Preserve configured sync artifacts

Author: Miskler

README.md

@@ -114,6 +114,8 @@ The repository ships the sync workflow template used by generated package reposi
 
 The generated `source-sync` workflow stays thin: it checks out this repository as the generator logic, reads the consumer repo's `.msra` source tree, generates the artifact into a staging tree, syncs the generated tree into `main`, validates the result, and pushes the commit so the target repo's `publish.yml` can run on `push` to `main`.
 
+If a project defines `[app.sync].preserved_target_paths` in its MSRA source, the generated sync step preserves those repo-specific runtime artifacts. For the FixPrice layout that means `tests/__snapshots__` stays in `main` and is not rewritten by the generator.
+
 Before validation, the generated workflow installs the target project's `requirements-dev.txt` so the validation step runs with the generated dependencies available.
 
 The source and target branches come from the repo-specific sync config, so the generated workflow itself stays thin.

docs/msra-repo-b-sync.md

@@ -8,6 +8,7 @@
 - этот workflow запускается вручную через `workflow_dispatch`
 - внутри job он сам checkout-ит repo с логикой генератора
 - затем читает `repo B/source`, генерирует артефакт и пушит его в `repo B/main`
+- при синке сохраняет repo-specific runtime artifacts, перечисленные в `[app.sync].preserved_target_paths` source-MSTRA, например `tests/__snapshots__`
 - после `push` в `main` автоматически стартует `publish.yml`
 
 ## Почему workflow должен лежать в `main`
@@ -113,6 +114,19 @@ jobs:
 
           target = Path("target")
           generated = Path("generated")
+          preserved_root = Path("preserved")
+          preserve_paths = ["tests/__snapshots__"]
+
+          for relative_path in preserve_paths:
+              source_path = target / relative_path
+              if not source_path.exists():
+                  raise RuntimeError(f'Preserved target path "{relative_path}" does not exist.')
+              destination_path = preserved_root / relative_path
+              destination_path.parent.mkdir(parents=True, exist_ok=True)
+              if source_path.is_dir():
+                  shutil.copytree(source_path, destination_path)
+              else:
+                  shutil.copy2(source_path, destination_path)
 
           for child in list(target.iterdir()):
               if child.name == ".git":
@@ -128,6 +142,15 @@ jobs:
                   shutil.copytree(item, destination, dirs_exist_ok=True)
               else:
                   shutil.copy2(item, destination)
+
+          for relative_path in preserve_paths:
+              source_path = preserved_root / relative_path
+              destination_path = target / relative_path
+              if source_path.is_dir():
+                  shutil.copytree(source_path, destination_path, dirs_exist_ok=True)
+              else:
+                  destination_path.parent.mkdir(parents=True, exist_ok=True)
+                  shutil.copy2(source_path, destination_path)
           PY
 
       - name: Validate generated project

msra_codegen/github_workflows.py

@@ -14,12 +14,16 @@ def build_github_workflows_context(project: dict[str, Any], package_name: str) -
     tests_config = github_config["tests"]
     publish_config = github_config["publish"]
     source_sync_config = github_config["source_sync"]
+    app_sync_config = project.get("app", {}).get("sync", {})
     package_name_slug = package_name.replace("_", "-")
     source_msra_path = str(Path(project["source_path"]).name)
 
     tests_python_version = str(tests_config["python_version"])
     publish_python_version = str(publish_config["python_version"])
     source_sync_python_version = str(source_sync_config["python_version"])
+    preserved_target_paths = app_sync_config.get("preserved_target_paths", [])
+    if not isinstance(preserved_target_paths, list):
+        raise RuntimeError("app.sync.preserved_target_paths must be a list.")
 
     return {
         "package_name": package_name,
@@ -82,6 +86,7 @@ def build_github_workflows_context(project: dict[str, Any], package_name: str) -
             "commit_user_name": str(source_sync_config["commit_user_name"]),
             "commit_user_email": str(source_sync_config["commit_user_email"]),
             "source_msra_path": source_msra_path,
+            "preserved_target_paths": [str(item) for item in preserved_target_paths],
         },
         "makefile": {
             "package_name": package_name,

msra_codegen/project_model.py

@@ -107,6 +107,7 @@ def get_assignment(table: dict[str, Any] | None, key: str, default: Any = None)
         "humanize": get_plain_value(get_assignment(app_table, "humanize", False)),
         "block_images": bool(get_plain_value(get_assignment(app_table, "block_images", False))),
         "disallow_headless": bool(get_plain_value(get_assignment(app_table, "disallow_headless", False))),
+        "sync": {},
         "abstractions": [],
     }
     abstractions_value = get_plain_value(get_assignment(app_table, "abstractions", []))
@@ -169,6 +170,22 @@ def get_assignment(table: dict[str, Any] | None, key: str, default: Any = None)
             "script": warmup_script,
         }
 
+    sync_table = get_table(["app", "sync"])
+    if sync_table:
+        preserved_target_paths_value = get_plain_value(
+            get_assignment(sync_table, "preserved_target_paths", [])
+        )
+        if not isinstance(preserved_target_paths_value, list):
+            raise TypeError("app.sync.preserved_target_paths must be a list of strings.")
+        preserved_target_paths: list[str] = []
+        for item in preserved_target_paths_value:
+            if not isinstance(item, str):
+                raise TypeError("app.sync.preserved_target_paths entries must be strings.")
+            text = item.strip()
+            if text:
+                preserved_target_paths.append(text)
+        app["sync"] = {"preserved_target_paths": preserved_target_paths}
+
     variable_tables = [
         table
         for table in tables

msra_codegen/templates/github/workflows/source-sync.yml.tpl

@@ -67,6 +67,19 @@ jobs:
 
           target = Path("target")
           generated = Path("generated")
+          preserved_root = Path("preserved")
+          preserve_paths = {{ source_sync.preserved_target_paths | tojson }}
+
+          for relative_path in preserve_paths:
+              source_path = target / relative_path
+              if not source_path.exists():
+                  raise RuntimeError(f'Preserved target path "{relative_path}" does not exist.')
+              destination_path = preserved_root / relative_path
+              destination_path.parent.mkdir(parents=True, exist_ok=True)
+              if source_path.is_dir():
+                  shutil.copytree(source_path, destination_path)
+              else:
+                  shutil.copy2(source_path, destination_path)
 
           for child in list(target.iterdir()):
               if child.name == ".git":
@@ -82,6 +95,17 @@ jobs:
                   shutil.copytree(item, destination, dirs_exist_ok=True)
               else:
                   shutil.copy2(item, destination)
+
+          for relative_path in preserve_paths:
+              source_path = preserved_root / relative_path
+              destination_path = target / relative_path
+              if source_path.is_dir():
+                  shutil.copytree(source_path, destination_path, dirs_exist_ok=True)
+              else:
+                  destination_path.parent.mkdir(parents=True, exist_ok=True)
+                  shutil.copy2(source_path, destination_path)
+
+          shutil.rmtree(preserved_root, ignore_errors=True)
           PY
 
       - name: Install target project dependencies

tests/codegen.test.js

@@ -845,6 +845,42 @@ test("python codegen generates both bundled msra documents without failing", ()
   }
 });
 
+test("source sync workflow preserves target artifacts from app.sync", () => {
+  const repoRoot = path.resolve(__dirname, "..");
+  const workDir = mkdtempSync(path.join(os.tmpdir(), "msra-source-sync-preserve-"));
+  const inputPath = path.join(workDir, "sync.msra");
+  const outputDir = path.join(workDir, "generated");
+  const pythonReleasesFixturePath = createPythonReleasesApiFixture(workDir);
+  const source = normalizeNewlines(readFileSync(path.join(repoRoot, "examples", "example", "example.msra"), "utf8"))
+    .replace('package_name="ozon_api"', 'package_name="sync_api"')
+    .replace(
+      /\n\[app\.warmup\]\n/,
+      '\n[app.sync]\npreserved_target_paths=["tests/__snapshots__"]\n\n[app.warmup]\n',
+    );
+
+  try {
+    writeFileSync(inputPath, source, "utf8");
+    writeFileSync(
+      path.join(workDir, "warmup.py"),
+      readFileSync(path.join(repoRoot, "examples", "example", "warmup.py"), "utf8"),
+      "utf8",
+    );
+    const result = spawnSync("python", ["-m", "msra_codegen", "generate", inputPath, "-o", outputDir], {
+      cwd: repoRoot,
+      env: buildCodegenPythonPath(pythonReleasesFixturePath),
+      encoding: "utf8",
+    });
+    assert.strictEqual(result.status, 0, result.stderr || result.stdout);
+
+    const sourceSyncWorkflowText = normalizeNewlines(
+      readFileSync(path.join(outputDir, ".github", "workflows", "source-sync.yml"), "utf8"),
+    );
+    assert.match(sourceSyncWorkflowText, /preserve_paths = \["tests\/__snapshots__"\]/);
+  } finally {
+    rmSync(workDir, { recursive: true, force: true });
+  }
+});
+
 test("python codegen builds the TODO/2 abstractions project", () => {
   const repoRoot = path.resolve(__dirname, "..");
   const workDir = mkdtempSync(path.join(os.tmpdir(), "msra-abstractions-build-"));

vscode-extension/lsp/assignment-schema.js

@@ -677,6 +677,9 @@ const TABLE_SCHEMAS = [
     on_error_screenshot_path: SCREENSHOT_PATH_SPEC,
     timeout_ms: integerAtLeast(0),
   }),
+  makeFixedSchema(exactPath(["app", "sync"]), {
+    preserved_target_paths: arrayOf(STRING),
+  }),
   makeFixedSchema(exactPath(["app", "variables", "*"]), {
     types: arrayOf(VARIABLE_TYPE_ITEM_SPEC),
     description: STRINGISH,

vscode-extension/lsp/path-schema.js

@@ -7,6 +7,7 @@ const TABLE_ROOT_SEGMENTS = new Set([
 
 const APP_CHILD_SEGMENTS = new Set([
   "warmup",
+  "sync",
   "defaults",
   "variables",
   "prefixes",
@@ -275,6 +276,9 @@ function validateAppPath(segments, index) {
   if (segment.value === "warmup" && !segment.quoted) {
     return validateLeafNamespace(segments, index, `"${renderPath(segments, index + 1)}"`);
   }
+  if (segment.value === "sync" && !segment.quoted) {
+    return validateLeafNamespace(segments, index, `"${renderPath(segments, index + 1)}"`);
+  }
   if (segment.value === "defaults" && !segment.quoted) {
     return validateDefaultsNamespace(segments, index + 1);
   }
@@ -296,8 +300,8 @@ function validateAppPath(segments, index) {
   return invalidPath(
     segments,
     index,
-    `Invalid child table "${renderSegment(segment)}" under "app" in "${renderPath(segments)}". Expected "warmup", "defaults", "variables", "prefixes", "regexes", "groups", or "func".`,
-    ["warmup", "defaults", "variables", "prefixes", "regexes", "groups", "func"],
+    `Invalid child table "${renderSegment(segment)}" under "app" in "${renderPath(segments)}". Expected "warmup", "sync", "defaults", "variables", "prefixes", "regexes", "groups", or "func".`,
+    ["warmup", "sync", "defaults", "variables", "prefixes", "regexes", "groups", "func"],
   );
 }