From 441d555caf68b73d22c5d36bdcbfb20a2558569d Mon Sep 17 00:00:00 2001
From: RoyirinRosmontis <support@rrrosmontis.icu>
Date: Mon, 3 Aug 2026 13:30:42 +0800
Subject: [PATCH] =?UTF-8?q?=E6=96=B0=E5=B2=97=E4=BD=8D=E5=8F=91=E7=8E=B0?=
 =?UTF-8?q?=EF=BC=9A=E5=A2=9E=E9=95=BF=E7=8E=87=E6=94=B9=E5=8F=8C=E5=B9=B4?=
 =?UTF-8?q?=E7=AA=97=E5=8F=A3=E5=8F=A3=E5=BE=84=EF=BC=88=E4=BF=AE=E5=A4=8D?=
 =?UTF-8?q?=E6=88=90=E7=86=9F=E5=B2=97=E4=BD=8D=E8=A2=AB=E8=AF=AF=E6=9D=80?=
 =?UTF-8?q?=EF=BC=89=EF=BC=8C=E6=B5=8B=E8=AF=95=E5=AF=B9=E9=BD=90=E5=9B=9B?=
 =?UTF-8?q?=E6=8C=87=E6=A0=87=E9=98=88=E5=80=BC?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 app/discover/cluster.py   | 9 ++++++---
 tests/test_persistence.py | 4 ++--
 2 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/app/discover/cluster.py b/app/discover/cluster.py
index 352f34d..82e1262 100644
--- a/app/discover/cluster.py
+++ b/app/discover/cluster.py
@@ -143,9 +143,12 @@ def discover_new_jobs(
             continue
         years = Counter(r.publish_date.year for r in group)
         latest_year = max(years)
-        prev_year_n = years.get(latest_year - 1, 0)
-        # 前一年没有基数视为“从无到有”，增长率视为达标
-        growth = years[latest_year] / prev_year_n if prev_year_n else float("inf")
+        # 增长率：最近两年合计 / 之前两年合计（从无到有视为达标）。
+        # 按年同比会在岗位成熟后回落（如 2026/2025=1.43），
+        # 双年窗口更能刻画"正在兴起"的全周期趋势（与方案"增长率≥50%"一致）。
+        recent2 = years[latest_year] + years.get(latest_year - 1, 0)
+        before2 = years.get(latest_year - 2, 0) + years.get(latest_year - 3, 0)
+        growth = recent2 / before2 if before2 else float("inf")
         if growth < min_growth:
             continue
         # 差异度：与最近既有岗位质心的距离，过近说明只是既有岗位的变体
diff --git a/tests/test_persistence.py b/tests/test_persistence.py
index a3565aa..dc7753c 100644
--- a/tests/test_persistence.py
+++ b/tests/test_persistence.py
@@ -172,7 +172,7 @@ def test_discover_candidate_carries_support_evidence():
     from app.models import ExtractionResult
 
     records, extractions = [], {}
-    for i in range(9):
+    for i in range(15):  # ≥ MIN_CLUSTER_SIZE(15)，5 家企业 ≥ MIN_COMPANIES(5)
         r = JDRecord(jid=f"c{i}", job_title="提示词工程师", company=f"公司{i % 5}",
                      industry="人工智能", publish_date=date(2025, i % 6 + 1, 1),
                      text="负责大模型提示词设计与优化", source_id=f"t.jsonl#{i + 1}")
@@ -183,7 +183,7 @@ def test_discover_candidate_carries_support_evidence():
     cands = cluster.discover_new_jobs(records, extractions, known_titles=set())
     assert len(cands) == 1
     c = cands[0]
-    assert len(c.support_source_ids) == 9
+    assert len(c.support_source_ids) == 15
     assert min(c.support_windows) == "2025Q1"
     assert c.industries == ["人工智能"]