add logging and dockerfile

.gitignore

@@ -166,5 +166,4 @@ cython_debug/
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
 
-/data
 .env

Dockerfile

@@ -1,8 +1,8 @@
 # syntax=docker/dockerfile:1
 
-FROM python:3.9-slim-buster
+FROM python:3.10-slim-buster
 
-ENV POETRY_VERSION=1.4 \
+ENV POETRY_VERSION=1.6 \
     POETRY_VIRTUALENVS_CREATE=false
 
 # Install poetry
@@ -18,4 +18,6 @@ RUN poetry install --no-interaction --no-ansi --no-root --no-dev
 # Copy Python code to the Docker image
 COPY pypi_llm /code/pypi_llm/
 
+ENV PYTHONPATH=/code
+
 CMD [ "python", "pypi_llm/foo.py"]

README.md

@@ -1,55 +1,20 @@
 # pypi-llm
 
-[![Release](https://img.shields.io/github/v/release/fpgmaas/pypi-llm)](https://img.shields.io/github/v/release/fpgmaas/pypi-llm)
-[![Build status](https://img.shields.io/github/actions/workflow/status/fpgmaas/pypi-llm/main.yml?branch=main)](https://github.com/fpgmaas/pypi-llm/actions/workflows/main.yml?query=branch%3Amain)
-[![codecov](https://codecov.io/gh/fpgmaas/pypi-llm/branch/main/graph/badge.svg)](https://codecov.io/gh/fpgmaas/pypi-llm)
-[![Commit activity](https://img.shields.io/github/commit-activity/m/fpgmaas/pypi-llm)](https://img.shields.io/github/commit-activity/m/fpgmaas/pypi-llm)
-[![License](https://img.shields.io/github/license/fpgmaas/pypi-llm)](https://img.shields.io/github/license/fpgmaas/pypi-llm)
+https://drive.google.com/file/d/1huR7-VD3AieBRCcQyRX9MWbPLMb_czjq/view?usp=sharing
 
-This is a template repository for Python projects that use Poetry for their dependency management.
+# setup
 
-- **Github repository**: <https://github.com/fpgmaas/pypi-llm/>
-- **Documentation** <https://fpgmaas.github.io/pypi-llm/>
-
-## Getting started with your project
-
-First, create a repository on GitHub with the same name as this project, and then run the following commands:
-
-```bash
-git init -b main
-git add .
-git commit -m "init commit"
-git remote add origin git@github.com:fpgmaas/pypi-llm.git
-git push -u origin main
 ```
-
-Finally, install the environment and the pre-commit hooks with
-
-```bash
-make install
+docker build -t pypi-llm .
 ```
 
-You are now ready to start development on your project!
-The CI/CD pipeline will be triggered when you open a pull request, merge to main, or when you create a new release.
-
-To finalize the set-up for publishing to PyPi or Artifactory, see [here](https://fpgmaas.github.io/cookiecutter-poetry/features/publishing/#set-up-for-pypi).
-For activating the automatic documentation with MkDocs, see [here](https://fpgmaas.github.io/cookiecutter-poetry/features/mkdocs/#enabling-the-documentation-on-github).
-To enable the code coverage reports, see [here](https://fpgmaas.github.io/cookiecutter-poetry/features/codecov/).
-
-## Releasing a new version
-
-- Create an API Token on [Pypi](https://pypi.org/).
-- Add the API Token to your projects secrets with the name `PYPI_TOKEN` by visiting [this page](https://github.com/fpgmaas/pypi-llm/settings/secrets/actions/new).
-- Create a [new release](https://github.com/fpgmaas/pypi-llm/releases/new) on Github.
-- Create a new tag in the form `*.*.*`.
-
-For more details, see [here](https://fpgmaas.github.io/cookiecutter-poetry/features/cicd/#how-to-trigger-a-release).
-
----
-
-Repository initiated with [fpgmaas/cookiecutter-poetry](https://github.com/fpgmaas/cookiecutter-poetry).
-
----
+```
+docker run --rm \
+  --env-file .env \
+  -v $(pwd)/data:/code/data \
+  pypi-llm \
+  python /code/pypi_llm/scripts/1_download_dataset.py
+```
 
 ## total
 

data/.gitignore

@@ -0,0 +1,4 @@
+# Ignore everything in this directory
+*
+# Except this file
+!.gitignore

frontend/app/components/InfoBox.tsx

@@ -0,0 +1,30 @@
+import React from "react";
+
+interface InfoBoxProps {
+  infoBoxVisible: boolean;
+}
+
+const InfoBox: React.FC<InfoBoxProps> = ({ infoBoxVisible }) => {
+  if (!infoBoxVisible) return null;
+
+  return (
+    <div className="w-3/5 bg-white p-6 rounded-lg shadow-lg mt-4">
+      <h2 className="text-2xl font-bold mb-2">How does this work?</h2>
+      <p className="text-gray-700">
+        This application allows you to search for Python packages on PyPi using
+        natural language. An example query would be "a package that creates
+        plots and beautiful visualizations".
+      </p>
+      <br />
+      <p className="text-gray-700">
+        Once you click search, your query will be matched against the summary
+        and the first part of the description of all PyPi packages with more
+        than 50 weekly downloads. The results are then scored based on their
+        similarity and their number of weekly downloads, and the thirty best
+        results are displayed in the table below.
+      </p>
+    </div>
+  );
+};
+
+export default InfoBox;

frontend/app/page.tsx

@@ -1,53 +1,28 @@
 "use client";
 
 import { useState } from "react";
-import axios from "axios";
-import SearchResultsTable from "../components/SearchResultsTable";
+import { handleSearch, sortResults } from "./utils/search";
+import SearchResultsTable from "./components/SearchResultsTable";
+import InfoBox from "./components/InfoBox";
 import { ClipLoader } from "react-spinners";
 
-export default function Home() {
-  const [text, setText] = useState("");
-  const [results, setResults] = useState([]);
-  const [sortField, setSortField] = useState("weekly_downloads");
-  const [sortDirection, setSortDirection] = useState("desc");
-  const [loading, setLoading] = useState(false);
-  const [error, setError] = useState("");
-  const [infoBoxVisible, setInfoBoxVisible] = useState(false);
-
-  const handleSearch = async () => {
-    setLoading(true);
-    setError("");
-    try {
-      const response = await axios.post(
-        "http://localhost:8000/search",
-        {
-          query: text,
-        },
-        {
-          headers: {
-            "Content-Type": "application/json",
-          },
-        },
-      );
-      const fetchedResults = response.data.matches;
-      setResults(sortResults(fetchedResults, sortField, sortDirection));
-    } catch (error) {
-      setError("Error fetching search results.");
-      console.error("Error fetching search results:", error);
-    } finally {
-      setLoading(false);
-    }
-  };
+interface Match {
+  name: string;
+  similarity: number;
+  weekly_downloads: number;
+  summary: string;
+}
 
-  const sortResults = (data, field, direction) => {
-    return [...data].sort((a, b) => {
-      if (a[field] < b[field]) return direction === "asc" ? -1 : 1;
-      if (a[field] > b[field]) return direction === "asc" ? 1 : -1;
-      return 0;
-    });
-  };
+export default function Home() {
+  const [text, setText] = useState<string>("");
+  const [results, setResults] = useState<Match[]>([]);
+  const [sortField, setSortField] = useState<string>("similarity");
+  const [sortDirection, setSortDirection] = useState<string>("desc");
+  const [loading, setLoading] = useState<boolean>(false);
+  const [error, setError] = useState<string>("");
+  const [infoBoxVisible, setInfoBoxVisible] = useState<boolean>(false);
 
-  const handleSort = (field) => {
+  const handleSort = (field: string) => {
     const direction =
       sortField === field && sortDirection === "asc" ? "desc" : "asc";
     setSortField(field);
@@ -72,7 +47,16 @@ export default function Home() {
         ></textarea>
         <button
           className="w-[250px] p-2 border rounded bg-blue-500 text-white hover:bg-blue-700 focus:outline-none focus:ring-2 focus:ring-blue-500"
-          onClick={handleSearch}
+          onClick={() =>
+            handleSearch(
+              text,
+              sortField,
+              sortDirection,
+              setResults,
+              setLoading,
+              setError,
+            )
+          }
         >
           Search
         </button>
@@ -91,20 +75,7 @@ export default function Home() {
         </button>
       </div>
 
-      {infoBoxVisible && (
-        <div className="w-3/5 bg-white p-6 rounded-lg shadow-lg mt-4">
-          <h2 className="text-2xl font-bold mb-2">How does this work?</h2>
-          <p className="text-gray-700">
-            This application allows you to search for Python packages on PyPi
-            using natural language. So an example query would be "a package that
-            creates plots and beautiful visualizations". Once you click search,
-            your query will be matched against the summary and the first part of
-            the description of all PyPi packages with more than 50 weekly
-            downloads, and the 50 most similar results will be displayed in a
-            table below.
-          </p>
-        </div>
-      )}
+      <InfoBox infoBoxVisible={infoBoxVisible} />
 
       {results.length > 0 && (
         <div className="w-full flex justify-center mt-6">

frontend/app/utils/search.ts

@@ -0,0 +1,52 @@
+import axios from "axios";
+
+interface Match {
+  name: string;
+  similarity: number;
+  weekly_downloads: number;
+  summary: string;
+}
+
+export const handleSearch = async (
+  query: string,
+  sortField: string,
+  sortDirection: string,
+  setResults: React.Dispatch<React.SetStateAction<Match[]>>,
+  setLoading: React.Dispatch<React.SetStateAction<boolean>>,
+  setError: React.Dispatch<React.SetStateAction<string>>,
+) => {
+  setLoading(true);
+  setError("");
+  try {
+    const response = await axios.post(
+      "http://localhost:8000/search",
+      {
+        query: query,
+      },
+      {
+        headers: {
+          "Content-Type": "application/json",
+        },
+      },
+    );
+    const fetchedResults: Match[] = response.data.matches;
+    setResults(sortResults(fetchedResults, sortField, sortDirection));
+  } catch (error) {
+    setError("Error fetching search results.");
+    console.error("Error fetching search results:", error);
+  } finally {
+    setLoading(false);
+  }
+};
+
+export const sortResults = (
+  data: Match[],
+  field: string,
+  direction: string,
+): Match[] => {
+  return [...data].sort((a, b) => {
+    if (a[field] < b[field]) return direction === "asc" ? -1 : 1;
+    if (a[field] > b[field]) return direction === "asc" ? 1 : -1;
+    return 0;
+  });
+};

pypi_llm/api/main.py

@@ -1,3 +1,5 @@
+import logging
+
 import polars as pl
 from dotenv import load_dotenv
 from fastapi import FastAPI
@@ -6,9 +8,12 @@
 from sentence_transformers import SentenceTransformer
 
 from pypi_llm.config import Config
+from pypi_llm.utils.logging import setup_logging
 from pypi_llm.utils.score_calculator import calculate_score
 from pypi_llm.vector_database import VectorDatabaseInterface
 
+setup_logging()
+
 app = FastAPI()
 
 load_dotenv()
@@ -55,12 +60,20 @@ class SearchResponse(BaseModel):
 
 @app.post("/search/", response_model=SearchResponse)
 async def search(query: QueryModel):
+    """
+    Search for the packages whose summary and description have the highest similarity to the query.
+    We take the top_k * 2 most similar packages, and then calculate weighted score based on the similarity and weekly downloads.
+    The top_k packages with the highest score are returned.
+    """
+
+    logging.info(f"Searching for similar projects. Query: '{query.query}'")
     df_matches = vector_database_interface.find_similar(query.query, top_k=query.top_k * 2)
     df_matches = df_matches.join(df, how="left", on="name")
 
+    logging.info("Found similar projects. Calculating the weighted scores and filtering...")
     df_matches = calculate_score(df_matches)
     df_matches = df_matches.sort("score", descending=True)
     df_matches = df_matches.head(query.top_k)
 
-    print("sending")
+    logging.info("Returning the results...")
     return SearchResponse(matches=df_matches.to_dicts())