feat(api): implement 30-minute cache for PDF link requests

.gitignore

@@ -1,3 +1,4 @@
 /target
 pdf
 csv
+pdf.json

Cargo.lock

@@ -30,6 +30,50 @@ dependencies = [
  "zerocopy",
 ]
 
+[[package]]
+name = "askama"
+version = "0.12.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b79091df18a97caea757e28cd2d5fda49c6cd4bd01ddffd7ff01ace0c0ad2c28"
+dependencies = [
+ "askama_derive",
+ "askama_escape",
+ "humansize",
+ "num-traits",
+ "percent-encoding",
+]
+
+[[package]]
+name = "askama_derive"
+version = "0.12.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "19fe8d6cb13c4714962c072ea496f3392015f0989b1a2847bb4b2d9effd71d83"
+dependencies = [
+ "askama_parser",
+ "basic-toml",
+ "mime",
+ "mime_guess",
+ "proc-macro2",
+ "quote",
+ "serde",
+ "syn 2.0.76",
+]
+
+[[package]]
+name = "askama_escape"
+version = "0.10.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "619743e34b5ba4e9703bba34deac3427c72507c7159f5fd030aea8cac0cfe341"
+
+[[package]]
+name = "askama_parser"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "acb1161c6b64d1c3d83108213c2a2533a342ac225aabd0bda218278c2ddb00c0"
+dependencies = [
+ "nom",
+]
+
 [[package]]
 name = "async-trait"
 version = "0.1.81"
@@ -129,6 +173,15 @@ version = "0.22.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "72b3254f16251a8381aa12e40e3c4d2f0199f8c6508fbecb9d91f575e0fbb8c6"
 
+[[package]]
+name = "basic-toml"
+version = "0.1.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "823388e228f614e9558c6804262db37960ec8821856535f5c3f59913140558f8"
+dependencies = [
+ "serde",
+]
+
 [[package]]
 name = "bitflags"
 version = "2.6.0"
@@ -582,6 +635,12 @@ dependencies = [
  "pin-project-lite",
 ]
 
+[[package]]
+name = "http-range-header"
+version = "0.4.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "08a397c49fec283e3d6211adbe480be95aae5f304cfb923e9970e08956d5168a"
+
 [[package]]
 name = "httparse"
 version = "1.9.4"
@@ -594,6 +653,15 @@ version = "1.0.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "df3b46402a9d5adb4c86a0cf463f42e19994e3ee891101b1841f30a545cb49a9"
 
+[[package]]
+name = "humansize"
+version = "2.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6cb51c9a029ddc91b07a787f1d86b53ccfa49b0e86688c946ebe8d3555685dd7"
+dependencies = [
+ "libm",
+]
+
 [[package]]
 name = "hyper"
 version = "1.4.1"
@@ -715,6 +783,12 @@ version = "0.2.158"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "d8adc4bb1803a324070e64a98ae98f38934d91957a99cfb3a43dcbc01bc56439"
 
+[[package]]
+name = "libm"
+version = "0.2.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4ec2a862134d2a7d32d7983ddcdd1c4923530833c9f2ea1a44fc5fa473989058"
+
 [[package]]
 name = "linux-raw-sys"
 version = "0.4.14"
@@ -775,6 +849,22 @@ version = "0.3.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "6877bb514081ee2a7ff5ef9de3281f14a4dd4bceac4c09388074a6b5df8a139a"
 
+[[package]]
+name = "mime_guess"
+version = "2.0.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f7c44f8e672c00fe5308fa235f821cb4198414e1c77935c1ab6948d3fd78550e"
+dependencies = [
+ "mime",
+ "unicase",
+]
+
+[[package]]
+name = "minimal-lexical"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "68354c5c6bd36d73ff3feceb05efa59b6acb7626617f4962be322a825e61f79a"
+
 [[package]]
 name = "miniz_oxide"
 version = "0.7.4"
@@ -819,6 +909,25 @@ version = "1.0.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "650eef8c711430f1a879fdd01d4745a7deea475becfb90269c06775983bbf086"
 
+[[package]]
+name = "nom"
+version = "7.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d273983c5a657a70a3e8f2a01329822f3b8c8172b73826411a55751e404a0a4a"
+dependencies = [
+ "memchr",
+ "minimal-lexical",
+]
+
+[[package]]
+name = "num-traits"
+version = "0.2.19"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "071dfc062690e90b734c0b2273ce72ad0ffa95f0c74596bc250dcfd960262841"
+dependencies = [
+ "autocfg",
+]
+
 [[package]]
 name = "object"
 version = "0.36.4"
@@ -882,6 +991,7 @@ dependencies = [
 name = "orario-scolastico-itet"
 version = "0.1.0"
 dependencies = [
+ "askama",
  "axum",
  "csv",
  "failure",
@@ -892,6 +1002,7 @@ dependencies = [
  "serde_json",
  "sha1",
  "tokio",
+ "tower-http",
 ]
 
 [[package]]
@@ -1679,6 +1790,31 @@ dependencies = [
  "tracing",
 ]
 
+[[package]]
+name = "tower-http"
+version = "0.5.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1e9cd434a998747dd2c4276bc96ee2e0c7a2eadf3cae88e52be55a05fa9053f5"
+dependencies = [
+ "bitflags",
+ "bytes",
+ "futures-util",
+ "http",
+ "http-body",
+ "http-body-util",
+ "http-range-header",
+ "httpdate",
+ "mime",
+ "mime_guess",
+ "percent-encoding",
+ "pin-project-lite",
+ "tokio",
+ "tokio-util",
+ "tower-layer",
+ "tower-service",
+ "tracing",
+]
+
 [[package]]
 name = "tower-layer"
 version = "0.3.3"
@@ -1723,6 +1859,15 @@ version = "1.17.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "42ff0bf0c66b8238c6f3b578df37d0b7848e55df8577b3f74f92a69acceeb825"
 
+[[package]]
+name = "unicase"
+version = "2.7.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f7d2d4dafb69621809a81864c9c1b864479e1235c0dd4e199924b9742439ed89"
+dependencies = [
+ "version_check",
+]
+
 [[package]]
 name = "unicode-bidi"
 version = "0.3.15"

Cargo.toml

@@ -4,6 +4,7 @@ version = "0.1.0"
 edition = "2021"
 
 [dependencies]
+askama = "0.12.1"
 axum = "0.7.5"
 csv = "1.3.0"
 failure = "0.1.8"
@@ -14,3 +15,4 @@ serde = { version = "1.0.209", features = ["derive"] }
 serde_json = "1.0.127"
 sha1 = "0.10.6"
 tokio = { version = "1.40.0", features = ["full"] }
+tower-http = { version = "0.5.2", features = ["fs"] }

Dockerfile

@@ -0,0 +1,34 @@
+FROM rust:1.80 AS builder
+
+WORKDIR /code
+
+COPY Cargo.toml Cargo.lock ./
+RUN mkdir src && echo "fn main() {}" > src/main.rs
+RUN cargo build --release
+
+COPY src src
+COPY templates templates
+RUN cargo install --path .
+
+FROM debian:bookworm-slim AS runtime
+
+RUN apt-get update && \
+  apt-get install -y libssl-dev curl python3 python3-venv && \
+  apt-get clean && \
+  rm -rf /var/lib/apt/lists/*
+
+
+WORKDIR /app
+COPY --from=builder /usr/local/cargo/bin/orario-scolastico-itet .
+
+RUN python3 -m venv pyenv && \
+  bash -c "source pyenv/bin/activate"
+RUN ./pyenv/bin/pip install pdfplumber
+
+ENV PATH $PATH:/app/pyenv/bin
+
+COPY static static
+COPY pdf2csv.py .
+
+EXPOSE 3000
+CMD ["./orario-scolastico-itet"]

docker-compose.yaml

@@ -0,0 +1,6 @@
+services:
+  web:
+    build: .
+    restart: unless-stopped
+    ports:
+      - "3000:3000"

src/api/mod.rs

@@ -28,6 +28,16 @@ pub fn get_routes() -> Router {
 }
 
 async fn get_pdf_links() -> Result<Vec<String>, reqwest::Error> {
+    if let Ok(metadata) = fs::metadata("pdf.json") {
+        let last_modified = metadata.modified().unwrap();
+        let now = std::time::SystemTime::now();
+        let diff = now.duration_since(last_modified).unwrap().as_secs();
+        if diff < 30 * 60 {
+            let data = fs::read_to_string("pdf.json").unwrap();
+            return Ok(serde_json::from_str(&data).unwrap());
+        }
+    }
+
     let response = reqwest::get("https://cassandroferminervi.edu.it/orario-scolastico/")
         .await?
         .text()
@@ -41,6 +51,9 @@ async fn get_pdf_links() -> Result<Vec<String>, reqwest::Error> {
         result.push(element.attr("href").unwrap().to_owned());
     }
 
+    let data = serde_json::to_string(&result).unwrap();
+    fs::write("pdf.json", data).unwrap();
+
     Ok(result)
 }
 
@@ -96,7 +109,7 @@ pub struct Header {
     weekdays: HashMap<String, Range<usize>>,
 }
 
-pub async fn get_class(Query(params): Query<FilterByTeacherQuery>) -> impl IntoResponse {
+pub async fn get_teacher(Query(params): Query<FilterByTeacherQuery>) -> impl IntoResponse {
     let filename = match download_pdf(params.id).await {
         Ok(x) => x,
         Err(_) => {
@@ -138,7 +151,7 @@ pub struct FilterByClassQuery {
     classe: String,
 }
 
-pub async fn get_teacher(Query(params): Query<FilterByClassQuery>) -> impl IntoResponse {
+pub async fn get_class(Query(params): Query<FilterByClassQuery>) -> impl IntoResponse {
     let filename = match download_pdf(params.id).await {
         Ok(x) => x,
         Err(_) => {
@@ -202,6 +215,8 @@ pub fn get_header(record: &StringRecord) -> Header {
             .position(|x| !x.is_empty())
             .unwrap_or(record.len());
 
+        let field = field.replace("ì", "i").replace(" ", "").to_uppercase();
+
         if field == "DOCENTE" {
             docente = i as u8;
         } else if field == "LUNEDI" {
@@ -250,12 +265,7 @@ pub async fn get_teachers(Query(params): Query<GetTeachersQuery>) -> impl IntoRe
     for record in rdr.records() {
         let record = match record {
             Ok(x) => x,
-            Err(_) => {
-                return (
-                    StatusCode::UNPROCESSABLE_ENTITY,
-                    Json(json!({ "errore": "Formato non supportato"})),
-                )
-            }
+            Err(_) => continue,
         };
 
         if let Some(teacher) = record.get(header.teacher as usize) {
@@ -304,7 +314,10 @@ pub async fn get_classes(Query(params): Query<GetClassesQuery>) -> impl IntoResp
     let mut classes: Vec<HashMap<char, HashSet<String>>> = vec![HashMap::new(); 5];
 
     for record in rdr.records() {
-        let record = record.unwrap();
+        let record = match record {
+            Ok(x) => x,
+            Err(_) => continue,
+        };
         for (i, cell) in record.iter().enumerate() {
             for (_, range) in header.weekdays.clone() {
                 if range.contains(&i) && is_valid_class(cell) {

src/frontend/mod.rs

@@ -0,0 +1,23 @@
+use askama::Template;
+use axum::{
+    http::StatusCode,
+    response::{Html, IntoResponse},
+    routing::get,
+};
+use tower_http::services::ServeDir;
+
+pub fn get_routes() -> axum::Router {
+    axum::Router::new()
+        .route("/", get(index))
+        .nest_service("/static", ServeDir::new("static"))
+}
+
+#[derive(Template)]
+#[template(path = "index.html")]
+struct IndexTemplate {}
+
+async fn index() -> impl IntoResponse {
+    let template = IndexTemplate {};
+    let html = template.render().unwrap();
+    (StatusCode::OK, Html(html).into_response())
+}

src/main.rs

@@ -1,10 +1,13 @@
 use std::net::SocketAddr;
 
 mod api;
+mod frontend;
 
 #[tokio::main]
 async fn main() {
-    let app = axum::Router::new().merge(api::get_routes());
+    let app = axum::Router::new()
+        .merge(api::get_routes())
+        .merge(frontend::get_routes());
 
     let addr = SocketAddr::from(([0, 0, 0, 0], 3000));
     let listener = tokio::net::TcpListener::bind(&addr).await.unwrap();

static/index.js

@@ -0,0 +1,234 @@
+function load_links() {
+  fetch("/api/pdf")
+    .then((response) => response.json())
+    .then((data) => {
+      let i = 0;
+      data.links.forEach((el) => {
+        const select = document.getElementById("links");
+        const option = document.createElement("option");
+        option.value = i;
+        option.text = el.split("/").pop().split(".pdf")[0];
+        select.appendChild(option);
+        i++;
+      });
+      document.getElementById("links-label").removeAttribute("aria-busy");
+    })
+    .catch((error) => {
+      alert("Errore nel caricamento dei dati, riprova più tardi.");
+      console.error("Error:", error);
+    });
+}
+
+function load_teachers(id) {
+  fetch(`/api/professori?id=${id}`)
+    .then((response) => response.json())
+    .then((data) => {
+      sessionStorage.setItem("teachers", JSON.stringify(data));
+      let i = 0;
+      const list = document.getElementById("teachers");
+      list.innerHTML = "";
+      data.forEach((el) => {
+        const option = document.createElement("option");
+        option.value = el;
+        list.appendChild(option);
+        i++;
+      });
+      document.getElementById("teacher-label").removeAttribute("aria-busy");
+    })
+    .catch((error) => {
+      alert("Errore nel caricamento dei dati, riprova più tardi.");
+      console.error("Error:", error);
+    });
+}
+
+function load_classes(id) {
+  fetch(`/api/classi?id=${id}`)
+    .then((response) => response.json())
+    .then((data) => {
+      const select = document.getElementById("year");
+      for (let i = 0; i < select.children.length; i++) {
+        if (!select.children[i].hasAttribute("disabled")) {
+          select.removeChild(select.children[i]);
+          i--;
+        }
+      }
+      for (let i = 0; i < data.length; i++) {
+        const option = document.createElement("option");
+        option.value = i;
+        option.text = i + 1;
+        select.appendChild(option);
+      }
+      sessionStorage.setItem("classes", JSON.stringify(data));
+      document.getElementById("class-label").removeAttribute("aria-busy");
+    })
+    .catch((error) => {
+      alert("Errore nel caricamento dei dati, riprova più tardi.");
+      console.error("Error:", error);
+    });
+}
+
+function make_teacher_table(id, teacher) {
+  fetch(`/api/professore?id=${id}&professore=${teacher}`)
+    .then((response) => response.json())
+    .then((data) => {
+      let table = document.getElementById("table");
+      let tbody = document.createElement("tbody");
+      let week = [
+        "Lunedì",
+        "Martedì",
+        "Mercoledì",
+        "Giovedì",
+        "Venerdì",
+        "Sabato",
+      ];
+      week.forEach((weekday) => {
+        let tr = document.createElement("tr");
+        let th = document.createElement("th");
+        th.scope = "row";
+        th.textContent = weekday;
+        tr.appendChild(th);
+        data[weekday].forEach((teacher_class) => {
+          let td = document.createElement("td");
+          td.textContent = teacher_class;
+          tr.appendChild(td);
+        });
+        tbody.appendChild(tr);
+      });
+      table.appendChild(tbody);
+      document.getElementById("table").removeAttribute("aria-busy");
+    })
+    .catch((error) => {
+      alert("Errore nel caricamento dei dati, riprova più tardi.");
+      console.error("Error:", error);
+    });
+}
+
+function make_student_table(id, student_class) {
+  fetch(`/api/classe?id=${id}&classe=${student_class}`)
+    .then((response) => response.json())
+    .then((data) => {
+      let table = document.getElementById("table");
+      let tbody = document.createElement("tbody");
+      let week = [
+        "Lunedì",
+        "Martedì",
+        "Mercoledì",
+        "Giovedì",
+        "Venerdì",
+        "Sabato",
+      ];
+      week.forEach((weekday) => {
+        let tr = document.createElement("tr");
+        let th = document.createElement("th");
+        th.scope = "row";
+        th.textContent = weekday;
+        tr.appendChild(th);
+        data[weekday].forEach((student_class) => {
+          let td = document.createElement("td");
+          td.textContent = student_class;
+          tr.appendChild(td);
+        });
+        tbody.appendChild(tr);
+      });
+      table.appendChild(tbody);
+      document.getElementById("table").removeAttribute("aria-busy");
+    })
+    .catch((error) => {
+      alert("Errore nel caricamento dei dati, riprova più tardi.");
+      console.error("Error:", error);
+    });
+}
+
+function select_pdf() {
+  Cookies.set("id", document.getElementById("links").value);
+  document.getElementById("form-type").removeAttribute("hidden");
+}
+
+function select_type() {
+  let id = document.getElementById("links").value;
+  let value = document.getElementById("type").value;
+  if (value === "teacher") {
+    load_teachers(id);
+    document.getElementById("form-teacher").removeAttribute("hidden");
+    document.getElementById("form-classes").setAttribute("hidden", "");
+  } else if (value === "student") {
+    load_classes(id);
+    document.getElementById("form-classes").removeAttribute("hidden");
+    document.getElementById("form-teacher").setAttribute("hidden", "");
+  }
+}
+
+function load_section() {
+  let data = JSON.parse(sessionStorage.getItem("classes"));
+  let year = document.getElementById("year").value;
+  const select = document.getElementById("section");
+  for (let i = 0; i < select.children.length; i++) {
+    if (!select.children[i].hasAttribute("disabled")) {
+      select.removeChild(select.children[i]);
+      i--;
+    }
+  }
+  Object.keys(data[year]).forEach((el) => {
+    const option = document.createElement("option");
+    option.value = el;
+    option.text = el;
+    select.appendChild(option);
+  });
+}
+
+function load_major() {
+  let data = JSON.parse(sessionStorage.getItem("classes"));
+  let year = document.getElementById("year").value;
+  let section = document.getElementById("section").value;
+  let select = document.getElementById("major");
+  for (let i = 0; i < select.children.length; i++) {
+    if (!select.children[i].hasAttribute("disabled")) {
+      select.removeChild(select.children[i]);
+      i--;
+    }
+  }
+  data[year][section].forEach((el) => {
+    const option = document.createElement("option");
+    option.value = el;
+    option.text = el;
+    select.appendChild(option);
+  });
+}
+
+function select_teacher() {
+  let teacher = document.getElementById("teacher").value;
+  let teachers = JSON.parse(sessionStorage.getItem("teachers"));
+  if (!teachers.includes(teacher)) {
+    return false;
+  }
+  Cookies.set("type", "teacher");
+  Cookies.set("teacher", teacher);
+}
+
+function select_class() {
+  Cookies.set("type", "student");
+  let student_class = document.getElementById("year").value;
+  student_class += document.getElementById("section").value;
+  student_class += document.getElementById("major").value;
+  Cookies.set("class", student_class);
+}
+
+function delete_cookies() {
+  Cookies.remove("id");
+  Cookies.remove("type");
+  Cookies.remove("teacher");
+  Cookies.remove("class");
+}
+
+if (Cookies.get("id") && Cookies.get("type")) {
+  if (Cookies.get("type") === "teacher" && Cookies.get("teacher")) {
+    make_teacher_table(Cookies.get("id"), Cookies.get("teacher"));
+    document.getElementById("dashboard").removeAttribute("hidden");
+  } else if (Cookies.get("type") === "student" && Cookies.get("class")) {
+    make_student_table(Cookies.get("id"), Cookies.get("class"));
+    document.getElementById("dashboard").removeAttribute("hidden");
+  }
+} else {
+  document.getElementById("forms").removeAttribute("hidden");
+  load_links();
+}

templates/index.html

@@ -0,0 +1,81 @@
+<!doctype html>
+<html lang="it">
+  <head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1" />
+    <title>Orario scolastico</title>
+    <link
+      href="https://cdn.jsdelivr.net/npm/@picocss/pico@2/css/pico.min.css"
+      rel="stylesheet"
+    />
+    <script src="https://cdn.jsdelivr.net/npm/js-cookie@3.0.5/dist/js.cookie.min.js"></script>
+    <script src="/static/index.js" defer></script>
+  </head>
+  <body>
+    <main class="container">
+      <div id="forms" hidden>
+        <form id="form-pdf">
+          <fieldset>
+            <label id="links-label" aria-busy="true"
+              >Quale orario vuoi usare?</label
+            >
+            <select id="links" onchange="select_pdf()" required>
+              <option selected disabled value="none">
+                Seleziona un orario
+              </option>
+            </select>
+          </fieldset>
+        </form>
+        <form id="form-type" hidden>
+          <formfield>
+            <label>Sei un professore o uno studente?</label>
+            <select id="type" onchange="select_type()">
+              <option selected disabled value="none">
+                Seleziona una categoria
+              </option>
+              <option value="teacher">Professore</option>
+              <option value="student">Studente</option>
+            </select>
+          </formfield>
+        </form>
+        <form id="form-teacher" onsubmit="return select_teacher()" hidden>
+          <formfield>
+            <label id="teacher-label" aria-busy="true"
+              >Seleziona il tuo cognome</label
+            >
+            <input id="teacher" type="text" list="teachers" />
+            <datalist id="teachers"></datalist>
+          </formfield>
+          <input type="submit" value="Continua" />
+        </form>
+        <form id="form-classes" hidden>
+          <formfield>
+            <label id="class-label" aria-busy="true"
+              >Seleziona la tua classe</label
+            >
+            <select id="year" onchange="load_section()">
+              <option selected disabled value="none">Seleziona l'anno</option>
+            </select>
+            <select id="section" onchange="load_major()">
+              <option selected disabled value="none">
+                Seleziona la sezione
+              </option>
+            </select>
+            <select id="major" onchange="select_class()">
+              <option selected disabled value="none">
+                Seleziona l'indirizzo
+              </option>
+            </select>
+            <input type="submit" value="Continua" />
+          </formfield>
+        </form>
+      </div>
+      <div id="dashboard" hidden>
+        <table id="table" class="striped" aria-busy="true"></table>
+        <form>
+          <input type="submit" onclick="delete_cookies()" value="Reimposta" />
+        </form>
+      </div>
+    </main>
+  </body>
+</html>