我不太擅长 sql,我需要你的帮助。
我在 Oracle DB 中有一个包含示例数据的表,如下所示:
| DATE_TIME | TYPE | LAT | LON | MSG |
| :------------------------ | :--- | :---- | :---- | :---- |
| 12.11.20 10:04.00.0000000 | A | 10.03 | 11.15 | |
| 12.11.20 10:05.00.0000000 | A | 10.04 | 11.14 | |
| 12.11.20 10:06.00.0000000 | A | 10.05 | 11.13 | |
| 12.11.20 11:18.00.0000000 | B | 20.45 | 40.58 | |
| 12.11.20 11:19.00.0000000 | B | 20.46 | 40.59 | |
| 12.11.20 11:20.00.0000000 | B | 20.47 | 40.60 | |
我有一个选择,它给了我START_TIME、END_TIME、TYPE、START_LAT、START_LON、END_LAT、END_LON、START_MSG、END_MSG:
SELECT
TO_CHAR(START_TIME, 'DD.MM.YYYY HH24:MI') AS START_TIME,
TO_CHAR(END_TIME, 'DD.MM.YYYY HH24:MI') AS END_TIME,
TYPE,
START_LAT,
START_LON,
END_LAT,
END_LON,
START_MSG,
END_MSG
FROM TABLE_1
MATCH_RECOGNIZE (
PARTITION BY TYPE
ORDER BY DATE_TIME
MEASURES
FIRST(DATE_TIME) AS START_TIME,
LAST(DATE_TIME) AS END_TIME,
FIRST(LAT) AS START_LAT,
FIRST(LON) AS START_LON,
LAST(LAT) AS END_LAT,
LAST(LON) AS END_LON,
FIRST(MSG) AS START_MSG,
LAST(MSG) AS END_MSG
ONE ROW PER MATCH
PATTERN ( adjacent_minutes * last_minute )
DEFINE adjacent_minutes AS NEXT(DATE_TIME) <= LAST(DATE_TIME) + INTERVAL '5' MINUTE
);
此选择的输出:
| START_TIME | END_TIME | TYPE | START_LAT | START_LON | END_LAT | END_LON | START_MSG | END_MSG |
| :------------- | :------------- | :--- | :-------- | :-------- | :------ | :------ | :-------- | :------ |
| 12.11.20 10:04 | 12.11.20 10:06 | A | 10.03 | 11.15 | 10.05 | 11.13 | | |
| 12.11.20 11:18 | 12.11.20 11:20 | B | 20.45 | 40.58 | 20.47 | 40.60 | | |
现在,当例如START_LAT介于 10 和 11 之间并且START_LON介于 11 和 12 之间时,我需要将值为“国家 xy”的数据插入 START_MSG 列。对于END_MSG也是如此。
我怎样才能做到这一点?
抱歉我的英语不好;)。
非常感谢您!