streaming_sql_features stringclasses 2
values | sql_dialect stringclasses 1
value | final_sql_query stringclasses 2
values | clarification_summary stringclasses 2
values | query_revision_summary stringclasses 2
values | conversation_transcript stringclasses 2
values | conversation_turn_count int64 6 6 | user_requirements_summary stringclasses 2
values |
|---|---|---|---|---|---|---|---|
Event-time processing; upstream rowtime attribute and watermark; five-minute tumbling windows; grouped count and sum; HAVING threshold on an aggregate; settled-status filtering. | Apache Flink SQL | SELECT
window_start,
window_end,
merchant_id,
COUNT(*) AS settled_payment_count,
SUM(amount) AS settled_amount
FROM TABLE(
TUMBLE(TABLE payment_events, DESCRIPTOR(event_time), INTERVAL '5' MINUTES)
)
WHERE status = 'SETTLED'
GROUP BY window_start, window_end, merchant_id
HAVING SUM(amount) >= 10000.00; | The engine is Apache Flink SQL, and event_time has an upstream 20-second watermark. The initial minimum of 25 settled payments was replaced by a minimum settled amount of 10000 per window. The settled-only filter remains in effect. | The initial query used a minimum count of 25 settled payments per merchant window. After the user replaced that criterion with a minimum settled amount of 10000, the final query changed its HAVING clause to SUM(amount) >= 10000.00 while retaining the five-minute TUMBLE window and settled-only filter. | User: Build a streaming query that reports payment count and total amount by merchant in five-minute event-time windows.
Assistant: Which SQL engine should I target, and what are the event-time column, source fields, and any status or threshold filters?
User: Use Apache Flink SQL. The payment_events table has merchant_... | 6 | Produce five-minute event-time aggregates by merchant from payment_events, reporting the count and sum of SETTLED payments. Return only merchant windows with settled_amount of at least 10000.00. Amounts share a consistent denomination, and event_time is an upstream-defined rowtime attribute. |
Event-time processing; upstream rowtime attribute and watermark; hopping windows with a two-minute slide and ten-minute size; filtered streaming aggregation; COUNT(DISTINCT) with nullable visitor IDs; HAVING threshold. | Apache Flink SQL | SELECT
window_start,
window_end,
page_id,
COUNT(*) AS view_count,
COUNT(DISTINCT visitor_id) AS unique_visitors
FROM TABLE(
HOP(TABLE page_view_stream, DESCRIPTOR(event_time), INTERVAL '2' MINUTES, INTERVAL '10' MINUTES)
)
WHERE is_internal = FALSE
GROUP BY window_start, window_end, page_id
HAVING COUNT(DIS... | The query targets Apache Flink SQL. event_time is an upstream-defined rowtime attribute with a 30-second watermark, and visitor_id may be NULL. The user changed the initial tumbling-window cadence to overlapping ten-minute windows with a two-minute slide. Non-internal anonymous views count toward total views but not di... | The first query used a ten-minute TUMBLE window. The user revised the cadence to overlapping ten-minute windows advancing every two minutes, so the final query uses HOP with a two-minute slide and ten-minute size. The internal-traffic filter, anonymous-view counting behavior, and distinct-visitor threshold are retained... | User: I need a streaming report that helps identify popular pages. For each page, show view volume and the number of distinct visitors, and exclude internal traffic.
Assistant: Which SQL engine and window cadence should I use? Please also clarify the event-time and watermark setup, how to treat anonymous views, and whe... | 6 | Report non-internal page views and distinct visitors per page in ten-minute event-time windows that advance every two minutes. Include anonymous views in view_count, exclude their NULL visitor IDs from the distinct count, and return only windows with at least 20 distinct visitors. |
Streaming SQL Multi-Turn Dialogue and Query Refinement Dataset
This dataset contains multi-turn text conversations in which users describe streaming analytics needs, add constraints, and iteratively refine SQL queries. It covers streaming concepts such as windowing, time semantics, and aggregation, with records that can include the full transcript, requirement and clarification summaries, the final query, SQL dialect, and revision history. It supports training models to interpret requirements through dialogue, generate streaming SQL, and revise queries based on feedback.
Technical Specifications
| Field | Type | Description |
|---|---|---|
| sql_dialect | string | The streaming SQL dialect or engine identified or assumed for the final query, when available. |
| final_sql_query | string | The final streaming SQL query produced or revised during the conversation, including relevant windowing and time semantics. |
| clarification_summary | string | A summary of ambiguities identified and clarified during the conversation, including the resulting decisions. |
| query_revision_summary | string | A summary of how the SQL query changed across iterations and which user-provided constraints motivated those changes. |
| streaming_sql_features | string | A concise list of streaming concepts used or requested, such as event time, watermarks, windows, joins, state, and aggregations. |
| conversation_transcript | string | The complete original conversation, preserving turn order and the user's requests, constraints, clarifications, and assistant responses. |
| conversation_turn_count | integer | The number of conversational turns represented in the original transcript. |
| user_requirements_summary | string | A concise summary of the analytics goal, requested outputs, and explicit user constraints derived from the conversation. |
Compliance Statement
| Authorization Type | CC-BY-NC-SA 4.0 (Attribution–NonCommercial–ShareAlike) |
| Commercial Use | Requires exclusive subscription or authorization contract (monthly or per-invocation charging) |
| Privacy and Anonymization | No PII, no real company names, simulated scenarios follow industry standards |
| Compliance System | Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs |
Source & Contact
- Downloads last month
- 4