https://github.com/perfringis/droplo-recruitment-task
droplo-recruitment-task
https://github.com/perfringis/droplo-recruitment-task
Last synced: 3 months ago
JSON representation
droplo-recruitment-task
- Host: GitHub
- URL: https://github.com/perfringis/droplo-recruitment-task
- Owner: perfringis
- Created: 2024-11-29T10:49:32.000Z (7 months ago)
- Default Branch: main
- Last Pushed: 2024-12-03T20:57:40.000Z (7 months ago)
- Last Synced: 2025-01-30T03:25:02.213Z (5 months ago)
- Language: JavaScript
- Size: 94.7 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# droplo-recruitment-task
## Task description
Oto Twoje zadanie rekrutacyjne:
W pliku image_processor.js znajduje sie klasa ImageProcessorTest jej jedynym zadaniem jest wczytanie pliku CSV iraz
wygenerowanie miniaturek do pobranych zdjec z linkow i zapisanie ich do bazy danych mongodb.
Struktura pliku csv wyglada nastepujaco:
```
index, id, url
0, 42fb7bba-7f84-433e-88e5-cb20e6243cac, https://placeholder.com/image1.jpg
```
Dla uproszczenia mozemy zlozyc ze pliki wieksze niz 120mb powinny byc odrzucane.
Twoim zadaniem bedzie zastanowienie sie nad kodem tej klasy i wypisanie potencjanych bledow w kodzie oraz usprawnien.
## How to run project?
1. Clone repository:
```
git clone [email protected]:perfringis/droplo-recruitment-task.git
```
2. Configure `.env` file based on `.env.dev` template.
3. Run docker configuration for development stage:
```
docker-compose -f docker-compose-dev.yml up
```
4. Run project by:
```
npm/yarn/pnpm run start
```
5. Run test by:
```
npm/yarn/pnpm run test
```
## What is missing?
- The proper project structure
- The instruction to run the project
- The .env file config for running the app
- The data.csv file with data
- Missing unit test for the methods of the `ImageProcessor`.
## What is wrong?
- The index.js file has a lot of responsibilities. The script should be split into smaller ones.
- useNewUrlParser and useUnifiedTopology are deprecated.
```js
mongoose.connect(process.env.MONGO_URI, { useNewUrlParser: true, useUnifiedTopology: true });
```
- The Program never runs/executes any csv file.
- Lack of `start` method invocation from `ImageProcessorTest` class.
- The path to `data/data.csv` could not be resolved. Additionally, template literals are not needed.
- The `__dirname` is pointing to the specific directory and that implementation is "fragile" to any changes.
- The `global.gc` will never invoke. Moreover, custom cleaning the memory should be investigated via `--inspect`.
- Instead of using `console.info` or `console.error`. The project should use logging system such as `winston`.
- The `batchSize` parameter for `processChunk` function is unused.
- Missing secure way of getting information from `.env` file.
- Mongoose connection in the `index.js` file.
- The `DEFAULT_BATCH_SIZE` is not dynamic type of information. Each time app has to be restarted to change that value.
- The implementation of `start` method creates own `id` in mongo database.
- The `readFileSync` is invoking action in synchronous way.
- `ImageProcessor` class has a lot of responsibilities. The code should be slit into smaller ones.
## What have I done?
- I moved schema and model to separate directories for better readability.
- I moved `ImageProcessorTest` class to separate directory.
- Introduced `prettier` for better readability of the code.
- Removed `global.gc`.
- Introduced `winston` logging system.
- Remove `batchSize` parameter from `processChunk` function.
- Installed `dotenv` lib for secure reading variables.
- Moved `dotenv` config to separate file.
- Moved `mongoose` connection to separate file.
- Restructure `.env` file in case of mongo variables.
- Removing `DEFAULT_BATCH_SIZE` variable and add parameter to start method with default start value.
- Rename from `id` to `_id`.
- Replaced `readFileSync` with async version of method.
- Created `File` class implementation.
- Created `CSVParser` class.
- Added docker container setup for testing.
- Overall refactor of the code.