Fun/bettergi-scripts-list
1
0
Fork 0
mirror of https://github.com/babalae/bettergi-scripts-list.git synced 2026-03-30 05:49:51 +08:00
Code Issues Packages Projects Releases Wiki Activity

js：采集cd管理 (#2526)

Browse Source 
修正分均效率计算
This commit is contained in:
mno
2025-12-19 09:15:28 +08:00
committed by GitHub
parent 33d7dbca7c
commit 1219c3f85c
2 changed files with 34 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files

62
repo/js/采集cd管理/main.js
View File

@@ -931,8 +931,6 @@ let underWater = false;
await genshin.returnMainUi();
try {
const filePaths = groupFiles.map(f => f.fullPath);
/* ================== 提前计算分均效率（所有模式通用） ================== */
// 0) 解析优先关键词
const priorityKeywords = settings.priorityTags
@@ -955,8 +953,8 @@ let underWater = false;
}
// 2) 先计算一次基础效率（未知路线先标 -1）
filePaths.forEach(p => {
const fullName = basename(p);
groupFiles.forEach(p => {
const fullName = basename(p.fullPath);
const obj = cdMap.get(fullName);
let avgEff = -1; // 先标记为“未知”
@@ -973,35 +971,40 @@ let underWater = false;
p._efficiency = avgEff; // 已知路线存真实效率，未知路线存 -1
});
// 3) 计算默认效率（分位值）
const knownEff = filePaths
// 3) 计算默认效率（分位值 & 临界值 取最大）
const knownEff = groupFiles
.map(p => p._efficiency)
.filter(e => e >= 0) // 只保留已知路线
.filter(e => e >= 0) // 只保留已知路线
.sort((a, b) => a - b);
const userThreshold = Number(settings[`pathGroup${i}thresholdEfficiency`]) || 0;
let defaultEff;
if (knownEff.length === 0) {
// 一条已知路线都没有 → 回退到老逻辑
defaultEff = Number(settings[`pathGroup${i}thresholdEfficiency`]) || 0;
// 一条已知路线都没有 → 直接用临界值
defaultEff = userThreshold;
} else {
// 按配置的分位取默认效率
// 按配置的分位取效率
const rawPct = settings.defaultEffPercentile;
const pct = Math.max(0, Math.min(1, rawPct === "" ? 0.5 : Number(rawPct)));
const idx = Math.ceil(pct * knownEff.length) - 1;
defaultEff = knownEff[Math.max(0, idx)];
const percentileEff = knownEff[Math.max(0, idx)];
// 关键改动：与临界值取最大
defaultEff = Math.max(percentileEff, userThreshold);
}
// 4) 把 -1 的未知路线替换成默认效率
filePaths.forEach(p => {
groupFiles.forEach(p => {
if (p._efficiency === -1) p._efficiency = defaultEff;
});
// 5) 计算全局最大效率值（已含默认效率）
const maxEff = Math.max(...filePaths.map(p => p._efficiency), 0);
const maxEff = Math.max(...groupFiles.map(p => p._efficiency), 0);
// 6) 优先关键词加分（逻辑不变）
filePaths.forEach(p => {
const fullName = basename(p);
groupFiles.forEach(p => {
const fullName = basename(p.fullPath);
const obj = cdMap.get(fullName);
const itemHit = obj?.history?.some(log =>
@@ -1009,7 +1012,7 @@ let underWater = false;
priorityKeywords.some(key => item.includes(key))
)
);
const pathHit = priorityKeywords.some(key => p.includes(key));
const pathHit = priorityKeywords.some(key => p.fullPath.includes(key));
const descHit = priorityKeywords.some(key => (p.description || '').includes(key));
if (itemHit || pathHit || descHit) {
@@ -1020,9 +1023,9 @@ let underWater = false;
/* ================== 排序分支 ================== */
switch (settings.sortMode) {
case "优先最早刷新，将优先执行最早刷新的路线":
filePaths.sort((a, b) => {
const nameA = basename(a);
const nameB = basename(b);
groupFiles.sort((a, b) => {
const nameA = basename(a.fullPath);
const nameB = basename(b.fullPath);
const timeA = cdMap.has(nameA) ? new Date(cdMap.get(nameA).cdTime) : new Date(0);
const timeB = cdMap.has(nameB) ? new Date(cdMap.get(nameB).cdTime) : new Date(0);
return timeA - timeB; // 越早刷新越靠前
@@ -1031,7 +1034,7 @@ let underWater = false;
case "优先最高效率，将优先执行最高分均拾取物的路线":
// 直接复用提前算好的 _efficiency
filePaths.sort((a, b) => (b._efficiency || 0) - (a._efficiency || 0));
groupFiles.sort((a, b) => (b._efficiency || 0) - (a._efficiency || 0));
break;
default:
@@ -1039,8 +1042,8 @@ let underWater = false;
break;
}
for (const filePath of filePaths) {
const fileName = basename(filePath).replace('.json', '');
for (const filePath of groupFiles) {
const fileName = basename(filePath.fullPath).replace('.json', '');
const fullName = fileName + '.json';
const targetObj = cdMap.get(fullName);
const nextCD = targetObj ? new Date(targetObj.cdTime) : new Date(0);
@@ -1054,8 +1057,8 @@ let underWater = false;
let doSkip = false;
for (const kw of disableArray) {
if (filePath.includes(kw)) {
log.info(`路径文件 ${filePath} 包含禁用关键词 "${kw}"，跳过任务 ${fileName}`);
if (filePath.fullPath.includes(kw)) {
log.info(`路径文件 ${filePath.fullPath} 包含禁用关键词 "${kw}"，跳过任务 ${fileName}`);
doSkip = true; break;
}
}
@@ -1116,19 +1119,20 @@ let underWater = false;
const pickupTask = recognizeAndInteract();
runnedAnyPath = true;
log.info(`当前进度：路径组${i} ${folder} ${fileName} 为第 ${filePaths.indexOf(filePath) + 1}/${filePaths.length} 个`);
if (!underWater && filePath.includes('枫丹水下')) {
log.info(`当前进度：路径组${i} ${folder} ${fileName} 为第 ${groupFiles.indexOf(filePath) + 1}/${groupFiles.length} 个`);
log.info(`当前路线分均效率为 ${(filePath._efficiency ?? 0).toFixed(2)}`);
if (!underWater && filePath.fullPath.includes('枫丹水下')) {
await pathingScript.runFile("assets/A00-塞洛海原（学习螃蟹技能）.json");
underWater = true;
}
if (underWater && !filePath.includes('枫丹水下')) {
if (underWater && !filePath.fullPath.includes('枫丹水下')) {
underWater = false;
}
try {
state.runPickupLog = []; // 新路线开始前清空
await pathingScript.runFile(filePath);
await pathingScript.runFile(filePath.fullPath);
} catch (error) {
log.error(`路径文件 ${filePath} 不存在或执行失败: ${error}`);
log.error(`路径文件 ${filePath.fullPath} 不存在或执行失败: ${error}`);
continue;
}

2
repo/js/采集cd管理/manifest.json
View File

@@ -1,7 +1,7 @@
{
"manifest_version": 1,
"name": "采集cd管理",
"version": "2.6.3",
"version": "2.6.6",
"bgi_version": "0.44.8",
"description": "仅面对会操作文件和读readme的用户，基于文件夹操作自动管理采集路线的cd，会按照路径组的顺序依次运行，直到指定的时间，并会按照给定的cd类型，自动跳过未刷新的路线",
"saved_files": [