Kevin Hu
commited on
Commit
·
a3da325
1
Parent(s):
3fb798a
Fix raptor resuable issue. (#4063)
Browse files### What problem does this PR solve?
#4045
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
- api/db/services/document_service.py +10 -0
- api/db/services/task_service.py +19 -15
- graphrag/utils.py +1 -1
api/db/services/document_service.py
CHANGED
@@ -344,6 +344,8 @@ class DocumentService(CommonService):
|
|
344 |
old[k] = v
|
345 |
|
346 |
dfs_update(d.parser_config, config)
|
|
|
|
|
347 |
cls.update_by_id(id, {"parser_config": d.parser_config})
|
348 |
|
349 |
@classmethod
|
@@ -432,6 +434,11 @@ class DocumentService(CommonService):
|
|
432 |
|
433 |
|
434 |
def queue_raptor_tasks(doc):
|
|
|
|
|
|
|
|
|
|
|
435 |
def new_task():
|
436 |
nonlocal doc
|
437 |
return {
|
@@ -443,6 +450,9 @@ def queue_raptor_tasks(doc):
|
|
443 |
}
|
444 |
|
445 |
task = new_task()
|
|
|
|
|
|
|
446 |
bulk_insert_into_db(Task, [task], True)
|
447 |
task["type"] = "raptor"
|
448 |
assert REDIS_CONN.queue_product(SVR_QUEUE_NAME, message=task), "Can't access Redis. Please check the Redis' status."
|
|
|
344 |
old[k] = v
|
345 |
|
346 |
dfs_update(d.parser_config, config)
|
347 |
+
if not config.get("raptor") and d.parser_config.get("raptor"):
|
348 |
+
del d.parser_config["raptor"]
|
349 |
cls.update_by_id(id, {"parser_config": d.parser_config})
|
350 |
|
351 |
@classmethod
|
|
|
434 |
|
435 |
|
436 |
def queue_raptor_tasks(doc):
|
437 |
+
chunking_config = DocumentService.get_chunking_config(doc["id"])
|
438 |
+
hasher = xxhash.xxh64()
|
439 |
+
for field in sorted(chunking_config.keys()):
|
440 |
+
hasher.update(str(chunking_config[field]).encode("utf-8"))
|
441 |
+
|
442 |
def new_task():
|
443 |
nonlocal doc
|
444 |
return {
|
|
|
450 |
}
|
451 |
|
452 |
task = new_task()
|
453 |
+
for field in ["doc_id", "from_page", "to_page"]:
|
454 |
+
hasher.update(str(task.get(field, "")).encode("utf-8"))
|
455 |
+
task["digest"] = hasher.hexdigest()
|
456 |
bulk_insert_into_db(Task, [task], True)
|
457 |
task["type"] = "raptor"
|
458 |
assert REDIS_CONN.queue_product(SVR_QUEUE_NAME, message=task), "Can't access Redis. Please check the Redis' status."
|
api/db/services/task_service.py
CHANGED
@@ -34,15 +34,17 @@ from rag.utils.redis_conn import REDIS_CONN
|
|
34 |
from api import settings
|
35 |
from rag.nlp import search
|
36 |
|
|
|
37 |
def trim_header_by_lines(text: str, max_length) -> str:
|
38 |
len_text = len(text)
|
39 |
if len_text <= max_length:
|
40 |
return text
|
41 |
for i in range(len_text):
|
42 |
if text[i] == '\n' and len_text - i <= max_length:
|
43 |
-
return text[i+1:]
|
44 |
return text
|
45 |
|
|
|
46 |
class TaskService(CommonService):
|
47 |
model = Task
|
48 |
|
@@ -73,10 +75,10 @@ class TaskService(CommonService):
|
|
73 |
]
|
74 |
docs = (
|
75 |
cls.model.select(*fields)
|
76 |
-
|
77 |
-
|
78 |
-
|
79 |
-
|
80 |
)
|
81 |
docs = list(docs.dicts())
|
82 |
if not docs:
|
@@ -111,7 +113,7 @@ class TaskService(CommonService):
|
|
111 |
]
|
112 |
tasks = (
|
113 |
cls.model.select(*fields).order_by(cls.model.from_page.asc(), cls.model.create_time.desc())
|
114 |
-
|
115 |
)
|
116 |
tasks = list(tasks.dicts())
|
117 |
if not tasks:
|
@@ -131,18 +133,18 @@ class TaskService(CommonService):
|
|
131 |
cls.model.select(
|
132 |
*[Document.id, Document.kb_id, Document.location, File.parent_id]
|
133 |
)
|
134 |
-
|
135 |
-
|
136 |
File2Document,
|
137 |
on=(File2Document.document_id == Document.id),
|
138 |
join_type=JOIN.LEFT_OUTER,
|
139 |
)
|
140 |
-
|
141 |
File,
|
142 |
on=(File2Document.file_id == File.id),
|
143 |
join_type=JOIN.LEFT_OUTER,
|
144 |
)
|
145 |
-
|
146 |
Document.status == StatusEnum.VALID.value,
|
147 |
Document.run == TaskStatus.RUNNING.value,
|
148 |
~(Document.type == FileType.VIRTUAL.value),
|
@@ -212,8 +214,8 @@ def queue_tasks(doc: dict, bucket: str, name: str):
|
|
212 |
if doc["parser_id"] == "paper":
|
213 |
page_size = doc["parser_config"].get("task_page_size", 22)
|
214 |
if doc["parser_id"] in ["one", "knowledge_graph"] or not do_layout:
|
215 |
-
page_size = 10**9
|
216 |
-
page_ranges = doc["parser_config"].get("pages") or [(1, 10**5)]
|
217 |
for s, e in page_ranges:
|
218 |
s -= 1
|
219 |
s = max(0, s)
|
@@ -257,7 +259,8 @@ def queue_tasks(doc: dict, bucket: str, name: str):
|
|
257 |
if task["chunk_ids"]:
|
258 |
chunk_ids.extend(task["chunk_ids"].split())
|
259 |
if chunk_ids:
|
260 |
-
settings.docStoreConn.delete({"id": chunk_ids}, search.index_name(chunking_config["tenant_id"]),
|
|
|
261 |
DocumentService.update_by_id(doc["id"], {"chunk_num": ck_num})
|
262 |
|
263 |
bulk_insert_into_db(Task, tsks, True)
|
@@ -271,7 +274,8 @@ def queue_tasks(doc: dict, bucket: str, name: str):
|
|
271 |
|
272 |
|
273 |
def reuse_prev_task_chunks(task: dict, prev_tasks: list[dict], chunking_config: dict):
|
274 |
-
idx = bisect.bisect_left(prev_tasks, task.get("from_page", 0),
|
|
|
275 |
if idx >= len(prev_tasks):
|
276 |
return 0
|
277 |
prev_task = prev_tasks[idx]
|
@@ -286,4 +290,4 @@ def reuse_prev_task_chunks(task: dict, prev_tasks: list[dict], chunking_config:
|
|
286 |
task["progress_msg"] += "reused previous task's chunks."
|
287 |
prev_task["chunk_ids"] = ""
|
288 |
|
289 |
-
return len(task["chunk_ids"].split())
|
|
|
34 |
from api import settings
|
35 |
from rag.nlp import search
|
36 |
|
37 |
+
|
38 |
def trim_header_by_lines(text: str, max_length) -> str:
|
39 |
len_text = len(text)
|
40 |
if len_text <= max_length:
|
41 |
return text
|
42 |
for i in range(len_text):
|
43 |
if text[i] == '\n' and len_text - i <= max_length:
|
44 |
+
return text[i + 1:]
|
45 |
return text
|
46 |
|
47 |
+
|
48 |
class TaskService(CommonService):
|
49 |
model = Task
|
50 |
|
|
|
75 |
]
|
76 |
docs = (
|
77 |
cls.model.select(*fields)
|
78 |
+
.join(Document, on=(cls.model.doc_id == Document.id))
|
79 |
+
.join(Knowledgebase, on=(Document.kb_id == Knowledgebase.id))
|
80 |
+
.join(Tenant, on=(Knowledgebase.tenant_id == Tenant.id))
|
81 |
+
.where(cls.model.id == task_id)
|
82 |
)
|
83 |
docs = list(docs.dicts())
|
84 |
if not docs:
|
|
|
113 |
]
|
114 |
tasks = (
|
115 |
cls.model.select(*fields).order_by(cls.model.from_page.asc(), cls.model.create_time.desc())
|
116 |
+
.where(cls.model.doc_id == doc_id)
|
117 |
)
|
118 |
tasks = list(tasks.dicts())
|
119 |
if not tasks:
|
|
|
133 |
cls.model.select(
|
134 |
*[Document.id, Document.kb_id, Document.location, File.parent_id]
|
135 |
)
|
136 |
+
.join(Document, on=(cls.model.doc_id == Document.id))
|
137 |
+
.join(
|
138 |
File2Document,
|
139 |
on=(File2Document.document_id == Document.id),
|
140 |
join_type=JOIN.LEFT_OUTER,
|
141 |
)
|
142 |
+
.join(
|
143 |
File,
|
144 |
on=(File2Document.file_id == File.id),
|
145 |
join_type=JOIN.LEFT_OUTER,
|
146 |
)
|
147 |
+
.where(
|
148 |
Document.status == StatusEnum.VALID.value,
|
149 |
Document.run == TaskStatus.RUNNING.value,
|
150 |
~(Document.type == FileType.VIRTUAL.value),
|
|
|
214 |
if doc["parser_id"] == "paper":
|
215 |
page_size = doc["parser_config"].get("task_page_size", 22)
|
216 |
if doc["parser_id"] in ["one", "knowledge_graph"] or not do_layout:
|
217 |
+
page_size = 10 ** 9
|
218 |
+
page_ranges = doc["parser_config"].get("pages") or [(1, 10 ** 5)]
|
219 |
for s, e in page_ranges:
|
220 |
s -= 1
|
221 |
s = max(0, s)
|
|
|
259 |
if task["chunk_ids"]:
|
260 |
chunk_ids.extend(task["chunk_ids"].split())
|
261 |
if chunk_ids:
|
262 |
+
settings.docStoreConn.delete({"id": chunk_ids}, search.index_name(chunking_config["tenant_id"]),
|
263 |
+
chunking_config["kb_id"])
|
264 |
DocumentService.update_by_id(doc["id"], {"chunk_num": ck_num})
|
265 |
|
266 |
bulk_insert_into_db(Task, tsks, True)
|
|
|
274 |
|
275 |
|
276 |
def reuse_prev_task_chunks(task: dict, prev_tasks: list[dict], chunking_config: dict):
|
277 |
+
idx = bisect.bisect_left(prev_tasks, (task.get("from_page", 0), task.get("digest", "")),
|
278 |
+
key=lambda x: (x.get("from_page", 0), x.get("digest", "")))
|
279 |
if idx >= len(prev_tasks):
|
280 |
return 0
|
281 |
prev_task = prev_tasks[idx]
|
|
|
290 |
task["progress_msg"] += "reused previous task's chunks."
|
291 |
prev_task["chunk_ids"] = ""
|
292 |
|
293 |
+
return len(task["chunk_ids"].split())
|
graphrag/utils.py
CHANGED
@@ -78,7 +78,7 @@ def get_llm_cache(llmnm, txt, history, genconf):
|
|
78 |
bin = REDIS_CONN.get(k)
|
79 |
if not bin:
|
80 |
return
|
81 |
-
return bin
|
82 |
|
83 |
|
84 |
def set_llm_cache(llmnm, txt, v: str, history, genconf):
|
|
|
78 |
bin = REDIS_CONN.get(k)
|
79 |
if not bin:
|
80 |
return
|
81 |
+
return bin
|
82 |
|
83 |
|
84 |
def set_llm_cache(llmnm, txt, v: str, history, genconf):
|