我想在Oracle中更改一周的开始日期，我需要从每日/每小时数据计算每周的一些报告。
通过大量Google搜索，我只获得NLS_TERROTORY选项，我不知道如何设置为TUESDAY。即使有领土，那么未来如何改变到不同的日子。提前谢谢！
答案 0 :(得分：3)
您可以创建一个包，以包含与新周开始相关的自定义函数函数：
注意：一年中的周数是基于包含一周的第一个开始日（即星期二）的一年中的第一周 - 因此，如果1月1日是星期三，那么一年的第一周将是从1月7日开始）。如果您需要不同的逻辑，则需要修改WEEK_OF_YEAR函数。
Oracle 11g R2架构设置：
CREATE PACKAGE change_week_start IS
week_start CONSTANT VARCHAR2(9) := 'TUESDAY';
FUNCTION TRUNC_TO_WEEK_START(
in_date IN DATE
) RETURN DATE;
FUNCTION DAY_OF_WEEK(
in_date IN DATE
) RETURN NUMBER;
FUNCTION WEEK_OF_YEAR(
in_date IN DATE
) RETURN NUMBER;
FUNCTION TO_CHAR_WEEK(
in_date IN DATE
) RETURN VARCHAR2;
END;
/
CREATE PACKAGE BODY change_week_start IS
FUNCTION TRUNC_TO_WEEK_START(
in_date IN DATE
) RETURN DATE
IS
BEGIN
RETURN NEXT_DAY( TRUNC( in_date ) - 7, week_start );
END;
FUNCTION DAY_OF_WEEK(
in_date IN DATE
) RETURN NUMBER
IS
BEGIN
RETURN ( TRUNC( in_date ) - TRUNC_TO_WEEK_START( in_date ) ) + 1;
END;
FUNCTION WEEK_OF_YEAR(
in_date IN DATE
) RETURN NUMBER
IS
BEGIN
RETURN TRUNC(
(
in_date
-
TRUNC_TO_WEEK_START(
TRUNC( TRUNC_TO_WEEK_START( in_date ), 'YYYY' ) + 6
)
) / 7
) + 1;
END;
FUNCTION TO_CHAR_WEEK(
in_date IN DATE
) RETURN VARCHAR2
IS
BEGIN
RETURN TO_CHAR( TRUNC_TO_WEEK_START( in_date ), 'FMYYYY' )
|| '-W' || TO_CHAR( WEEK_OF_YEAR( in_date ), 'FM00' )
|| '-' || DAY_OF_WEEK( in_date );
END;
END;
/
查询1 ：
SELECT value,
CHANGE_WEEK_START.TO_CHAR_WEEK( value ) AS week,
TO_CHAR( value, 'DAY' ) AS day
FROM (
SELECT TRUNC( SYSDATE, 'YYYY' ) + LEVEL - 1 AS value
FROM DUAL
CONNECT BY LEVEL <= 14
)
<强> Results ：
| VALUE | WEEK | DAY |
|----------------------|------------|-----------|
| 2018-01-01T00:00:00Z | 2017-W52-7 | MONDAY |
| 2018-01-02T00:00:00Z | 2018-W01-1 | TUESDAY |
| 2018-01-03T00:00:00Z | 2018-W01-2 | WEDNESDAY |
| 2018-01-04T00:00:00Z | 2018-W01-3 | THURSDAY |
| 2018-01-05T00:00:00Z | 2018-W01-4 | FRIDAY |
| 2018-01-06T00:00:00Z | 2018-W01-5 | SATURDAY |
| 2018-01-07T00:00:00Z | 2018-W01-6 | SUNDAY |
| 2018-01-08T00:00:00Z | 2018-W01-7 | MONDAY |
| 2018-01-09T00:00:00Z | 2018-W02-1 | TUESDAY |
| 2018-01-10T00:00:00Z | 2018-W02-2 | WEDNESDAY |
| 2018-01-11T00:00:00Z | 2018-W02-3 | THURSDAY |
| 2018-01-12T00:00:00Z | 2018-W02-4 | FRIDAY |
| 2018-01-13T00:00:00Z | 2018-W02-5 | SATURDAY |
| 2018-01-14T00:00:00Z | 2018-W02-6 | SUNDAY |
答案 1 :(得分：1)
我们欧洲人星期一是本周的第一天。因此，如果您将NLS_TERRITORY设置为欧洲的国家/地区（例如英国，或者 - 为什么不 - 克罗地亚），您将获得＆＃34; 1＆＃34;作为结果。这是一个简短的演示：
y= list(word)
words = y
similar = [[item[0] for item in model.wv.most_similar(word) if item[1] > 0.7] for word in words]
similarity_matrix = pd.DataFrame({'Root_Word': words, 'Similar_Words': similar})
similarity_matrix = similarity_matrix[['Root_Word', 'Similar_Words']]
similarity_matrix['Unlist_Root']=similarity_matrix['Root_Word'].apply(lambda x: ', '.join(x))
similarity_matrix['Unlist_Similar']=similarity_matrix['Similar_Words'].apply(lambda x: ', '.join(x))
similarity_matrix=similarity_matrix.drop(['Root_Word','Similar_Words'],1)
similarity_matrix.columns=['Root_Word','Similar_Words']
[编辑] 自定义功能，可返回星期几：
SQL> ALTER SESSION SET NLS_TERRITORY = 'UNITED KINGDOM';
Session altered.
SQL> select to_char(sysdate, 'D') european from dual;
E
-
1
SQL>
SQL> ALTER SESSION SET NLS_TERRITORY = 'AMERICA';
Session altered.
SQL> select to_char(sysdate, 'D') american from dual;
A
-
2
SQL> ALTER SESSION SET NLS_TERRITORY = 'CROATIA';
Session altered.
SQL> select to_char(sysdate, 'D') croatian from dual;
C
-
1
SQL>
答案 2 :(得分：1)
使用此PL / SQL块，您可以获得一周中所有可能的日期：
BEGIN
FOR aLang IN (SELECT * FROM V$NLS_VALID_VALUES WHERE parameter = 'TERRITORY' ORDER BY VALUE) LOOP
EXECUTE IMMEDIATE 'ALTER SESSION SET NLS_TERRITORY = '''||aLang.VALUE||'''';
DBMS_OUTPUT.PUT_LINE(aLang.VALUE || ' -> ' || TO_CHAR(TRUNC(SYSDATE, 'D'), 'Day'));
end loop;
end;
ALBANIA -> Monday
ALGERIA -> Saturday
AMERICA -> Sunday
ARGENTINA -> Monday
AUSTRALIA -> Monday
AUSTRIA -> Monday
AZERBAIJAN -> Monday
BAHRAIN -> Saturday
BANGLADESH -> Friday
BELARUS -> Monday
BELGIUM -> Monday
BRAZIL -> Sunday
BULGARIA -> Monday
CANADA -> Sunday
CATALONIA -> Monday
CHILE -> Monday
CHINA -> Sunday
CIS -> Monday
COLOMBIA -> Sunday
COSTA RICA -> Sunday
CROATIA -> Monday
CYPRUS -> Monday
CZECH REPUBLIC -> Monday
CZECHOSLOVAKIA -> Monday
DENMARK -> Monday
DJIBOUTI -> Saturday
ECUADOR -> Monday
EGYPT -> Saturday
EL SALVADOR -> Sunday
ESTONIA -> Monday
FINLAND -> Monday
FRANCE -> Monday
FYR MACEDONIA -> Monday
GERMANY -> Monday
GREECE -> Monday
GUATEMALA -> Sunday
HONG KONG -> Sunday
HUNGARY -> Monday
ICELAND -> Monday
INDIA -> Sunday
INDONESIA -> Monday
IRAQ -> Saturday
IRELAND -> Monday
ISRAEL -> Sunday
ITALY -> Monday
JAPAN -> Sunday
JORDAN -> Saturday
KAZAKHSTAN -> Monday
KOREA -> Sunday
KUWAIT -> Saturday
LATVIA -> Monday
LEBANON -> Saturday
LIBYA -> Saturday
LITHUANIA -> Monday
LUXEMBOURG -> Monday
MACEDONIA -> Monday
MALAYSIA -> Sunday
MAURITANIA -> Saturday
MEXICO -> Monday
MOROCCO -> Saturday
NEW ZEALAND -> Monday
NICARAGUA -> Monday
NORWAY -> Monday
OMAN -> Saturday
PANAMA -> Sunday
PERU -> Sunday
PHILIPPINES -> Sunday
POLAND -> Monday
PORTUGAL -> Sunday
PUERTO RICO -> Sunday
QATAR -> Saturday
ROMANIA -> Monday
RUSSIA -> Monday
SAUDI ARABIA -> Saturday
SERBIA AND MONTENEGRO -> Monday
SINGAPORE -> Sunday
SLOVAKIA -> Monday
SLOVENIA -> Monday
SOMALIA -> Saturday
SOUTH AFRICA -> Sunday
SPAIN -> Monday
SUDAN -> Saturday
SWEDEN -> Monday
SWITZERLAND -> Monday
SYRIA -> Saturday
TAIWAN -> Sunday
THAILAND -> Sunday
THE NETHERLANDS -> Monday
TUNISIA -> Saturday
TURKEY -> Monday
UKRAINE -> Monday
UNITED ARAB EMIRATES -> Saturday
UNITED KINGDOM -> Monday
UZBEKISTAN -> Monday
VENEZUELA -> Sunday
VIETNAM -> Sunday
YEMEN -> Saturday
YUGOSLAVIA -> Monday
你知道，星期二的第一天没有领土，所以你必须建立自己的功能。 Littlefoot的解决方案应该运作良好。
答案 3 :(得分：0)
这是我自己在2015年推出的3个视图的解决方案，可以配置为动态运行任何一年。
- 计算一整年的所有日子。
CREATE OR REPLACE VIEW CUSTOM_DAYS_AND_WEEKS
AS
SELECT DAYS.*,
CASE
WHEN ((EXTRACT(YEAR FROM DAYS.WEEK_START) <>EXTRACT(YEAR FROM DAYS.WEEK_END))
AND (((TRUNC(WEEK_END,'YYYY')-1)-WEEK_START)>=3))
THEN TO_CHAR(WEEK_START,'YYYY')
WHEN ((EXTRACT(YEAR FROM DAYS.WEEK_START) <>EXTRACT(YEAR FROM DAYS.WEEK_END))
AND (WEEK_START-(TRUNC(WEEK_START,'YYYY'))>=3))
THEN TO_CHAR(WEEK_END,'YYYY')
ELSE TO_CHAR(WEEK_START,'YYYY')
END AS YEAR_NO
FROM
(SELECT "A1"."DD" "DD",
TO_CHAR("A1"."DD",'DY') "DAY",
DECODE(TO_CHAR("A1"."DD",'DY'),'TUE', "A1"."DD",NEXT_DAY("A1"."DD" -7, 'TUESDAY')) WEEK_START,
DECODE(TO_CHAR("A1"."DD",'DY'),'MON',"A1"."DD",NEXT_DAY("A1"."DD" , 'MONDAY')) WEEK_END
FROM
(SELECT DD+(level-1) "DD"
FROM
(SELECT TO_DATE('01012015','DDMMYYYY') "DD" FROM "SYS"."DUAL" "A2"
)
CONNECT BY (DD+(level-1))< TO_DATE('01012016','DDMMYYYY')
) "A1"
) DAYS ;
- 对当前年度和下一年的所有周数进行分组（如果年末有的话）
CREATE OR REPLACE VIEW CUSTOM_WEEKS AS
SELECT T.YEAR_NO,
T.WEEK_START,
T.WEEK_END,
SUM(1) S
FROM CUSTOM_DAYS_AND_WEEKS T WHERE YEAR_NO>=2015
group by T.YEAR_NO, T.WEEK_START, T.WEEK_END
ORDER BY 1, 2;
- 周开始日期，结束日期和周数
CREATE OR REPLACE VIEW CUSTOM_WEEKS_NO AS
SELECT T.YEAR_NO,
T.WEEK_START,
T.WEEK_END,
DECODE (MOD(ROWNUM,(SELECT COUNT(*)+1 FROM CUSTOM_WEEKS T WHERE T.YEAR_NO=2015)),0,1,MOD(ROWNUM,(SELECT COUNT(*)+1 FROM CUSTOM_WEEKS T WHERE T.YEAR_NO=2015))) WEEK_NO
FROM CUSTOM_WEEKS T;
- 检查创建的视图
SELECT * FROM CUSTOM_DAYS_AND_WEEKS;
SELECT * FROM CUSTOM_WEEKS;
- 周的所有日期，包括周开始日期，结束日期和周数
SELECT D_W.*, W.WEEK_NO
FROM CUSTOM_DAYS_AND_WEEKS D_W,
CUSTOM_WEEKS_NO W
WHERE D_W.WEEK_START=W.WEEK_START
AND D_W.WEEK_END=W.WEEK_END ORDER BY DD DESC;