clean the repo

1 files changed, 102 insertions, 72 deletions

diff --git a/src/n2000.py b/src/n2000.py
index 357f744..a0741a0 100644
--- a/src/n2000.py
+++ b/src/n2000.py
@@ -1,96 +1,126 @@
+import argparse
+import json
 import re
+import xml.etree.ElementTree as ET
+
 import requests
 
+from kml_utils import add_multi_polygon, add_placemark, new_document, write_kml
+
+MAP_URL = "https://www.natura2000.fr/carte-natura2000"
+POPUP_URL_TEMPLATE = (
+    "https://www.natura2000.fr/leaflet-ajax-popup/node/{entity_id}"
+    "/pop_up_n2000_field/und?_wrapper_format=drupal_ajax"
+)
+SITE_URL_TEMPLATE = "https://www.natura2000.fr/site-natura/{slug}"
 
-reponse = requests.get("https://www.natura2000.fr/carte-natura2000")
-regex = r"\"entity_id\":\"(\d+)\""
-matches = re.findall(regex, reponse.text)
 
+def get_entity_ids(map_url: str = MAP_URL) -> list[str]:
+    """Fetch the Natura 2000 map page and extract every site's numeric entity id.
 
-kml_file = open("natura_2000.kml", "w", encoding="utf-8")
-kml_file.write("""<?xml version="1.0" encoding="UTF-8"?>
-<kml xmlns="http://www.opengis.net/kml/2.2">
-  <Document>
-    <name>Natura 2000</name>
-""")
+    Args:
+        map_url (str, optional): URL of the Natura 2000 map page. Defaults to MAP_URL.
+    Returns:
+        list[str]: Entity ids found on the page.
+    """
+    response = requests.get(map_url)
+    return re.findall(r"\"entity_id\":\"(\d+)\"", response.text)
 
 
-for i, match in enumerate(matches):
-    url = f"https://www.natura2000.fr/leaflet-ajax-popup/node/{match}/pop_up_n2000_field/und?_wrapper_format=drupal_ajax"
-    response = requests.get(url)
+def get_site_slug(entity_id: str) -> str | None:
+    """Resolve an entity id to its site page URL slug via the popup endpoint.
 
-    regex = r"href=\\u0022\\/site-natura\\/(.*?)\\u0022"
-    match = re.search(regex, response.text)
+    Args:
+        entity_id (str): Numeric entity id, as found by get_entity_ids.
+    Returns:
+        str | None: The site URL slug, or None if it could not be found.
+    """
+    response = requests.get(POPUP_URL_TEMPLATE.format(entity_id=entity_id))
+    match = re.search(r"href=\\u0022\\/site-natura\\/(.*?)\\u0022", response.text)
+    return match.group(1) if match else None
 
-    if not match:
-        print(f"No match found for entity_id {match}.")
-        continue
 
-    site_url = f"https://www.natura2000.fr/site-natura/{match.group(1)}"
+def get_site_details(site_url: str) -> tuple[str, list[list[tuple[float, float]]]] | None:
+    """Fetch a site page and extract its name and outer polygon rings.
+
+    Args:
+        site_url (str): URL of the site page.
+    Returns:
+        tuple[str, list[list[tuple[float, float]]]] | None: (name, rings), where each ring is
+        a list of (longitude, latitude) points, or None if the name, geometry type, or
+        coordinates could not be found on the page.
+    """
     response = requests.get(site_url)
 
-    regex = r"<title>(.*?) \| Natura 2000</title>"
-    match = re.search(regex, response.text)
-    if not match:
+    name_match = re.search(r"<title>(.*?) \| Natura 2000</title>", response.text)
+    if not name_match:
         print(f"No name found for {site_url}.")
-        continue
-    name = match.group(1)
+        return None
+    name = name_match.group(1)
 
-    regex = r"\"type\":\"(multipolygon|polygon)\""
-    match = re.search(regex, response.text)
-    if not match:
+    type_match = re.search(r"\"type\":\"(multipolygon|polygon)\"", response.text)
+    if not type_match:
         print(f"No geometry type found for {site_url}.")
-        continue
-    geometry_type = match.group(1)
-    
+        return None
+    geometry_type = type_match.group(1)
+
     if geometry_type == "polygon":
-        regex = r"\"points\":(\[\[.*?\]\])"
-    elif geometry_type == "multipolygon":
-        regex = r"\"points\":(\[\[\[.*?\]\]\])"
+        points_regex = r"\"points\":(\[\[.*?\]\])"
     else:
-        print(f"Unsupported geometry type {geometry_type} for {site_url}.")
-        continue
+        points_regex = r"\"points\":(\[\[\[.*?\]\]\])"
 
-    match = re.search(regex, response.text)
-    if not match:
+    points_match = re.search(points_regex, response.text)
+    if not points_match:
         print(f"No coordinates found for {site_url}.")
-        continue
-    polygons = eval(match.group(1))
-    
+        return None
+
+    polygons = json.loads(points_match.group(1))
     if geometry_type == "polygon":
-        # Wrap single polygon in a list to treat it as multipolygon
-        polygons = [polygons]
+        polygons = [polygons]  # normalize to a list of polygons, like multipolygon
+
+    rings = [[(point["lon"], point["lat"]) for point in polygon[0]] for polygon in polygons]
+    return name, rings
+
+
+def scrape_natura2000(map_url: str = MAP_URL) -> ET.Element:
+    """Scrape natura2000.fr and build a KML tree with one Placemark per site.
+
+    Args:
+        map_url (str, optional): URL of the Natura 2000 map page. Defaults to MAP_URL.
+    Returns:
+        ET.Element: The root <kml> element.
+    """
+    root, document = new_document("Natura 2000")
+
+    entity_ids = get_entity_ids(map_url)
+    for i, entity_id in enumerate(entity_ids):
+        slug = get_site_slug(entity_id)
+        if slug is None:
+            print(f"No site URL found for entity_id {entity_id}.")
+            continue
+
+        site_url = SITE_URL_TEMPLATE.format(slug=slug)
+        details = get_site_details(site_url)
+        if details is None:
+            continue
+        name, rings = details
+
+        placemark = add_placemark(document, name=name, description=f"{name}<br/>{site_url}")
+        add_multi_polygon(placemark, rings)
+
+        print(f"Processed: {name} ({i + 1:5d}/{len(entity_ids)})")
+
+    return root
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Scrape natura2000.fr and export sites to KML")
+    parser.add_argument("-o", "--output", default="natura_2000.kml", help="Path to the output KML file")
+    args = parser.parse_args()
 
-    kml_file.write(f"""
-        <Placemark>
-          <name>{name}</name>
-          <description>
-            {name}<br/>{site_url}
-          </description>
-          <MultiGeometry>
-          """)
+    root = scrape_natura2000()
+    write_kml(root, args.output)
 
-    for polygon in polygons:
-        kml_file.write("""
-          <Polygon>
-            <outerBoundaryIs>
-              <LinearRing>
-                <coordinates>
-        """)
-        for point in polygon[0]:
-            kml_file.write(f"{point['lon']},{point['lat']},0\n")
-        kml_file.write("""
-                    </coordinates>
-                </LinearRing>
-                </outerBoundaryIs>
-            </Polygon>
-            """)
-    kml_file.write(f"""
-        </MultiGeometry>
-    </Placemark>
-    """)
-    
-    print(f"Processed: {name} ({i+1: 5d}/{len(matches)})")
 
-kml_file.write("""  </Document>
-</kml>""")
+if __name__ == "__main__":
+    main()