Allowing omitting the "T" separator from a datetime string

Our customers have issue using spark to create a dataframe on a
datetime column, because spark interpolates the provided datetime
string, but drops the "T" separator when filling the range.

It seems that omitting the "T" character is fairly acceptable, and
ISO 8601 does allow it to be omitted in a few cases. This is a
very simple patch to allow it.

Signed-off-by: Rivers Zhang <[email protected]>

Author: riverszhang89

db/types.c

@@ -7202,7 +7202,7 @@ static int _isISO8601(const char *str, int len)
         } else if (str[i] < '0' || str[i] > '9')
             return 1;
         if (!isalnum(str[i]) && str[i] != '-' && str[i] != ':' &&
-            str[i] != '.') {
+            str[i] != '.' && str[i] != ' ') {
             /* invalid character in datetime string */
             return -1;
         }
@@ -7262,7 +7262,7 @@ static int _isISO8601(const char *str, int len)
     /* (month<1 || month>12)*/                                                 \
                                                                                \
     /*get day*/                                                                \
-    day = getInt(in, len, &offset, &ltoken, 1, 2, "-T", &skipped);             \
+    day = getInt(in, len, &offset, &ltoken, 1, 2, "-T ", &skipped);            \
     if (!ltoken)                                                               \
         return CONV_WRONG_MDAY;                                                \
     /*(day <1 || !is_valid_days(year, month, day))*/                           \

tests/datetime.test/t25.expected

@@ -0,0 +1,2 @@
+(rows inserted=1)
+(d="1970-01-01T000000.000 UTC")

tests/datetime.test/t25.sql

@@ -0,0 +1,6 @@
+SET TIMEZONE UTC
+DROP TABLE IF EXISTS t25
+CREATE TABLE t25(d datetime)$$
+INSERT INTO t25 VALUES("1970-01-01 00:00:00 UTC")
+SELECT * FROM t25 WHERE d = "1970-01-01 00:00:00 UTC"
+DROP TABLE t25