rosetta-cli 3.0.4__tar.gz → 3.0.5__tar.gz

{rosetta_cli-3.0.4/rosetta_cli.egg-info → rosetta_cli-3.0.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rosetta-cli
-Version: 3.0.4
+Version: 3.0.5
 Summary: Rosetta CLI for publishing knowledge base content to RAGFlow
 Author-email: Rosetta Team <rosetta-support@griddynamics.com>
 License-Expression: Apache-2.0

{rosetta_cli-3.0.4 → rosetta_cli-3.0.5}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "rosetta-cli"
-version = "3.0.4"
+version = "3.0.5"
 description = "Rosetta CLI for publishing knowledge base content to RAGFlow"
 readme = "README.md"
 requires-python = ">=3.12"

{rosetta_cli-3.0.4 → rosetta_cli-3.0.5}/rosetta_cli/cli.py

@@ -378,7 +378,8 @@ Frontmatter Metadata (publish flow):
             chunk_method=config.chunk_method,
             parser_config=config.parser_config,
             page_size=config.page_size,
-            timeout=config.timeout
+            dataset_default=config.dataset_default,
+            timeout=config.timeout,
         )
         
         return execute_command(args.command, args, client, config)

{rosetta_cli-3.0.4 → rosetta_cli-3.0.5}/rosetta_cli/ragflow_client.py

@@ -116,7 +116,8 @@ class RAGFlowClient:
         embedding_model: str | None = None,
         chunk_method: str = "naive",
         parser_config: JsonDict | None = None,
-        page_size: int = 1000
+        page_size: int = 1000,
+        dataset_default: str = "aia",
     ):
         """
         Initialize RAGFlow client.
@@ -130,6 +131,7 @@ class RAGFlowClient:
             chunk_method: Chunking method (default: naive)
             parser_config: Parser configuration dict for chunk_method settings
             page_size: Default page size for list operations (default: 1000)
+            dataset_default: Default dataset name when no release-specific or explicit dataset name is available
             
         Raises:
             ValueError: If api_key or base_url is empty
@@ -147,6 +149,7 @@ class RAGFlowClient:
         self.chunk_method = chunk_method
         self.parser_config = parser_config or {}
         self.page_size = page_size
+        self.dataset_default = dataset_default
 
         # Initialize RAGFlow SDK client
         self._client = RAGFlow(api_key=api_key, base_url=base_url, version=version)
@@ -469,26 +472,38 @@ class RAGFlowClient:
         # Dataset doesn't exist, create it (gated by dry_run)
         return self.create_dataset(name, description, dry_run=dry_run)
     
-    def _resolve_dataset_name(self, template: str, release: str | None) -> str:
+    def _resolve_dataset_name(
+        self,
+        dataset_template: str,
+        release: str | None,
+        dataset_name: str | None,
+    ) -> str:
         """
-        Resolve dataset name from template.
+        Resolve dataset name using release + dataset_template, explicit dataset_name or default.
         
         Args:
-            template: Name template (e.g., "aia-{release}")
+            dataset_template: Name template (e.g., "aia-{release}")
             release: Release identifier (e.g., "r1")
+            dataset_name: Explicit dataset name, if provided (e.g., "aia-r1")
             
         Returns:
             Resolved dataset name
             
         Examples:
-            >>> _resolve_dataset_name("aia-{release}", "r1")
+            >>> _resolve_dataset_name("aia-{release}", "r1", "aia-r2")
             "aia-r1"
-            >>> _resolve_dataset_name("aia", None)
-            "aia"
+            >>> _resolve_dataset_name("aia-{release}", None, "aia-r2")
+            "aia-r2"
         """
-        if release and "{release}" in template:
-            return template.format(release=release)
-        return template
+        if release and "{release}" in dataset_template:
+            resolved_name = dataset_template.format(release=release)
+        else:
+            resolved_name = dataset_name or self.dataset_default
+
+        if "{release}" in resolved_name:
+            raise ValueError(f"Unresolved release placeholder in dataset name: {resolved_name}")
+
+        return resolved_name
     
     def _build_title_with_tags(self, tags: list[str], filename: str) -> str:
         """
@@ -521,7 +536,7 @@ class RAGFlowClient:
         OPTIMIZED: Now accepts pre-read content to avoid redundant file I/O.
         
         This method:
-        1. Resolves dataset name from template + release
+        1. Resolves the target dataset name
         2. Ensures dataset exists
         3. Builds title with tag prefixes
         4. Checks if document exists (by ims_doc_id)
@@ -529,6 +544,12 @@ class RAGFlowClient:
         6. Deletes existing document if changed
         7. Uploads new document with metadata
         
+        Dataset name resolution precedence:
+        1. If metadata.release is present, and dataset_template contains "{release}",
+           resolve dataset_template with the release.
+        2. Otherwise, use dataset_name when provided.
+        3. Otherwise, fall back to dataset_default.
+
         Args:
             file_path: Path to file (for filename, backward compatibility)
             metadata: Document metadata with pre-calculated hash
@@ -567,10 +588,7 @@ class RAGFlowClient:
         actual_hash = metadata.content_hash
         
         # Resolve dataset name
-        resolved_name = self._resolve_dataset_name(
-            dataset_template if "{release}" in dataset_template else (dataset_name or dataset_template),
-            metadata.release
-        )
+        resolved_name = self._resolve_dataset_name(dataset_template, metadata.release, dataset_name)
         
         # Ensure dataset exists (dry_run gates the underlying create_dataset call)
         dataset = self._ensure_dataset(

{rosetta_cli-3.0.4 → rosetta_cli-3.0.5}/rosetta_cli/services/dataset_service.py

@@ -27,14 +27,27 @@ class DatasetService:
     
     def resolve_dataset_name(self, args_dataset: Optional[str]) -> tuple[Optional[str], bool]:
         """
-        Resolve dataset name with auto-detection.
-        
+        Resolve which dataset to act on.
+
+        How the name is picked:
+          1. --dataset given      -> use it               (False)
+          2. else list server datasets, keep those starting with the template
+             prefix ("aia-" from "aia-{release}")
+          3. exactly one match    -> use it               (True)
+          4. several matches      -> give up, ask user    (None, False)
+          5. no matches           -> use dataset_default  (True)
+
         Args:
             args_dataset: Dataset name from arguments (can be None)
-        
+
         Returns:
             Tuple of (dataset_name: str or None, auto_detected: bool)
             Returns (None, False) if resolution fails
+
+            auto_detected answers "did the caller supply this name?", not "did
+            we find it on the server". False only for step 1. True for steps 3
+            and 5 alike: the caller supplied nothing and this method chose.
+            True on step 5 is correct.
         """
         # Explicit dataset provided
         if args_dataset:

{rosetta_cli-3.0.4 → rosetta_cli-3.0.5/rosetta_cli.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rosetta-cli
-Version: 3.0.4
+Version: 3.0.5
 Summary: Rosetta CLI for publishing knowledge base content to RAGFlow
 Author-email: Rosetta Team <rosetta-support@griddynamics.com>
 License-Expression: Apache-2.0