mirror of https://github.com/jackwener/wx-cli.git
feat(sns): export linux moments images via CDN
zzgz
parent
8b3f63deea
commit
c203ae6ad3
|
|
@ -215,14 +215,17 @@ wx sns-search "婚礼" --user "李四" --since 2023-01-01
|
|||
|
||||
朋友圈数据只覆盖你本地刷到过的帖子（微信 app 按需下载）。
|
||||
|
||||
本地已加载的朋友圈图片可从 `cache/<YYYY-MM>/Sns/Img` 解密导出：
|
||||
已加载的朋友圈图片可导出到本地目录：
|
||||
|
||||
```bash
|
||||
wx sns-extract -o ./sns-images --month 2026-06 -n 100 --overwrite
|
||||
wx sns-extract -o ./sns-images --month 2026-06 --xor-key 0xe7 --json
|
||||
```
|
||||
|
||||
Windows 微信 4.x 的朋友圈 V2 缓存图复用聊天图片 AES key，但 SNS 尾部 XOR byte 实测为 `0xe7`。保持微信运行，并先在桌面微信里打开对应朋友圈或大图，命令才能导出本地缓存里已有的图片。
|
||||
- Windows 微信 4.x：从 `cache/<YYYY-MM>/Sns/Img` 解本地 V2 缓存图，复用聊天图片 AES key，SNS 尾部 XOR byte 实测为 `0xe7`。保持微信运行，便于 wx-cli 扫内存提取 V2 AES image key。
|
||||
- Linux/XWeChat：从 `sns.db` 的 media URL 重新下载 CDN 图片，用 `url_key` 经 WxIsaac64 生成密钥流后全量 XOR，并裁到 JPEG EOI。需要系统有 `curl`、`node`，并设置 `WX_SNS_WASM_DIR` 指向包含 `wasm_video_decode.js` 和 `wasm_video_decode.wasm` 的目录。
|
||||
|
||||
命令只处理本地朋友圈数据库已经记录的 media；如果要大图，先在桌面微信里打开对应朋友圈或大图，让 XML 中的 token/key 可用。
|
||||
|
||||
### 公众号文章
|
||||
|
||||
|
|
|
|||
9
SKILL.md
9
SKILL.md
|
|
@ -240,16 +240,19 @@ wx sns-search "婚礼" --user "李四" --since 2023-01-01 -n 50
|
|||
|
||||
> 只保存你本地刷到过的朋友圈（微信 app 按需下载）。没刷到过的帖子不在本地，任何命令都拿不到。
|
||||
|
||||
#### 朋友圈 / SNS 缓存图片导出
|
||||
#### 朋友圈 / SNS 图片导出
|
||||
|
||||
Windows 微信 4.x 已加载过的朋友圈图片会缓存到 `xwechat_files/<wxid>/cache/<YYYY-MM>/Sns/Img`。这类 V2 文件可复用聊天图片的 AES image key provider 解密，但 SNS 尾部 raw 区的 XOR byte 实测为 `0xe7`，不是聊天图片的默认值。
|
||||
`sns-extract` 按平台自动选择导出方式：
|
||||
|
||||
- Windows 微信 4.x：已加载过的朋友圈图片会缓存到 `xwechat_files/<wxid>/cache/<YYYY-MM>/Sns/Img`。这类 V2 文件可复用聊天图片的 AES image key provider 解密，但 SNS 尾部 raw 区的 XOR byte 实测为 `0xe7`，不是聊天图片的默认值。
|
||||
- Linux/XWeChat：从 `sns.db` 里的 media URL 重新下载 CDN 图片，用 `url_key` 经 WxIsaac64 生成密钥流后全量 XOR，并裁到 JPEG EOI。需要 `curl`、`node`，并设置 `WX_SNS_WASM_DIR` 指向包含 `wasm_video_decode.js` 和 `wasm_video_decode.wasm` 的目录。
|
||||
|
||||
```bash
|
||||
wx sns-extract -o ./sns-images --month 2026-06 -n 100 --overwrite
|
||||
wx sns-extract -o ./sns-images --month 2026-06 --xor-key 0xe7 --json
|
||||
```
|
||||
|
||||
使用时保持 `Weixin.exe` 运行，便于 wx-cli 扫内存提取 V2 AES image key。命令只导出本地缓存里已经加载过的图；如果要大图，先在桌面微信里打开对应朋友圈或大图，再运行导出。若导出的文件花屏或半张图，优先尝试用 `--xor-key` 覆盖 SNS tail XOR byte。
|
||||
Windows 使用时保持 `Weixin.exe` 运行，便于 wx-cli 扫内存提取 V2 AES image key。命令只处理本地朋友圈数据库已经记录的 media；如果要大图，先在桌面微信里打开对应朋友圈或大图。Windows 若导出的文件花屏或半张图，优先尝试用 `--xor-key` 覆盖 SNS tail XOR byte。
|
||||
|
||||
### 公众号文章
|
||||
|
||||
|
|
|
|||
|
|
@ -7,6 +7,8 @@ use serde_json::{json, Value};
|
|||
use std::collections::{HashMap, HashSet};
|
||||
use std::fs;
|
||||
use std::path::{Path, PathBuf};
|
||||
#[cfg(target_os = "linux")]
|
||||
use std::process::Command;
|
||||
use std::sync::{Arc, OnceLock};
|
||||
|
||||
use super::cache::{CacheMode, DbCache};
|
||||
|
|
@ -4609,6 +4611,14 @@ pub async fn q_sns_extract(
|
|||
overwrite: bool,
|
||||
xor_key: u8,
|
||||
) -> Result<Value> {
|
||||
#[cfg(target_os = "linux")]
|
||||
{
|
||||
let _ = xor_key;
|
||||
return q_sns_extract_cdn(db, output_dir, month, limit, overwrite).await;
|
||||
}
|
||||
|
||||
#[cfg(not(target_os = "linux"))]
|
||||
{
|
||||
let db_dir = db.db_dir().to_path_buf();
|
||||
let output_dir = PathBuf::from(output_dir);
|
||||
let month = month.map(str::to_string);
|
||||
|
|
@ -4715,8 +4725,303 @@ pub async fn q_sns_extract(
|
|||
}))
|
||||
})
|
||||
.await?
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(target_os = "linux")]
|
||||
pub async fn q_sns_extract_cdn(
|
||||
db: &DbCache,
|
||||
output_dir: &str,
|
||||
month: Option<&str>,
|
||||
limit: usize,
|
||||
overwrite: bool,
|
||||
) -> Result<Value> {
|
||||
let path = db.get("sns/sns.db").await?.context("无法解密 sns.db")?;
|
||||
let output_dir = PathBuf::from(output_dir);
|
||||
let month = month.map(str::to_string);
|
||||
|
||||
tokio::task::spawn_blocking(move || -> Result<Value> {
|
||||
fs::create_dir_all(&output_dir)
|
||||
.with_context(|| format!("创建输出目录失败: {}", output_dir.display()))?;
|
||||
let tmp_dir = output_dir.join(".wx-sns-tmp");
|
||||
fs::create_dir_all(&tmp_dir)
|
||||
.with_context(|| format!("创建临时目录失败: {}", tmp_dir.display()))?;
|
||||
|
||||
let conn = Connection::open(&path)?;
|
||||
let mut stmt = conn.prepare("SELECT tid, user_name, content FROM SnsTimeLine ORDER BY tid DESC")?;
|
||||
let rows = stmt.query_map([], |row| Ok((
|
||||
row.get::<_, i64>(0)?,
|
||||
row.get::<_, String>(1).unwrap_or_default(),
|
||||
row.get::<_, String>(2).unwrap_or_default(),
|
||||
)))?;
|
||||
|
||||
let max = limit.max(1);
|
||||
let mut exported = Vec::new();
|
||||
let mut failed = Vec::new();
|
||||
let mut skipped_existing = 0usize;
|
||||
let mut skipped_no_key = 0usize;
|
||||
let mut skipped_month = 0usize;
|
||||
let mut scanned_posts = 0usize;
|
||||
let mut scanned_media = 0usize;
|
||||
|
||||
for row in rows {
|
||||
if exported.len() >= max {
|
||||
break;
|
||||
}
|
||||
scanned_posts += 1;
|
||||
if scanned_posts > SNS_MAX_SCAN {
|
||||
eprintln!(
|
||||
"[sns_extract] scan 超过硬上限 {}，结果可能不完整。建议加 --month 缩小范围。",
|
||||
SNS_MAX_SCAN
|
||||
);
|
||||
break;
|
||||
}
|
||||
|
||||
let (tid, uname, content) = row?;
|
||||
let post = parse_post_xml(tid, &uname, &content);
|
||||
if let Some(month) = month.as_deref() {
|
||||
let post_month = fmt_time(post.create_time, "%Y-%m");
|
||||
if post_month != month {
|
||||
skipped_month += post.media.len();
|
||||
continue;
|
||||
}
|
||||
}
|
||||
|
||||
for (idx, media) in post.media.iter().enumerate() {
|
||||
if exported.len() >= max {
|
||||
break;
|
||||
}
|
||||
scanned_media += 1;
|
||||
let Some(obj) = media.as_object() else {
|
||||
skipped_no_key += 1;
|
||||
continue;
|
||||
};
|
||||
let url = obj.get("url").and_then(Value::as_str).unwrap_or_default();
|
||||
let token = obj.get("url_token").and_then(Value::as_str).unwrap_or_default();
|
||||
let key = obj.get("url_key").and_then(Value::as_str).unwrap_or_default();
|
||||
if url.is_empty() || token.is_empty() || key.is_empty() {
|
||||
skipped_no_key += 1;
|
||||
continue;
|
||||
}
|
||||
|
||||
let stem = sns_cdn_output_stem(tid, idx, obj);
|
||||
let out_path = output_dir.join(format!("{stem}.jpg"));
|
||||
if out_path.exists() && !overwrite {
|
||||
skipped_existing += 1;
|
||||
continue;
|
||||
}
|
||||
|
||||
let raw_path = tmp_dir.join(format!("{stem}.enc"));
|
||||
let headers_path = tmp_dir.join(format!("{stem}.headers"));
|
||||
let dl_url = sns_cdn_download_url(url, token);
|
||||
|
||||
let result = (|| -> Result<()> {
|
||||
download_sns_cdn_media(&dl_url, &raw_path, &headers_path)?;
|
||||
decrypt_sns_cdn_media(key, &raw_path, &out_path)?;
|
||||
Ok(())
|
||||
})();
|
||||
|
||||
match result {
|
||||
Ok(()) => {
|
||||
let _ = fs::remove_file(&raw_path);
|
||||
let _ = fs::remove_file(&headers_path);
|
||||
let output_size = fs::metadata(&out_path).map(|m| m.len()).unwrap_or(0);
|
||||
exported.push(json!({
|
||||
"source": dl_url,
|
||||
"output": out_path.display().to_string(),
|
||||
"format": "jpg",
|
||||
"decoder": "sns_cdn_isaac64",
|
||||
"output_size": output_size,
|
||||
"tid": tid,
|
||||
"media_index": idx,
|
||||
"author_username": post.author_username,
|
||||
"time": fmt_time(post.create_time, "%Y-%m-%d %H:%M"),
|
||||
"md5": obj.get("md5").and_then(Value::as_str).unwrap_or_default(),
|
||||
"width": obj.get("width").and_then(Value::as_i64).unwrap_or_default(),
|
||||
"height": obj.get("height").and_then(Value::as_i64).unwrap_or_default(),
|
||||
}));
|
||||
}
|
||||
Err(e) => {
|
||||
failed.push(json!({
|
||||
"url": url,
|
||||
"tid": tid,
|
||||
"media_index": idx,
|
||||
"error": e.to_string(),
|
||||
}));
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
let total_exported = exported.len();
|
||||
let total_failed = failed.len();
|
||||
|
||||
Ok(json!({
|
||||
"output_dir": output_dir.display().to_string(),
|
||||
"month": month,
|
||||
"method": "linux_sns_cdn_isaac64",
|
||||
"exported": exported,
|
||||
"total_exported": total_exported,
|
||||
"skipped_existing": skipped_existing,
|
||||
"skipped_no_key": skipped_no_key,
|
||||
"skipped_month": skipped_month,
|
||||
"scanned_posts": scanned_posts,
|
||||
"scanned_media": scanned_media,
|
||||
"failed": failed,
|
||||
"total_failed": total_failed,
|
||||
}))
|
||||
})
|
||||
.await?
|
||||
}
|
||||
|
||||
#[cfg(target_os = "linux")]
|
||||
fn sns_cdn_download_url(url: &str, token: &str) -> String {
|
||||
let mut out = if let Some(rest) = url.strip_prefix("http://") {
|
||||
format!("https://{rest}")
|
||||
} else {
|
||||
url.to_string()
|
||||
};
|
||||
if out.contains('?') {
|
||||
out.push_str("&token=");
|
||||
} else {
|
||||
out.push_str("?token=");
|
||||
}
|
||||
out.push_str(token);
|
||||
out.push_str("&idx=1");
|
||||
out
|
||||
}
|
||||
|
||||
#[cfg(target_os = "linux")]
|
||||
fn sns_cdn_output_stem(tid: i64, idx: usize, media: &serde_json::Map<String, Value>) -> String {
|
||||
let md5 = media
|
||||
.get("md5")
|
||||
.and_then(Value::as_str)
|
||||
.filter(|v| !v.is_empty())
|
||||
.unwrap_or("no-md5");
|
||||
format!("sns_{}_{}_{}", tid, idx, md5)
|
||||
.chars()
|
||||
.map(|c| if c.is_ascii_alphanumeric() || c == '_' || c == '-' { c } else { '_' })
|
||||
.collect()
|
||||
}
|
||||
|
||||
#[cfg(target_os = "linux")]
|
||||
fn download_sns_cdn_media(url: &str, raw_path: &Path, headers_path: &Path) -> Result<()> {
|
||||
let status = Command::new("curl")
|
||||
.arg("-L")
|
||||
.arg("--fail")
|
||||
.arg("--http1.1")
|
||||
.arg("-A")
|
||||
.arg("MicroMessenger Client")
|
||||
.arg("-H")
|
||||
.arg("Accept: */*")
|
||||
.arg("-D")
|
||||
.arg(headers_path)
|
||||
.arg("-o")
|
||||
.arg(raw_path)
|
||||
.arg("-sS")
|
||||
.arg(url)
|
||||
.status()
|
||||
.context("执行 curl 下载朋友圈图片失败；请确认系统已安装 curl")?;
|
||||
if !status.success() {
|
||||
anyhow::bail!("curl 下载失败: {}", status);
|
||||
}
|
||||
Ok(())
|
||||
}
|
||||
|
||||
#[cfg(target_os = "linux")]
|
||||
fn decrypt_sns_cdn_media(decode_key: &str, raw_path: &Path, out_path: &Path) -> Result<()> {
|
||||
let (wasm_js, wasm_bin) = sns_wasm_paths()?;
|
||||
let script = r#"
|
||||
const fs = require('fs');
|
||||
const [decodeKey, rawPath, outPath, jsPath, wasmPath] = process.argv.slice(1);
|
||||
global.self = { location: { href: 'file:///wx-cli-sns-worker.html' } };
|
||||
global.window = global;
|
||||
global.VTS_WASM_URL = 'wasm_video_decode.wasm';
|
||||
global.MAX_HEAP_SIZE = 33554432;
|
||||
global.Module = { wasmBinary: fs.readFileSync(wasmPath) };
|
||||
global.keystreamData = null;
|
||||
global.wasm_isaac_generate = function(ptr, size) {
|
||||
const wasmArray = new Uint8Array(Module.HEAPU8.buffer, ptr, size);
|
||||
keystreamData = new Uint8Array(Array.from(wasmArray).reverse());
|
||||
};
|
||||
eval(fs.readFileSync(jsPath, 'utf8'));
|
||||
function waitReady() {
|
||||
return new Promise((resolve, reject) => {
|
||||
const timer = setInterval(() => {
|
||||
if (global.Module && Module.WxIsaac64) {
|
||||
clearInterval(timer);
|
||||
resolve();
|
||||
}
|
||||
}, 50);
|
||||
setTimeout(() => {
|
||||
clearInterval(timer);
|
||||
reject(new Error('WxIsaac64 WASM load timeout'));
|
||||
}, 10000);
|
||||
});
|
||||
}
|
||||
(async () => {
|
||||
await waitReady();
|
||||
const raw = fs.readFileSync(rawPath);
|
||||
const genLen = Math.ceil(raw.length / 8) * 8;
|
||||
const decryptor = new Module.WxIsaac64(decodeKey);
|
||||
decryptor.generate(genLen);
|
||||
decryptor.delete();
|
||||
if (!keystreamData || keystreamData.length < raw.length) {
|
||||
throw new Error('keystream too short');
|
||||
}
|
||||
const out = Buffer.alloc(raw.length);
|
||||
for (let i = 0; i < raw.length; i++) out[i] = raw[i] ^ keystreamData[i];
|
||||
const eoi = out.lastIndexOf(Buffer.from([0xff, 0xd9]));
|
||||
const finalOut = eoi >= 0 ? out.subarray(0, eoi + 2) : out;
|
||||
if (!(finalOut[0] === 0xff && finalOut[1] === 0xd8 && finalOut[2] === 0xff)) {
|
||||
throw new Error('decrypted data is not JPEG');
|
||||
}
|
||||
fs.writeFileSync(outPath, finalOut);
|
||||
})().catch((err) => {
|
||||
console.error(err && err.stack ? err.stack : String(err));
|
||||
process.exit(1);
|
||||
});
|
||||
"#;
|
||||
|
||||
let output = Command::new("node")
|
||||
.arg("-e")
|
||||
.arg(script)
|
||||
.arg(decode_key)
|
||||
.arg(raw_path)
|
||||
.arg(out_path)
|
||||
.arg(&wasm_js)
|
||||
.arg(&wasm_bin)
|
||||
.output()
|
||||
.context("执行 node 解密朋友圈图片失败；请确认系统已安装 node")?;
|
||||
if !output.status.success() {
|
||||
let stderr = String::from_utf8_lossy(&output.stderr);
|
||||
anyhow::bail!("Isaac64 解密失败: {}", stderr.trim());
|
||||
}
|
||||
Ok(())
|
||||
}
|
||||
|
||||
#[cfg(target_os = "linux")]
|
||||
fn sns_wasm_paths() -> Result<(PathBuf, PathBuf)> {
|
||||
let mut roots = Vec::new();
|
||||
if let Ok(root) = std::env::var("WX_SNS_WASM_DIR") {
|
||||
roots.push(PathBuf::from(root));
|
||||
}
|
||||
roots.push(PathBuf::from("wechat_files"));
|
||||
|
||||
for root in roots {
|
||||
let js = root.join("wasm_video_decode.js");
|
||||
let wasm = root.join("wasm_video_decode.wasm");
|
||||
if js.is_file() && wasm.is_file() {
|
||||
return Ok((js, wasm));
|
||||
}
|
||||
}
|
||||
|
||||
anyhow::bail!(
|
||||
"找不到 WxIsaac64 WASM 文件；请设置 WX_SNS_WASM_DIR 指向包含 wasm_video_decode.js 和 wasm_video_decode.wasm 的目录"
|
||||
)
|
||||
}
|
||||
|
||||
#[cfg(not(target_os = "linux"))]
|
||||
fn sns_img_roots(wxchat_base: &Path, month: Option<&str>) -> Result<Vec<PathBuf>> {
|
||||
let cache_root = wxchat_base.join("cache");
|
||||
if let Some(month) = month {
|
||||
|
|
@ -4737,6 +5042,7 @@ fn sns_img_roots(wxchat_base: &Path, month: Option<&str>) -> Result<Vec<PathBuf>
|
|||
Ok(roots)
|
||||
}
|
||||
|
||||
#[cfg(not(target_os = "linux"))]
|
||||
fn collect_sns_img_files(dir: &Path, out: &mut Vec<PathBuf>) -> Result<()> {
|
||||
for entry in fs::read_dir(dir)? {
|
||||
let path = entry?.path();
|
||||
|
|
@ -4749,6 +5055,7 @@ fn collect_sns_img_files(dir: &Path, out: &mut Vec<PathBuf>) -> Result<()> {
|
|||
Ok(())
|
||||
}
|
||||
|
||||
#[cfg(not(target_os = "linux"))]
|
||||
fn sns_output_stem(path: &Path) -> String {
|
||||
let parent = path
|
||||
.parent()
|
||||
|
|
|
|||
Loading…
Reference in New Issue