mirror of
https://github.com/zkldi/Tachi.git
synced 2026-10-12 01:08:29 +03:00
feat: IIDX SP12 (and 11 sorta) tierlist update rerunner
This commit is contained in:
GRIM657
@@ -0,0 +1,366 @@
|
||||
import { log } from "#log";
|
||||
import * as cheerio from "cheerio";
|
||||
import { type AnyNode } from "domhandler";
|
||||
import { type SEEDS_ChartDocument, type SEEDS_SongDocument } from "tachi-common";
|
||||
import { ReadCollection, WriteCollection } from "../../util";
|
||||
|
||||
if (require.main !== module) {
|
||||
throw new Error(`This is a script. It should be ran directly from the command line with bun.`);
|
||||
}
|
||||
|
||||
const URL_L12_BASE =
|
||||
"https://docs.google.com/spreadsheets/u/0/d/e/2PACX-1vSUdp6iuEzE8Z5AL1hkoxzLexp89nJnLQMmICm6_MC0_UjCp1ImZFzabcZkvCpK7mcWvm_2t6iYoJRg/pubhtml/sheet?headers=false&gid=";
|
||||
const URL_L11_BASE =
|
||||
"https://docs.google.com/spreadsheets/u/0/d/e/2PACX-1vSLnjRLdYQ53QgdmKoCvW9zVSgHjBCcGf8iTeEgjiSia4i_4YwR9isSfhxC2kSAq_wWygd_9xGORhk0/pubhtml/sheet?headers=false&gid=";
|
||||
|
||||
const TIER_CONFIGS = [
|
||||
{
|
||||
url: URL_L12_BASE + "1184656976",
|
||||
level: 12,
|
||||
gauge: "normal",
|
||||
format: "columnwise",
|
||||
},
|
||||
{
|
||||
url: URL_L12_BASE + "1277599511",
|
||||
level: 12,
|
||||
gauge: "hard",
|
||||
format: "columnwise",
|
||||
},
|
||||
// The 11 sheet is out of date - trying to find a better one.
|
||||
// {
|
||||
// url: URL_L11_BASE + "411493762",
|
||||
// level: 11,
|
||||
// gauge: "normal",
|
||||
// format: "rowwise",
|
||||
// },
|
||||
// {
|
||||
// url: URL_L11_BASE + "1585306050",
|
||||
// level: 11,
|
||||
// gauge: "hard",
|
||||
// format: "rowwise",
|
||||
// },
|
||||
];
|
||||
|
||||
type songInTier = {
|
||||
title: string;
|
||||
difficulty: string;
|
||||
kojinsa: boolean;
|
||||
};
|
||||
|
||||
const KOJINSA_CLASS = "s5";
|
||||
|
||||
function editDistance(a: string, b: string) {
|
||||
const matrix = [];
|
||||
for (let i = 0; i <= b.length; i++) {
|
||||
matrix[i] = [i];
|
||||
}
|
||||
for (let j = 0; j <= a.length; j++) {
|
||||
matrix[0][j] = j;
|
||||
}
|
||||
for (let i = 1; i <= b.length; i++) {
|
||||
for (let j = 1; j <= a.length; j++) {
|
||||
if (b.charAt(i - 1) === a.charAt(j - 1)) {
|
||||
matrix[i][j] = matrix[i - 1][j - 1];
|
||||
} else {
|
||||
matrix[i][j] = Math.min(
|
||||
matrix[i - 1][j - 1] + 1,
|
||||
matrix[i][j - 1] + 1,
|
||||
matrix[i - 1][j] + 1,
|
||||
);
|
||||
}
|
||||
}
|
||||
}
|
||||
return matrix[b.length][a.length];
|
||||
}
|
||||
|
||||
function parseTitleAndDiff(text: string) {
|
||||
// L12 sheet uses [L]/[H]; L11 sheet uses (L)/(H).
|
||||
if (text.endsWith("[L]") || text.endsWith("(L)")) {
|
||||
return { title: text.slice(0, -3).trim(), difficulty: "LEGGENDARIA" };
|
||||
}
|
||||
if (text.endsWith("[H]") || text.endsWith("(H)")) {
|
||||
return { title: text.slice(0, -3).trim(), difficulty: "HYPER" };
|
||||
}
|
||||
return { title: text.trim(), difficulty: "ANOTHER" };
|
||||
}
|
||||
|
||||
// NFKC folds fullwidth ASCII/digits to halfwidth (１ -> 1, ＋ -> +); the regex
|
||||
// passes then collapse curly quotes which NFKC leaves as stylistic variants.
|
||||
function normalize(s: string) {
|
||||
return s.normalize("NFKC").replace(/[“”]/g, '"').replace(/[‘’]/g, "'");
|
||||
}
|
||||
|
||||
function findMatchingSong(title: string, songsData: SEEDS_SongDocument<"iidx">[]) {
|
||||
const waveDashed = title.replaceAll("～", "〜");
|
||||
const normalizedClean = normalize(waveDashed);
|
||||
|
||||
for (const song of songsData) {
|
||||
if (normalize(song.title) === normalizedClean) {
|
||||
return song;
|
||||
}
|
||||
}
|
||||
|
||||
const lowerTitle = normalizedClean.toLowerCase();
|
||||
let bestMatch = null;
|
||||
let bestDistance = Infinity;
|
||||
const maxDistance = 4;
|
||||
|
||||
for (const song of songsData) {
|
||||
const distance = editDistance(lowerTitle, normalize(song.title).toLowerCase().trim());
|
||||
if (distance <= maxDistance && distance < bestDistance) {
|
||||
bestDistance = distance;
|
||||
bestMatch = song;
|
||||
}
|
||||
}
|
||||
|
||||
if (bestMatch === null) {
|
||||
return null;
|
||||
}
|
||||
if (bestDistance > 0) {
|
||||
log.info(`Fuzzy match: "${title}" -> "${bestMatch.title}" (distance: ${bestDistance})`);
|
||||
}
|
||||
return bestMatch;
|
||||
}
|
||||
|
||||
function isRedSpan(node: AnyNode) {
|
||||
if (node.type !== "tag" || node.name !== "span") {
|
||||
return false;
|
||||
}
|
||||
const style = node.attribs?.style || "";
|
||||
return /color\s*:\s*#ff0000/i.test(style);
|
||||
}
|
||||
|
||||
// Walks a <td>, returning [{title, difficulty, kojinsa}] split on <br>.
|
||||
// Any text encountered inside a red <span> marks its segment kojinsa=true.
|
||||
function parseCell(cellEl: AnyNode) {
|
||||
const titles: songInTier[] = [];
|
||||
let buffer = "";
|
||||
let bufferIsRed = false;
|
||||
|
||||
function flush() {
|
||||
const t = buffer.trim();
|
||||
if (t.length > 0) {
|
||||
const { title, difficulty } = parseTitleAndDiff(t);
|
||||
titles.push({ title: title, difficulty: difficulty, kojinsa: bufferIsRed });
|
||||
}
|
||||
buffer = "";
|
||||
bufferIsRed = false;
|
||||
}
|
||||
|
||||
function walk(nodes: AnyNode[], redContext: boolean) {
|
||||
for (const node of nodes) {
|
||||
if (node.type === "text") {
|
||||
buffer += node.data;
|
||||
if (redContext) {
|
||||
bufferIsRed = true;
|
||||
}
|
||||
continue;
|
||||
}
|
||||
if (node.type !== "tag") {
|
||||
continue;
|
||||
}
|
||||
if (node.name === "br") {
|
||||
flush();
|
||||
continue;
|
||||
}
|
||||
if (isRedSpan(node)) {
|
||||
walk(node.children, true);
|
||||
continue;
|
||||
}
|
||||
walk(node.children, redContext);
|
||||
}
|
||||
}
|
||||
if (cellEl.type === "tag") {
|
||||
walk(cellEl.children, false);
|
||||
}
|
||||
flush();
|
||||
return titles;
|
||||
}
|
||||
|
||||
async function fetchSheet(url: string) {
|
||||
log.info(`Fetching ${url}`);
|
||||
const res = await fetch(url);
|
||||
if (!res.ok) {
|
||||
throw new Error(`HTTP ${res.status} fetching ${url}`);
|
||||
}
|
||||
return res.text();
|
||||
}
|
||||
|
||||
function cleanTierName(text: string) {
|
||||
return text.trim().replaceAll("＋", "+");
|
||||
}
|
||||
|
||||
// L12 layout: thead shim, then one tier-header row + one data row in tbody.
|
||||
// Each data <td> holds <br>-separated titles; red <span>s mark 個人差.
|
||||
function extractTiersColumnwise(html: string) {
|
||||
const $ = cheerio.load(html);
|
||||
const rows = $("tr").toArray();
|
||||
if (rows.length < 3) {
|
||||
throw new Error(`Expected at least 3 rows, got ${rows.length}`);
|
||||
}
|
||||
|
||||
const headerCells = $(rows[1]).find("td").toArray();
|
||||
const tierCells = $(rows[2]).find("td").toArray();
|
||||
|
||||
const colCount = Math.min(headerCells.length, tierCells.length);
|
||||
if (headerCells.length !== tierCells.length) {
|
||||
log.info(
|
||||
`Warning: header has ${headerCells.length} cols, data has ${tierCells.length} cols`,
|
||||
);
|
||||
}
|
||||
|
||||
const tiers = [];
|
||||
for (let k = 0; k < colCount; k++) {
|
||||
const tierName = cleanTierName($(headerCells[k]).text());
|
||||
const songsInTier = parseCell(tierCells[k]);
|
||||
tiers.push({ tier: tierName, songsInTier });
|
||||
}
|
||||
return tiers;
|
||||
}
|
||||
|
||||
// L11 layout: thead shim, then in tbody: title row, tier-header row, then one
|
||||
// song per cell per row. Empty cell = tier ran out of songs. 個人差 cells
|
||||
// carry class "s5" on the <td>.
|
||||
function extractTiersRowwise(html: string) {
|
||||
const $ = cheerio.load(html);
|
||||
const rows = $("tr").toArray();
|
||||
if (rows.length < 4) {
|
||||
throw new Error(`Expected at least 4 rows, got ${rows.length}`);
|
||||
}
|
||||
|
||||
const headerCells = $(rows[2]).find("td").toArray();
|
||||
const tiers = headerCells.map((c) => ({
|
||||
tier: cleanTierName($(c).text()),
|
||||
songsInTier: [] as songInTier[],
|
||||
}));
|
||||
|
||||
for (let k = 3; k < rows.length; k++) {
|
||||
const cells = $(rows[k]).find("td").toArray();
|
||||
const colCount = Math.min(tiers.length, cells.length);
|
||||
for (let col = 0; col < colCount; col++) {
|
||||
const cell = $(cells[col]);
|
||||
const text = cell.text().trim();
|
||||
if (text.length === 0) {
|
||||
continue;
|
||||
}
|
||||
const cls = cell.attr("class") || "";
|
||||
const kojinsa = cls.split(/\s+/).includes(KOJINSA_CLASS);
|
||||
const { title, difficulty } = parseTitleAndDiff(text);
|
||||
tiers[col].songsInTier.push({ title: title, difficulty: difficulty, kojinsa: kojinsa });
|
||||
}
|
||||
}
|
||||
return tiers;
|
||||
}
|
||||
|
||||
function applyTiers(
|
||||
chartsData: SEEDS_ChartDocument<"iidx-sp">[],
|
||||
songsData: SEEDS_SongDocument<"iidx">[],
|
||||
level: number,
|
||||
tiers: {
|
||||
tier: string;
|
||||
songsInTier: songInTier[];
|
||||
}[],
|
||||
field: "hcTier" | "ncTier",
|
||||
) {
|
||||
let updated = 0;
|
||||
let notFound = 0;
|
||||
|
||||
function tierToValue(tier: string) {
|
||||
switch (tier) {
|
||||
case "S+":
|
||||
return level + 1.2;
|
||||
case "S":
|
||||
return level + 1.0;
|
||||
case "A+":
|
||||
return level + 0.9;
|
||||
case "A":
|
||||
return level + 0.8;
|
||||
case "B+":
|
||||
return level + 0.7;
|
||||
case "B":
|
||||
return level + 0.6;
|
||||
case "C":
|
||||
return level + 0.4;
|
||||
case "D":
|
||||
return level + 0.2;
|
||||
case "E":
|
||||
return level + 0.0;
|
||||
case "F":
|
||||
return level - 0.2;
|
||||
default:
|
||||
return level;
|
||||
}
|
||||
}
|
||||
|
||||
function isInAc(versions: string[]) {
|
||||
for (const version of ["33", "32", "31", "30"]) {
|
||||
if (versions.includes(version)) return true;
|
||||
}
|
||||
return false;
|
||||
}
|
||||
|
||||
for (const { tier, songsInTier } of tiers) {
|
||||
if (!tier || tier === "未定" || tier === "超個人差") {
|
||||
continue;
|
||||
}
|
||||
for (const { title, difficulty, kojinsa } of songsInTier) {
|
||||
const songMatch = findMatchingSong(title, songsData);
|
||||
if (!songMatch) {
|
||||
log.warn(`No match: ${title}`);
|
||||
notFound++;
|
||||
continue;
|
||||
}
|
||||
let chartFound = false;
|
||||
for (const chart of chartsData) {
|
||||
if (
|
||||
chart.songID === songMatch.id &&
|
||||
chart.difficulty === difficulty &&
|
||||
chart.data["2dxtraSet"] === null &&
|
||||
isInAc(chart.versions)
|
||||
) {
|
||||
chart.data[field] = {
|
||||
individualDifference: kojinsa,
|
||||
text: level + tier,
|
||||
value: tierToValue(tier),
|
||||
};
|
||||
updated++;
|
||||
chartFound = true;
|
||||
break;
|
||||
}
|
||||
}
|
||||
if (!chartFound) {
|
||||
log.warn(`Chart not found: ${songMatch.title} (SP ${difficulty})`);
|
||||
notFound++;
|
||||
continue;
|
||||
}
|
||||
}
|
||||
}
|
||||
return { updated, notFound };
|
||||
}
|
||||
|
||||
async function main() {
|
||||
const chartsData = ReadCollection("charts-iidx-sp.json");
|
||||
const songsData = ReadCollection("songs-iidx.json");
|
||||
|
||||
for (const config of TIER_CONFIGS) {
|
||||
log.info(`Processing L${config.level} ${config.gauge}`);
|
||||
const html = await fetchSheet(config.url);
|
||||
const tiers =
|
||||
config.format === "rowwise" ? extractTiersRowwise(html) : extractTiersColumnwise(html);
|
||||
log.info(`Parsed ${tiers.length} tiers: ${tiers.map((t) => t.tier).join(", ")}`);
|
||||
const { updated, notFound } = applyTiers(
|
||||
chartsData,
|
||||
songsData,
|
||||
config.level,
|
||||
tiers,
|
||||
config.gauge === "normal" ? "ncTier" : "hcTier",
|
||||
);
|
||||
log.info(`L${config.level} ${config.gauge}: ${updated} updated, ${notFound} not found`);
|
||||
}
|
||||
WriteCollection("charts-iidx-sp.json", chartsData);
|
||||
}
|
||||
|
||||
main().catch((err) => {
|
||||
console.error(`ERROR: ${err.message}`);
|
||||
process.exit(1);
|
||||
});
|
||||
Reference in New Issue
Block a user