element.chen/ai-g3sb-backman2.0
1
0
Fork 0
Code Pull Requests Activity
Files
be15b512a42909f6496bdfc4e8b24a7c9bdd6334
ai-g3sb-backman2.0/backend/utils/__pycache__/dialog_context.cpython-312.pyc
T

54 lines
7.2 KiB
Plaintext
Raw Normal View History

Enhance dialog context handling for Text2SQL queries by integrating session history. Introduce new methods for summarizing previous assistant messages and determining if the last interaction was a data query. Update environment configuration for embedding options and improve error handling in SQL generation. Add user-facing delivery messages for successful SQL execution. This update supports more coherent follow-up questions and improves user experience in conversational interactions.
2026-04-14 18:02:12 +08:00
Ë
Implement follow-up detection in dialog context to prevent context pollution in multi-topic conversations. Introduce is_likely_follow_up function to determine when to inject session history based on user input. Adjust _load_session_text2sql_context to conditionally include context based on follow-up status, enhancing SQL generation accuracy. Update relevant API endpoints to pass user text for improved context handling.
2026-04-15 17:33:07 +08:00
ßAßiRãóì—dZddlmZddlZddlZddlZddlmZmZm Z m
Z
eje «Z
dZdZdZej"dej$«Zdd „Zdd
Enhance dialog context handling for Text2SQL queries by integrating session history. Introduce new methods for summarizing previous assistant messages and determining if the last interaction was a data query. Update environment configuration for embedding options and improve error handling in SQL generation. Add user-facing delivery messages for successful SQL execution. This update supports more coherent follow-up questions and improves user experience in conversational interactions.
2026-04-14 18:02:12 +08:00
„Zdd „Zdd „Zed
Implement follow-up detection in dialog context to prevent context pollution in multi-topic conversations. Introduce is_likely_follow_up function to determine when to inject session history based on user input. Adjust _load_session_text2sql_context to conditionally include context based on follow-up status, enhancing SQL generation accuracy. Update relevant API endpoints to pass user text for improved context handling.
2026-04-15 17:33:07 +08:00
dœ dd„Zy)uŠ
从 Lite NL 会�消�构造 Text2SQL �用的上文摘�，并判断上一轮是�为数�查询（用于续问�图分类）。
é)Ú annotationsN)ÚAnyÚDictÚListÚTupleixi„iLu'(å†�|继续|å�Œæ ·|沿用|还是|照旧|刚æ‰�|上é�¢|上一æ�¡|上一个|上次|上述|这个SQL|è¿™æ�¡SQL|按上é�¢|按刚æ‰�|在此基础上|å�Œä¸€å�£å¾„|å�Œå�£å¾„|按之å‰�|改æˆ�|改为|加上|加一下|筛一下|过滤一下|补充|补一下|追加|优化一下|调整一下|æ�¢æˆ�|æ�¢ä¸º)có´—|xsdj«}|syt|«dkrtj|«ryt tj|««S)uæ
仅基于用户本轮文本�“是�续问�的轻�判断。
True -> �许注入少�上文（最近 1~2 轮），用于指代消解/沿用�径
False -> 新�题，�用上文，��上下文污染
ÚFé
T)ÚstripÚlenÚ_FOLLOW_UP_PATTERNÚsearchÚbool)Ú user_textÚts úbackend/utils/dialog_context.pyÚis_likely_follow_uprsO€ð
Š�b×ÑÓ!€AÙ Øä
Enhance dialog context handling for Text2SQL queries by integrating session history. Introduce new methods for summarizing previous assistant messages and determining if the last interaction was a data query. Update environment configuration for embedding options and improve error handling in SQL generation. Add user-facing delivery messages for successful SQL execution. This update supports more coherent follow-up questions and improves user experience in conversational interactions.
2026-04-14 18:02:12 +08:00
ˆ1ƒv�‚|Ô*×1Ñ1°!Ô4ØÜ Ô"×)Ñ)¨!Ó,Ó -Ð-ócó|—|xsdj«}t|«|kr|S|d|dz
Implement follow-up detection in dialog context to prevent context pollution in multi-topic conversations. Introduce is_likely_follow_up function to determine when to inject session history based on user input. Adjust _load_session_text2sql_context to conditionally include context based on follow-up status, enhancing SQL generation accuracy. Update relevant API endpoints to pass user text for improved context handling.
2026-04-15 17:33:07 +08:00
j«dzS)Nr éu…)r r Úrstrip)ÚsÚmax_lens rÚ _truncater)sB€Ø
Enhance dialog context handling for Text2SQL queries by integrating session history. Introduce new methods for summarizing previous assistant messages and determining if the last interaction was a data query. Update environment configuration for embedding options and improve error handling in SQL generation. Add user-facing delivery messages for successful SQL execution. This update supports more coherent follow-up questions and improves user experience in conversational interactions.
2026-04-14 18:02:12 +08:00
Šˆb�‰Ó€AÜ
ˆ1ƒv�ÒØˆØ ˆ]ˆw˜‰{Ð × "Ñ "Ó $ uÑ ,Ð,rcóÚ—|xsdj«}|sy tj|«}|j
d«xsi}t|j
d«t«r7|dj«r$dt |dj«t
Implement follow-up detection in dialog context to prevent context pollution in multi-topic conversations. Introduce is_likely_follow_up function to determine when to inject session history based on user input. Adjust _load_session_text2sql_context to conditionally include context based on follow-up status, enhancing SQL generation accuracy. Update relevant API endpoints to pass user text for improved context handling.
2026-04-15 17:33:07 +08:00
«zSt|j
d«t«r#|j
Enhance dialog context handling for Text2SQL queries by integrating session history. Introduce new methods for summarizing previous assistant messages and determining if the last interaction was a data query. Update environment configuration for embedding options and improve error handling in SQL generation. Add user-facing delivery messages for successful SQL execution. This update supports more coherent follow-up questions and improves user experience in conversational interactions.
2026-04-14 18:02:12 +08:00
d«xsdj«nd}|r®dt |t«zg}|j
d«r|jd«d D]b}|j
|«}t|t«sŒ%|j«sŒ6|jd
t |j«d «z«nd j|«S|j
d
«xsij
d
«}t d|›�t
«S#tj$rt |t
«cYSwxYw)u:å°†è�½åº“çš„ assistant JSON 转为一å°�段å�¯è¯»æ‘˜è¦�。r Ú
branch_resultÚansweru [助手] Úsqlu
Implement follow-up detection in dialog context to prevent context pollution in multi-topic conversations. Introduce is_likely_follow_up function to determine when to inject session history based on user input. Adjust _load_session_text2sql_context to conditionally include context based on follow-up status, enhancing SQL generation accuracy. Update relevant API endpoints to pass user text for improved context handling.
2026-04-15 17:33:07 +08:00
[上轮 SQL] Úfollow_up_requiredu:[状æ€�] 上轮为库探针0行，需补充æ�¡ä»¶å�Žé‡�é—®)Údb_empty_feedbackÚ sql_explainu [说明] iôÚ
Úintentu[助手] intent=) r ÚjsonÚloadsÚJSONDecodeErrorrÚ_MAX_ASSISTANT_TEXT_CHARSÚgetÚ
Enhance dialog context handling for Text2SQL queries by integrating session history. Introduce new methods for summarizing previous assistant messages and determining if the last interaction was a data query. Update environment configuration for embedding options and improve error handling in SQL generation. Add user-facing delivery messages for successful SQL execution. This update supports more coherent follow-up questions and improves user experience in conversational interactions.
2026-04-14 18:02:12 +08:00
isinstanceÚstrÚ_MAX_ASSISTANT_SQL_CHARSÚappendÚjoin) ÚcontentÚrawÚdataÚbrrÚpartsÚkeyÚvr#s rÚsummarize_assistant_nl_payloadr50s—€à Š=�b×
Ñ
Ó
!€CÙ Øð9Ü#Ÿz™z¨#›ˆð
�‰�/Ó "Ò ( b€BÜ�"—&‘&˜Ó"¤CÔ(¨R°©\×-?Ñ-?Ô-AØœY r¨(¡|×'9Ñ'9Ó';Ô=VÓWÑWÐWä+5°b·f±f¸U³mÄSÔ+Iˆ2�6‰6�%‹=Ò ˜B×
%Ñ
%Ô
'Èr€CÙ
Ø ¤9¨SÔ2JÓ#KÑKÐLˆØ
Implement follow-up detection in dialog context to prevent context pollution in multi-topic conversations. Introduce is_likely_follow_up function to determine when to inject session history based on user input. Adjust _load_session_text2sql_context to conditionally include context based on follow-up status, enhancing SQL generation accuracy. Update relevant API endpoints to pass user text for improved context handling.
2026-04-15 17:33:07 +08:00
�6‰6Ð&Ô 'Ø �L‰LÐUÔ VÛ7ˆCØ—‘�s“ ˆAܘ!œSÕ! a§g¡g¥iØ— ‘ ˜[¬9°Q·W±W³YÀÓ+DÑDÔEÙð 8ð
Enhance dialog context handling for Text2SQL queries by integrating session history. Introduce new methods for summarizing previous assistant messages and determining if the last interaction was a data query. Update environment configuration for embedding options and improve error handling in SQL generation. Add user-facing delivery messages for successful SQL execution. This update supports more coherent follow-up questions and improves user experience in conversational interactions.
2026-04-14 18:02:12 +08:00
�y‰y˜ÓÐà�h‰h�xÓ Ò& B×
Implement follow-up detection in dialog context to prevent context pollution in multi-topic conversations. Introduce is_likely_follow_up function to determine when to inject session history based on user input. Adjust _load_session_text2sql_context to conditionally include context based on follow-up status, enhancing SQL generation accuracy. Update relevant API endpoints to pass user text for improved context handling.
2026-04-15 17:33:07 +08:00
+Ñ
+¨HÓ
Enhance dialog context handling for Text2SQL queries by integrating session history. Introduce new methods for summarizing previous assistant messages and determining if the last interaction was a data query. Update environment configuration for embedding options and improve error handling in SQL generation. Add user-facing delivery messages for successful SQL execution. This update supports more coherent follow-up questions and improves user experience in conversational interactions.
2026-04-14 18:02:12 +08:00
5€FÜ Ð'¨ xÐ0Ô2KÓ LÐLøô) × Ñ ò9ܘÔ7Ó8Ò8ð9ús™GÇ#G*Ç)G*có$—t|xsg«D]ä}|jd«dk7rŒ|jd«xsdj«}|sy tj|«}|jd«xsijd«}|dk(ry|jd «xsi}t
Implement follow-up detection in dialog context to prevent context pollution in multi-topic conversations. Introduce is_likely_follow_up function to determine when to inject session history based on user input. Adjust _load_session_text2sql_context to conditionally include context based on follow-up status, enhancing SQL generation accuracy. Update relevant API endpoints to pass user text for improved context handling.
2026-04-15 17:33:07 +08:00
|jd
«t«r#|jd
«xsdj«nd}t|«cSy#tj
$rYywxYw) u^最å�Žä¸€æ�¡ assistant 消æ�¯æ˜¯å�¦ä¸ºæ•°æ�®æŸ¥è¯¢åˆ†æ”¯ï¼ˆå�« SQL 或 DATA_QUERY intent）。ÚroleÚ assistantr.r Fr#Ú
Copy Permalink