提交

scripts/compare_crm_sql_models.py

@@ -0,0 +1,303 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+from __future__ import annotations
+
+import argparse
+import dataclasses
+import html
+import re
+from pathlib import Path
+from typing import Dict, List, Tuple
+
+
+@dataclasses.dataclass(frozen=True)
+class Column:
+    name: str
+    raw_def: str  # definition text after column name, normalized
+
+
+@dataclasses.dataclass(frozen=True)
+class Table:
+    name: str
+    columns: Dict[str, Column]
+
+
+CREATE_TABLE_RE = re.compile(r"^\s*CREATE\s+TABLE\s+`(?P<name>crm_[^`]+)`\s*\(\s*$", re.IGNORECASE)
+COL_RE = re.compile(r"^\s*`(?P<col>[^`]+)`\s+(?P<rest>.+)$")
+
+
+def _normalize_ws(s: str) -> str:
+    s = s.strip()
+    s = re.sub(r"\s+", " ", s)
+    return s
+
+
+def parse_crm_tables(sql_text: str) -> Dict[str, Table]:
+    """
+    Parse Navicat-style MySQL DDL and extract CREATE TABLE blocks for crm_*.
+    Only captures column lines beginning with backticked column names.
+    """
+    lines = sql_text.splitlines()
+    tables: Dict[str, Table] = {}
+
+    i = 0
+    while i < len(lines):
+        m = CREATE_TABLE_RE.match(lines[i])
+        if not m:
+            i += 1
+            continue
+
+        table_name = m.group("name")
+        i += 1
+
+        cols: Dict[str, Column] = {}
+        while i < len(lines):
+            line = lines[i].rstrip("\n")
+            # End of CREATE TABLE block
+            if re.search(r"^\s*\)\s*ENGINE=.*;\s*$", line, flags=re.IGNORECASE):
+                break
+
+            cm = COL_RE.match(line)
+            if cm:
+                col_name = cm.group("col").strip()
+                rest = cm.group("rest").strip()
+                # strip trailing comma
+                if rest.endswith(","):
+                    rest = rest[:-1].rstrip()
+                cols[col_name] = Column(name=col_name, raw_def=_normalize_ws(rest))
+
+            i += 1
+
+        tables[table_name] = Table(name=table_name, columns=cols)
+
+        # move past ENGINE line if present
+        while i < len(lines) and not lines[i].strip().endswith(";"):
+            i += 1
+        if i < len(lines):
+            i += 1
+
+    return tables
+
+
+def diff_tables(
+    left: Dict[str, Table],
+    right: Dict[str, Table],
+) -> Tuple[List[str], List[str], Dict[str, Dict[str, Tuple[str | None, str | None]]]]:
+    """
+    Returns:
+    - left_only_tables
+    - right_only_tables
+    - per_table_column_diffs: {table: {col: (left_def, right_def)}}
+      where any of left_def/right_def can be None (missing), or both present but different.
+    """
+    left_only = sorted(set(left.keys()) - set(right.keys()))
+    right_only = sorted(set(right.keys()) - set(left.keys()))
+
+    common = sorted(set(left.keys()) & set(right.keys()))
+    per_table: Dict[str, Dict[str, Tuple[str | None, str | None]]] = {}
+
+    for t in common:
+        lcols = left[t].columns
+        rcols = right[t].columns
+
+        all_cols = sorted(set(lcols.keys()) | set(rcols.keys()))
+        diffs: Dict[str, Tuple[str | None, str | None]] = {}
+        for c in all_cols:
+            ldef = lcols.get(c).raw_def if c in lcols else None
+            rdef = rcols.get(c).raw_def if c in rcols else None
+            if ldef != rdef:
+                diffs[c] = (ldef, rdef)
+        if diffs:
+            per_table[t] = diffs
+
+    return left_only, right_only, per_table
+
+
+def _h(s: str) -> str:
+    return html.escape(s, quote=True)
+
+
+def render_html_report(
+    left_name: str,
+    right_name: str,
+    left_only_tables: List[str],
+    right_only_tables: List[str],
+    per_table_column_diffs: Dict[str, Dict[str, Tuple[str | None, str | None]]],
+) -> str:
+    diff_table_count = len(per_table_column_diffs)
+    diff_col_count = sum(len(v) for v in per_table_column_diffs.values())
+
+    parts: List[str] = []
+    parts.append("<!doctype html>")
+    parts.append('<html lang="zh-CN">')
+    parts.append("<head>")
+    parts.append('<meta charset="utf-8" />')
+    parts.append('<meta name="viewport" content="width=device-width,initial-scale=1" />')
+    parts.append("<title>CRM 表结构差异报告</title>")
+    parts.append(
+        """
+<style>
+  :root{
+    --bg:#0b1220;
+    --panel:#0f1b2d;
+    --panel2:#0c1626;
+    --text:#e5e7eb;
+    --muted:#94a3b8;
+    --border:#1f2a3b;
+    --danger:#ef4444;
+    --danger-bg:rgba(239,68,68,.12);
+    --ok:#10b981;
+  }
+  body{margin:0;background:var(--bg);color:var(--text);font:14px/1.5 -apple-system,BlinkMacSystemFont,"Segoe UI",Roboto,Helvetica,Arial,"PingFang SC","Hiragino Sans GB","Microsoft YaHei",sans-serif}
+  .wrap{max-width:1200px;margin:0 auto;padding:24px}
+  .header{display:flex;align-items:flex-end;justify-content:space-between;gap:16px;margin-bottom:16px}
+  .title{font-size:20px;font-weight:700}
+  .sub{color:var(--muted);font-size:12px}
+  .card{background:linear-gradient(180deg,var(--panel),var(--panel2));border:1px solid var(--border);border-radius:12px;padding:16px;margin:12px 0}
+  .grid{display:grid;grid-template-columns:1fr 1fr;gap:12px}
+  @media (max-width: 900px){.grid{grid-template-columns:1fr}}
+  .kpi{display:flex;gap:12px;flex-wrap:wrap}
+  .kpi .item{background:rgba(255,255,255,.03);border:1px solid var(--border);border-radius:10px;padding:10px 12px}
+  .kpi .num{font-weight:800;font-size:18px}
+  .kpi .lab{color:var(--muted);font-size:12px;margin-top:2px}
+  h2{margin:0 0 10px 0;font-size:16px}
+  ul{margin:0;padding-left:18px}
+  a{color:#60a5fa;text-decoration:none}
+  a:hover{text-decoration:underline}
+  table{width:100%;border-collapse:collapse;border:1px solid var(--border);border-radius:10px;overflow:hidden}
+  th,td{padding:10px 10px;border-bottom:1px solid var(--border);vertical-align:top}
+  th{background:rgba(255,255,255,.04);text-align:left;color:#cbd5e1;font-size:12px;letter-spacing:.02em}
+  tr:last-child td{border-bottom:none}
+  .diff-row td{background:var(--danger-bg)}
+  .diff{color:var(--danger);font-weight:700}
+  .mono{font-family:ui-monospace,SFMono-Regular,Menlo,Monaco,Consolas,"Liberation Mono","Courier New",monospace;white-space:pre-wrap;word-break:break-word}
+  .pill{display:inline-block;padding:2px 8px;border-radius:999px;border:1px solid var(--border);color:var(--muted);font-size:12px}
+  .toplinks{display:flex;gap:10px;flex-wrap:wrap}
+</style>
+""".strip()
+    )
+    parts.append("</head>")
+    parts.append("<body>")
+    parts.append('<div class="wrap">')
+    parts.append('<div class="header">')
+    parts.append('<div>')
+    parts.append('<div class="title">CRM 表结构差异报告</div>')
+    parts.append(
+        f'<div class="sub">仅比较 <span class="pill">crm_*</span> 表的字段定义（类型/长度、NULL、DEFAULT、COMMENT 等属性）。</div>'
+    )
+    parts.append("</div>")
+    parts.append(
+        f'<div class="sub">{_h(left_name)} vs {_h(right_name)}</div>'
+    )
+    parts.append("</div>")
+
+    parts.append('<div class="card kpi">')
+    parts.append(
+        f'<div class="item"><div class="num">{len(left_only_tables)}</div><div class="lab">{_h(left_name)} 独有表</div></div>'
+    )
+    parts.append(
+        f'<div class="item"><div class="num">{len(right_only_tables)}</div><div class="lab">{_h(right_name)} 独有表</div></div>'
+    )
+    parts.append(
+        f'<div class="item"><div class="num">{diff_table_count}</div><div class="lab">同名但字段有差异的表</div></div>'
+    )
+    parts.append(
+        f'<div class="item"><div class="num">{diff_col_count}</div><div class="lab">字段差异条目</div></div>'
+    )
+    parts.append("</div>")
+
+    parts.append('<div class="card grid">')
+    parts.append("<div>")
+    parts.append(f"<h2>{_h(left_name)} 比 {_h(right_name)} 多的表</h2>")
+    if left_only_tables:
+        parts.append("<ul>")
+        for t in left_only_tables:
+            parts.append(f"<li class='mono'>{_h(t)}</li>")
+        parts.append("</ul>")
+    else:
+        parts.append(f"<div class='sub'>无</div>")
+    parts.append("</div>")
+    parts.append("<div>")
+    parts.append(f"<h2>{_h(right_name)} 比 {_h(left_name)} 多的表</h2>")
+    if right_only_tables:
+        parts.append("<ul>")
+        for t in right_only_tables:
+            parts.append(f"<li class='mono'>{_h(t)}</li>")
+        parts.append("</ul>")
+    else:
+        parts.append(f"<div class='sub'>无</div>")
+    parts.append("</div>")
+    parts.append("</div>")
+
+    if per_table_column_diffs:
+        parts.append('<div class="card">')
+        parts.append("<h2>同名表字段差异（红色为差异）</h2>")
+        parts.append('<div class="toplinks sub">')
+        for t in sorted(per_table_column_diffs.keys()):
+            parts.append(f"<a href='#{_h(t)}' class='mono'>{_h(t)}</a>")
+        parts.append("</div>")
+        parts.append("</div>")
+
+    for t in sorted(per_table_column_diffs.keys()):
+        diffs = per_table_column_diffs[t]
+        parts.append(f"<div class='card' id='{_h(t)}'>")
+        parts.append(f"<h2 class='mono'>{_h(t)}</h2>")
+        parts.append("<table>")
+        parts.append("<thead>")
+        parts.append(
+            f"<tr><th style='width:220px'>字段</th><th>{_h(left_name)}</th><th>{_h(right_name)}</th></tr>"
+        )
+        parts.append("</thead>")
+        parts.append("<tbody>")
+        for col in sorted(diffs.keys()):
+            ldef, rdef = diffs[col]
+            ltxt = ldef if ldef is not None else "(缺失)"
+            rtxt = rdef if rdef is not None else "(缺失)"
+            parts.append("<tr class='diff-row'>")
+            parts.append(f"<td class='mono diff'>{_h(col)}</td>")
+            parts.append(f"<td class='mono diff'>{_h(ltxt)}</td>")
+            parts.append(f"<td class='mono diff'>{_h(rtxt)}</td>")
+            parts.append("</tr>")
+        parts.append("</tbody>")
+        parts.append("</table>")
+        parts.append("</div>")
+
+    parts.append("</div>")
+    parts.append("</body>")
+    parts.append("</html>")
+    return "\n".join(parts)
+
+
+def main() -> int:
+    p = argparse.ArgumentParser(description="Compare crm_* tables between two SQL dumps and output an HTML report.")
+    p.add_argument("--left", required=True, help="Left SQL file path (e.g. shanxia2.0.sql)")
+    p.add_argument("--right", required=True, help="Right SQL file path (e.g. sx.sql)")
+    p.add_argument("--out", required=True, help="Output HTML path")
+    p.add_argument("--left-name", default="shanxia2.0", help="Label for left side")
+    p.add_argument("--right-name", default="crm(sx)", help="Label for right side")
+    args = p.parse_args()
+
+    left_text = Path(args.left).read_text(encoding="utf-8", errors="ignore")
+    right_text = Path(args.right).read_text(encoding="utf-8", errors="ignore")
+
+    left_tables = parse_crm_tables(left_text)
+    right_tables = parse_crm_tables(right_text)
+
+    left_only, right_only, per_table = diff_tables(left_tables, right_tables)
+    report = render_html_report(args.left_name, args.right_name, left_only, right_only, per_table)
+
+    out_path = Path(args.out)
+    out_path.parent.mkdir(parents=True, exist_ok=True)
+    out_path.write_text(report, encoding="utf-8")
+
+    print(f"Parsed tables: left={len(left_tables)} right={len(right_tables)}")
+    print(f"Left-only tables: {len(left_only)} Right-only tables: {len(right_only)}")
+    print(f"Tables with column diffs: {len(per_table)}; diff items: {sum(len(v) for v in per_table.values())}")
+    print(f"Wrote HTML: {out_path}")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())
+