N'essayez pas de créer une colonne par jour ; créez simplement une table avec des colonnes pour l'emplacement, la date et pour chaque statistique (c'est-à-dire infecté, récupéré, mort, etc.), puis si vous avez besoin de les faire pivoter, faites-le dans une requête (ou dans n'importe quelle application de niveau intermédiaire [c'est-à-dire PHP, Java, .net] que vous utilisez pour accéder à la base de données).
Quelque chose comme :
CREATE TABLE Regions(
id VARCHAR2(6)
CONSTRAINT regions__id__pk PRIMARY KEY,
parent_id VARCHAR2(6)
CONSTRAINT regions_parent__fk REFERENCES Regions ( id ),
name VARCHAR2(50)
CONSTRAINT regions__name__nn NOT NULL
CONSTRAINT regions__name__u UNIQUE,
latitude NUMBER
CONSTRAINT regions__lat__nn NOT NULL,
longitude NUMBER
CONSTRAINT regions__long__nn NOT NULL,
CONSTRAINT regions__id__chk CHECK (
( parent_id IS NULL AND REGEXP_LIKE( id, '^[A-Z]{2}$' ) )
OR ( parent_id IS NOT NULL AND REGEXP_LIKE( id, '^[A-Z]{2}-[A-Z0-9]{1,3}$' ) )
)
);
COMMENT ON COLUMN Regions.id IS 'ISO 3166-2 Alpha-2 Country Code or ISO 3166-2 Province Code';
COMMENT ON COLUMN Regions.name IS 'ISO 3166-2 English Short Name.';
COMMENT ON COLUMN Regions.latitude IS 'Latitude of the region''s main city.';
COMMENT ON COLUMN Regions.longitude IS 'Longitude of the region''s main city.';
CREATE TABLE Virus_Statistics(
id NUMBER(20,0)
GENERATED ALWAYS AS IDENTITY
CONSTRAINT virus_statistics__id__pk PRIMARY KEY,
location VARCHAR2(6)
CONSTRAINT virus_statistics__loc__nn NOT NULL
CONSTRAINT virus_statistics__loc__fk REFERENCES Regions ( id ),
datetime DATE
CONSTRAINT virus_statistics__dt__nn NOT NULL
CONSTRAINT virus_statistics__dt__chk CHECK ( datetime = TRUNC( datetime ) ),
infected NUMBER(10,0),
recovered NUMBER(10,0),
dead NUMBER(10,0),
CONSTRAINT virus_statistics__loc__dt__u UNIQUE ( location, datetime )
);
Ensuite, vous pouvez entrer vos données. Par exemple, les régions seraient :
INSERT INTO Regions ( id, parent_id, name, latitude, longitude )
SELECT 'TH', NULL, 'Thailand', 15.00000, 101.00000 FROM DUAL UNION ALL
SELECT 'JP', NULL, 'Japan', 36.00000, 138.00000 FROM DUAL UNION ALL
SELECT 'SG', NULL, 'Singapore', 1.28333, 103.83333 FROM DUAL UNION ALL
SELECT 'NP', NULL, 'Nepal', 28.16667, 84.25000 FROM DUAL UNION ALL
SELECT 'MY', NULL, 'Malaysia', 2.50000, 112.50000 FROM DUAL UNION ALL
SELECT 'CA', NULL, 'Canada', 45.42472, - 75.69500 FROM DUAL UNION ALL
SELECT 'CA-BC', 'CA', 'British Columbia', 48.40733, -123.32977 FROM DUAL;
Et les 3 premières colonnes de données seraient :
INSERT INTO Virus_Statistics ( location, datetime, infected, recovered, dead )
SELECT 'TH', DATE '2020-01-22', 2 AS i, 0 AS r, 0 AS d FROM DUAL UNION ALL
SELECT 'TH', DATE '2020-01-23', 3, 0, 0 FROM DUAL UNION ALL
SELECT 'TH', DATE '2020-01-24', 5, 0, 0 FROM DUAL UNION ALL
SELECT 'JP', DATE '2020-01-22', 2, 0, 0 FROM DUAL UNION ALL
SELECT 'JP', DATE '2020-01-23', 1, 0, 0 FROM DUAL UNION ALL
SELECT 'JP', DATE '2020-01-24', 2, 0, 0 FROM DUAL UNION ALL
SELECT 'SG', DATE '2020-01-22', 0, 0, 0 FROM DUAL UNION ALL
SELECT 'SG', DATE '2020-01-23', 1, 0, 0 FROM DUAL UNION ALL
SELECT 'SG', DATE '2020-01-24', 3, 0, 0 FROM DUAL UNION ALL
SELECT 'NP', DATE '2020-01-22', 0, 0, 0 FROM DUAL UNION ALL
SELECT 'NP', DATE '2020-01-23', 0, 0, 0 FROM DUAL UNION ALL
SELECT 'NP', DATE '2020-01-24', 0, 0, 0 FROM DUAL UNION ALL
SELECT 'MY', DATE '2020-01-22', 0, 0, 0 FROM DUAL UNION ALL
SELECT 'MY', DATE '2020-01-23', 0, 0, 0 FROM DUAL UNION ALL
SELECT 'MY', DATE '2020-01-24', 0, 0, 0 FROM DUAL UNION ALL
SELECT 'CA-BC', DATE '2020-01-22', 0, 0, 0 FROM DUAL UNION ALL
SELECT 'CA-BC', DATE '2020-01-23', 0, 0, 0 FROM DUAL UNION ALL
SELECT 'CA-BC', DATE '2020-01-24', 0, 0, 0 FROM DUAL;
Ensuite, si vous voulez le sortir sous forme de colonnes par jour, utilisez un PIVOT :
SELECT *
FROM (
SELECT name,
latitude,
longitude,
datetime,
infected
FROM Virus_Statistics v
INNER JOIN Regions r
ON ( r.id = v.location )
)
PIVOT (
MAX( infected )
FOR datetime IN (
DATE '2020-01-22' AS "2020-01-22",
DATE '2020-01-23' AS "2020-01-23",
DATE '2020-01-24' AS "2020-01-24"
)
)
Qui sort :
db<>violon ici
