter_behave/models/tags.py

@@ -9,3 +9,4 @@ class TagType:
     TAGS = 'Tags='
     NAMESPACE = 'NameSpace='
     CLASSNAME = 'ClassName='
+    LAYER = 'Layer='

testit-adapter-behave/src/testit_adapter_behave/scenario_parser.py

@@ -57,6 +57,7 @@ def parse_scenario(scenario):
         'links': [],
         'labels': [],
         'tags': [],
+        'layer': None,
         'workItemsID': [],
         "externalKey": get_scenario_name(scenario)
         # TODO: Add to python-commons
@@ -89,6 +90,9 @@ def parse_scenario(scenario):
         # TODO: Fix in python-commons to "workItemIds"
         executable_test['workItemsID'] = tags[TagType.WORK_ITEM_IDS]
 
+    if TagType.LAYER in tags and tags[TagType.LAYER]:
+        executable_test['layer'] = tags[TagType.LAYER]
+
     return executable_test
 
 

testit-adapter-behave/src/testit_adapter_behave/tags_parser.py

@@ -15,6 +15,7 @@ def parse_test_tags(tags):
         TagType.TAGS: [],
         TagType.NAMESPACE: None,
         TagType.CLASSNAME: None,
+        TagType.LAYER: None,
     }
 
     for tag in tags:
@@ -52,6 +53,9 @@ def parse_test_tags(tags):
         elif TagType.CLASSNAME in tag:
             parsed_tags[TagType.CLASSNAME] = __parse_space_in_tag(tag[len(TagType.CLASSNAME):])
 
+        elif TagType.LAYER in tag:
+            parsed_tags[TagType.LAYER] = __parse_space_in_tag(tag[len(TagType.LAYER):])
+
     return parsed_tags
 
 

testit-adapter-behave/src/testit_adapter_behave/utils.py

@@ -151,6 +151,7 @@ def convert_executable_test_to_test_result_model(executable_test: dict) -> TestR
         .set_result_links(executable_test['resultLinks'])\
         .set_labels(executable_test['labels']) \
         .set_tags(executable_test['tags'])\
+        .set_layer(executable_test.get('layer'))\
         .set_work_item_ids(executable_test['workItemsID'])\
         .set_message(executable_test['message'])\
         .set_external_key(executable_test['externalKey'])

testit-adapter-behave/tests/test_behave_layer.py

@@ -0,0 +1,8 @@
+from testit_adapter_behave.models.tags import TagType
+from testit_adapter_behave.tags_parser import parse_test_tags
+
+
+def test_parse_layer_tag():
+    parsed = parse_test_tags(['Layer=API'])
+
+    assert parsed[TagType.LAYER] == 'API'

testit-adapter-nose/setup.py

@@ -1,6 +1,6 @@
 from setuptools import setup, find_packages
 
-VERSION = "5.1.6"
+VERSION = "5.2.0"
 
 setup(
     name='testit-adapter-nose',

testit-adapter-nose/src/testit_adapter_nose/utils.py

@@ -86,6 +86,7 @@ def form_test(item, top_level_directory):
         'links': __get_links_from(item, function),
         'labels': __get_labels_from(item, function),
         'tags': __get_tags_from(item, function),
+        'layer': __get_layer_from(item, function),
         'workItemsID': __get_work_item_ids_from(item, function),
         'message': None,
         'externalKey': __get_fullname(function, top_level_directory)
@@ -242,6 +243,16 @@ def __get_tags_from(item, function) -> List[str]:
     return tags
 
 
+def __get_layer_from(item, function):
+    test_layer = __search_attribute(function, 'test_layer')
+    if not test_layer:
+        return None
+
+    return str(collect_parameters_in_string_attribute(
+        test_layer,
+        get_all_parameters(item)))
+
+
 def __get_work_item_ids_from(item, function):
     test_workitems_id = __search_attribute(function, 'test_workitems_id')
 
@@ -405,6 +416,7 @@ def convert_executable_test_to_test_result_model(executable_test: dict) -> TestR
         .set_result_links(executable_test['resultLinks'])\
         .set_labels(executable_test['labels'])\
         .set_tags(executable_test['tags'])\
+        .set_layer(executable_test.get('layer'))\
         .set_work_item_ids(executable_test['workItemsID'])\
         .set_message(executable_test['message'])\
         .set_external_key(executable_test['externalKey'])

testit-adapter-pytest/setup.py

@@ -1,6 +1,6 @@
 from setuptools import find_packages, setup
 
-VERSION = "5.1.6"
+VERSION = "5.2.0"
 
 setup(
     name='testit-adapter-pytest',

testit-adapter-pytest/src/testit_adapter_pytest/models/executable_test.py

@@ -29,5 +29,6 @@ class ExecutableTest:
     labels = attrib(default=Factory(list))
     tags = attrib(default=Factory(list))
     work_item_ids = attrib(default=Factory(list))
+    layer = attrib(default=None)
     message = attrib(default=None)
     node_id = attrib(default=None)