From 5e08b8c53e804036b0ec66abc3a3e5202ef4605a Mon Sep 17 00:00:00 2001
From: HuangHai <10402852@qq.com>
Date: Fri, 27 Jun 2025 14:35:03 +0800
Subject: [PATCH] 'commit'

---
 dsRag/ElasticSearch/T6_XiangLiangQuery.py     |  53 +++++++++++-------
 dsRag/Util/EsSearchUtil.py                    |   2 +-
 .../__pycache__/EsSearchUtil.cpython-310.pyc  | Bin 0 -> 2627 bytes
 3 files changed, 33 insertions(+), 22 deletions(-)
 create mode 100644 dsRag/Util/__pycache__/EsSearchUtil.cpython-310.pyc

diff --git a/dsRag/ElasticSearch/T6_XiangLiangQuery.py b/dsRag/ElasticSearch/T6_XiangLiangQuery.py
index 07a27b4e..4d6a8466 100644
--- a/dsRag/ElasticSearch/T6_XiangLiangQuery.py
+++ b/dsRag/ElasticSearch/T6_XiangLiangQuery.py
@@ -35,23 +35,11 @@ def init_es_pool():
     return es_pool
 
 
-# 将文本转换为嵌入向量
-def text_to_embedding(text):
-    words = jieba.lcut(text)  # 使用 jieba 分词
-    print(f"文本: {text}, 分词结果: {words}")
-    try:
-        embeddings = [model[word] for word in words if word in model]
-        logger.info(f"有效词向量数量: {len(embeddings)}")
-        if embeddings:
-            avg_embedding = sum(embeddings) / len(embeddings)
-            logger.info(f"生成的平均向量: {avg_embedding[:5]}...")  # 打印前 5 维
-            return avg_embedding
-        else:
-            logger.warning("未找到有效词，返回零向量")
-            return [0.0] * model.vector_size
-    except Exception as e:
-        logger.error(f"向量转换失败: {str(e)}")
-        return [0.0] * model.vector_size
+# 导入EsSearchUtil
+from Util.EsSearchUtil import EsSearchUtil
+
+# 初始化EsSearchUtil
+es_search_util = EsSearchUtil(ES_CONFIG)
 
 
 def main():
@@ -60,6 +48,7 @@ def main():
     
     # 测试查询
     query = "小学数学中有哪些模型"
+    query_tags = ["MATH_1"]  # 默认搜索标签，可修改
     print(f"\n=== 开始执行查询 ===")
     print(f"原始查询文本: {query}")
 
@@ -69,7 +58,7 @@ def main():
         # 向量搜索
         print("\n=== 向量搜索阶段 ===")
         print("1. 文本分词和向量化处理中...")
-        query_embedding = text_to_embedding(query)
+        query_embedding = es_search_util.text_to_embedding(query)
         print(f"2. 生成的查询向量维度: {len(query_embedding)}")
         print(f"3. 前5维向量值: {query_embedding[:5]}")
         
@@ -79,7 +68,18 @@ def main():
             body={
                 "query": {
                     "script_score": {
-                        "query": {"match_all": {}},
+                        "query": {
+                            "bool": {
+                                "should": [
+                                    {
+                                        "terms": {
+                                            "tags.tags": query_tags
+                                        }
+                                    }
+                                ],
+                                "minimum_should_match": 1
+                            }
+                        },
                         "script": {
                             "source": "double score = cosineSimilarity(params.query_vector, 'embedding'); return score >= 0 ? score : 0",
                             "params": {"query_vector": query_embedding}
@@ -98,8 +98,19 @@ def main():
             index=ES_CONFIG['index_name'],
             body={
                 "query": {
-                    "match": {
-                        "user_input": query
+                    "bool": {
+                        "must": [
+                            {
+                                "match": {
+                                    "user_input": query
+                                }
+                            },
+                            {
+                                "terms": {
+                                    "tags.tags": query_tags
+                                }
+                            }
+                        ]
                     }
                 },
                 "size": 5
diff --git a/dsRag/Util/EsSearchUtil.py b/dsRag/Util/EsSearchUtil.py
index 5be19c8a..c0e436df 100644
--- a/dsRag/Util/EsSearchUtil.py
+++ b/dsRag/Util/EsSearchUtil.py
@@ -11,7 +11,7 @@ class EsSearchUtil:
         self.es_config = es_config
         self.es_conn = Elasticsearch(
             hosts=es_config['hosts'],
-            basic_auth=(es_config['username'], es_config['password']),
+            basic_auth=es_config['basic_auth'],
             verify_certs=False
         )
 
diff --git a/dsRag/Util/__pycache__/EsSearchUtil.cpython-310.pyc b/dsRag/Util/__pycache__/EsSearchUtil.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..b0a8cb27fee52a7b6c3b705436dac01b565a8512
GIT binary patch
literal 2627
zcmb7G&2Jk;6yKTsu;Z-LG!$qnQC1-2nnqC%oLt%>(ZVf4R0zqcST;M`W>eei&g_;t
zawQ}ZHB}QD1QY?GO$t?2LHUr7QmI88LE=A{3kPbuIrWaz!kZa;W2YBbYiHlQdGqn!
z@Auv~%I9+$Jm(L*GyTVqru`*4{f`cvSKwt6&=8HFidIEsgoKPMxQyjjFB`9GL?_0a
zMvM<|*(CUwRx~$Yj#flGf3)JUIP_WIQGa^U7wY~e-Q);p1FejR1`8vC=d`k}mMt4(
zlvpG)hhXthk|jeRSs>-eFi07)o#e>~NLjKCwjSg5QFcsj_f{NMeB8I4kWF}Bg_k`C
zO%pXWSBp^-5wr_>1n-j0Ea{-f#Dw*%<a030Io+qV@Kx)^#n$5G0ms_QE1jROwbrk<
z=GRlxQr)9o)d`sEN3|2-$&xcLXJdY;^UL?GTc5S&Z$G%Z*jk)#Eq^;5v6ziJ2@7b=
zs|Mpv-DB*%h>~$9tdZcXE3}<k7t&3*DW~w<l*dBf^^$n{wW3Z9I5QQla{Ej`!xIhH
z4`|FLi#i`l4@5;gct-VWMa&HrR8G(gXm4s%u;cHEBc<bny&ch0@OgtqGl3t2^rm<6
zxQOoY{!mZU8$9c}VJ(bZ*Miy1hQ=~6vfzi2{c!fl0sEfq07l~Ef)x^@Js`x|D&Vt*
z&fyr(=&J?+dDUFg&go4Z@WGQCzk^lL;^s^kOnKa>_({xht<Ej#)ksw3hWK(Tjw&IG
ziv|U@Yut!~voSYCY#F!CfL{@nKC5O}*lZUzZXCHmbt)hvteqSKH50z&&?qv{?4ALK
z9y8w%LG*<?y+>e@Qt%LTn%XK558TrQ?L#d@g7!@vzUB%h)(Q^MIisn~pcxFUPBK7L
zqlrm&Nw_~W$!+G-upYb2k7!UHs(P_M?Ru5U?7JjNrYeCW2b_bBAF;3&91E*q#iL=|
z*eeOjCQc^-ZMZ4~<4z%Uwop9a&>&7|t-HjbgU){EaJN%(_H(;`DiFwGQ9}KITWV#{
zn9>3U<QaKxlf^f6Zpt2g2EqW`3uglW=V|EBtk77hXh=R5O^KQ`r6P0rHnnwn*m5rn
zQoj|+yz<UfONu?%ZfJ4{W1uFy#v}OsW970yR3Plb+1&%k4on{aRDhcT>OcoT0<fTk
zItJAQ6TOK-_yJ^wgv^|z&w?E{B~m;mk{z<9p2SqlhwK&Z)8|DaauF1Qv75ecb#7Sl
zPPtYDOD0%np)3OTgeyA-T<J3&a6~vEl20a{ERK?yO}vI7)npP%8MxVmBwP`79w)!S
zjI<cFFWk7lw%A@@Z!IsVlK9~6LTmY>jq{gOcVGMR`S!|py|K+Q)~%HiyON{E+gGpO
z|KXaNk#5m76l5tm_kaDQ_2re$osZjBR@$Hb2-|GhDdLtKRafub{-$$dNv*5wJ73JH
z+M7-RwvKdhrU6STG2^xjR<~Eg^ab#i?uF(tK2i~Q`l4t=^!tiPe!4M5L(-+*ei)Sa
z*clsP9q@Uyg=z!l+)`a%Ak%w<oFerSbbwmu6s9><OPcx`U;{b~x*c8Qluld4aOFw4
zs3(8I)YSXVovZh5zu%4ghu2$|zD|e!H#qY4b>l`IFzUwqr#r1H7l92072yKZ84=rp
z(_Xmj6qJ*N2X`N1#D536AKdil^eOO>?tlieC`FI%6AE<^sOmGUvPYg-)|V|P%4|Fg
zvm|!etPR;^BOA{?F_2qomB}KaJH-x|j>AXEL*i}}x3}x^oa<I2l2n9jyYA`4tEdqw
zZcJ4I2M?>N!6vh$B<?8Mt3PZ6dRZ-tvUbkQjpZ^U+vFHrjQODdPnf**sj%i|EveQU
cbhNilpVR3EJo~ab>mdYYOq3>D?1IMdzjoy6ivR!s

literal 0
HcmV?d00001