From 3e852e551b41f4c52045b35a6f8ea7645dd16000 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Moritz=20H=C3=B6lting?=
 <87192362+moritz-hoelting@users.noreply.github.com>
Date: Wed, 8 May 2024 15:26:12 +0200
Subject: [PATCH] add chapter "how it is made" to mensa-upb-cli project

---
 src/content/projects/mensa-upb-cli/index.md | 64 +++++++++++++++++++++
 1 file changed, 64 insertions(+)

diff --git a/src/content/projects/mensa-upb-cli/index.md b/src/content/projects/mensa-upb-cli/index.md
index 6abe7eb..0078b60 100644
--- a/src/content/projects/mensa-upb-cli/index.md
+++ b/src/content/projects/mensa-upb-cli/index.md
@@ -46,3 +46,67 @@ It works by parsing the command-line arguments with [clap](https://crates.io/cra
     ```bash
     mensa-upb-cli -p student
     ```
+
+## How it is made
+
+My university has multiple cafeterias and a website with the menu of each one. I did not like checking multiple pages when choosing what and where to eat.
+
+Therefore I decided to build an application that would make this process easier. My solution had 4 steps:
+
+1. Read the user input
+2. Fetch the data
+3. Filter the data based on user input
+4. Output the data in a readable way
+
+### 1. Reading user input
+
+For reading the cli arguments, I choose [clap](https://crates.io/crates/clap) which is a popular library and very easy to use by deriving the traits.
+
+```rust
+#[derive(Parser)]
+#[command(author, version, about, long_about = None)]
+struct Cli {
+    /// Choose the mensa
+    #[arg(short, long, value_enum, default_values_t = [Mensa::Forum, Mensa::Academica])]
+    mensa: Vec<Mensa>,
+    /// Choose the price level
+    #[arg(short, long)]
+    price_level: Option<PriceLevel>,
+    /// Choose how many days in the future to fetch
+    #[arg(short, long)]
+    days_ahead: Option<u64>,
+    /// Filter by extras
+    #[arg(short, long)]
+    extras: Vec<String>,
+}
+```
+
+### 2. Fetch the data
+
+Because there is no API for our cafeteria, I had to scrape the website. The tools I used are [reqwest](https://crates.io/crates/reqwest) for fetching the html and [scraper](https://crates.io/crates/scraper) for extracting the required information from the html.
+
+### 3. Filter the data based on user input
+
+Filtering is done at multiple places:
+
+-   only fetch and parse the pages of the cafeterias requested and the day selected
+-   filter which price level to show (student, employee, guest)
+-   filter which meals match the selected extra (vegetarian, vegan)
+
+### 4. Output the data in a readable way
+
+For readability, I choose to display the meals in a table. Also I wanted to group the meals by categorie (main dishes, side dishes and desserts). A nice library I found for printing tables to the terminal is [comfy-table](https://crates.io/crates/comfy-table). It allows customizing the borders and alignment of each cell.
+
+```rust
+let mut desserts_row = Row::new();
+desserts_row.add_cell(
+    Cell::from("Desserts")
+        .set_alignment(CellAlignment::Center)
+        .add_attribute(comfy_table::Attribute::Underlined)
+        .add_attribute(comfy_table::Attribute::OverLined),
+);
+```
+
+## Source Code
+
+If you want to take a look at the source code, it is available on my [GitHub](https://github.com/moritz-hoelting/mensa-upb-cli). You could even try to adapt the web requesting and scraping to the cafeteria you go to.