panyy
/
UnisKB
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

refactor: optimize document matching logic for tag conditions

v3.2
CaptainB 2025-10-23 17:43:28 +08:00
parent 27b74580a4
commit 4cda73ba57
1 changed files with 43 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

67
apps/application/flow/step_node/search_document_node/impl/base_search_document_node.py
View File

@ -121,44 +121,63 @@ class BaseSearchDocumentNode(ISearchDocumentStepNode):
compare_type = condition['compare'] compare_type = condition['compare']
# 构建查询条件 # 构建查询条件
if compare_type == 'contain': if compare_type == 'not_contain':
q_filter = Q(tag__key=tag_key, tag__value__icontains=field_value) # 反向查询:找出包含该标签的文档,然后排除
elif compare_type == 'eq': exclude_docs = set(QuerySet(DocumentTag).filter(
q_filter = Q(tag__key=tag_key, tag__value=field_value) document_id__in=matched_doc_ids,
elif compare_type == 'not_contain': tag__key=tag_key,
q_filter = ~Q(tag__key=tag_key, tag__value__icontains=field_value) tag__value__icontains=field_value
).values_list('document_id', flat=True).distinct())
matched_doc_ids = matched_doc_ids - exclude_docs
else: else:
continue if compare_type == 'contain':
q_filter = Q(tag__key=tag_key, tag__value__icontains=field_value)
elif compare_type == 'eq':
q_filter = Q(tag__key=tag_key, tag__value=field_value)
else:
continue
# 单次查询获取符合条件的文档 # 单次查询获取符合条件的文档
tag_docs = set(QuerySet(DocumentTag).filter( tag_docs = set(QuerySet(DocumentTag).filter(
document_id__in=matched_doc_ids document_id__in=matched_doc_ids
).filter(q_filter).values_list('document_id', flat=True).distinct()) ).filter(q_filter).values_list('document_id', flat=True).distinct())
matched_doc_ids = matched_doc_ids.intersection(tag_docs) matched_doc_ids = matched_doc_ids.intersection(tag_docs)
return matched_doc_ids return matched_doc_ids
else: else:
# OR逻辑:使用一次查询完成 # OR逻辑
q_objects = Q() matched_docs = set()
for condition in search_condition_list: for condition in search_condition_list:
tag_key = condition['key'] tag_key = condition['key']
field_value = self.workflow_manage.generate_prompt(condition['value']) field_value = self.workflow_manage.generate_prompt(condition['value'])
compare_type = condition['compare'] compare_type = condition['compare']
if compare_type == 'contain': if compare_type == 'not_contain':
q_objects |= Q(tag__key=tag_key, tag__value__icontains=field_value) # 反向查询:找出包含该标签的文档,然后用全集减去
elif compare_type == 'eq': exclude_docs = set(QuerySet(DocumentTag).filter(
q_objects |= Q(tag__key=tag_key, tag__value=field_value) document_id__in=document_id_list,
elif compare_type == 'not_contain': tag__key=tag_key,
q_objects |= ~Q(tag__key=tag_key, tag__value__icontains=field_value) tag__value__icontains=field_value
).values_list('document_id', flat=True).distinct())
# 一次查询获取所有匹配的文档 matched_docs = matched_docs.union(set(document_id_list) - exclude_docs)
matched_docs = set(QuerySet(DocumentTag).filter( else:
document_id__in=document_id_list if compare_type == 'contain':
).filter(q_objects).values_list('document_id', flat=True).distinct()) q_filter = Q(tag__key=tag_key, tag__value__icontains=field_value)
elif compare_type == 'eq':
q_filter = Q(tag__key=tag_key, tag__value=field_value)
else:
continue
docs = set(QuerySet(DocumentTag).filter(
document_id__in=document_id_list
).filter(q_filter).values_list('document_id', flat=True).distinct())
matched_docs = matched_docs.union(docs)
return matched_docs return matched_docs