README.md

@@ -161,6 +161,52 @@ A short summary in our own words; it is not legal advice.
 - **Commercial use** of scraped data invites expensive disputes even where the scraping itself was allowed.
 - **Be a good guest:** prefer an official API when there is one, read the site's `robots.txt` to see what it allows, and send at most one request a minute. This project fetches one page per run and identifies itself with a `User-Agent`.
 
+### Project: 100 movies to watch
+
+The assignment: scrape Empire's "100 Greatest Movies" and write `movies.txt` with the titles numbered from `1)` upwards. The code is `infrastructure/empire_movies.py` (parsing), `domain/movie.py` (reverse and number) and `infrastructure/movie_file.py` (writing). `--source wayback` (default) reads the Internet Archive snapshot of May 2020 that the assignment pins, `--source live` reads the current Empire page and `--source fixture` reads the saved headings in `tests/fixtures/`.
+
+**1. Choose the page** (the saved headings of the page)
+
+![Step 1: the page](docs/images/movies-1-page.png)
+
+**2. Inspect the markup** (the 2020 page has `<h3 class="title">`; the current page has `<h2><strong>` headings that start with a rank)
+
+```bash
+python -m scrape_movies movies markup
+```
+
+![Step 2: markup](docs/images/movies-2-markup.png)
+
+**3. Titles in page order** (`select("h3.title")` and a list comprehension over `get_text()`; the page counts down from 100)
+
+```bash
+python -m scrape_movies movies titles
+```
+
+![Step 3: titles](docs/images/movies-3-titles.png)
+
+**4. Reverse the list** (`titles[::-1]`, so the list counts up from 1)
+
+```bash
+python -m scrape_movies movies reversed
+```
+
+![Step 4: reversed](docs/images/movies-4-reversed.png)
+
+**5. Write `movies.txt`** (a `with open(...)` block, one title per line)
+
+```bash
+python -m scrape_movies movies write
+```
+
+![Step 5: write](docs/images/movies-5-write.png)
+
+**6. The result**
+
+![Step 6: movies.txt](docs/images/movies-6-file.png)
+
+The committed [`movies.txt`](./movies.txt) was produced with `--source fixture` on 2026-10-28 from the current Empire list, so its ranking differs from the 2020 list in the lecture video. The Internet Archive answered `429 Too Many Requests` while this project was built, so the snapshot is read live only; its `h3.title` layout is covered by a hand-written sample in the tests.
+
 ## 🧪 Tests
 
 ```bash

tools/capture_page.py

@@ -1,56 +1,59 @@
-"""@brief Save a screenshot of a web page with a headless Chromium browser.
+"""@brief Save a screenshot of a web page with a headless Chromium browser.
-
+
-A headless browser draws into memory, so no desktop window is opened and no
+A headless browser draws into memory, so no desktop window is opened and no
-mouse or keyboard is used. It needs Edge, Chrome or Chromium on the machine.
+mouse or keyboard is used. It needs Edge, Chrome or Chromium on the machine.
-"""
+"""
-
+
-import shutil
+import shutil
-import subprocess
+import subprocess
-from pathlib import Path
+from pathlib import Path
-
+
-CANDIDATES = (
+CANDIDATES = (
-    r"C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe",
+    r"C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe",
-    r"C:\Program Files\Microsoft\Edge\Application\msedge.exe",
+    r"C:\Program Files\Microsoft\Edge\Application\msedge.exe",
-    r"C:\Program Files\Google\Chrome\Application\chrome.exe",
+    r"C:\Program Files\Google\Chrome\Application\chrome.exe",
-    "google-chrome",
+    "google-chrome",
-    "chromium",
+    "chromium",
-    "chromium-browser",
+    "chromium-browser",
-)
+)
-WINDOW = "1280,1000"
+WINDOW = "1280,1000"
-
+
-
+
-def find_browser() -> str:
+def find_browser() -> str:
-    """@brief Locate a Chromium-based browser.
+    """@brief Locate a Chromium-based browser.
-
+
-    @return The path or command name of the browser.
+    @return The path or command name of the browser.
-    @throws FileNotFoundError When none is installed.
+    @throws FileNotFoundError When none is installed.
-    """
+    """
-    for candidate in CANDIDATES:
+    for candidate in CANDIDATES:
-        if Path(candidate).exists() or shutil.which(candidate):
+        if Path(candidate).exists() or shutil.which(candidate):
-            return candidate
+            return candidate
-    raise FileNotFoundError("Install Edge, Chrome or Chromium to capture pages.")
+    raise FileNotFoundError("Install Edge, Chrome or Chromium to capture pages.")
-
+
-
+
-def capture_page(url: str, target: Path) -> Path:
+def capture_page(url: str, target: Path) -> Path:
-    """@brief Render a page and save it as a PNG.
+    """@brief Render a page and save it as a PNG.
-
+
-    @param url The page address.
+    @param url The page address.
-    @param target The PNG file to write.
+    @param target The PNG file to write.
-    @return The path of the PNG.
+    @return The path of the PNG.
-    """
+    """
-    target.parent.mkdir(parents=True, exist_ok=True)
+    target.parent.mkdir(parents=True, exist_ok=True)
-    subprocess.run(
+    browser = find_browser()
-        [
+    for mode in ("new", "old"):
-            find_browser(),
+        result = subprocess.run(
-            "--headless=new",
+            [
-            "--disable-gpu",
+                browser,
-            "--hide-scrollbars",
+                f"--headless={mode}",
-            f"--window-size={WINDOW}",
+                "--disable-gpu",
-            f"--screenshot={target}",
+                "--hide-scrollbars",
-            url,
+                f"--window-size={WINDOW}",
-        ],
+                f"--screenshot={target}",
-        check=True,
+                url,
-        capture_output=True,
+            ],
-        timeout=120,
+            capture_output=True,
-    )
+            timeout=120,
-    return target
+        )
+        if result.returncode == 0 and target.exists():
+            return target
+    raise RuntimeError(f"The browser could not capture {url}")

tools/make_screenshots.py

@@ -29,10 +29,20 @@ STEPS: dict[str, list[str]] = {
     "hn-4-links": ["hacker-news", "links"],
     "hn-5-upvotes": ["hacker-news", "upvotes"],
     "hn-6-top": ["hacker-news", "top"],
+    "movies-2-markup": ["movies", "markup", "--source", "fixture"],
+    "movies-3-titles": ["movies", "titles", "--source", "fixture"],
+    "movies-4-reversed": ["movies", "reversed", "--source", "fixture"],
+    "movies-5-write": ["movies", "write", "--source", "fixture"],
 }
+## Screenshots of a file's first lines: image name, file and number of lines.
+FILES: dict[str, tuple[str, int]] = {"movies-6-file": ("movies.txt", 15)}
 ## Screenshots of real web pages: image name and the address of the page.
 PAGES: dict[str, str] = {
     "hn-1-page": "https://appbrewery.github.io/news.ycombinator.com/",
+    # The live page opens a consent wall, so the saved headings are drawn instead.
+    "movies-1-page": (
+        ROOT / "tests" / "fixtures" / "empire_live_excerpt.html"
+    ).as_uri(),
 }
 
 
@@ -63,6 +73,13 @@ def main() -> None:
     for name, args in STEPS.items():
         if name.startswith(prefix):
             print(capture(name, args))
+    for name, (file_name, count) in FILES.items():
+        if name.startswith(prefix):
+            text = (ROOT / file_name).read_text(encoding="utf-8")
+            target = IMAGES / f"{name}.png"
+            shown = os.linesep.join(text.splitlines()[:count])
+            render(f"head -n {count} {file_name}", shown, target)
+            print(target)
     for name, url in PAGES.items():
         if name.startswith(prefix):
             print(capture_page(url, IMAGES / f"{name}.png"))