mailops 0.2.2__tar.gz → 0.2.4__tar.gz

{mailops-0.2.2 → mailops-0.2.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mailops
-Version: 0.2.2
+Version: 0.2.4
 Summary: 邮件操作工具箱 — 下载、统计、搜索、转发、附件管理、发送
 Author: Song Wupei
 License-Expression: MIT
@@ -71,6 +71,10 @@ mailops send -t me@163.com -s "日报" -b "详见附件" -a report.pdf
 # 执行工作流
 mailops workflow run 163-invoice-to-qq
 
+# 邮件自动分类归档
+mailops organize --dry-run          # 预览
+mailops organize -a 163             # 执行
+
 # 指定邮箱
 mailops -m qq list
 mailops -m 163 list
@@ -131,6 +135,31 @@ smtp:
 
 用于 `mailops workflow` 命令，定义自动化的同步→下载→转发流程。
 
+### 邮件分类规则 (`~/.config/aerc/mail-rules.yml`)
+
+用于 `mailops organize` 命令，按规则自动分类/归档邮件。示例：
+
+```yaml
+accounts:
+  "163":
+    maildir: ~/Mail/163
+    rules:
+      - name: github
+        match:
+          from: "(?i)@github\\.com"
+          list_id: "(?i)github"
+          match_any: true
+        action:
+          move: GitHub
+      - name: archive-old
+        match:
+          older_than: 7d
+        action:
+          move: "Archive/{year}/{month}"
+```
+
+参考 `config/mail-rules.example.yml`。
+
 ## 命令一览
 
 | 命令 | 说明 |
@@ -142,6 +171,7 @@ smtp:
 | `stats` | 邮件统计（总量、发件人排行、附件分布） |
 | `download` | 下载附件和正文（本地扫描，不先同步） |
 | `download-docs` | 按关键词搜索 PDF 下载（原 maildl.py） |
+| `organize` | 📬 按规则自动分类归档（GitHub → GitHub/, 旧邮件 → Archive/） |
 | `send` | 发送邮件（支持多附件、抄送） |
 | `forward` | 转发已下载文档 |
 | `fetch` | 同步 IMAP → Maildir |
@@ -156,6 +186,7 @@ smtp:
 |--------|--------|
 | `maildl.py` | `mailops download-docs` |
 | `mailfwd.py` | `mailops forward` |
+| `sync163.py` | `mailops fetch` (IMAP → Maildir) + `mailops organize` (分类) |
 | `sync163.py` | `mailops fetch -m 163` |
 | `run-wf.py` | `mailops workflow` |
 | `mailx` (系统命令) | `mailops send` |

{mailops-0.2.2 → mailops-0.2.4}/README.md

@@ -50,6 +50,10 @@ mailops send -t me@163.com -s "日报" -b "详见附件" -a report.pdf
 # 执行工作流
 mailops workflow run 163-invoice-to-qq
 
+# 邮件自动分类归档
+mailops organize --dry-run          # 预览
+mailops organize -a 163             # 执行
+
 # 指定邮箱
 mailops -m qq list
 mailops -m 163 list
@@ -110,6 +114,31 @@ smtp:
 
 用于 `mailops workflow` 命令，定义自动化的同步→下载→转发流程。
 
+### 邮件分类规则 (`~/.config/aerc/mail-rules.yml`)
+
+用于 `mailops organize` 命令，按规则自动分类/归档邮件。示例：
+
+```yaml
+accounts:
+  "163":
+    maildir: ~/Mail/163
+    rules:
+      - name: github
+        match:
+          from: "(?i)@github\\.com"
+          list_id: "(?i)github"
+          match_any: true
+        action:
+          move: GitHub
+      - name: archive-old
+        match:
+          older_than: 7d
+        action:
+          move: "Archive/{year}/{month}"
+```
+
+参考 `config/mail-rules.example.yml`。
+
 ## 命令一览
 
 | 命令 | 说明 |
@@ -121,6 +150,7 @@ smtp:
 | `stats` | 邮件统计（总量、发件人排行、附件分布） |
 | `download` | 下载附件和正文（本地扫描，不先同步） |
 | `download-docs` | 按关键词搜索 PDF 下载（原 maildl.py） |
+| `organize` | 📬 按规则自动分类归档（GitHub → GitHub/, 旧邮件 → Archive/） |
 | `send` | 发送邮件（支持多附件、抄送） |
 | `forward` | 转发已下载文档 |
 | `fetch` | 同步 IMAP → Maildir |
@@ -135,6 +165,7 @@ smtp:
 |--------|--------|
 | `maildl.py` | `mailops download-docs` |
 | `mailfwd.py` | `mailops forward` |
+| `sync163.py` | `mailops fetch` (IMAP → Maildir) + `mailops organize` (分类) |
 | `sync163.py` | `mailops fetch -m 163` |
 | `run-wf.py` | `mailops workflow` |
 | `mailx` (系统命令) | `mailops send` |

{mailops-0.2.2 → mailops-0.2.4}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "mailops"
-version = "0.2.2"
+version = "0.2.4"
 description = "邮件操作工具箱 — 下载、统计、搜索、转发、附件管理、发送"
 readme = "README.md"
 license = "MIT"

{mailops-0.2.2 → mailops-0.2.4}/src/mailops/cli.py

@@ -526,6 +526,50 @@ def config():
     print(f"\n  SMTP 默认: {smtp.get('default', '未配置')}")
     print(f"\n  输出目录: {get_output_dir()}")
 
+# ═══════════════════════════════════════════════════════════════
+# organize — 邮件自动分类归档
+# ═══════════════════════════════════════════════════════════════
+
+@cli.command()
+@click.option("-a", "--account", "account_name", default=None,
+              help="指定账号（默认所有）")
+@click.option("--dry-run", is_flag=True, help="预览模式，不实际移动")
+@click.option("--rules", "rules_path", default=None,
+              type=click.Path(exists=True), help="规则文件路径")
+@click.pass_context
+def organize(ctx, account_name, dry_run, rules_path):
+    """📬 按规则自动分类/归档邮件。
+
+    根据 ~/.config/aerc/mail-rules.yml 中定义的规则，
+    将邮件移动到对应文件夹（如 GitHub → GitHub/, 旧邮件 → Archive/）。
+
+    示例：
+      mailops organize --dry-run        # 预览所有账号
+      mailops organize -a 163           # 只处理 163
+      mailops organize -a 163 --dry-run # 预览 163 的分类结果
+    """
+    from .organizer import organize as run_organize, load_rules
+
+    rp = Path(rules_path) if rules_path else None
+
+    if account_name:
+        result = run_organize(account_name, rules_path=rp, dry_run=dry_run)
+        if dry_run:
+            print(f"\n📋 预览完成，共 {result['moved']} 封将被移动")
+        else:
+            print(f"\n✅ 完成，移动 {result['moved']} 封邮件")
+    else:
+        all_rules = load_rules(rp)
+        total_moved = 0
+        for name in all_rules:
+            result = run_organize(name, rules_path=rp, dry_run=dry_run)
+            total_moved += result["moved"]
+        if dry_run:
+            print(f"\n📋 预览完成，共 {total_moved} 封将被移动")
+        else:
+            print(f"\n✅ 完成，移动 {total_moved} 封邮件")
+
+
 # ═══════════════════════════════════════════════════════════════
 # 入口
 # ═══════════════════════════════════════════════════════════════

mailops-0.2.4/src/mailops/organizer.py

@@ -0,0 +1,318 @@
+"""Maildir 邮件自动分类引擎。
+
+根据 YAML 规则配置，对本地 Maildir 中的邮件进行自动分类/归档。
+
+用法:
+    from mailops.organizer import organize
+    organize("163", dry_run=True)   # 预览
+    organize("163")                 # 执行
+"""
+
+import email
+import os
+import re
+import shutil
+from datetime import datetime, timedelta
+from email.policy import default
+from pathlib import Path
+from typing import Optional
+
+
+DEFAULT_RULES_PATH = Path.home() / ".config" / "aerc" / "mail-rules.yml"
+
+
+def _parse_date(date_str: str) -> Optional[datetime]:
+    """尽量从 Date 头解析时间。"""
+    if not date_str:
+        return None
+    try:
+        # email.utils.parsedate_to_datetime 是标准方法
+        from email.utils import parsedate_to_datetime
+        return parsedate_to_datetime(date_str)
+    except Exception:
+        pass
+    # 尝试常见格式
+    for fmt in [
+        "%a, %d %b %Y %H:%M:%S %z",
+        "%d %b %Y %H:%M:%S %z",
+        "%a, %d %b %Y %H:%M:%S",
+        "%d %b %Y %H:%M:%S",
+    ]:
+        try:
+            return datetime.strptime(date_str.strip(), fmt)
+        except ValueError:
+            continue
+    return None
+
+
+def _parse_older_than(spec: str) -> timedelta:
+    """解析 '7d', '2w', '1m' 等时间字符串。"""
+    spec = spec.strip().lower()
+    m = re.match(r"(\d+)\s*(d|w|m)", spec)
+    if not m:
+        raise ValueError(f"无法解析 older_than 值: {spec!r}，格式如 7d, 2w, 1m")
+    n = int(m.group(1))
+    unit = m.group(2)
+    if unit == "d":
+        return timedelta(days=n)
+    elif unit == "w":
+        return timedelta(weeks=n)
+    elif unit == "m":
+        return timedelta(days=n * 30)
+    return timedelta(days=n)
+
+
+def load_rules(config_path: Optional[Path] = None) -> dict:
+    """加载邮件分类规则配置。
+
+    返回格式: {"account_name": {"maildir": ..., "rules": [...]}}
+    """
+    import yaml
+
+    path = Path(config_path) if config_path else DEFAULT_RULES_PATH
+    if not path.exists():
+        raise FileNotFoundError(
+            f"规则配置文件不存在: {path}\n"
+            f"请从示例创建: mailops/config/mail-rules.example.yml"
+        )
+    with open(path) as f:
+        cfg = yaml.safe_load(f)
+    return cfg.get("accounts", {})
+
+
+def _parse_mail_headers(filepath: Path) -> Optional[dict]:
+    """快速解析邮件头部（只读前 8KB）。
+
+    返回包含 uid, from_addr, subject, date, list_id, message_id 的字典。
+    """
+    try:
+        with open(filepath, "rb") as f:
+            raw = f.read(8192)
+    except OSError:
+        return None
+
+    try:
+        msg = email.message_from_bytes(raw, policy=default)
+    except Exception:
+        return None
+
+    fname = filepath.name
+    uid = fname.split(":2,")[0] if ":2," in fname else fname
+
+    return {
+        "uid": uid,
+        "fname": fname,
+        "filepath": str(filepath),
+        "from_addr": str(msg.get("From", "")),
+        "to_addr": str(msg.get("To", "")),
+        "subject": str(msg.get("Subject", "")),
+        "date": str(msg.get("Date", "")),
+        "list_id": str(msg.get("List-ID", "")),
+        "message_id": str(msg.get("Message-ID", "")).strip("<>"),
+        "parsed_date": _parse_date(str(msg.get("Date", ""))),
+    }
+
+
+def _match_rule(headers: dict, rule: dict) -> bool:
+    """检查一封邮件是否匹配某条规则。"""
+    match = rule.get("match", {})
+    if not match:
+        return False
+
+    match_any = match.get("match_any", False)
+
+    # 收集各字段匹配结果
+    results = {}
+
+    # From 匹配
+    from_pat = match.get("from")
+    if from_pat:
+        results["from"] = bool(
+            re.search(from_pat, headers.get("from_addr", ""))
+        )
+
+    # List-ID 匹配
+    lid_pat = match.get("list_id")
+    if lid_pat:
+        results["list_id"] = bool(
+            re.search(lid_pat, headers.get("list_id", ""), re.IGNORECASE)
+        )
+
+    # Subject 匹配
+    subj_pat = match.get("subject")
+    if subj_pat:
+        results["subject"] = bool(
+            re.search(subj_pat, headers.get("subject", ""), re.IGNORECASE)
+        )
+
+    # older_than: 邮件日期早于 N 天前
+    older_than = match.get("older_than")
+    if older_than:
+        parsed = headers.get("parsed_date")
+        if parsed:
+            delta = _parse_older_than(str(older_than))
+            results["older_than"] = parsed < (datetime.now().astimezone() - delta)
+        else:
+            results["older_than"] = False
+
+    if not results:
+        return False
+
+    if match_any:
+        return any(results.values())
+    else:
+        return all(results.values())
+
+
+def _expand_path(template: str, headers: dict) -> str:
+    """展开路径模板中的变量。
+
+    支持: {year}, {month}, {day}, {week} 等。
+    """
+    dt = headers.get("parsed_date") or datetime.now()
+    return template.format(
+        year=dt.year,
+        month=f"{dt.month:02d}",
+        day=f"{dt.day:02d}",
+        week=f"W{dt.isocalendar()[1]:02d}",
+    )
+
+
+def move_message(filepath: Path, maildir_base: Path, dest: str) -> Optional[str]:
+    """将邮件移动到目标 Maildir 文件夹。
+
+    Args:
+        filepath: 源邮件路径
+        maildir_base: Maildir 根目录 (如 ~/Mail/163)
+        dest: 目标文件夹名 (如 "GitHub" 或 "Archive/2026/07")
+
+    Returns:
+        目标文件路径，失败返回 None
+    """
+    dest_dir = (maildir_base / dest).expanduser().resolve()
+    dest_new = dest_dir / "new"
+    dest_cur = dest_dir / "cur"
+    dest_tmp = dest_dir / "tmp"
+
+    for d in (dest_new, dest_cur, dest_tmp):
+        d.mkdir(parents=True, exist_ok=True)
+
+    # 保持原文件名移入 new/
+    target = dest_new / filepath.name
+
+    try:
+        shutil.move(str(filepath), str(target))
+        return str(target)
+    except OSError:
+        return None
+
+
+def ensure_folders(maildir_base: Path, rules: list):
+    """根据规则预创建目标文件夹，确保 aerc 侧边栏可见。
+
+    对静态路径（如 GitHub）直接创建完整的 Maildir 结构；
+    对模板路径（如 Archive/{year}/{month}）创建根目录。
+    """
+    for rule in rules:
+        dest = rule.get("action", {}).get("move", "")
+        if not dest:
+            continue
+        # 取模板路径的根目录（第一个 / 之前的部分）
+        root_folder = dest.split("/")[0].split("{")[0].rstrip("/")
+        if not root_folder:
+            continue
+        target = (maildir_base / root_folder).expanduser().resolve()
+        for sub in ("cur", "new", "tmp"):
+            (target / sub).mkdir(parents=True, exist_ok=True)
+
+
+def organize(account_name: str, rules_path: Optional[Path] = None,
+             dry_run: bool = False) -> dict:
+    """对指定账号执行邮件分类。
+
+    Returns:
+        {"account": str, "scanned": int, "moved": int, "moves": [...]}
+    """
+    all_rules = load_rules(rules_path)
+
+    if account_name not in all_rules:
+        known = ", ".join(all_rules.keys())
+        raise KeyError(f"未找到账号 '{account_name}'，可用: {known}")
+
+    acct_cfg = all_rules[account_name]
+    maildir_str = acct_cfg.get("maildir", f"~/Mail/{account_name}")
+    maildir_base = Path(maildir_str).expanduser().resolve()
+    rules = acct_cfg.get("rules", [])
+
+    if not rules:
+        print(f"  [{account_name}] 无分类规则，跳过")
+        return {"account": account_name, "scanned": 0, "moved": 0, "moves": []}
+
+    # 预创建目标文件夹（确保 aerc 侧边栏可见）
+    ensure_folders(maildir_base, rules)
+
+    inbox = maildir_base / "INBOX"
+    if not inbox.is_dir():
+        inbox = maildir_base  # 可能 INBOX 就是根目录
+
+    moves = []
+    scanned = 0
+
+    for sub in ("new", "cur"):
+        scan_dir = inbox / sub
+        if not scan_dir.is_dir():
+            continue
+        for f in sorted(scan_dir.iterdir()):
+            if not f.is_file() or f.name.startswith("."):
+                continue
+
+            scanned += 1
+            headers = _parse_mail_headers(f)
+            if headers is None:
+                continue
+
+            for rule in rules:
+                if _match_rule(headers, rule):
+                    action = rule.get("action", {})
+                    dest_template = action.get("move", "")
+                    if not dest_template:
+                        continue
+
+                    dest = _expand_path(dest_template, headers)
+
+                    if dry_run:
+                        moves.append({
+                            "uid": headers["uid"],
+                            "from": str(f),
+                            "to": str((maildir_base / dest / "new" / f.name).expanduser()),
+                            "rule": rule.get("name", ""),
+                            "subject": headers["subject"][:80],
+                        })
+                    else:
+                        result = move_message(f, maildir_base, dest)
+                        if result:
+                            moves.append({
+                                "uid": headers["uid"],
+                                "from": str(f),
+                                "to": result,
+                                "rule": rule.get("name", ""),
+                                "subject": headers["subject"][:80],
+                            })
+                    break  # 命中第一条规则后停止
+
+    prefix = "[DRY RUN] " if dry_run else ""
+    if moves:
+        print(f"{prefix}[{account_name}] 扫描 {scanned} 封，匹配 {len(moves)} 封:")
+        for m in moves:
+            print(f"  [{m['rule']}] {m['subject']}")
+            if not dry_run:
+                print(f"    → {m['to']}")
+    else:
+        print(f"{prefix}[{account_name}] 扫描 {scanned} 封，无匹配")
+
+    return {
+        "account": account_name,
+        "scanned": scanned,
+        "moved": len(moves),
+        "moves": moves,
+    }

{mailops-0.2.2 → mailops-0.2.4}/src/mailops.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mailops
-Version: 0.2.2
+Version: 0.2.4
 Summary: 邮件操作工具箱 — 下载、统计、搜索、转发、附件管理、发送
 Author: Song Wupei
 License-Expression: MIT
@@ -71,6 +71,10 @@ mailops send -t me@163.com -s "日报" -b "详见附件" -a report.pdf
 # 执行工作流
 mailops workflow run 163-invoice-to-qq
 
+# 邮件自动分类归档
+mailops organize --dry-run          # 预览
+mailops organize -a 163             # 执行
+
 # 指定邮箱
 mailops -m qq list
 mailops -m 163 list
@@ -131,6 +135,31 @@ smtp:
 
 用于 `mailops workflow` 命令，定义自动化的同步→下载→转发流程。
 
+### 邮件分类规则 (`~/.config/aerc/mail-rules.yml`)
+
+用于 `mailops organize` 命令，按规则自动分类/归档邮件。示例：
+
+```yaml
+accounts:
+  "163":
+    maildir: ~/Mail/163
+    rules:
+      - name: github
+        match:
+          from: "(?i)@github\\.com"
+          list_id: "(?i)github"
+          match_any: true
+        action:
+          move: GitHub
+      - name: archive-old
+        match:
+          older_than: 7d
+        action:
+          move: "Archive/{year}/{month}"
+```
+
+参考 `config/mail-rules.example.yml`。
+
 ## 命令一览
 
 | 命令 | 说明 |
@@ -142,6 +171,7 @@ smtp:
 | `stats` | 邮件统计（总量、发件人排行、附件分布） |
 | `download` | 下载附件和正文（本地扫描，不先同步） |
 | `download-docs` | 按关键词搜索 PDF 下载（原 maildl.py） |
+| `organize` | 📬 按规则自动分类归档（GitHub → GitHub/, 旧邮件 → Archive/） |
 | `send` | 发送邮件（支持多附件、抄送） |
 | `forward` | 转发已下载文档 |
 | `fetch` | 同步 IMAP → Maildir |
@@ -156,6 +186,7 @@ smtp:
 |--------|--------|
 | `maildl.py` | `mailops download-docs` |
 | `mailfwd.py` | `mailops forward` |
+| `sync163.py` | `mailops fetch` (IMAP → Maildir) + `mailops organize` (分类) |
 | `sync163.py` | `mailops fetch -m 163` |
 | `run-wf.py` | `mailops workflow` |
 | `mailx` (系统命令) | `mailops send` |

{mailops-0.2.2 → mailops-0.2.4}/src/mailops.egg-info/SOURCES.txt

@@ -9,6 +9,7 @@ src/mailops/config.py
 src/mailops/downloader.py
 src/mailops/forwarder.py
 src/mailops/operations.py
+src/mailops/organizer.py
 src/mailops/sync.py
 src/mailops/workflow.py
 src/mailops.egg-info/PKG-INFO

mailops-0.2.4/src/sync163.py

@@ -0,0 +1,226 @@
+#!/usr/bin/env python3
+"""163 邮箱 → 本地 Maildir 同步（快速版，使用本地索引避免重复下载）。
+
+用法: python3 sync163.py
+      python3 sync163.py --limit 50    # 只同步最近 50 封
+"""
+
+import imaplib
+import json
+import os
+import re
+import ssl
+import sys
+import time
+from pathlib import Path
+
+HOST = "imap.163.com"
+USER = "songwupei@163.com"
+MAILDIR = Path.home() / "Mail" / "163" / "INBOX"
+INDEX_FILE = Path.home() / ".cache" / "aerc" / "sync163-index.json"
+PASS_ENV = "MAIL163_SMTP_PASS_AERC"
+
+
+def load_index():
+    """加载已知 Message-ID → 文件名 的索引。"""
+    if INDEX_FILE.exists():
+        try:
+            return json.loads(INDEX_FILE.read_text())
+        except (json.JSONDecodeError, OSError):
+            pass
+    return {}
+
+
+def save_index(idx):
+    """保存索引到磁盘。"""
+    INDEX_FILE.parent.mkdir(parents=True, exist_ok=True)
+    INDEX_FILE.write_text(json.dumps(idx))
+
+
+def extract_message_id(msg_bytes):
+    """从邮件字节中快速提取 Message-ID（只读头部，不解码整个邮件）。"""
+    # 只取头部（到第一个空行）
+    header_end = msg_bytes.find(b"\r\n\r\n")
+    if header_end == -1:
+        header_end = msg_bytes.find(b"\n\n")
+    if header_end == -1:
+        header_end = len(msg_bytes)
+    header = msg_bytes[:header_end].decode("utf-8", errors="replace")
+
+    # 提取 Message-ID
+    m = re.search(r"(?im)^Message-ID:\s*(.+)", header)
+    if m:
+        return m.group(1).strip().strip("<>")
+    return None
+
+
+def backfill_index(known):
+    """首次运行：从已有 Maildir 文件重建索引，避免重复下载。"""
+    existing = 0
+    for sub in ["cur", "new"]:
+        folder = MAILDIR / sub
+        if not folder.exists():
+            continue
+        for fpath in folder.iterdir():
+            if not fpath.is_file():
+                continue
+            try:
+                # 只读前 8KB 足以包含所有头部
+                mid = extract_message_id(fpath.read_bytes()[:8192])
+                if mid and mid not in known:
+                    known[mid] = fpath.name
+                    existing += 1
+            except OSError:
+                pass
+    if existing:
+        print(f"从本地 Maildir 重建索引: +{existing} 封")
+    return known
+
+
+def sync(limit=None):
+    # ---- 密码 ----
+    pass_file = Path.home() / ".config" / "aerc" / "163-pass"
+    if pass_file.exists():
+        password = pass_file.read_text().strip()
+    else:
+        password = os.environ.get(PASS_ENV, "")
+    if not password:
+        print(f"错误: 未找到密码（环境变量 {PASS_ENV} 未设置，且 {pass_file} 不存在）")
+        return
+
+    # ---- 初始化 Maildir ----
+    for sub in ["cur", "new", "tmp"]:
+        (MAILDIR / sub).mkdir(parents=True, exist_ok=True)
+
+    # ---- 加载索引（首次运行自动从已有文件重建） ----
+    known = load_index()
+    if not known:
+        known = backfill_index(known)
+        save_index(known)
+
+    # ---- 连接 ----
+    ctx = ssl.create_default_context()
+    with imaplib.IMAP4_SSL(HOST, 993, ssl_context=ctx) as conn:
+        conn.login(USER, password)
+
+        # 163.com 必须的 ID 命令
+        try:
+            conn.socket().send(
+                b'a001 ID ("name" "mail-dl" "version" "1.0")\r\n'
+            )
+            conn.socket().recv(1024)
+        except Exception:
+            pass
+
+        conn.select("INBOX")
+
+        # ---- 搜索所有邮件 ----
+        status, data = conn.search(None, "ALL")
+        if status != "OK":
+            print("搜索失败")
+            return
+
+        all_ids = data[0].split()
+        if not all_ids:
+            print("收件箱为空")
+            return
+
+        total = len(all_ids)
+        print(f"服务器共 {total} 封邮件，索引中已记录 {len(known)} 封")
+
+        # ---- 先通过只下载 HEADER 来获取 Message-ID（轻量级） ----
+        # 批量获取所有邮件的 Message-ID，用于判断哪些是新的
+        ids = list(reversed(all_ids))
+        if limit:
+            ids = ids[-limit:]
+
+        # 分批处理，每批 200 封
+        BATCH = 200
+        to_fetch = []  # (num, msg_id) pairs that we need to download
+        seen_msgids = set()
+
+        for i in range(0, len(ids), BATCH):
+            batch = ids[i : i + BATCH]
+            id_range = f"{batch[0].decode()}:{batch[-1].decode()}"
+            status, hdr_data = conn.fetch(
+                id_range, "(BODY.PEEK[HEADER.FIELDS (MESSAGE-ID)])"
+            )
+            if status != "OK":
+                continue
+
+            # 解析每个 HEADER 响应
+            current_num = None
+            for item in hdr_data:
+                if isinstance(item, tuple):
+                    # item[0] 是 "1 (BODY[HEADER...]" 之类的标记
+                    # item[1] 是头部内容
+                    header_bytes = item[1]
+                    if header_bytes:
+                        msg_id = extract_message_id(header_bytes)
+                        if msg_id and msg_id not in known and msg_id not in seen_msgids:
+                            # 从 item[0] 提取序列号
+                            marker = item[0].decode() if isinstance(
+                                item[0], bytes
+                            ) else str(item[0])
+                            m = re.match(r"(\d+)", marker)
+                            if m:
+                                to_fetch.append((m.group(1).encode(), msg_id))
+                                seen_msgids.add(msg_id)
+
+        # ---- 只下载新邮件 ----
+        # 对 IMAP 重新 select（因为上面的 fetch 可能改变了状态）
+        # 实际上不需要，IMAP sequence numbers 在一次 select 内是稳定的
+
+        new_count = 0
+        for num_bytes, msg_id in to_fetch:
+            status, msg_data = conn.fetch(num_bytes, "(RFC822)")
+            if status != "OK":
+                continue
+
+            msg_bytes = msg_data[0][1]
+            if msg_bytes is None:
+                continue
+
+            # 写入 Maildir
+            ts = int(time.time())
+            fname = (
+                f"{ts}.{num_bytes.decode()}.{new_count}.{os.uname().nodename}"
+            )
+            fpath = MAILDIR / "new" / fname
+            fpath.write_bytes(msg_bytes)
+
+            # 更新索引
+            known[msg_id] = fname
+
+            new_count += 1
+            # 快速预览主题
+            subject = "(无主题)"
+            try:
+                s = re.search(
+                    rb"(?im)^Subject:\s*(.+)", msg_bytes[:4096]
+                )
+                if s:
+                    subject = s.group(1).decode("utf-8", errors="replace").strip()
+            except Exception:
+                pass
+            print(f"  [{new_count}] {subject}")
+
+    # ---- 保存索引 ----
+    save_index(known)
+
+    # ---- 清理索引中已经不存在的文件 ----
+    stale = [mid for mid, fname in known.items() if not (MAILDIR / "new" / fname).exists() and not (MAILDIR / "cur" / fname).exists()]
+    if stale:
+        for mid in stale:
+            del known[mid]
+        save_index(known)
+        print(f"  (清理了 {len(stale)} 条过期索引记录)")
+
+    print(f"\n同步完成，新增 {new_count} 封 → {MAILDIR}")
+
+
+if __name__ == "__main__":
+    limit = None
+    if len(sys.argv) > 1 and sys.argv[1] == "--limit":
+        limit = int(sys.argv[2]) if len(sys.argv) > 2 else 50
+    sync(limit)

mailops-0.2.2/src/sync163.py

@@ -1,104 +0,0 @@
-#!/usr/bin/env python3
-"""163 邮箱 → 本地 Maildir 同步（Python 版 mbsync，解决 IMAP ID 问题）。
-
-用法: python3 sync163.py
-      python3 sync163.py --limit 50    # 只同步最近 50 封
-"""
-
-import email
-import imaplib
-import os
-import ssl
-import sys
-import time
-from pathlib import Path
-
-HOST = "imap.163.com"
-USER = "songwupei@163.com"
-MAILDIR = Path.home() / "Mail" / "163" / "INBOX"
-PASS_ENV = "MAIL163_SMTP_PASS_AERC"
-
-
-def sync(limit=None):
-    pass_file = Path.home() / ".config" / "aerc" / "163-pass"
-    if pass_file.exists():
-        password = pass_file.read_text().strip()
-    else:
-        password = os.environ.get(PASS_ENV, "")
-    if not password:
-        print(f"错误: 未找到密码（环境变量 {PASS_ENV} 未设置，且 {pass_file} 不存在）")
-        return
-
-    for sub in ["cur", "new", "tmp"]:
-        (MAILDIR / sub).mkdir(parents=True, exist_ok=True)
-
-    ctx = ssl.create_default_context()
-    with imaplib.IMAP4_SSL(HOST, 993, ssl_context=ctx) as conn:
-        conn.login(USER, password)
-        # 163.com 必须的 ID 命令
-        try:
-            conn.socket().send(
-                b'a001 ID ("name" "mail-dl" "version" "1.0")\r\n')
-            conn.socket().recv(1024)
-        except Exception:
-            pass
-        conn.select("INBOX")
-        status, data = conn.search(None, "ALL")
-        if status != "OK":
-            print("搜索失败")
-            return
-
-        all_ids = data[0].split()
-        if not all_ids:
-            print("收件箱为空")
-            return
-
-        ids = list(reversed(all_ids))
-        if limit:
-            ids = ids[-limit:]
-
-        new_count = 0
-        for num in ids:
-            status, msg_data = conn.fetch(num, "(RFC822 FLAGS)")
-            if status != "OK":
-                continue
-
-            msg_bytes = msg_data[0][1]
-            if msg_bytes is None:
-                continue
-
-            msg = email.message_from_bytes(msg_bytes)
-            msg_id = msg.get("Message-ID")
-            if msg_id is None:
-                msg_id = f"{int(time.time())}.{int(num)}@unknown"
-            # strip brackets from Message-ID if present
-            mid = str(msg_id).strip("<>").encode()
-
-            # 用 Message-ID 去重
-            existing = False
-            for sub in ["cur", "new"]:
-                for f in (MAILDIR / sub).iterdir():
-                    if f.is_file() and mid in f.read_bytes():
-                        existing = True
-                        break
-                if existing:
-                    break
-            if existing:
-                continue
-
-            # 写入 Maildir new/
-            ts = int(time.time())
-            fname = f"{ts}.{num}.{os.uname().nodename}"
-            fpath = MAILDIR / "new" / fname
-            fpath.write_bytes(msg_bytes)
-            new_count += 1
-            print(f"  [{new_count}] {msg.get('Subject', '(无主题)')}")
-
-    print(f"\n同步完成，新增 {new_count} 封 → {MAILDIR}")
-
-
-if __name__ == "__main__":
-    limit = None
-    if len(sys.argv) > 1 and sys.argv[1] == "--limit":
-        limit = int(sys.argv[2]) if len(sys.argv) > 2 else 50
-    sync(limit)