Julian Tabel
001ee98bb5
Replace all pokebase API calls with local JSON file reads from the PokeAPI/api-data git submodule, removing the network dependency and pokebase package. The script now runs with stdlib Python only. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
492 lines
17 KiB
Python
492 lines
17 KiB
Python
"""Fetch game data from local PokeAPI submodule and write static JSON seed files.
|
|
|
|
Reads from the PokeAPI/api-data git submodule at data/pokeapi/ — no network
|
|
access or container needed. Only uses Python stdlib.
|
|
|
|
Usage:
|
|
python -m app.seeds.fetch_pokeapi
|
|
|
|
Requires the submodule to be initialized:
|
|
git submodule update --init
|
|
"""
|
|
|
|
import json
|
|
import re
|
|
import sys
|
|
from pathlib import Path
|
|
|
|
REPO_ROOT = Path(__file__).parents[4] # backend/src/app/seeds -> repo root
|
|
POKEAPI_DIR = REPO_ROOT / "data" / "pokeapi" / "data" / "api" / "v2"
|
|
DATA_DIR = Path(__file__).parent / "data"
|
|
|
|
|
|
def load_resource(endpoint: str, resource_id: int) -> dict:
|
|
"""Load a PokeAPI resource from the local submodule data."""
|
|
path = POKEAPI_DIR / endpoint / str(resource_id) / "index.json"
|
|
with open(path) as f:
|
|
return json.load(f)
|
|
|
|
|
|
def extract_id(url: str) -> int:
|
|
"""Extract the numeric ID from a PokeAPI URL (absolute or relative)."""
|
|
return int(url.rstrip("/").split("/")[-1])
|
|
|
|
|
|
# Game definitions
|
|
VERSION_GROUPS = {
|
|
"firered-leafgreen": {
|
|
"versions": ["firered", "leafgreen"],
|
|
"generation": 3,
|
|
"region": "kanto",
|
|
"region_id": 1,
|
|
"games": {
|
|
"firered": {
|
|
"name": "Pokemon FireRed",
|
|
"slug": "firered",
|
|
"release_year": 2004,
|
|
},
|
|
"leafgreen": {
|
|
"name": "Pokemon LeafGreen",
|
|
"slug": "leafgreen",
|
|
"release_year": 2004,
|
|
},
|
|
},
|
|
},
|
|
"emerald": {
|
|
"versions": ["emerald"],
|
|
"generation": 3,
|
|
"region": "hoenn",
|
|
"region_id": 3,
|
|
"games": {
|
|
"emerald": {
|
|
"name": "Pokemon Emerald",
|
|
"slug": "emerald",
|
|
"release_year": 2005,
|
|
},
|
|
},
|
|
},
|
|
"heartgold-soulsilver": {
|
|
"versions": ["heartgold", "soulsilver"],
|
|
"generation": 4,
|
|
"region": "johto",
|
|
"region_id": 2,
|
|
"games": {
|
|
"heartgold": {
|
|
"name": "Pokemon HeartGold",
|
|
"slug": "heartgold",
|
|
"release_year": 2010,
|
|
},
|
|
"soulsilver": {
|
|
"name": "Pokemon SoulSilver",
|
|
"slug": "soulsilver",
|
|
"release_year": 2010,
|
|
},
|
|
},
|
|
},
|
|
}
|
|
|
|
# Encounter methods to include (excludes gift, legendary-only, etc.)
|
|
INCLUDED_METHODS = {
|
|
"walk",
|
|
"surf",
|
|
"old-rod",
|
|
"good-rod",
|
|
"super-rod",
|
|
"rock-smash",
|
|
"headbutt",
|
|
}
|
|
|
|
# Collect all pokemon dex numbers across games
|
|
all_pokemon_dex: set[int] = set()
|
|
|
|
|
|
def clean_location_name(name: str) -> str:
|
|
"""Convert PokeAPI location slug to a clean display name.
|
|
|
|
e.g. 'kanto-route-1' -> 'Route 1'
|
|
'pallet-town' -> 'Pallet Town'
|
|
"""
|
|
for prefix in [
|
|
"kanto-", "johto-", "hoenn-", "sinnoh-",
|
|
"unova-", "kalos-", "alola-", "galar-",
|
|
]:
|
|
if name.startswith(prefix):
|
|
name = name[len(prefix):]
|
|
break
|
|
|
|
name = name.replace("-", " ").title()
|
|
name = re.sub(r"Route (\d+)", r"Route \1", name)
|
|
return name
|
|
|
|
|
|
def clean_area_name(area_name: str, location_name: str) -> str | None:
|
|
"""Extract meaningful area suffix, or None if it's the default area."""
|
|
if area_name.startswith(location_name):
|
|
suffix = area_name[len(location_name):].strip("-").strip()
|
|
if not suffix or suffix == "area":
|
|
return None
|
|
return suffix.replace("-", " ").title()
|
|
return area_name.replace("-", " ").title()
|
|
|
|
|
|
def get_encounters_for_area(area_id: int, version_name: str) -> list[dict]:
|
|
"""Get encounter data for a location area, filtered by version."""
|
|
area = load_resource("location-area", area_id)
|
|
encounters = []
|
|
|
|
for pe in area["pokemon_encounters"]:
|
|
pokemon_url = pe["pokemon"]["url"]
|
|
dex_num = extract_id(pokemon_url)
|
|
pokemon_name = pe["pokemon"]["name"]
|
|
|
|
for vd in pe["version_details"]:
|
|
if vd["version"]["name"] != version_name:
|
|
continue
|
|
|
|
for enc in vd["encounter_details"]:
|
|
method = enc["method"]["name"]
|
|
if method not in INCLUDED_METHODS:
|
|
continue
|
|
|
|
encounters.append({
|
|
"pokemon_name": pokemon_name,
|
|
"national_dex": dex_num,
|
|
"method": method,
|
|
"chance": enc["chance"],
|
|
"min_level": enc["min_level"],
|
|
"max_level": enc["max_level"],
|
|
})
|
|
|
|
return encounters
|
|
|
|
|
|
def aggregate_encounters(raw_encounters: list[dict]) -> list[dict]:
|
|
"""Aggregate encounter rates by pokemon + method (sum chances across level ranges)."""
|
|
agg: dict[tuple[int, str], dict] = {}
|
|
|
|
for enc in raw_encounters:
|
|
key = (enc["national_dex"], enc["method"])
|
|
if key not in agg:
|
|
agg[key] = {
|
|
"national_dex": enc["national_dex"],
|
|
"pokemon_name": enc["pokemon_name"],
|
|
"method": enc["method"],
|
|
"encounter_rate": 0,
|
|
"min_level": enc["min_level"],
|
|
"max_level": enc["max_level"],
|
|
}
|
|
agg[key]["encounter_rate"] += enc["chance"]
|
|
agg[key]["min_level"] = min(agg[key]["min_level"], enc["min_level"])
|
|
agg[key]["max_level"] = max(agg[key]["max_level"], enc["max_level"])
|
|
|
|
result = list(agg.values())
|
|
for r in result:
|
|
r["encounter_rate"] = min(r["encounter_rate"], 100)
|
|
|
|
return sorted(result, key=lambda x: (-x["encounter_rate"], x["pokemon_name"]))
|
|
|
|
|
|
def process_version(version_name: str, vg_info: dict) -> list[dict]:
|
|
"""Process all locations for a specific game version."""
|
|
print(f"\n--- Processing {version_name} ---")
|
|
|
|
region = load_resource("region", vg_info["region_id"])
|
|
location_refs = list(region["locations"])
|
|
|
|
# For HGSS, also include Kanto locations
|
|
if version_name in ("heartgold", "soulsilver"):
|
|
kanto = load_resource("region", 1)
|
|
location_refs = location_refs + list(kanto["locations"])
|
|
|
|
print(f" Found {len(location_refs)} locations")
|
|
|
|
routes = []
|
|
order = 1
|
|
|
|
for loc_ref in location_refs:
|
|
loc_name = loc_ref["name"]
|
|
loc_id = extract_id(loc_ref["url"])
|
|
display_name = clean_location_name(loc_name)
|
|
|
|
location = load_resource("location", loc_id)
|
|
areas = location["areas"]
|
|
if not areas:
|
|
continue
|
|
|
|
all_encounters: list[dict] = []
|
|
area_specific: dict[str, list[dict]] = {}
|
|
|
|
for area_ref in areas:
|
|
area_id = extract_id(area_ref["url"])
|
|
area_slug = area_ref["name"]
|
|
area_suffix = clean_area_name(area_slug, loc_name)
|
|
|
|
encounters = get_encounters_for_area(area_id, version_name)
|
|
if not encounters:
|
|
continue
|
|
|
|
if area_suffix and len(areas) > 1:
|
|
area_specific[area_suffix] = encounters
|
|
else:
|
|
all_encounters.extend(encounters)
|
|
|
|
# Area-specific encounters become separate routes
|
|
if area_specific:
|
|
for area_suffix, area_encs in area_specific.items():
|
|
aggregated = aggregate_encounters(area_encs)
|
|
if aggregated:
|
|
route_name = f"{display_name} ({area_suffix})"
|
|
for enc in aggregated:
|
|
all_pokemon_dex.add(enc["national_dex"])
|
|
routes.append({
|
|
"name": route_name,
|
|
"order": order,
|
|
"encounters": aggregated,
|
|
})
|
|
order += 1
|
|
|
|
if all_encounters:
|
|
aggregated = aggregate_encounters(all_encounters)
|
|
if aggregated:
|
|
for enc in aggregated:
|
|
all_pokemon_dex.add(enc["national_dex"])
|
|
routes.append({
|
|
"name": display_name,
|
|
"order": order,
|
|
"encounters": aggregated,
|
|
})
|
|
order += 1
|
|
|
|
print(f" Routes with encounters: {len(routes)}")
|
|
total_enc = sum(len(r["encounters"]) for r in routes)
|
|
print(f" Total encounter entries: {total_enc}")
|
|
|
|
return routes
|
|
|
|
|
|
def fetch_pokemon_data(dex_numbers: set[int]) -> list[dict]:
|
|
"""Fetch Pokemon name/type data for all collected dex numbers."""
|
|
print(f"\n--- Fetching {len(dex_numbers)} Pokemon ---")
|
|
|
|
pokemon_list = []
|
|
dex_sorted = sorted(dex_numbers)
|
|
|
|
for i, dex in enumerate(dex_sorted, 1):
|
|
poke = load_resource("pokemon", dex)
|
|
types = [t["type"]["name"] for t in poke["types"]]
|
|
pokemon_list.append({
|
|
"national_dex": dex,
|
|
"name": poke["name"].title().replace("-", " "),
|
|
"types": types,
|
|
"sprite_url": f"https://raw.githubusercontent.com/PokeAPI/sprites/master/sprites/pokemon/{dex}.png",
|
|
})
|
|
|
|
if i % 50 == 0 or i == len(dex_sorted):
|
|
print(f" Fetched {i}/{len(dex_sorted)}")
|
|
|
|
return sorted(pokemon_list, key=lambda x: x["national_dex"])
|
|
|
|
|
|
def flatten_evolution_chain(chain: dict, seeded_dex: set[int]) -> list[dict]:
|
|
"""Recursively flatten a PokeAPI evolution chain into (from, to) pairs."""
|
|
pairs = []
|
|
from_dex = int(chain["species"]["url"].rstrip("/").split("/")[-1])
|
|
|
|
for evo in chain.get("evolves_to", []):
|
|
to_dex = int(evo["species"]["url"].rstrip("/").split("/")[-1])
|
|
|
|
for detail in evo["evolution_details"]:
|
|
trigger = detail["trigger"]["name"]
|
|
min_level = detail.get("min_level")
|
|
item = detail.get("item")
|
|
if item:
|
|
item = item["name"]
|
|
held_item = detail.get("held_item")
|
|
if held_item:
|
|
held_item = held_item["name"]
|
|
|
|
# Collect other conditions as a string
|
|
conditions = []
|
|
if detail.get("min_happiness"):
|
|
conditions.append(f"happiness >= {detail['min_happiness']}")
|
|
if detail.get("min_affection"):
|
|
conditions.append(f"affection >= {detail['min_affection']}")
|
|
if detail.get("min_beauty"):
|
|
conditions.append(f"beauty >= {detail['min_beauty']}")
|
|
if detail.get("time_of_day"):
|
|
conditions.append(detail["time_of_day"])
|
|
if detail.get("known_move"):
|
|
conditions.append(f"knows {detail['known_move']['name']}")
|
|
if detail.get("known_move_type"):
|
|
conditions.append(f"knows {detail['known_move_type']['name']}-type move")
|
|
if detail.get("location"):
|
|
conditions.append(f"at {detail['location']['name']}")
|
|
if detail.get("party_species"):
|
|
conditions.append(f"with {detail['party_species']['name']} in party")
|
|
if detail.get("party_type"):
|
|
conditions.append(f"with {detail['party_type']['name']}-type in party")
|
|
if detail.get("gender") is not None:
|
|
conditions.append("female" if detail["gender"] == 1 else "male")
|
|
if detail.get("needs_overworld_rain"):
|
|
conditions.append("raining")
|
|
if detail.get("turn_upside_down"):
|
|
conditions.append("turn upside down")
|
|
if detail.get("trade_species"):
|
|
conditions.append(f"trade for {detail['trade_species']['name']}")
|
|
if detail.get("relative_physical_stats") is not None:
|
|
stat_map = {1: "atk > def", -1: "atk < def", 0: "atk = def"}
|
|
conditions.append(stat_map.get(detail["relative_physical_stats"], ""))
|
|
|
|
condition = ", ".join(conditions) if conditions else None
|
|
|
|
if from_dex in seeded_dex and to_dex in seeded_dex:
|
|
pairs.append({
|
|
"from_national_dex": from_dex,
|
|
"to_national_dex": to_dex,
|
|
"trigger": trigger,
|
|
"min_level": min_level,
|
|
"item": item,
|
|
"held_item": held_item,
|
|
"condition": condition,
|
|
})
|
|
|
|
# Recurse into further evolutions
|
|
pairs.extend(flatten_evolution_chain(evo, seeded_dex))
|
|
|
|
return pairs
|
|
|
|
|
|
def fetch_evolution_data(seeded_dex: set[int]) -> list[dict]:
|
|
"""Fetch evolution chains from local PokeAPI data for all seeded pokemon."""
|
|
print(f"\n--- Fetching evolution chains ---")
|
|
|
|
# First, get the evolution chain URL for each pokemon species
|
|
chain_ids: set[int] = set()
|
|
dex_sorted = sorted(seeded_dex)
|
|
|
|
for i, dex in enumerate(dex_sorted, 1):
|
|
species = load_resource("pokemon-species", dex)
|
|
chain_url = species["evolution_chain"]["url"]
|
|
chain_id = extract_id(chain_url)
|
|
chain_ids.add(chain_id)
|
|
if i % 50 == 0 or i == len(dex_sorted):
|
|
print(f" Species fetched: {i}/{len(dex_sorted)}")
|
|
|
|
print(f" Found {len(chain_ids)} unique evolution chains")
|
|
|
|
# Fetch each chain and flatten
|
|
all_pairs: list[dict] = []
|
|
seen: set[tuple[int, int, str]] = set()
|
|
|
|
for chain_id in sorted(chain_ids):
|
|
chain = load_resource("evolution-chain", chain_id)
|
|
pairs = flatten_evolution_chain(chain["chain"], seeded_dex)
|
|
for p in pairs:
|
|
key = (p["from_national_dex"], p["to_national_dex"], p["trigger"])
|
|
if key not in seen:
|
|
seen.add(key)
|
|
all_pairs.append(p)
|
|
|
|
print(f" Total evolution pairs: {len(all_pairs)}")
|
|
return sorted(all_pairs, key=lambda x: (x["from_national_dex"], x["to_national_dex"]))
|
|
|
|
|
|
def apply_evolution_overrides(evolutions: list[dict]) -> None:
|
|
"""Apply overrides from evolution_overrides.json if it exists."""
|
|
overrides_path = DATA_DIR / "evolution_overrides.json"
|
|
if not overrides_path.exists():
|
|
return
|
|
|
|
with open(overrides_path) as f:
|
|
overrides = json.load(f)
|
|
|
|
# Remove entries
|
|
for removal in overrides.get("remove", []):
|
|
evolutions[:] = [
|
|
e for e in evolutions
|
|
if not (e["from_national_dex"] == removal["from_dex"]
|
|
and e["to_national_dex"] == removal["to_dex"])
|
|
]
|
|
|
|
# Add entries
|
|
for addition in overrides.get("add", []):
|
|
evolutions.append({
|
|
"from_national_dex": addition["from_dex"],
|
|
"to_national_dex": addition["to_dex"],
|
|
"trigger": addition.get("trigger", "level-up"),
|
|
"min_level": addition.get("min_level"),
|
|
"item": addition.get("item"),
|
|
"held_item": addition.get("held_item"),
|
|
"condition": addition.get("condition"),
|
|
})
|
|
|
|
# Modify entries
|
|
for mod in overrides.get("modify", []):
|
|
for e in evolutions:
|
|
if (e["from_national_dex"] == mod["from_dex"]
|
|
and e["to_national_dex"] == mod["to_dex"]):
|
|
for key, value in mod.get("set", {}).items():
|
|
e[key] = value
|
|
|
|
# Re-sort
|
|
evolutions.sort(key=lambda x: (x["from_national_dex"], x["to_national_dex"]))
|
|
print(f" Applied overrides: {len(evolutions)} pairs after overrides")
|
|
|
|
|
|
def write_json(filename: str, data):
|
|
path = DATA_DIR / filename
|
|
with open(path, "w") as f:
|
|
json.dump(data, f, indent=2)
|
|
print(f" -> {path}")
|
|
|
|
|
|
def main():
|
|
if not POKEAPI_DIR.is_dir():
|
|
print(
|
|
f"Error: PokeAPI data not found at {POKEAPI_DIR}\n"
|
|
"Initialize the submodule with: git submodule update --init",
|
|
file=sys.stderr,
|
|
)
|
|
sys.exit(1)
|
|
|
|
DATA_DIR.mkdir(parents=True, exist_ok=True)
|
|
|
|
# Build games.json
|
|
games = []
|
|
for vg_info in VERSION_GROUPS.values():
|
|
for game_info in vg_info["games"].values():
|
|
games.append({
|
|
"name": game_info["name"],
|
|
"slug": game_info["slug"],
|
|
"generation": vg_info["generation"],
|
|
"region": vg_info["region"],
|
|
"release_year": game_info["release_year"],
|
|
})
|
|
|
|
write_json("games.json", games)
|
|
print(f"Wrote {len(games)} games to games.json")
|
|
|
|
# Process each version
|
|
for vg_info in VERSION_GROUPS.values():
|
|
for ver_name in vg_info["versions"]:
|
|
routes = process_version(ver_name, vg_info)
|
|
write_json(f"{ver_name}.json", routes)
|
|
|
|
# Fetch all Pokemon data
|
|
pokemon = fetch_pokemon_data(all_pokemon_dex)
|
|
write_json("pokemon.json", pokemon)
|
|
print(f"\nWrote {len(pokemon)} Pokemon to pokemon.json")
|
|
|
|
# Fetch evolution chains
|
|
evolutions = fetch_evolution_data(all_pokemon_dex)
|
|
apply_evolution_overrides(evolutions)
|
|
write_json("evolutions.json", evolutions)
|
|
print(f"\nWrote {len(evolutions)} evolution pairs to evolutions.json")
|
|
|
|
print("\nDone! JSON files written to seeds/data/")
|
|
print("Review route ordering and curate as needed.")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|