Hentning af bestemte strenge i en streng ved hjælp af Oracle SQL

Du kan bruge regexp_substr at opdele dine inputdata i linjer og derefter se efter de relevante strenge, for eksempel:

SQL> SELECT regexp_substr(line, 'aXYZApple[^,]*') subtxt
  2    FROM (SELECT regexp_substr(:x, '[^|]*\|', 1, rownum + 1) line
  3             FROM dual
  4           CONNECT BY LEVEL <= length(:x) - length(REPLACE(:x, '|', '')))
  5   WHERE regexp_substr(line || ',', '[^,]*,', 1, 2) = 'OU=Managed,'
  6     AND line LIKE '%aXYZApple%';

SUBTXT
--------------------------------------------------------------------------------
aXYZApple-Au
aXYZApple-Readonly
aXYZApple-Write

Her er en lille forklaring. Du skal gennemgå forespørgslen trin for trin.

Den indre del af forespørgslen vil gennemgå dine data (for hver |). ):

SQL> SELECT regexp_substr(:x, '[^|]*\|', 1, rownum + 1) line
  2    FROM dual
  3  CONNECT BY LEVEL <= length(:x) - length(REPLACE(:x, '|', ''));

LINE
--------------------------------------------------------------------------------
 CN=aXYZApple-Au,OU=Managed,OU=Groups,DC=core,DC=dir,DC=abc,DC=com |
 CN=31107427,OU=Distribution Lists,OU=Shared Mailboxes,DC=core,DC=dir,DC=abc,DC=
 CN=ea90045052,OU=Groups,OU=eProfile,DC=core,DC=dir,DC=abc,DC=com |
 CN=S0901448,OU=Distribution Lists,OU=Shared Mailboxes,DC=core,DC=dir,DC=abc,DC=
 CN=00900887,OU=Distribution Lists,OU=Shared Mailboxes,DC=core,DC=dir,DC=abc,DC=
 CN=NSMMMM,OU=LRP,OU=Groups,DC=core,DC=dir,DC=abc,DC=com |
 CN=aXYZApple-Readonly,OU=Managed,OU=Groups,DC=core,DC=dir,DC=abc,DC=com |
 CN=WWSWW-Au,OU=LRP,OU=Groups,DC=core,DC=dir,DC=abc,DC=com |
 CN=aLogical_RW,OU=Managed,OU=Groups,DC=core,DC=dir,DC=abc,DC=com |
 CN=aXYZApple-Write,OU=Managed,OU=Groups,DC=core,DC=dir,DC=abc,DC=com |

Du ville derefter loope for OU=Managed streng i anden position:

SQL> SELECT regexp_substr(line || ',', '[^,]*,', 1, 2) second_part
  2    FROM (SELECT regexp_substr(:x, '[^|]*\|', 1, rownum + 1) line
  3             FROM dual
  4           CONNECT BY LEVEL <= length(:x) - length(REPLACE(:x, '|', '')));

SECOND_PART
--------------------------------------------------------------------------------
OU=Managed,
OU=Distribution Lists,
OU=Groups,
OU=Distribution Lists,
OU=Distribution Lists,
OU=LRP,
OU=Managed,
OU=LRP,
OU=Managed,
OU=Managed,

Til sidst skal du vælge den relevante del med en sidste regexp_substr .