backend/__init__.py

@@ -0,0 +1,52 @@
+import importlib
+import os
+import traceback
+
+import matplotlib
+from apiflask import APIBlueprint, APIFlask
+from flask_cors import CORS
+
+matplotlib.use("agg")
+
+cors = CORS()
+api_bp = APIBlueprint("api", __name__, url_prefix="/api/v1")
+dataset_path: str | None = None
+
+
+class Config:
+    SECRET_KEY = "secret!"
+    SEND_FILE_MAX_AGE_DEFAULT = -1
+
+
+def create_app():
+    global dataset_path
+
+    # Create and configure app
+    app = APIFlask(
+        "MAI Service",
+        title="MAI Service API",
+        docs_path="/",
+        version="1.0",
+        static_folder="",
+        template_folder="",
+    )
+    app.config.from_object(Config)
+
+    dataset_path = os.path.join(app.instance_path, "dataset")
+    os.makedirs(dataset_path, exist_ok=True)
+
+    @app.errorhandler(Exception)
+    def my_error_processor(error):
+        traceback.print_exception(error)
+        return {"message": str(error), "detail": "No details"}, 500
+
+    # Import custom REST methods
+    importlib.import_module("backend.api")
+
+    # Enable REST API
+    app.register_blueprint(api_bp)
+
+    # Enable app extensions
+    cors.init_app(app)
+
+    return app

backend/api.py

@@ -0,0 +1,57 @@
+from apiflask import FileSchema, Schema, fields
+from flask import send_file
+
+from backend import api_bp, dataset_path
+from backend.service import Service
+
+
+class FileUpload(Schema):
+    file = fields.File(required=True)
+
+
+class ColumnInfoDto(Schema):
+    datatype = fields.String()
+    items = fields.List(fields.String())
+
+
+class TableColumnDto(Schema):
+    name = fields.String()
+    datatype = fields.String()
+    items = fields.List(fields.String())
+
+
+service = Service(dataset_path)
+
+
+@api_bp.post("/dataset")
+@api_bp.input(FileUpload, location="files")
+def upload_dataset(files_data):
+    uploaded_file = files_data["file"]
+    return service.upload_dataset(uploaded_file)
+
+
+@api_bp.get("/dataset")
+def get_all_datasets():
+    return service.get_all_datasets()
+
+
+@api_bp.get("/dataset/<string:name>")
+@api_bp.output(TableColumnDto(many=True))
+def get_dataset_info(name: str):
+    return service.get_dataset_info(name)
+
+
+@api_bp.get("/dataset/<string:name>/<string:column>")
+@api_bp.output(ColumnInfoDto)
+def get_column_info(name: str, column: str):
+    return service.get_column_info(name, column)
+
+
+@api_bp.get("/dataset/draw/hist/<string:name>/<string:column>")
+@api_bp.output(
+    FileSchema(type="string", format="binary"), content_type="image/png", example=""
+)
+def get_dataset_hist(name: str, column: str):
+    data = service.get_hist(name, column)
+    data.seek(0)
+    return send_file(data, download_name=f"{name}.hist.png", mimetype="image/png")

backend/service.py

@@ -0,0 +1,59 @@
+import io
+import os
+import pathlib
+from typing import BinaryIO, Dict, List
+
+import pandas as pd
+from matplotlib.figure import Figure
+from werkzeug.datastructures import FileStorage
+from werkzeug.utils import secure_filename
+
+
+class Service:
+    def __init__(self, dataset_path: str | None) -> None:
+        if dataset_path is None:
+            raise Exception("Dataset path is not defined")
+        self.__path: str = dataset_path
+
+    def __get_dataset(self, filename: str) -> pd.DataFrame:
+        full_file_name = os.path.join(self.__path, secure_filename(filename))
+        return pd.read_csv(full_file_name)
+
+    def upload_dataset(self, file: FileStorage) -> str:
+        if file.filename is None:
+            raise Exception("Dataset upload error")
+        file_name: str = file.filename
+        full_file_name = os.path.join(self.__path, secure_filename(file_name))
+        file.save(full_file_name)
+        return file_name
+
+    def get_all_datasets(self) -> List[str]:
+        return [file.name for file in pathlib.Path(self.__path).glob("*.csv")]
+
+    def get_dataset_info(self, filename) -> List[Dict]:
+        dataset = self.__get_dataset(filename)
+        dataset_info = []
+        for column in dataset.columns:
+            items = dataset[column].astype(str)
+            column_info = {
+                "name": column,
+                "datatype": dataset.dtypes[column],
+                "items": items,
+            }
+            dataset_info.append(column_info)
+        return dataset_info
+
+    def get_column_info(self, filename, column) -> Dict:
+        dataset = self.__get_dataset(filename)
+        datatype = dataset.dtypes[column]
+        items = sorted(dataset[column].astype(str).unique())
+        return {"datatype": datatype, "items": items}
+
+    def get_hist(self, filename, column) -> BinaryIO:
+        dataset = self.__get_dataset(filename)
+        bytes = io.BytesIO()
+        plot: Figure | None = dataset.plot.hist(column=[column], bins=80).get_figure()
+        if plot is None:
+            raise Exception("Can't create hist plot")
+        plot.savefig(bytes, dpi=300, format="png")
+        return bytes

lab_1/README.md

@@ -0,0 +1,2 @@
+# AIM-PIbd-32-Kuzin-P-S
+

lab_1/requirements.txt

@@ -0,0 +1,32 @@
+asttokens==2.4.1
+colorama==0.4.6
+comm==0.2.2
+debugpy==1.8.5
+decorator==5.1.1
+executing==2.1.0
+ipykernel==6.29.5
+ipython==8.27.0
+jedi==0.19.1
+jupyter_client==8.6.2
+jupyter_core==5.7.2
+matplotlib-inline==0.1.7
+nest-asyncio==1.6.0
+numpy==2.1.1
+packaging==24.1
+pandas==2.2.2
+parso==0.8.4
+platformdirs==4.3.3
+prompt_toolkit==3.0.47
+psutil==6.0.0
+pure_eval==0.2.3
+Pygments==2.18.0
+python-dateutil==2.9.0.post0
+pytz==2024.2
+pywin32==306
+pyzmq==26.2.0
+six==1.16.0
+stack-data==0.6.3
+tornado==6.4.1
+traitlets==5.14.3
+tzdata==2024.1
+wcwidth==0.2.13