aisberg 0.1.0__py3-none-any.whl → 0.2.0__py3-none-any.whl

aisberg/modules/models.py

@@ -25,7 +25,7 @@ class AbstractModelsModule(ABC):
             ValueError: If no collections are found.
             Exception: If there is an error fetching the collections.
         """
-        pass
+        ...
 
     @abstractmethod
     def get(self, model_id: str) -> Model:
@@ -40,7 +40,7 @@ class AbstractModelsModule(ABC):
         Raises:
             ValueError: If the specified model is not found.
         """
-        pass
+        ...
 
     @abstractmethod
     def is_available(self, model_id: str) -> bool:
@@ -52,7 +52,7 @@ class AbstractModelsModule(ABC):
         Returns:
             bool: True if the model is available, False otherwise.
         """
-        pass
+        ...
 
     @staticmethod
     def _get_model_by_id(models: List[Model], model_id: str) -> Model | None:

aisberg/modules/s3.py

@@ -0,0 +1,316 @@
+import os
+import io
+import sys
+import mimetypes
+import logging
+from datetime import datetime
+from abc import ABC, abstractmethod
+from typing import Optional, Dict, Any, Union
+
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.INFO)
+
+
+# -------- Progress Bar --------
+class ProgressPercentage:
+    def __init__(self, file_size, object_name, bar_length=30, verbose=False):
+        """
+        Initialize the ProgressPercentage class.
+        """
+        self._file_size = file_size / 1024 / 1024  # MB
+        self._seen_so_far = 0
+        self._object_name = object_name
+        self._bar_length = bar_length
+        self._verbose = verbose
+
+    def __call__(self, bytes_amount):
+        if not self._verbose:
+            return
+
+        self._seen_so_far += bytes_amount / 1024 / 1024  # MB
+        percentage = (self._seen_so_far / self._file_size) * 100
+        bar_filled_length = int(self._bar_length * self._seen_so_far // self._file_size)
+        bar = "█" * bar_filled_length + "-" * (self._bar_length - bar_filled_length)
+        sys.stdout.write(
+            f"\r{self._object_name}: |{bar}| {self._seen_so_far:.2f} MB / "
+            f"{self._file_size:.2f} MB ({percentage:.2f}%)"
+        )
+        sys.stdout.flush()
+        if self._seen_so_far >= self._file_size:
+            sys.stdout.write("\n")
+            sys.stdout.flush()
+
+
+# -------- Base S3Module --------
+class BaseS3Module(ABC):
+    """
+    Base abstraite pour modules S3.
+    """
+
+    def __init__(
+        self,
+        access_key_id: str,
+        secret_access_key: str,
+        endpoint_url: Optional[str] = None,
+        region_name: str = "fr-par",
+        verbose: bool = False,
+    ):
+        self._args = dict(
+            aws_access_key_id=access_key_id,
+            aws_secret_access_key=secret_access_key,
+            endpoint_url=endpoint_url,
+            region_name=region_name,
+        )
+        self._verbose = verbose
+
+    @abstractmethod
+    def _get_client(self):
+        """Returns the S3 client."""
+        ...
+
+    @staticmethod
+    def _prepare_metadata(file, metadata=None):
+        """Prepare metadata for the file.
+
+        Args:
+            file (str | io.BytesIO): path to the file or a BytesIO object.
+            metadata (dict, optional): Additional metadata to enrich the file information.
+
+        Returns:
+            tuple: (file_size (int), metadata (dict)) with enriched file information.
+        """
+        if metadata is None:
+            metadata = {}
+
+        if isinstance(file, str):
+            file_size = os.path.getsize(file)
+            file_type, _ = mimetypes.guess_type(file)
+            file_creation_time = datetime.fromtimestamp(
+                os.path.getctime(file)
+            ).isoformat()
+            file_name = os.path.basename(file)
+            metadata.update(
+                {
+                    "FileName": file_name,
+                    "FileSize": f"{file_size / (1024 * 1024):.2f} MB",
+                    "FileType": file_type or "application/octet-stream",
+                    "CreationTime": file_creation_time,
+                }
+            )
+        else:
+            file.seek(0, io.SEEK_END)
+            file_size = file.tell()
+            file.seek(0)
+            metadata.update(
+                {
+                    "FileSize": f"{file_size / (1024 * 1024):.2f} MB",
+                    "FileType": "application/octet-stream",
+                    "CreationTime": datetime.now().isoformat(),
+                }
+            )
+        return file_size, metadata
+
+    @staticmethod
+    def _get_object_name(self, file, object_name):
+        return object_name or (file if isinstance(file, str) else "default_object_name")
+
+    @abstractmethod
+    def upload_file(
+        self,
+        file: Union[str, io.BytesIO],
+        bucket_name: str,
+        object_name: Optional[str] = None,
+        metadata: Optional[Dict[str, str]] = None,
+    ) -> Any:
+        """
+        Uploads a file to the S3 bucket.
+
+        Args:
+            file (str or BytesIO): The file to upload. Can be a path to a file or a BytesIO object.
+            bucket_name (str): The name of the S3 bucket.
+            object_name (str, optional): The object name in S3. Defaults to filename if not provided.
+            metadata (dict, optional): Additional metadata for the object.
+
+        Returns:
+            Any: True if the upload succeeded, else False or raises an exception.
+        """
+        ...
+
+    @abstractmethod
+    def download_file(
+        self,
+        bucket_name: str,
+        object_name: str,
+        save_to: Optional[str] = None,
+    ) -> Any:
+        """
+        Downloads a file from the S3 bucket.
+
+        Args:
+            bucket_name (str): The name of the S3 bucket.
+            object_name (str): The object name in S3.
+            save_to (str, optional): Local path to save the file. If not provided, returns file content as BytesIO.
+
+        Returns:
+            Any: True if the file was saved locally, else a BytesIO object with file content, or False on failure.
+        """
+        ...
+
+    @abstractmethod
+    def delete_file(self, bucket_name: str, object_name: str) -> Any:
+        """
+        Deletes a file from the S3 bucket.
+
+        Args:
+            bucket_name (str): The name of the S3 bucket.
+            object_name (str): The object name in S3.
+
+        Returns:
+            Any: True if the file was deleted, False otherwise.
+        """
+        ...
+
+    @abstractmethod
+    def list_files(self, bucket_name: str) -> Any:
+        """
+        Lists all files in the specified S3 bucket.
+
+        Args:
+            bucket_name (str): The name of the S3 bucket.
+
+        Returns:
+            Any: List of object names (keys) in the bucket, or an empty list on failure.
+        """
+        ...
+
+    @abstractmethod
+    def list_buckets(self) -> Any:
+        """
+        Lists all buckets accessible by the credentials.
+
+        Returns:
+            Any: List of bucket names, or an empty list on failure.
+        """
+        ...
+
+
+# -------- Sync --------
+class SyncS3Module(BaseS3Module):
+    def _get_client(self):
+        try:
+            import boto3
+
+            return boto3.client("s3", **self._args)
+        except ImportError:
+            logger.error(
+                "boto3 is not installed. Please install it to use S3 functionalities."
+            )
+            raise ImportError("boto3 is required for S3 operations.")
+        except Exception as e:
+            logger.error(f"Failed to create S3 client: {e}")
+            raise RuntimeError("Failed to create S3 client.") from e
+
+    def upload_file(self, file, bucket_name, object_name=None, metadata=None) -> bool:
+        s3 = self._get_client()
+        object_name = self._get_object_name(file, object_name)
+        file_size, metadata = self._prepare_metadata(file, metadata)
+        progress_callback = ProgressPercentage(
+            file_size, object_name, verbose=self._verbose
+        )
+        from boto3.s3.transfer import TransferConfig
+
+        config = TransferConfig(use_threads=False)
+        extra_args = {"Metadata": metadata}
+
+        try:
+            if isinstance(file, str):
+                s3.upload_file(
+                    file,
+                    bucket_name,
+                    object_name,
+                    Config=config,
+                    Callback=progress_callback,
+                    ExtraArgs=extra_args,
+                )
+            else:
+                s3.upload_fileobj(
+                    file,
+                    bucket_name,
+                    object_name,
+                    Config=config,
+                    Callback=progress_callback,
+                    ExtraArgs=extra_args,
+                )
+        except Exception as e:
+            logger.error(f"Upload failed: {e}")
+            return False
+        return True
+
+    def download_file(self, bucket_name, object_name, save_to=None):
+        s3 = self._get_client()
+        from boto3.s3.transfer import TransferConfig
+
+        try:
+            obj = s3.head_object(Bucket=bucket_name, Key=object_name)
+            file_size = obj["ContentLength"]
+            progress_callback = ProgressPercentage(
+                file_size, object_name, verbose=self._verbose
+            )
+            config = TransferConfig(use_threads=False)
+
+            if self._verbose:
+                print(
+                    f"Downloading {object_name} from bucket {bucket_name} "
+                    f"({file_size / (1024 * 1024):.2f} MB)"
+                )
+
+            if save_to:
+                s3.download_file(
+                    bucket_name,
+                    object_name,
+                    save_to,
+                    Config=config,
+                    Callback=progress_callback,
+                )
+                return True
+            else:
+                file_obj = io.BytesIO()
+                s3.download_fileobj(
+                    bucket_name,
+                    object_name,
+                    file_obj,
+                    Config=config,
+                    Callback=progress_callback,
+                )
+                file_obj.seek(0)
+                return file_obj
+        except Exception as e:
+            logger.error(f"Download failed: {e}")
+            return False
+
+    def delete_file(self, bucket_name, object_name):
+        s3 = self._get_client()
+        try:
+            s3.delete_object(Bucket=bucket_name, Key=object_name)
+            return True
+        except Exception as e:
+            logger.error(f"Delete failed: {e}")
+            return False
+
+    def list_files(self, bucket_name):
+        s3 = self._get_client()
+        try:
+            response = s3.list_objects_v2(Bucket=bucket_name)
+            return [obj["Key"] for obj in response.get("Contents", [])]
+        except Exception as e:
+            logger.error(f"List files failed: {e}")
+            return []
+
+    def list_buckets(self):
+        s3 = self._get_client()
+        try:
+            response = s3.list_buckets()
+            return [bucket["Name"] for bucket in response.get("Buckets", [])]
+        except Exception as e:
+            logger.error(f"List buckets failed: {e}")
+            return []

aisberg/modules/workflows.py

@@ -28,7 +28,7 @@ class AbstractWorkflowsModule(ABC):
         Raises:
             Exception: If there is an error fetching the workflows.
         """
-        pass
+        ...
 
     @abstractmethod
     def details(self, workflow_id: str) -> WorkflowDetails:
@@ -44,7 +44,7 @@ class AbstractWorkflowsModule(ABC):
             ValueError: If the workflow with the specified ID is not found.
             Exception: If there is an error fetching the workflow details.
         """
-        pass
+        ...
 
     @abstractmethod
     def run(self, workflow_id: str, data: dict = None, stream: bool = False):
@@ -67,7 +67,7 @@ class AbstractWorkflowsModule(ABC):
             ValueError: If the workflow with the specified ID is not found.
             Exception: If there is an error running the workflow.
         """
-        pass
+        ...
 
 
 class SyncWorkflowsModule(SyncModule, AbstractWorkflowsModule):

{aisberg-0.1.0.dist-info → aisberg-0.2.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aisberg
-Version: 0.1.0
+Version: 0.2.0
 Summary: Aisberg SDK for Python - A simple and powerful SDK to interact with the Aisberg API
 Author: Free Pro
 Author-email: Mathis Lambert <mathis.lambert@freepro.com>
@@ -16,6 +16,7 @@ License-File: LICENSE
 Requires-Dist: httpx>=0.28.1
 Requires-Dist: pydantic>=2.11.7
 Requires-Dist: pydantic-settings>=2.10.1
+Requires-Dist: boto3>=1.38.44
 Provides-Extra: dev
 Requires-Dist: pytest>=8.4.1; extra == "dev"
 Requires-Dist: pytest-asyncio>=1.0.0; extra == "dev"
@@ -47,6 +48,7 @@ conversational LLM workflows, collections, embeddings, and more.
 - **Environment-based configuration** (supports `.env` files and system environment variables)
 - **Context manager support** for easy resource management
 - **Custom tool registration**: Easily extend LLM capabilities with your own functions
+- **Document Parsing**: Parse documents into structured data (e.g., JSON, CSV, PNG, PDF, etc.)
 
 ---
 
@@ -77,6 +79,15 @@ AISBERG_API_KEY=...
 AISBERG_BASE_URL=https://url
 ```
 
+In order to use the Document Parsing feature, you also need to set the `S3` credentials - ask the FreePro team for
+these:
+
+```env
+S3_ACCESS_KEY_ID=...
+S3_SECRET_ACCESS_KEY=...
+S3_ENDPOINT=https://s3.endpoint
+```
+
 ### 2. **Synchronous Usage**
 
 ```python
@@ -127,6 +138,7 @@ asyncio.run(main())
 * `client.models` — Model discovery & info
 * `client.workflows` — Workflow management & execution
 * `client.tools` — Register and execute tools for LLM tool calls
+* `client.documents` — Document parsing and management
 
 Each module is available both in the sync and async clients with similar APIs.
 
@@ -173,6 +185,9 @@ client = AisbergClient(
 * `AISBERG_API_KEY`
 * `AISBERG_BASE_URL`
 * `AISBERG_TIMEOUT` (optional)
+* `S3_ACCESS_KEY_ID` (for document parsing)(optional)
+* `S3_SECRET_ACCESS_KEY` (for document parsing)(optional)
+* `S3_ENDPOINT` (for document parsing)(optional)
 
 ### **Using in a Context Manager**
 
@@ -208,5 +223,3 @@ For enterprise/commercial use, please contact [Mathis Lambert](mailto:mathis.lam
 ## Support
 
 For support, bug reports, or feature requests, please contact your technical representative.
-
----

{aisberg-0.1.0.dist-info → aisberg-0.2.0.dist-info}/RECORD

@@ -1,43 +1,46 @@
 aisberg/__init__.py,sha256=jMX3F2Fh5JobAkRMKGAcOpmNdoxSthgR4v_oM-Yw1l0,141
-aisberg/async_client.py,sha256=wxwkRUjxFbBlcj3D0aFspCVX5aACCksVhG9ZPVYrtx4,3547
-aisberg/client.py,sha256=-UgM9pi-R_qeUmods8sWTxBVGM13pNUfLJv8Zq1vQto,3586
-aisberg/config.py,sha256=tYuaE8WQFyyGhXEgF3kKLX7wM4PDIv6vDoezXaryOhk,441
+aisberg/async_client.py,sha256=0hO2lzdP2Vcb4qV619aJLrsxeXmfYj5wo0mog1bda5Q,3816
+aisberg/client.py,sha256=W1i9mh-o7f283kz2s9B2nHBCegJGN0Lgf1q0lAMNx9I,3853
+aisberg/config.py,sha256=6qSucVFxCBD9TiISDh2bTt7QlrjwvptTwEtcbj4Wt8k,610
 aisberg/exceptions.py,sha256=T1ioitOczdkmW3-8XfPG0ta_O2WqzS1iQtzhDBucb5Y,515
 aisberg/utils.py,sha256=4wkrpC5gz8PJWw3kZqaRbd3KekrI37zlDC__JiAVitw,3552
 aisberg/abstract/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 aisberg/abstract/modules.py,sha256=aRmB21oiL9NCZsHUx8zB0Q8VvZAZBucsZxYv5l7JygY,1698
 aisberg/api/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-aisberg/api/async_endpoints.py,sha256=cYOmUGBnMrvj8zaiek_Cjut6tB3paEIlJRHjRbggQBc,8279
-aisberg/api/endpoints.py,sha256=Ni-boFbHpB4ZiCbBwLCzoy8bxRTie-LHmBddqbwVn7Q,8003
+aisberg/api/async_endpoints.py,sha256=UZhz7_JeGnkQ1AWE7gEvvOdbaBUGE1nofG0NddqRctg,10982
+aisberg/api/endpoints.py,sha256=ByVxBVxrdPo-_8G0vzb_PnrwkmOSPI09koBsHjzl3gg,10610
 aisberg/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 aisberg/models/chat.py,sha256=LcvOgQm9Aan7pS4S90h4lrdzENX-NebA9pAKYdcDLB8,3667
-aisberg/models/collections.py,sha256=gtcpyMfUVkfZkWP82AlsfNDT0mM_wREdjnR32i3DP20,806
+aisberg/models/collections.py,sha256=Je9THFx-2qfTgmd_AktRd2oi6YeCw-j3iLLtlUdmaJ0,1103
+aisberg/models/documents.py,sha256=-9jNPRqMQEhkBMtZxDv8Fe0r8X91cpgQCv0FqPL8qH8,916
 aisberg/models/embeddings.py,sha256=gBGXR3DgyBCWB9Vky-iowWh_2Md7wa57I3fWxIrl77I,2409
 aisberg/models/models.py,sha256=NcECFpeVjt9nzQEN8lggDy-SesDSDFgSAYhqWz0dupA,1206
-aisberg/models/requests.py,sha256=9QXt9COo3GfubLQR5AgajxaDEr-ZOHrdjp-j9nyZrlw,205
+aisberg/models/requests.py,sha256=oCihxhUBMMaFtSKitnachNbXfCak4NVSWviurXU0orA,297
 aisberg/models/token.py,sha256=p3-uxCZMHbe4yryQdH21iDnPh-etJWgvUDUpCXV2xhk,190
 aisberg/models/tools.py,sha256=dO8-O-tV9usakviEvWzT7Z-dG5s7EkVV8cg5CyoxZHg,1830
 aisberg/models/workflows.py,sha256=CiqJdHsSFrOo2DAYNvYFvFOmO0hyg0xUehktkn3maSg,1258
-aisberg/modules/__init__.py,sha256=i499QPWq-JB596dRVuBFNleYoldJshYXGmC4oed0f58,725
-aisberg/modules/chat.py,sha256=WrMvq0O5gdBggoGm_dJcJywsHJvlKKYBUQ5HgPipQuU,13822
-aisberg/modules/collections.py,sha256=kgV7zFHme6d2SXPKITJknxnk1yc6SoWz2LGnGoifon0,3947
-aisberg/modules/document.py,sha256=twloRFq1wRS4j6fzth77IdceAaSrdDULzu3hlTiIsrA,3902
-aisberg/modules/embeddings.py,sha256=wq7DlOXVQBP9-V5f4ytY4nnD60DAJ49v6JFlBuJwmEs,10470
-aisberg/modules/me.py,sha256=yN4-ZYVDLViwz1oXbDpF1_1gULWiBf5GL0wj9-Pp-H4,2145
-aisberg/modules/models.py,sha256=EIyWOkO2FaOYV-fNB-42-AEY2bob6yA1_4bqlwKs6JY,3048
+aisberg/modules/__init__.py,sha256=Gfag_zlHi-f7t0UN41ZhJu80gdTkaLXk9IOnV8ug6dQ,894
+aisberg/modules/chat.py,sha256=q_Ob7stzFNZgh1ltktOoHTEDfKhwH3g6KxfDqMNqyP4,14129
+aisberg/modules/collections.py,sha256=Spk9kt88BDwGU3R9wUf7XJzq_wtjnuPdDgvhEhayIU8,16273
+aisberg/modules/documents.py,sha256=YJhGyyy4noj2Dj5XHDOy8RMlseqP8BYkcRnfaft3dJ4,5930
+aisberg/modules/embeddings.py,sha256=UV4lzNTIvooGNisAIS8sUmZ0rAxs3Omq-bnPhlyqQ74,10627
+aisberg/modules/me.py,sha256=VcbXBvcCNPoCKYHjDLvevuZWTJjQx3LLBzPFYkZKlb8,2144
+aisberg/modules/models.py,sha256=IWa2ezrWS4uQQPzx-YBYrBkRgC_ZnMJCnxFB2Az0DkA,3045
+aisberg/modules/s3.py,sha256=2pupGSaHxu4LzxwhMbxgYFt9-ILLcjF74iYmLtpmwY4,10170
 aisberg/modules/tools.py,sha256=TnTIfjQuPhPImtWTjEy3UCmXeroPrYX3BRzTvT9nIkk,2424
-aisberg/modules/workflows.py,sha256=TGDriJYy__dRBwP-MK-8zl6mlzZs22-0_TUqYYTeVfs,5004
+aisberg/modules/workflows.py,sha256=AxyBwCnZkRdgKHUOAoYXMAJH6J7K-qaL1psZ-f-MKhI,5001
 aisberg/requests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 aisberg/requests/async_requests.py,sha256=4J6Gj6vY5AD9lXhSTdYdAKQIqsqH9m1uaKc7XLDKWx8,2849
 aisberg/requests/sync_requests.py,sha256=ov7GDgx4C-oQ9xEZ7VQsGVY1fnuqRpPwjS0H_yH1yb8,2839
-aisberg-0.1.0.dist-info/licenses/LICENSE,sha256=dCGgiGa4f14s5mzTYnj1wIZNqGRVG2bUYyyk7nzKlEo,406
+aisberg-0.2.0.dist-info/licenses/LICENSE,sha256=dCGgiGa4f14s5mzTYnj1wIZNqGRVG2bUYyyk7nzKlEo,406
 tests/integration/test_collections_integration.py,sha256=HSs-Mhh5JNx2V6EeKo--G3xZq7pMk3z3gLReX7FJidg,3883
 tests/unit/test_collections_sync.py,sha256=DyizG4MIYNtBOC9TlsFdLlH5vz7USA12K9Uydlcydoo,3199
 tmp/test.py,sha256=8H3GxLztspwyt3_wLPtQkwdiLl--qqAn8ZaEUsXbwDE,743
 tmp/test_async.py,sha256=BaY3MkOIDXsersjYgnfaI43kZ42KDj5Andq5u-yO82s,4168
-tmp/test_doc_parse.py,sha256=b41Tsa1VkgNwyKwKoDRLNd05sIFGVaJncJGlFGiwYWY,324
+tmp/test_collection.py,sha256=wV8AnxuaTsHxlzVPigG_qW6kr57lMkKH_bBm9eFfJ-I,1893
+tmp/test_doc_parse.py,sha256=G8-jpZY2cuauzUK9oEvEHjE7Dgjf6LlPFg9F8WyF49g,1188
 tmp/test_sync.py,sha256=87QSz728gvjTXz2JykjF0U_mNsunDYrWsXYDjkBybI8,4929
-aisberg-0.1.0.dist-info/METADATA,sha256=ETmG5wRMX4W_Xr_L8ZyAXyS2BgtQcOLiIIE03vYEatA,5053
-aisberg-0.1.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-aisberg-0.1.0.dist-info/top_level.txt,sha256=7_CasmLkyF_h1CVRncNuAjam-DMIeMu1Rev_LpVcWoA,18
-aisberg-0.1.0.dist-info/RECORD,,
+aisberg-0.2.0.dist-info/METADATA,sha256=BZzpCVk1OsZQEw9oomWeK8Q50BPVi_YeCvY4JXvPX4E,5604
+aisberg-0.2.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+aisberg-0.2.0.dist-info/top_level.txt,sha256=7_CasmLkyF_h1CVRncNuAjam-DMIeMu1Rev_LpVcWoA,18
+aisberg-0.2.0.dist-info/RECORD,,

tmp/test_collection.py

@@ -0,0 +1,65 @@
+from aisberg import AisbergClient, AisbergAsyncClient
+import json
+
+
+def test_coll():
+    client = AisbergClient()
+    # client.collections.delete('test_aisberg_sdk')
+    #
+    # colls = client.collections.create('test_aisberg_sdk', {
+    #     "chunks": [
+    #         "This is a test chunk 1",
+    #         "This is a test chunk 2",
+    #         "This is a test chunk 3"
+    #     ],
+    #     "metadata": [
+    #         {"source": "test1.txt", "author": "Author 1"},
+    #         {"source": "test2.txt", "author": "Author 2"},
+    #         {"source": "test3.txt", "author": "Author 3"}
+    #     ]
+    # })
+
+
+    points = client.collections.insert_points('test_aisberg_sdk', {
+        "chunks": [
+            "LKFJELZFKJZLEsdfsdfKFJZLKEJF chunk 1",
+            "LKFJELZFKJZsdfsdfLEKFJZLKEJF chunk 2",
+            "LKFJELZFsdfsdfsdfKJZLEKFJZLsdfsdfKEJF chunk 3"
+        ],
+        "metadata": [
+            {"source": "test1.txt", "author": "Author 1"},
+            {"source": "test2.txt", "author": "Author 2"},
+            {"source": "test3.txt", "author": "Author 3"}
+        ]
+    }).points
+
+    print(f"### Collection Created: {points}\n")
+
+    points_to_delete = []
+    for i in range(2):
+        points_to_delete.append(points[i].id)
+    print(f"### Points to delete: {points_to_delete}\n")
+    a = client.collections.delete_points('test_aisberg_sdk', points_to_delete)
+    print(f"### Points Deleted: {a}\n")
+
+    c = client.collections.clear('test_aisberg_sdk')
+    print(f"### Collection Cleared: {c}\n")
+
+
+
+
+async def async_test_coll():
+    client = AisbergAsyncClient()
+    await client.collections.delete('test_aisberg_sdk')
+
+    colls = await client.collections.create('test_aisberg_sdk', './tmp/test.json')
+
+    print(f"### Collection Created: {colls}\n")
+
+
+if __name__ == "__main__":
+    test_coll()
+    #
+    # import asyncio
+    #
+    # asyncio.run(async_test_coll())

tmp/test_doc_parse.py

@@ -1,12 +1,36 @@
-from aisberg import AisbergClient
-from aisberg.api.endpoints import parse_document
+from aisberg import AisbergClient, AisbergAsyncClient
+import json
 
-client = AisbergClient()
 
 def test_doc():
-    file_bytes = b"Hello, this is a test document."
-    doc = parse_document(client._client, (file_bytes, "test.txt"), source="test_source")
-    print(doc)
+    client = AisbergClient()
+    print(client.models.list())
+    stt = str(input("STT model to use (default: tts-1.5): ") or "tts-1.5")
+    docs = client.documents.parse(
+        ["./tmp/test.mp3", "./tmp/test.png"],
+        stt_model=stt,
+        vlm_model="pixtral-12b-2409",
+        input="Que vois tu en haut à droite de l'image ?"
+    )
+
+    print(f"### MP3 Content: {docs[0].content.data}\n")
+    # print(f"### PDF Content: {docs[1].content.data}\n")
+    # print(f"### JSON Content: {json.dumps(docs[2].content.data, indent=2)}\n")
+
+
+async def async_test_doc():
+    client = AisbergAsyncClient()
+    docs = await client.documents.parse(["./tmp/test.mp3", "./tmp/test.pdf", "./tmp/test.json"])
+
+    img = await client.documents.parse("./tmp/test.png", input="Que vois tu en haut à droite de l'image ?")
+
+    print(f"### MP3 Content: {docs[0].content.data}\n")
+    print(f"### PDF Content: {docs[1].content.data}\n")
+    print(f"### JSON Content: {json.dumps(docs[2].content.data, indent=2)}\n")
+
 
 if __name__ == "__main__":
-    test_doc()
+    test_doc()
+
+    # import asyncio
+    # asyncio.run(async_test_doc())