eotdl 2025.4.2__tar.gz → 2025.4.2.post2__tar.gz

{eotdl-2025.4.2 → eotdl-2025.4.2.post2}/PKG-INFO

@@ -1,23 +1,18 @@
 Metadata-Version: 2.4
 Name: eotdl
-Version: 2025.4.2
+Version: 2025.4.2.post2
 Summary: Earth Observation Training Data Lab
 Author-email: earthpulse <it@earthpulse.es>
 License-Expression: MIT
-Requires-Python: >=3.8
-Requires-Dist: geomet>=1.1.0
-Requires-Dist: geopandas>=0.13.2
-Requires-Dist: markdown>=3.7
-Requires-Dist: pydantic>=2.10.6
-Requires-Dist: pyjwt>=2.9.0
-Requires-Dist: pystac>=1.8.4
+Requires-Python: >=3.12
+Requires-Dist: geopandas>=1.0.1
+Requires-Dist: pydantic>=2.11.1
+Requires-Dist: pyjwt>=2.10.1
+Requires-Dist: pystac>=1.12.2
 Requires-Dist: python-frontmatter>=1.1.0
-Requires-Dist: pyyaml>=6.0.2
-Requires-Dist: rasterio>=1.3.11
-Requires-Dist: requests>=2.32.3
-Requires-Dist: sentinelhub>=3.11.1
+Requires-Dist: stac-geoparquet>=0.6.0
 Requires-Dist: tqdm>=4.67.1
-Requires-Dist: typer>=0.15.1
+Requires-Dist: typer>=0.15.2
 Description-Content-Type: text/markdown
 
 <p align="center">

eotdl-2025.4.2.post2/eotdl/__init__.py

@@ -0,0 +1 @@
+__version__ = "2025.04.02-2"

eotdl-2025.4.2.post2/eotdl/repos/DatasetsAPIRepo.py

@@ -0,0 +1,43 @@
+import requests
+
+from ..repos import APIRepo
+
+class DatasetsAPIRepo(APIRepo):
+    def __init__(self, url=None):
+        super().__init__(url)
+
+    def retrieve_datasets(self, name, limit):
+        url = self.url + "datasets"
+        if name is not None:
+            url += "?match=" + name
+        if limit is not None:
+            if name is None:
+                url += "?limit=" + str(limit)
+            else:
+                url += "&limit=" + str(limit)
+        response = requests.get(url)
+        return self.format_response(response)
+    
+    def retrieve_dataset(self, name):
+        response = requests.get(self.url + "datasets?name=" + name)
+        return self.format_response(response)
+    
+    def get_dataset_by_id(self, dataset_id):
+        response = requests.get(self.url + "datasets/" + dataset_id)
+        return self.format_response(response)
+    
+    def create_dataset(self, metadata, user):
+        response = requests.post(
+            self.url + "datasets",
+            json=metadata,
+            headers=self.generate_headers(user),
+        )
+        return self.format_response(response)
+
+    def complete_ingestion(self, dataset_id, version, size, user):
+        response = requests.post(
+            self.url + "datasets/complete/" + dataset_id,
+            json={"version": version, "size": size},
+            headers=self.generate_headers(user),
+        )
+        return self.format_response(response)

eotdl-2025.4.2.post2/eotdl/repos/FilesAPIRepo.py

@@ -0,0 +1,103 @@
+import requests
+import os
+
+from ..repos import APIRepo
+
+
+class FilesAPIRepo(APIRepo):
+    def __init__(self, url=None):
+        super().__init__(url)
+
+    def ingest_file(
+        self, file_path_or_bytes, file_name, dataset_or_model_id, user, endpoint, version=None
+    ):
+        url = self.url + f"{endpoint}/{dataset_or_model_id}"
+        if version is not None:
+            url += "?version=" + str(version)
+        # get a presigned url to upload the file directly to the bucket
+        reponse = requests.post(
+            url,
+            json={
+                "file_name": file_name,
+                # "file_size": files_size,
+                # "checksum": checksum
+            },
+            headers=self.generate_headers(user),
+        )
+        data, error = self.format_response(reponse)
+        if error:
+            raise Exception(error)
+        # ingest the file
+        error = None
+        try:
+            presigned_url = data["presigned_url"]
+            if isinstance(file_path_or_bytes, (str, bytes)):
+                if isinstance(file_path_or_bytes, str):
+                    # Handle file path
+                    with open(file_path_or_bytes, 'rb') as f:
+                        file_data = f.read()
+                else:
+                    # Handle bytes directly
+                    file_data = file_path_or_bytes
+                # Send file data to presigned URL
+                response = requests.put(presigned_url, data=file_data)
+                response.raise_for_status()
+            else:
+                raise TypeError("file_path_or_bytes must be either a file path string or bytes")
+        except Exception as e:
+            error = str(e)
+        return data, error
+
+    def stage_file(
+        self,
+        dataset_or_model_id,
+        file_name,
+        user,
+        path,
+        endpoint="datasets",
+        progress=False,
+    ):
+        url = self.url + f"{endpoint}/{dataset_or_model_id}/stage/{file_name}"
+        # if file_version is not None:
+        #     url += "?version=" + str(file_version)
+        return self.stage_file_url(url, path, user)
+
+
+    def stage_file_url(
+        self,
+        url,
+        path,
+        user,
+    ):
+        if '/stage/' in url:  # asset is in EOTDL (can do better...)
+            file_name = url.split("/stage/")[-1] 
+            reponse = requests.get(url, headers=self.generate_headers(user))
+            data, error = self.format_response(reponse)
+            if error:
+                raise Exception(error)
+            presigned_url = data["presigned_url"]
+        else:
+            file_name = url.split("//")[-1]
+            presigned_url = url
+        file_path = f"{path}/{file_name}"
+        for i in range(1, len(file_path.split("/")) - 1):
+            os.makedirs("/".join(file_path.split("/")[: i + 1]), exist_ok=True)
+        try:
+            response = requests.get(presigned_url)
+            response.raise_for_status()  # This will raise an HTTPError for 4XX and 5XX status codes
+            with open(file_path, 'wb') as f:
+                f.write(response.content)
+        except requests.exceptions.HTTPError as e:
+            raise Exception(f"Failed to stage file: {str(e)}")
+        except Exception as e:
+            raise Exception(f"Unexpected error while staging file: {str(e)}")
+        return file_path
+    
+    def generate_presigned_url(self, filename, dataset_or_model_id, user, endpoint="datasets"):
+        url = f"{self.url}{endpoint}/{dataset_or_model_id}/stage/{filename}"
+        reponse = requests.get(url, headers=self.generate_headers(user))
+        data, error = self.format_response(reponse)
+        if error:
+            # print("ERROR generate_presigned_url", error)
+            return None
+        return data["presigned_url"]

eotdl-2025.4.2.post2/eotdl/repos/ModelsAPIRepo.py

@@ -0,0 +1,41 @@
+import requests
+
+from ..repos import APIRepo
+
+class ModelsAPIRepo(APIRepo):
+    def __init__(self, url=None):
+        super().__init__(url)
+
+    def retrieve_models(self, name, limit):
+        url = self.url + "models"
+        if name is not None:
+            url += "?match=" + name
+        if limit is not None:
+            if name is None:
+                url += "?limit=" + str(limit)
+            else:
+                url += "&limit=" + str(limit)
+        response = requests.get(url)
+        return self.format_response(response)
+
+    def retrieve_model(self, name):
+        response = requests.get(self.url + "models?name=" + name)
+        return self.format_response(response)
+    
+    def create_model(self, metadata, user):
+        response = requests.post(
+            self.url + "models",
+            json=metadata,
+            headers=self.generate_headers(user),
+        )
+        return self.format_response(response)
+
+    def complete_ingestion(self, model_id, version, size, user):
+        response = requests.post(
+            self.url + "models/complete/" + model_id,
+            json={"version": version, "size": size},
+            headers=self.generate_headers(user),
+        )
+        return self.format_response(response)
+
+    

{eotdl-2025.4.2 → eotdl-2025.4.2.post2}/eotdl/repos/STACAPIRepo.py

@@ -1,5 +1,4 @@
 import requests
-import geopandas as gpd
 
 from ..repos import APIRepo
 

eotdl-2025.4.2.post2/pyproject.toml

@@ -0,0 +1,30 @@
+[project]
+name = "eotdl"
+version = "2025.04.02-2"
+description = "Earth Observation Training Data Lab"
+authors = [
+    {name = "earthpulse", email = "it@earthpulse.es"},
+]
+license = "MIT"
+readme = "README.md"
+requires-python = ">=3.12"
+dependencies = [
+    "geopandas>=1.0.1",
+    "pydantic>=2.11.1",
+    "pyjwt>=2.10.1",
+    "pystac>=1.12.2",
+    "python-frontmatter>=1.1.0",
+    "stac-geoparquet>=0.6.0",
+    "tqdm>=4.67.1",
+    "typer>=0.15.2",
+]
+
+[project.scripts]
+eotdl = "eotdl.cli:app"
+
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[tool.hatch.build.targets.wheel]
+packages = ["eotdl"]