AI/ragflow
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/infiniflow/ragflow.git synced 2025-08-13 18:55:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

Make infinity adapt (#4635)

Browse Source 
### What problem does this PR solve?

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
Kevin Hu 2025-01-24 17:45:04 +08:00 committed by GitHub
parent 9d717f0b6e
commit 71c132f76d
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 1131 additions and 1124 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2236
agent/templates/research_report.json
View File

File diff suppressed because one or more lines are too long

2
conf/infinity_mapping.json
View File

@ -30,7 +30,7 @@
"knowledge_graph_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace"}, "knowledge_graph_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace"},
"entities_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace"}, "entities_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace"},
"pagerank_fea": {"type": "integer", "default": 0}, "pagerank_fea": {"type": "integer", "default": 0},
"tag_feas": {"type": "integer", "default": 0}, "tag_feas": {"type": "varchar", "default": ""},
"important_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace"}, "important_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace"},
"from_entity_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace"}, "from_entity_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace"},

2
graphrag/utils.py
View File

@ -484,7 +484,7 @@ def update_nodes_pagerank_nhop_neighbour(tenant_id, kb_id, graph, n_hop):
chunk, chunk,
search.index_name(tenant_id), kb_id) search.index_name(tenant_id), kb_id)
else: else:
settings.docStoreConn.insert([{"id": chunk_id(chunk), **chunk}], search.index_name(tenant_id)) settings.docStoreConn.insert([{"id": chunk_id(chunk), **chunk}], search.index_name(tenant_id), kb_id)
def get_entity_type2sampels(idxnms, kb_ids: list): def get_entity_type2sampels(idxnms, kb_ids: list):

15
rag/utils/infinity_conn.py
View File

@ -299,7 +299,7 @@ class InfinityConnection(DocStoreConnection):
matchExpr.extra_options[k] = str(v) matchExpr.extra_options[k] = str(v)
logger.debug(f"INFINITY search MatchTextExpr: {json.dumps(matchExpr.__dict__)}") logger.debug(f"INFINITY search MatchTextExpr: {json.dumps(matchExpr.__dict__)}")
elif isinstance(matchExpr, MatchDenseExpr): elif isinstance(matchExpr, MatchDenseExpr):
if filter_cond and "filter" not in matchExpr.extra_options: if filter_fulltext and filter_cond and "filter" not in matchExpr.extra_options:
matchExpr.extra_options.update({"filter": filter_fulltext}) matchExpr.extra_options.update({"filter": filter_fulltext})
for k, v in matchExpr.extra_options.items(): for k, v in matchExpr.extra_options.items():
if not isinstance(v, str): if not isinstance(v, str):
@ -424,9 +424,11 @@ class InfinityConnection(DocStoreConnection):
assert "_id" not in d assert "_id" not in d
assert "id" in d assert "id" in d
for k, v in d.items(): for k, v in d.items():
if k in ["important_kwd", "question_kwd", "entities_kwd"]: if k in ["important_kwd", "question_kwd", "entities_kwd", "tag_kwd"]:
assert isinstance(v, list) assert isinstance(v, list)
d[k] = "###".join(v) d[k] = "###".join(v)
elif re.search(r"_feas$", k):
d[k] = json.dumps(v)
elif k == 'kb_id': elif k == 'kb_id':
if isinstance(d[k], list): if isinstance(d[k], list):
d[k] = d[k][0] # since d[k] is a list, but we need a str d[k] = d[k][0] # since d[k] is a list, but we need a str
@ -462,7 +464,12 @@ class InfinityConnection(DocStoreConnection):
del condition["exist"] del condition["exist"]
filter = equivalent_condition_to_str(condition) filter = equivalent_condition_to_str(condition)
for k, v in list(newValue.items()): for k, v in list(newValue.items()):
if k.endswith("_kwd") and isinstance(v, list): if k in ["important_kwd", "question_kwd", "entities_kwd", "tag_kwd"]:
assert isinstance(v, list)
newValue[k] = "###".join(v)
elif re.search(r"_feas$", k):
newValue[k] = json.dumps(v)
elif k.endswith("_kwd") and isinstance(v, list):
newValue[k] = " ".join(v) newValue[k] = " ".join(v)
elif k == 'kb_id': elif k == 'kb_id':
if isinstance(newValue[k], list): if isinstance(newValue[k], list):
@ -531,7 +538,7 @@ class InfinityConnection(DocStoreConnection):
v = res[fieldnm][i] v = res[fieldnm][i]
if isinstance(v, Series): if isinstance(v, Series):
v = list(v) v = list(v)
elif fieldnm in ["important_kwd", "question_kwd", "entities_kwd"]: elif fieldnm in ["important_kwd", "question_kwd", "entities_kwd", "tag_kwd"]:
assert isinstance(v, str) assert isinstance(v, str)
v = [kwd for kwd in v.split("###") if kwd] v = [kwd for kwd in v.split("###") if kwd]
elif fieldnm == "position_int": elif fieldnm == "position_int":