From 7b6bc7c126657ae943f3d37e5f19133d8bb187c5 Mon Sep 17 00:00:00 2001
From: zhizhijie <zhizhijie@users.noreply.gitee.com>
Date: 星期四, 17 九月 2026 14:54:44 +0800
Subject: [PATCH] fix(投资): 全部改为「按表头定位」——目标明细表不再写死列号 + 同名列按数据消歧
---
docs/投资/工具/格式探测.py | 77 +++++++++++++++++++++++++++++++++-----
1 files changed, 67 insertions(+), 10 deletions(-)
diff --git "a/docs/\346\212\225\350\265\204/\345\267\245\345\205\267/\346\240\274\345\274\217\346\216\242\346\265\213.py" "b/docs/\346\212\225\350\265\204/\345\267\245\345\205\267/\346\240\274\345\274\217\346\216\242\346\265\213.py"
index f23585c..a6ed4dd 100644
--- "a/docs/\346\212\225\350\265\204/\345\267\245\345\205\267/\346\240\274\345\274\217\346\216\242\346\265\213.py"
+++ "b/docs/\346\212\225\350\265\204/\345\267\245\345\205\267/\346\240\274\345\274\217\346\216\242\346\265\213.py"
@@ -193,10 +193,66 @@
return bool(re.match(r'^-?\d+(\.\d+)?$', norm(v)))
-def profile(cell, nrows, ncols, hits, city=None):
- namecol = None
- if '椤圭洰鍚嶇О' in hits:
- namecol = hits['椤圭洰鍚嶇О'][0][1]
+def _num_count(cell, nrows, col, start_row1):
+ """缁熻璇ュ垪鍦ㄨ〃澶磋浠ヤ笅鐨勬暟鍊煎崟鍏冩牸涓暟锛堢敤浜庡尯鍒嗗悓鍚嶅垪锛屽銆岃嚜寮�濮嬪缓璁剧疮璁″畬鎴愭姇璧勩�嶄笌銆屸�︽柊澧炲缓绛戦潰绉�嶏級銆�"""
+ n = 0
+ for r in range(start_row1, min(nrows, 800)):
+ v = cell(r, col)
+ if isinstance(v, (int, float)) and not isinstance(v, bool):
+ n += 1
+ return n
+
+
+def pick_columns(cell, nrows, ncols, merges=None, fields=None):
+ """鎸夎〃澶村叧閿瓧瀹氫綅鍒� 鈫� ({瀛楁: 0 鍩哄垪鍙穧, hits)銆�
+
+ 涓ょ骇琛ㄥご娑堟锛氳嫢鏌愬瓧娈电殑鍛戒腑钀藉湪**璺ㄥ鍒楃殑鍚堝苟鍗曞厓鏍�**閲岋紝璇存槑瀹冩槸缁勮〃澶达紝涓嶈兘鐩存帴褰撹瀛楁鐨勫垪锛�
+ 姝ゆ椂鍦ㄦ湰鍚堝苟鍖哄唴鎺掗櫎銆屽叾瀹冨瓧娈垫洿娣辩殑瀛愯〃澶淬�嶆墍鍗犵殑鍒楋紝鑻ュ彧鍓╁敮涓�涓�鍒楋紝鍒欒鍒楁墠鏄湰瀛楁鐨勫垪銆�
+ 渚嬶細閯傚窞銆婁紒涓氭湀鎶ユ眹鎬汇�婻5 缁勮〃澶淬�岃嚜骞村垵绱瀹屾垚鎶曡祫锛堜竾鍏冿級銆嶅悎骞� I5:K5锛�
+ 瀛愯〃澶� I=鏈勾璁″垝鎶曡祫銆丣=鍚堣銆並=鍏朵腑锛氭湰鏈堝畬鎴愭姇璧� 鈫�
+ 銆岃嚜骞村垵绱銆嶈惤鍦� J锛堜笌鎶曡祫绯荤粺 G 鍒椾竴鑷达級锛岃�屼笉鏄� I锛堥偅鏄湰骞磋鍒掓姇璧勶級銆�
+ """
+ hits = scan_fields(cell, nrows, ncols)
+ fields = list(fields) if fields else [k for k, _ in FIELD_PAT]
+ best = {}
+ for f in fields:
+ if f not in hits:
+ continue
+ uniq, seen = [], set()
+ for h in sorted(hits[f], key=lambda h: (h[0], colindex(h[1]))):
+ if h[1] in seen:
+ continue
+ seen.add(h[1]); uniq.append(h)
+ # 鍚屽悕鍒楁秷姝э細浼樺厛銆岃〃澶翠笅鏂圭湡鐨勬湁鏁板瓧銆嶇殑閭ｄ竴鍒椼��
+ # 渚嬶細姝︽眽鐗╂祦銆岃嚜寮�濮嬪缓璁剧疮璁″畬鎴愭姇璧勶紙涓囧厓锛夈�岺 涓庛�岃嚜寮�濮嬪缓璁剧疮璁℃柊澧炲缓绛戦潰绉�峆 鍚屽悕锛�
+ # 鍙湁 H 鍒楁湁鏁板瓧 鈫� 鍙� H锛堟寜鈥滃懡涓瘝鏈�鐭�濅細閿欏彇 P锛岄偅鏄缓绛戦潰绉級銆�
+ withnum = [h for h in uniq if _num_count(cell, nrows, colindex(h[1]), h[0]) > 0]
+ pool = withnum or uniq
+ best[f] = sorted(pool, key=lambda h: (h[0], colindex(h[1])))[0]
+ cols = {}
+ for f, h in best.items():
+ row1, letter, _txt = h
+ c0 = colindex(letter)
+ span = merge_span(merges, row1, c0)
+ if span and (span[1] - span[0]) > 1:
+ taken = set()
+ for f2, h2 in best.items():
+ if f2 == f or h2[0] <= row1:
+ continue
+ c2 = colindex(h2[1])
+ if span[0] <= c2 < span[1]:
+ taken.add(c2)
+ free = [c for c in range(span[0], span[1]) if c not in taken]
+ if len(free) == 1:
+ cols[f] = free[0]
+ continue
+ cols[f] = c0
+ return cols, hits
+
+
+def profile(cell, nrows, ncols, hits, city=None, merges=None):
+ cols, _ = pick_columns(cell, nrows, ncols, merges)
+ namecol = colname(cols['椤圭洰鍚嶇О']) if '椤圭洰鍚嶇О' in cols else None
data_start = data_end = None
project_rows = 0
groups = []
@@ -211,8 +267,8 @@
hi = segs[i + 1][0] if i + 1 < len(segs) else 10 ** 9
break
if namecol is not None:
- nc = colindex(namecol)
- anchors = [colindex(hits[k][0][1]) for k in ('鑷紑濮嬪缓璁剧疮璁�', '鑷勾鍒濈疮璁�', '鎬绘姇璧�') if k in hits]
+ nc = cols['椤圭洰鍚嶇О']
+ anchors = [cols[k] for k in ('鑷紑濮嬪缓璁剧疮璁�', '鑷勾鍒濈疮璁�', '鎬绘姇璧�') if k in cols]
for r in range(min(nrows, 800)):
if not (lo <= r + 1 < hi):
continue
@@ -235,7 +291,7 @@
data_start = data_start or r + 1
data_end = r + 1
project_rows += 1
- return dict(namecol=namecol, data_start=data_start, data_end=data_end,
+ return dict(namecol=namecol, cols=cols, data_start=data_start, data_end=data_end,
project_rows=project_rows, groups=groups[:6],
city=city or '', city_row=city_row or '', segments=segs[:8],
header_rows=sorted({h[0] for v in hits.values() for h in v}))
@@ -302,7 +358,8 @@
sel = pick_sheet(open_sheets(p), ym=ym)
if sel:
nm, nr, nc, cell, hits = sel
- pr = profile(cell, nr, nc, hits, city=file_city)
+ pr = profile(cell, nr, nc, hits, city=file_city,
+ merges=sheet_merges(p).get(nm, []))
rec.update(宸ヤ綔琛�=nm, 琛ㄥご琛�=','.join(map(str, pr['header_rows'])),
鏁版嵁璧�=pr['data_start'], 鏁版嵁姝�=pr['data_end'],
椤圭洰琛屾暟=pr['project_rows'], 鍚嶇О鍒�=pr['namecol'] or '',
@@ -310,8 +367,8 @@
甯傚窞=pr['city'] or (file_city or ''), 甯傚窞琛�=pr['city_row'],
甯傚窞娈�=';'.join('R%s:%s' % s for s in pr['segments']))
for k in KEY:
- if k in hits:
- rec[k + '鍒�'] = hits[k][0][1]
+ if k in pr['cols']:
+ rec[k + '鍒�'] = colname(pr['cols'][k])
except Exception as e:
rec['閿欒'] = '%s: %s' % (type(e).__name__, e)
rows.append(rec)
--
Gitblit v1.9.1