push

.env (example)

@@ -0,0 +1,2 @@
+GOOGLE_DEVELOPER_KEY=key
+GOOGLE_CSE_ID=key

.gitignore

@@ -0,0 +1,2 @@
+venv
+.env

Dockerfile

@@ -0,0 +1,6 @@
+FROM python:3.11-slim
+WORKDIR /app
+COPY . /app
+RUN pip install --no-cache-dir -r requirements.txt
+ENV FLASK_APP=app.py
+CMD ["python", "app.py"]

app.py

@@ -0,0 +1,47 @@
+from flask import Flask, render_template, request, send_file, jsonify, make_response
+from scraper import get_google_first_page
+import io, json, csv, yaml, os
+from dotenv import load_dotenv
+
+load_dotenv()
+
+app = Flask(__name__)
+
+@app.route("/", methods=["GET"])
+def index():
+    return render_template("index.html", results=[])
+
+@app.route("/search", methods=["POST"])
+def search():
+    query = request.form.get("q", "").strip()
+    if not query:
+        return render_template("index.html", error="Zadejte dotaz.")
+    try:
+        results = get_google_first_page(query)  # list of dicts
+    except Exception as e:
+        return render_template("index.html", error=f"Vyhledávání selhalo: {e}")
+
+    notice = None
+    if not results:
+        notice = (
+            "Nebyly nalezeny žádné výsledky/dosáhli jste limitu (5 sekund mezi požadavky)."
+        )
+    return render_template("results.html", query=query, results=results, notice=notice)
+
+@app.route("/export", methods=["POST"])
+def export():
+    data = request.get_json()
+    ext = data.get("format", "json")
+
+    results = data.get("results", [])
+    filename = f"results.{ext}"
+    if ext == "json":
+        
+        buf = io.BytesIO(json.dumps(results, ensure_ascii=False, indent=2).encode("utf-8"))
+        return send_file(buf, as_attachment=True, download_name=filename, mimetype="application/json")
+    else:
+        return jsonify({"error": "unsupported format"}), 400
+
+if __name__ == "__main__":
+    port = int(os.environ.get("PORT", 5000))
+    app.run(host="0.0.0.0", port=port, debug=True)

docker-compose.yml

@@ -0,0 +1,17 @@
+version: "3.8"
+services:
+  web:
+    build: .
+    ports:
+      - "8050:5000"
+    environment:
+      - FLASK_ENV=development
+      - SERPAPI_KEY=${SERPAPI_KEY:-}
+    volumes:
+      - .:/app
+    healthcheck:
+      test: ["CMD-SHELL", "pytest -q tests/test_scraper.py::test_get_google_first_page_maps_output || exit 1"]
+      interval: 1m
+      timeout: 20s
+      retries: 3
+      start_period: 20s

requirements.txt

@@ -0,0 +1,5 @@
+Flask
+requests
+PyYAML
+python-dotenv
+pytest

scraper.py

@@ -0,0 +1,102 @@
+import os
+import time
+from urllib.parse import urlparse
+
+import requests
+
+
+# --- Configurace ---
+API_ENDPOINT = "https://www.googleapis.com/customsearch/v1"
+DEFAULT_LOCALE = "cs"   # language for results
+DEFAULT_NUM = 10         # how many results to fetch (max 10 per API call)
+RATE_SECONDS = 5.0       # wait time between API requests
+last_api_call = 0.0      # remember when we last called the API
+
+
+def _throttle_api():
+    """
+    Pause if the last API call was too recent.
+
+    This is a super basic rate limiter: we allow one request every 5 seconds.
+    It helps to not run through your daily quota too fast.
+    """
+    global last_api_call
+    now = time.time()
+    wait = RATE_SECONDS - (now - last_api_call)
+
+    if wait > 0:
+        time.sleep(wait)
+    last_api_call = time.time()
+
+
+def favicon_from_link(link):
+    try:
+        u = urlparse(link)
+        scheme = u.scheme or "https"
+        netloc = u.netloc
+        if not netloc and u.path:
+            # Handle URLs without a scheme like: example.com/path
+            netloc = u.path.split("/")[0]
+        if not netloc:
+            return ""
+        return f"{scheme}://{netloc}/favicon.ico"
+    except Exception:
+        return ""
+
+
+def _cse_request(q, num=DEFAULT_NUM, hl=DEFAULT_LOCALE):
+    """Call the Custom Search API and return the JSON data.
+
+    Requires two env variables in your .env file:
+      - GOOGLE_DEVELOPER_KEY (your API key)
+      - GOOGLE_CSE_ID (your search engine ID)
+    """
+    api_key = os.environ.get("GOOGLE_DEVELOPER_KEY")
+    cse_id = os.environ.get("GOOGLE_CSE_ID")
+    if not api_key:
+        raise RuntimeError("GOOGLE_DEVELOPER_KEY není nastaven v .env")
+    if not cse_id:
+        raise RuntimeError("GOOGLE_CSE_ID (Programmable Search Engine ID) není nastaven v .env")
+
+    _throttle_api()
+    params = {
+        "key": api_key,
+        "cx": cse_id,
+        "q": q,
+        "num": min(max(num, 1), 10),  # API allows up to 10 per call
+        "hl": hl,
+        "safe": "off",
+    }
+    resp = requests.get(API_ENDPOINT, params=params, timeout=15)
+    if resp.status_code != 200:
+        # Try to extract a nice error message from API response
+        try:
+            data = resp.json()
+            msg = data.get("error", {}).get("message") or resp.text
+        except Exception:
+            msg = resp.text
+        raise RuntimeError(f"Google CSE API chyba ({resp.status_code}): {msg}")
+    return resp.json()
+
+
+def get_google_first_page(query):
+    """Return a list of results for a query using the Custom Search API.
+
+    Each item has: position, title, link, snippet, icon
+    """
+    data = _cse_request(query, num=DEFAULT_NUM, hl=DEFAULT_LOCALE)
+    items = data.get("items", []) or []
+
+    results = []
+    for i, item in enumerate(items, start=1):
+        link = item.get("link") or item.get("formattedUrl") or ""
+        title = item.get("title") or item.get("htmlTitle") or ""
+        snippet = item.get("snippet") or item.get("htmlSnippet") or ""
+        results.append({
+            "position": i,
+            "title": title,
+            "link": link,
+            "snippet": snippet,
+            "icon": favicon_from_link(link),
+        })
+    return results

templates/index.html

@@ -0,0 +1,47 @@
+<!doctype html>
+<html lang="cs">
+  <head>
+    <meta charset="utf-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <title>Vyhledávání Google — Inzio</title>
+    <!-- Bootstrap 5 CSS -->
+    <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.3/dist/css/bootstrap.min.css" rel="stylesheet" integrity="sha384-QWTKZyjpPEjISv5WaRU9OFeRpok6YctnYmDr5pNlyT2bRjXh0JMhjY6hW+ALEwIH" crossorigin="anonymous">
+  </head>
+  <body class="bg-light d-flex min-vh-100">
+    <main class="container my-auto py-5">
+      <div class="row justify-content-center">
+        <div class="col-12 col-md-10 col-lg-8">
+          <div class="text-center mb-4">
+            <h1 class="fw-semibold">Inzio Web Search</h1>
+            <p class="text-muted mb-0">Jednoduché vyhledávání na Googlu</p>
+          </div>
+
+          {% if error %}
+						<div class="alert alert-danger" role="alert">
+							{{ error }}
+						</div>
+          {% endif %}
+
+          <div class="card shadow-sm">
+            <div class="card-body p-4">
+              <form action="/search" method="POST" class="d-flex flex-column gap-3">
+                <div>
+                  <label for="q" class="form-label">Hledaný dotaz</label>
+                  <input type="text" class="form-control form-control-lg" id="q" name="q" placeholder="Zadejte dotaz…" required>
+                </div>
+                <div class="d-grid d-sm-flex gap-2">
+                  <button type="submit" class="btn btn-primary btn-lg">Hledat</button>
+                </div>
+              </form>
+            </div>
+            <div class="card-footer bg-white text-muted small">
+              Výsledky poskytuje <a href="https://developers.google.com/custom-search/v1/overview#search_engine_id" target="_blank" rel="noopener">Google Custom Search JSON API</a> · limit: 1 požadavek / 5&nbsp;s
+            </div>
+          </div>
+        </div>
+      </div>
+    </main>
+
+    <script src="https://cdn.jsdelivr.net/npm/bootstrap@5.3.3/dist/js/bootstrap.bundle.min.js" integrity="sha384-YvpcrYf0tY3lHB60NNkmXc5s9fDVZLESaAA55NDzOxhy9GkcIdslK1eN7N6jIeHz" crossorigin="anonymous"></script>
+  </body>
+</html>

templates/results.html

@@ -0,0 +1,90 @@
+<!doctype html>
+<html lang="cs">
+  <head>
+    <meta charset="utf-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <title>Výsledky pro "{{ query }}" - vontor.cz</title>
+    <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.3/dist/css/bootstrap.min.css" rel="stylesheet" integrity="sha384-QWTKZyjpPEjISv5WaRU9OFeRpok6YctnYmDr5pNlyT2bRjXh0JMhjY6hW+ALEwIH" crossorigin="anonymous">
+    <script>
+      function download(fmt) {
+        const payload = { format: fmt, results: JSON.parse(document.getElementById('json-data').textContent) };
+        fetch('/export', {
+          method: 'POST',
+          headers: {'Content-Type': 'application/json'},
+          body: JSON.stringify(payload)
+        }).then(r => {
+          if (r.ok) return r.blob();
+          return r.json().then(j=>{throw j});
+        }).then(blob=>{
+          const url = window.URL.createObjectURL(blob);
+          const a = document.createElement('a');
+          a.href = url;
+          a.download = 'results.' + fmt;
+          document.body.appendChild(a);
+          a.click();
+          a.remove();
+          window.URL.revokeObjectURL(url);
+        }).catch(e=>{ alert('Export failed: ' + JSON.stringify(e));});
+      }
+    </script>
+  </head>
+  <body class="bg-light min-vh-100 d-flex">
+    <main class="container my-auto py-4">
+      <div class="d-flex justify-content-between align-items-center mb-3">
+        <h1 class="h3 mb-0">Výsledky pro "{{ query }}"</h1>
+        <a class="btn btn-outline-secondary" href="/">Nové hledání</a>
+      </div>
+
+      {% if notice %}
+      <div class="alert alert-warning" role="alert">
+        {{ notice }}
+      </div>
+      {% endif %}
+
+      <div class="card shadow-sm mb-3">
+        <div class="card-body">
+          <div class="d-flex flex-wrap gap-2 justify-content-between align-items-center">
+            <div class="text-muted">Počet výsledků: <strong>{{ results|length }}</strong></div>
+            <div class="d-flex gap-2">
+              <button class="btn btn-sm btn-outline-primary" onclick="download('json')">Stáhnout JSON</button>
+            </div>
+          </div>
+        </div>
+      </div>
+
+      {% if results and results|length > 0 %}
+        <ol class="list-group list-group-numbered">
+          
+          {% for item in results %}
+            <li class="list-group-item">
+              <div class="d-flex gap-3 align-items-start">
+                {% if item.icon %}
+                  <img src="{{ item.icon }}" alt="" width="20" height="20" class="rounded mt-1" onerror="this.style.display='none'">
+                {% endif %}
+
+                <div class="flex-grow-1">
+                  <a class="fw-semibold" href="{{ item.link }}" target="_blank" rel="noopener noreferrer">{{ item.title }}</a>
+
+                  {% if r.snippet %}
+                    <small class="text-muted d-block">{{ r.snippet }}</small>
+                  {% endif %}
+
+                  <small class="text-break text-secondary">{{ r.link }}</small>
+                </div>
+              </div>
+            </li>
+          {% endfor %}
+
+        </ol>
+      {% else %}
+        <div class="alert alert-info" role="alert">
+          Žádné položky k zobrazení.
+        </div>
+      {% endif %}
+
+      <script id="json-data" type="application/json">{{ results|tojson }}</script>
+    </main>
+
+    <script src="https://cdn.jsdelivr.net/npm/bootstrap@5.3.3/dist/js/bootstrap.bundle.min.js" integrity="sha384-YvpcrYf0tY3lHB60NNkmXc5s9fDVZLESaAA55NDzOxhy9GkcIdslK1eN7N6jIeHz" crossorigin="anonymous"></script>
+  </body>
+</html>

tests/test_scraper.py

@@ -0,0 +1,78 @@
+import os
+import json
+import types
+import pytest
+
+import scraper
+
+
+def env_setup_test(monkeypatch):
+    """Set required env vars for the API client (helper used by tests)."""
+    monkeypatch.setenv("GOOGLE_DEVELOPER_KEY", "test-key")
+    monkeypatch.setenv("GOOGLE_CSE_ID", "test-cse")
+
+
+def mock_api_response_test(monkeypatch):
+    """Mock out requests.get to return a fixed API payload (helper)."""
+    class FakeResp:
+        status_code = 200
+        def __init__(self, data):
+            self._data = data
+            self.text = json.dumps(data)
+        def json(self):
+            return self._data
+
+    data = {
+        "items": [
+            {
+                "title": "Example Domain",
+                "link": "https://example.com/",
+                "snippet": "This domain is for use in illustrative examples.",
+            },
+            {
+                "title": "OpenAI",
+                "link": "https://openai.com/research",
+                "snippet": "Research from OpenAI.",
+            },
+        ]
+    }
+
+    def fake_get(url, params=None, timeout=15):
+        return FakeResp(data)
+
+    monkeypatch.setattr(scraper, "requests", types.SimpleNamespace(get=fake_get))
+
+
+def test_get_google_first_page_maps_output(monkeypatch):
+    env_setup_test(monkeypatch)
+    mock_api_response_test(monkeypatch)
+    # Avoid waiting for the throttle in tests
+    monkeypatch.setattr(scraper, "RATE_SECONDS", 0)
+    monkeypatch.setattr(scraper, "last_api_call", 0)
+
+    results = scraper.get_google_first_page("example query")
+
+    assert isinstance(results, list)
+    assert len(results) == 2
+
+    first = results[0]
+    assert first["position"] == 1
+    assert first["title"] == "Example Domain"
+    assert first["link"] == "https://example.com/"
+    assert first["snippet"].startswith("This domain is for use")
+    assert first["icon"] == "https://example.com/favicon.ico"
+
+    second = results[1]
+    assert second["position"] == 2
+    assert second["title"] == "OpenAI"
+    assert second["link"] == "https://openai.com/research"
+    assert second["icon"] == "https://openai.com/favicon.ico"
+
+
+def test_missing_env_raises(monkeypatch):
+    # Unset env to simulate missing configuration
+    monkeypatch.delenv("GOOGLE_DEVELOPER_KEY", raising=False)
+    monkeypatch.delenv("GOOGLE_CSE_ID", raising=False)
+
+    with pytest.raises(RuntimeError):
+        scraper.get_google_first_page("anything")