修正

identify_duplicates_detailed.py

@@ -0,0 +1,126 @@
+import psycopg2
+
+conn = psycopg2.connect(
+    host="192.168.0.61",
+    port=55432,
+    database="njts_acct",
+    user="njts_app",
+    password="njts_app2025"
+)
+
+cur = conn.cursor()
+
+print("=" * 120)
+print("【重复仕訳分析】按描述和金额分组")
+print("=" * 120)
+
+# 查找疑似重复的记录：相同描述和金额，但日期不同
+cur.execute("""
+    SELECT 
+        je.description,
+        COALESCE(SUM(jl.debit), 0) as debit_total,
+        COALESCE(SUM(jl.credit), 0) as credit_total,
+        COUNT(*) as cnt,
+        STRING_AGG(
+            je.journal_entry_id || '(' || TO_CHAR(je.entry_date, 'YYYY-MM-DD') || ')',
+            ' | ' 
+            ORDER BY je.entry_date
+        ) as entries
+    FROM journal_entries je
+    LEFT JOIN journal_lines jl ON je.journal_entry_id = jl.journal_entry_id
+    WHERE je.is_deleted = false
+      AND EXTRACT(YEAR FROM je.entry_date) = 2025
+      AND EXTRACT(MONTH FROM je.entry_date) IN (6, 7, 8)
+    GROUP BY je.description
+    HAVING COUNT(*) > 1
+    ORDER BY cnt DESC
+""")
+
+results = cur.fetchall()
+
+if results:
+    print("\n疑似重复记录（相同描述，多个月份）:\n")
+    for row in results:
+        description, debit, credit, cnt, entries = row
+        print(f"【{cnt}件】{description}")
+        print(f"  金額: 借={debit:>10,.0f}  貸={credit:>10,.0f}")
+        print(f"  仕訳: {entries}")
+        print()
+else:
+    print("未找到疑似重复的记录")
+
+# 查询6月的所有记录
+print("\n" + "=" * 120)
+print("【6月の全仕訳】")
+print("=" * 120)
+
+cur.execute("""
+    SELECT 
+        je.journal_entry_id,
+        je.entry_date,
+        je.description,
+        COALESCE(SUM(jl.debit), 0) as debit_total,
+        COALESCE(SUM(jl.credit), 0) as credit_total
+    FROM journal_entries je
+    LEFT JOIN journal_lines jl ON je.journal_entry_id = jl.journal_entry_id
+    WHERE je.is_deleted = false
+      AND EXTRACT(YEAR FROM je.entry_date) = 2025
+      AND EXTRACT(MONTH FROM je.entry_date) = 6
+    GROUP BY je.journal_entry_id, je.entry_date, je.description
+    ORDER BY je.journal_entry_id
+""")
+
+june_records = cur.fetchall()
+print(f"\n全{len(june_records)}件:")
+for row in june_records:
+    entry_id, entry_date, description, debit, credit = row
+    print(f"ID={entry_id:3d} | {entry_date} | 借={debit:>10,.0f}  貸={credit:>10,.0f} | {description[:50]}")
+
+# 查询只在7月、8月出现但6月没有的相同记录
+print("\n" + "=" * 120)
+print("【可能通过复制创建的记录（6月有，7月/8月也有相同金额）】")
+print("=" * 120)
+
+cur.execute("""
+    WITH june_records AS (
+        SELECT 
+            je.description,
+            COALESCE(SUM(jl.debit), 0) as debit_total,
+            COALESCE(SUM(jl.credit), 0) as credit_total
+        FROM journal_entries je
+        LEFT JOIN journal_lines jl ON je.journal_entry_id = jl.journal_entry_id
+        WHERE je.is_deleted = false
+          AND EXTRACT(YEAR FROM je.entry_date) = 2025
+          AND EXTRACT(MONTH FROM je.entry_date) = 6
+        GROUP BY je.description
+    )
+    SELECT 
+        je.journal_entry_id,
+        TO_CHAR(je.entry_date, 'YYYY-MM') as month,
+        je.description,
+        COALESCE(SUM(jl.debit), 0) as debit_total,
+        COALESCE(SUM(jl.credit), 0) as credit_total
+    FROM journal_entries je
+    LEFT JOIN journal_lines jl ON je.journal_entry_id = jl.journal_entry_id
+    INNER JOIN june_records jr ON je.description = jr.description 
+        AND COALESCE(SUM(jl.debit), 0) = jr.debit_total
+        AND COALESCE(SUM(jl.credit), 0) = jr.credit_total
+    WHERE je.is_deleted = false
+      AND EXTRACT(YEAR FROM je.entry_date) = 2025
+      AND EXTRACT(MONTH FROM je.entry_date) IN (7, 8)
+    GROUP BY je.journal_entry_id, je.entry_date, je.description
+    ORDER BY je.entry_date, je.journal_entry_id
+""")
+
+duplicates = cur.fetchall()
+
+if duplicates:
+    print(f"\n找到{len(duplicates)}件可能通过复制创建的记录:\n")
+    for row in duplicates:
+        entry_id, month, description, debit, credit = row
+        print(f"ID={entry_id:3d} | {month} | 借={debit:>10,.0f}  貸={credit:>10,.0f} | {description[:50]}")
+else:
+    print("\n未找到可能通过复制创建的记录")
+
+cur.close()
+conn.close()