# -*- coding: utf-8 -*-
# bruno-sync 全量对账（票 04 t9）：源码 tags 钥匙 vs 文档仓文件夹钥匙 双向差集
# 钥匙 = (METHOD, url, tag全路径/文档文件夹相对路径)
# 修正版：注解区扫描（@Operation 可在 mapping 之前或之后），覆盖四种盲区：
#   @PreAuthorize 嵌套括号 / 裸字符串 tags / 无括号 @PostMapping / @Operation 在 @PostMapping 之后
import os, sys, io, re
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')
REPO = r'D:\code\crm-backend-matt'
DOCS = r'D:\code\crm-api-docs'
MODS = ['crm-auth','crm-file','crm-lead','crm-opportunity','crm-rule','crm-preference','crm-dict']
BT = chr(96)
src = {}
for m in MODS:
    base = os.path.join(REPO, m, 'src', 'main', 'java')
    for dp, dns, fns in os.walk(base):
        for fn in sorted(fns):
            if not fn.endswith('Controller.java'):
                continue
            txt = open(os.path.join(dp, fn), encoding='utf-8').read()
            mt = re.search(r'@RequestMapping\(\s*"([^"]+)"', txt)
            cls_prefix = mt.group(1) if mt else ''
            ct = re.search(r'@Tag\((?:[^()]*?)name\s*=\s*"([^"]+)"', txt) or re.search(r'@Tag\(\s*"([^"]+)"', txt)
            cls_tag = ct.group(1) if ct else ''
            for mm in re.finditer(r'@(Get|Post|Put|Delete|Patch)Mapping(?:\(([^)]*)\))?', txt):
                verb, args = mm.group(1), mm.group(2) or ''
                pm = re.search(r'"([^"]*)"', args)
                path = pm.group(1) if pm else ''
                wstart = max(0, mm.start() - 800)
                pa = re.search(r'public\s', txt[mm.end():mm.end()+600])
                bend = mm.end() + (pa.start() if pa else 600)
                region = txt[wstart:bend]
                gs = re.findall(r'@Operation\((?:[^()]|\([^()]*\))*\)', region)
                g = gs[-1] if gs else ''
                mtags = []
                if g:
                    t = re.search(r'tags\s*=\s*(\{[^}]*\}|"[^"]+")', g, re.S)
                    if t:
                        mtags = re.findall(r'"([^"]+)"', t.group(1))
                tags = mtags if mtags else ([cls_tag] if cls_tag else [])
                url = cls_prefix + path
                for t in tags:
                    src.setdefault((verb.upper(), url, t), []).append(fn)

doc = {}
for dp, dns, fns in os.walk(DOCS):
    dns[:] = [d for d in dns if d.lower() != 'environments']
    rel = os.path.relpath(dp, DOCS).replace('\\', '/')
    tag = '' if rel == '.' else rel
    for fn in sorted(fns):
        if not fn.endswith('.bru') or fn in ('folder.bru', 'collection.bru'):
            continue
        txt = open(os.path.join(dp, fn), encoding='utf-8').read()
        mv = re.search(r'^[ \t]*(get|post|put|delete|patch)[ \t]*\{', txt, re.M | re.I)
        if not mv:
            continue
        verb = mv.group(1).upper()
        ul = re.search(r'url:[^\n]*', txt)
        if not ul:
            continue
        u = ul.group(0).split(':', 1)[1].strip()
        if u.startswith(BT) and u.endswith(BT):
            u = u[1:-1]
        u = re.sub(r'\{\{[^}]*\}\}', '', u).strip()
        u = re.sub(r'^https?://[^/\s]+', '', u).strip()
        u = re.sub(r'^/crm-api(?=/api/)', '', u).strip()
        u = u.split('?', 1)[0].strip()
        doc.setdefault((verb, u, tag), []).append(fn)

miss_doc = sorted(set(src) - set(doc))
miss_src = sorted(set(doc) - set(src))
print('src keys:', len(src), '; doc keys:', len(doc))
print('--- doc-has-src-missing', len(miss_doc), '---')
for k in miss_doc:
    print(' | '.join(k), ' <=', ','.join(src[k]))
print('--- src-has-doc-missing', len(miss_src), '---')
for k in miss_src:
    print(' | '.join(k), ' <=', ','.join(doc[k]))