Problem of null external Table spaces

Similar Questions

• problems with the external table to excel file

  Hi all
  I tried to use the statement below to create an external table.this table refers to an excel file.
  
  CREATE TABLE EMPFRAUD_TEST
  (
  SERIAL_NUM WHICH VARCHAR2 (10 BYTE),
  BRANCH_CODE VARCHAR2 (10 BYTE),
  BUSINESS_ADD VARCHAR2 (100 BYTE),
  VARCHAR2 (50 BYTE) REGION,
  DATE OF TRANSACTION_DATE_TIME,
  DATE OF REPORT_DATE_TIME,
  NO_OF_TRANS VARCHAR2 (4 BYTE),
  NUMBER OF AMOUNT,
  FRAUD_TYPE VARCHAR2 (25 BYTE),
  IMPACT_CATEGORY VARCHAR2 (10 BYTE)
  )
  EXTERNAL ORGANIZATION
  (TYPE ORACLE_LOADER
  THE DEFAULT DIRECTORY EXT_EMP_TEST
  ACCESS SETTINGS
  (records delimited by newline
  BadFile "empfraud%a.bad."
  logfile "empfraud%a.log".
  fields completed by «,»
  surrounded of possibly ' "'lrtrim"
  missing field values are null
  )
  LOCATION ("fraud.csv")
  )
  REJECT LIMIT UNLIMITED
  NOPARALLEL
  NOMONITORING;
  
  the problem is the following
  (1) when I run the query above the table will be created,
  but when I try to select in the table, an empty table display.
  When I checked the error log file, the following message was given.
  It was obtained from an oracle on unix Server database.
  
  'L_NUM '.
  KUP-04036: second enclosing delimiter not found
  KUP-04101: Save 71 rejected in the /home/oracle/ext_folder_test/fraud.csv file
  KUP-04021: field error for the field formatting ACCOUNT_KEY
  KUP-04036: second enclosing delimiter not found
  KUP-04101: Save 79 rejected in the /home/oracle/ext_folder_test/fraud.csv file
  KUP-04021: field formatting of error for the serial_num field which
  KUP-04036: second enclosing delimiter not found
  KUP-04101: Save 80 rejected in the /home/oracle/ext_folder_test/fraud.csv file
  column in treatment of TRANSACTION_DATE_TIME error in the row 1 for the /home/oracle/ext_folder_test/fraud.csv data file
  ORA-01858: a non-digit character was found here where was waiting for a digital
  column in treatment of TRANSACTION_DATE_TIME error in line 2 to the /home/oracle/ext_folder_test/fraud.csv data file
  ORA-01843: not one month valid
  column in treatment of TRANSACTION_DATE_TIME error on line 3 for the /home/oracle/ext_folder_test/fraud.csv data file
  ORA-01843: not one month valid
  column in treatment of TRANSACTION_DATE_TIME error in row 8 for the /home/oracle/ext_folder_test/fraud.csv data file
  ORA-01843: not one month valid
  column in treatment of TRANSACTION_DATE_TIME error in line 9 to the /home/oracle/ext_folder_test/fraud.csv data file
  ORA-01843: not one month valid
  column in treatment of TRANSACTION_DATE_TIME error in line 10 by /home/oracle/ext_folder_test/fraud.csv data file
  ORA-01843: not one month valid
  column in treatment of TRANSACTION_DATE_TIME error at line 11 to the /home/oracle/ext_folder_test/fraud.csv data file
  ORA-01858: a non-digit character was found here where was waiting for a digital
  column in treatment of TRANSACTION_DATE_TIME error in line 12 to the /home/oracle/ext_folder_test/fraud.csv data file
  ORA-01843: not one month valid
  column in treatment of TRANSACTION_DATE_TIME error in the 13th for the /home/oracle/ext_folder_test/fraud.csv data file row
  ORA-01843: not one month valid
  column in treatment of TRANSACTION_DATE_TIME error in line 14 to the /home/oracle/ext_folder_test/fraud.csv data file
  ORA-01843: not one month valid
  column in treatment of TRANSACTION_DATE_TIME error in row 15 for the /home/oracle/ext_folder_test/fraud.csv data file
  "ORA-01843: not one month valid."
  
  
  pls I need help to fix it fast
  Thank you
  
  concerning
  Abdulrahman olayide arpin
  
  NB:
  After the conversion to .csv format.
  HERE IS THE DATA THAT I'M TRYING TO ACCESS THE EXCEL FILE
  
  REGIONS, Transaction_Date_Time, Report_Date_Time, Account_Key (account number), BUSINESS OFFICE, Number_of_Transactions, Total_Amount (N)
  Ojo 1,162,9 Street, South of Lagos, various, 17 January 00, CHECK
  2,0238, "10 cyril Road, Enugu", Orient, 21/06/2006,23/12/10,020968765357 09867653920174,1,20000000
  3,0127, '261, Roland Rd, Asaba', Midwest, 22/12/2010,23/12/10, 1,'00160030006149, 6000000
  4,0519, 'just road, Onitsha.
  ', East, 12/03/2010,14/02/11,0896002416575,1,5000000.
  5,0519, 'just road, Onitsha.
  ", Orient, 03/12/2010,14/02/11,06437890134356,1,5000000.
  6 149, rue olayide, South NGN01492501036 of Lagos, 10/02/2010,17/02/11, 1,6108950
  7,0066, wale, mi - west,18/02/2011,18/02/10,'05590020002924,1,55157977.53
  8.66, john, Mid-west, 6787500, 1,'00660680054177 11/03/2010,14/03/09
  Waheed 9,0273, Biem, N/central, Jan 09 Dec/2010,01/04/11, Nile, 1, 14146040
  

• External table to read another encoding

  Hi all
  
  I am having some problems with the external table. My database coding is currently AL32UTF8. If my external table wants to read the txt file with the ZHS16GBK, the value is displayed not correctly when the external table reads the file. Y at - it means to display the value in the file txt correctly without changing the database coding and coding in the txt file.

  You can give a character defined in the external table definition.
  See for example: http://asktom.oracle.com/pls/apex/f?p=100:11:0:P11_QUESTION_ID:6611962171229 #10978420712036
  There is also the {forum: id = 732} Forum, where you could find/ask as well.

• Facing many problems on the creation of directory and an external Table

  Question:
  
  The strange thing is that if you look at the question 10-b page 3-41, it says:
  (page 3-41 "Oracle Database 10g SQL Fund. II Vol.1")
  
  Merge the data into the EMP_DATA table that is created in the lab last in the data in the table emp_hist. assume
  EMP_DATA external table data corresponds to the EMP_HIST of table, update the email column
  table EMP_HIST to match the row in the EMP_DATA table. If a row in the EMP_DATA table is not
  match, to be inserted in the tables of EMP_HIST lines are considered as corresponding whenever his first and
  family name are the same.
  
  For me, this issue is built wrong. First of all in the last lab we have not been asked to create EMP_DATA. Secondly, EMP_DATA is empty.
  Thirdly, this question asks us to merge into the table EMP_HIST while EMP_DATA is empty.
  
  Table EMP_HIST currently copied data from the employees table. Structure EMP_HIST:
  
  FIRST NAME VARCHAR2 (20)
  LAST_NAME NOT NULL VARCHAR2 (25)
  EMAIL NOT NULL VARCHAR2 (45)
  
  Anway, I did the merger as follows:
  
  merge into e emp_hist
  with the help of emp_data d
  on (e.first_name = d.first_name)
  When matched then
  game update
  Select = d.last_name,
  e.email = d.email
  When not matched then
  Insert values (d.first_name, d.last_name, d.email);
  
  I get this error:
  
  Error report:
  SQL error: ORA-29913: error in executing ODCIEXTTABLEOPEN legend
  ORA-29400: data cartridge error
  KUP-04040: file emp.dat in EMP_DIR not found
  ORA-06512: at "SYS." ORACLE_LOADER', line 19
  29913 00000 - "error in the execution of %s legend".
  * Cause: The execution of the specified legend caused an error.
  * Action: Examine the error messages take appropriate measures.
  
  On the other hand, I said I'm going to try this:
  
  merge into emp_data d
  using e emp_hist
  on (d.first_name = e.first_name)
  When matched then
  game update
  d.last_name = select,
  d.email = e.email
  When not matched then
  Insert values (e.first_name, select, e.email);
  
  I get this error because the external table is final once its creation is as far as I know:
  
  Error report:
  SQL error: ORA-30657: operation not supported on external organized table
  30657.0000 - "operation not supported on external-organized table".
  * Cause: User attempted on the operation on an external table which is
  not supported.
  * Action: Don't do that!
  
  **********************************
  
  I don't know what to do. I did my best, please help.
  
  Published by: user11164565 on July 27, 2009 02:43

  user11164565 wrote:
  NOTE: I did my best, I've done everything I can do, but the problem persists. Help, please

  I'll mention all the steps that I made it clear...

  I gave scott the following grants:

  grant create any directory Scott;
  grant read on the directory emp_dir to scott;

  -----------------------

  1. create a directory and its been created successfully:

  create or replace directory emp_dir
  like 'F:\emp_dir ';

  Then I made just to make sure that my directory is recognized the following:

  SELECT *.
  OF dba_directories;

  I found the drive among the results...

  OWNER NOM_REPERTOIRE
  ------------------------------ ------------------------------
  DIRECTORY_PATH
  --------------------------------------------------------------------------------
  SYS EMP_DIR
  F:\emp_dir

  SYS SUBDIR
  D:\oracle\product\10.2.0\db_1\demo\schema\order_entry\/2002/Sep

  SYS XMLDIR
  D:\oracle\product\10.2.0\db_1\demo\schema\order_entry\

  ----------------------
  2. I created an external table emp_data (the script is given by the text book): successfully done

  drop table emp_data;

  CREATE TABLE emp_data
  (first name VARCHAR2 (20))
  , last_name VARCHAR2 (20)
  , email VARCHAR2 (30)
  )
  EXTERNAL ORGANIZATION
  (
  TYPE oracle_loader
  Emp_dir default DIRECTORY
  ACCESS SETTINGS
  (
  RECORDS DELIMITED BY NEWLINE CHARACTERSET US7ASCII
  NOBADFILE
  NOLOGFILE
  FIELDS
  (first name POSITION (01:20) TANK)

  
  , last_name POSITION (22:41) TANK
  CHAR POSITION (43:72) by email)
  )
  LOCATION ('emp.dat'));

  --------------------------------

  3. I went to drive F:\ to see if emp_dir file exist or not! I did not see it. I checked the files hidden, nothing. Anyway, I ignored and did step 4.

  "Anyway, I ignored him... "

  and so the rest of your problems. I do not see in the steps you told that acually you created a directory ("folder") named "\emp_dir" on your f: drive. Anything that you create the database will actually create this directory on the operating system. Uttering a directory in Oracle, uttering an external table in Oracle, don't create pointers to Oracle objects will assume just actually exists.

• Access to an external table via the display in the UCM configuration manager problem

  Hi all

  I know that this question is already asked here: to access the external tables in DB / in UCM Applet configuration manager but the conversation has already been archived, and I can't comment there. I'm following approach given by William Phelps in above said archived thread.

  I am able to create dblink with data remote and created the view by using the linked table in my diagram of the Complutense University of Madrid. The problem that I face is that this point of view such as told by William is not visible in the configuration manager. Although I tried to give (create view and create all views) system diagram University Complutense of MADRID user privileges, but I'm not sure that it will also provide access rights. I am using the SQL Developer and working at the University Complutense of MADRID 11 g.

  All comment or suggestion will be a great help.

  Stéphane yapi

  The value of this configuration variable: EBRIncludeViewsInTableList = 1

  https://jonathanhult.com/blog/2013/11/use-database-view-WebCenter-content-schema-view/

  Jonathan

  http://jonathanhult.com

• problem of counting of returns to the clob line with external tables

  Hello
  
  I want to count the number of lines (x0A) in a clob loaded from an external table.
  This is the data of the file

  d               2010-06-21 13:25:23               2               1073807679               0               4               1               1               0               415               348                mtp       16387 ALIGN TIMER 2 EXPIRED iacSt=8 (Lnk 1)
  
  
  xx               2010-06-21 13:25:23               2               1073807679               0               4               1               1               0               415               348                mtp       16387 Sent SIOS (Lnk 1) (AlignNP in InitialAlign)

  That's how the loaded hex of the clob looks like:

  64 09 09 09 32 30 31 30 2D 30 36 2D 32 31 20 31 33 3A 32 35 3A 32 33 09 09 09 32 09 09 09 31 30 37 33 38 30 37 36 37 39 09 09 09 30 09 09 09 34 09 09 09 31 09 09 09 31 09 09 09 30 09 09 09 34 31 35 09 09 09 33 34 38 09 09 09 20 6D 74 70 20 20 20 20 20 20 20 31 36 33 38 37 20 41 4C 49 47 4E 20 54 49 4D 45 52 20 32 20 45 58 50 49 52 45 44 20 69 61 63 53 74 3D 38 20 28 4C 6E 6B 20 31 29 0A 0A 0A 78 78 09 09 09 32 30 31 30 2D 30 36 2D 32 31 20 31 33 3A 32 35 3A 32 33 09 09 09 32 09 09 09 31 30 37 33 38 30 37 36 37 39 09 09 09 30 09 09 09 34 09 09 09 31 09 09 09 31 09 09 09 30 09 09 09 34 31 35 09 09 09 33 34 38 09 09 09 20 6D 74 70 20 20 20 20 20 20 20 31 36 33 38 37 20 53 65 6E 74 20 53 49 4F 53 20 28 4C 6E 6B 20 31 29 20 28 41 6C 69 67 6E 4E 50 20 69 6E 20 49 6E 69 74 69 61 6C 41 6C 69 67 6E 29 0A 0A 0A 

  You can see that there is a total of 6 '0 a' or 6 new lines. However, I get the following results:

  select regexp_count(outputrow,'[[:cntrl:]]') control,regexp_count(outputrow,'[[:space:]]') space, regexp_count(outputrow,'[\n]') backslash_n, regexp_count(outputrow,'[\x0A]') x0A  
  from BULK_BAD_FILE_EXTERNAL b;
     CONTROL      SPACE BACKSLASH_N        X0A
  ---------- ---------- ----------- ----------
          72        104           9         22

  what I am doing wrong?

  11g is not yet I'd

  select length(outputrow) - length(replace(outputrow,chr(10))) num_of_rows
    from bulk_bad_file_external
  

  Concerning

  Etbin

• External table WHITE vs. NULL

  Hello
  
  I need help. I have create a portal with a load when the condition and most of the data appears in the external table except those with COL1 that blanks/null.
  First row is not in the external table that make sense.
  However, why second line isn't in the external table? How to put at the disposal of external table? I used load when col1 = BLANKS but no record.
  
  CSV file
  ABC, 1, 2, 3
  2,3,4
  VDB, 1, 2, 3
  Qwe, 1, 2, 3
  
  
  CREATE TABLE TEST
  (COL1 VARCHAR2 (50))
  COL2 VARCHAR2 (50).
  COL3 VARCHAR2 (50).
  COL4 VARCHAR2 (50))
  EXTERNAL ORGANIZATION
  (
  TYPE ORACLE_LOADER DEFAULT DIRECTORY 'INPUTS_DIR' (SETTINGS ACCESS
  RECORDS DELIMITED BY '\r\n' LOADING WHEN
  (
  Col1! = 'abc '.
  )
  NOBADFILE NODISCARDFILE COMPLETED FIELDS NOLOGFILE BY ',' OPTIONALLY
  BOX BY "" "
  AND ' "').
  LOCATION ("test1.csv")
  )
  REJECT LIMIT UNLIMITED;
  
  Thank you
  AL

  I have miissed OPTIONALLY ENCLOSED BY ' "', so that my solution will not load records as:"

  "", 5,6,7.

  Corrected table definition would be:

  {code}
  CREATE TABLE TEST
  (COL1 VARCHAR2 (50))
  COL2 VARCHAR2 (50).
  COL3 VARCHAR2 (50).
  COL4 VARCHAR2 (50))
  EXTERNAL ORGANIZATION
  (
  TYPE ORACLE_LOADER DEFAULT DIRECTORY ACCESS (SETTINGS TEMP
  RECORDS DELIMITED BY '\r\n' LOADING WHEN
  (
  Col1! = 'abc' OR (1 / 1) = ',' OR (1 / 3) = "" ","
  )
  NOBADFILE NODISCARDFILE COMPLETED FIELDS NOLOGFILE BY ',' OPTIONALLY
  BOX BY "" "
  AND ' "').
  LOCATION ("test1.csv")
  )
  REJECT LIMIT UNLIMITED;
  {code}

  SY.

• Problem with the format of the external tables

  I am using the external Tables feature in Oracle.
  
  The code I am using currently to create my table is (I cut down the list of fields for clarity).

  create table EMP_EXTERNAL2
  (HIREDATE DATE )
  organization external
  ( default directory EXT_TABLES
  access parameters
  ( records delimited by newline
  fields terminated by ',')
  location ('EMP.CSV') )
  
  REJECT LIMIT 0
  ;

  But I get rejected folders when I select in the table due to the format of the HIRE date. The examples I've found suggest using a date format, change the code to:

  create table EMP_EXTERNAL2
  (HIREDATE DATE 'mm/dd/yyyy')
  organization external
  ( default directory EXT_TABLES
  access parameters
  ( records delimited by newline
  fields terminated by ',')
  location ('EMP.CSV') )
  
  REJECT LIMIT 0
  ;

  When I run this I get ORA-00907: lack the closing parenthesis. What I'm doing wrong here (I'm under the code in SQL Developer incidentally)?

  Hello

  Try this: -.

  create table EMP_EXTERNAL2
  (HIREDATE DATE )
  organization external
  ( default directory EXT_TABLES
  access parameters
  ( records delimited by newline
  fields terminated by ','
  (
  HIREDATE DATE 'MM/DD/YYYY'
  ))
  location ('EMP.CSV') )
  
  REJECT LIMIT 0
  

  -Pavan Kumar N

• External table with the preprocessor file to display the list of files

  Hello

  in my db 11.2.0.2 I would create an external table with script preprocessor to show me the list of files in a directory.

  Preprocessor banally:

  #! / bin/bash

  CD/MyDir

  / bin/ls-l *.txt 2 >/dev/null

  The problem is the file name that contains white space (style windows in a Linux env), for ex:

  -rw - r - r - 1 oracle oinstall 920 9 Jun 17:37 File1.txt

  -rw - r - r - 1 oracle oinstall 72316 Jun 10 10:37 file GC output1.txt

  -rw - r - r - 1 oracle oinstall 72316 Jun 10 10:40 GC file output2.txt

  So I can't use FIELDS DELIMITED BY "". ""

  FIXED size? No, because the file size change the length of the lines.

  I tried to use awk in my preprocessos script, but I have the same problem:

  
  

  / bin/ls-l *.txt 2 >/dev/null | / bin/awk ' {printf "%s %.2d %s %s\n", $6, $ 7, $ 8, $9} '.

  
  

  Any ideas?
  

  Hello

  I used java to retrieve information about the files.

• ORA-29913 for external table on Windows 7

  Hello

  Please help with the following problem.

  I installed Oracle 11 g on Windows 7 Pro workstation and have created an external table.

  I'm getting an ORA-29913 when I select.

  What I did is given below.

  create the directory imp_files as 'C:\app\user\admin\orcl\dpdump\'-'C:\orcl_work '.
  ;

  drop table people
  ;
  create table people)
  first name varchar2 (250).
  VARCHAR2 (250) last_name,.
  hire_date date,
  number of salary
  )
  (external) Organization
  type oracle_loader
  the default directory data_pump_dir
  (settings) access
  records delimited by newline
  BadFile data_pump_dir: 'pers1%a_%p.bad' /*'pers.bad'*/
  data_pump_dir log file: "pers1%a_%p.log" /*'pers.log'*/
  fields termintated by ' |'
  missing field values are null
  (first_name, last_name, hire_date 'dd.mm.yyyy' format mask, salary)
  )
  location ("pers_table1.txt")
  )
  reject limit unlimited
  ;

  Hello

  you completed wrongly spelt...

  Try to fix that and try again.

  See you soon,.

  rich

• Pretreatment of an external table - table is empty at the end

  Hello

  Oracle Database 11 g Enterprise Edition Release 11.2.0.1.0 - 64 bit Production

  PL/SQL Release 11.2.0.1.0 - Production

  "CORE 11.2.0.1.0 Production."

  AMT for Solaris: Version 11.2.0.1.0 - Production

  NLSRTL Version 11.2.0.1.0 - Production

  Solaris 10 x 86

  I'm trying to create an external table that first pre-process a file and then the bed. The problem is that, in the end, the outer table is empty even if the read file is not empty.

  First, the table works gunzip on the 'employees.csv.gz' file located in/export/home/oracle/archives and then bed. The file employees.csv.gz exists in the specified location. This is the complete script:

  -This Directory keeps the archived files

  CREATE or REPLACE DIRECTORY arch_dir AS 'archives/export/home/oracle/archives";

  -This directory shows where the command gunzip

  CREATE or REPLACE the bin_dir AS DIRECTORY ' / usr/bin';

  CREATE TABLE emp_exadata

  (

  employee_id NUMBER (22, 0),

  first name VARCHAR2 (20).

  last_name VARCHAR2 (25).

  e-mail VARCHAR2 (25).

  Phone_Number VARCHAR2 (20).

  hire_date DATE,

  job_id VARCHAR2 (10),

  higher wages (22: 2),

  commission_pct NUMBER 22 (2)

  manager_id NUMBER (22, 0),

  department_id NUMBER (22, 0)

  )

  EXTERNAL ORGANIZATION

  (

  TYPE oracle_loader

  Arch_dir default DIRECTORY

  ACCESS SETTINGS

  (

  RECORDS delimited BY NEWLINE

  preprocessor bin_dir: "gunzip".

  end fields BY «»

  missing field VALUES are NULL

  (

  employe_id,

  first name,

  last_name,

  E-mail

  Phone_Number,

  hire_date CHAR date_format DATE mask "dd-mm-yyyy hh24:mi:ss."

  job_id,

  salary,

  commission_pct,

  manager_id,

  department_id

  )

  )

  LOCATION ("employees_exp.csv.gz")

  )

  REJECT LIMIT UNLIMITED;

  When I choose to emp_exadata the result set is empty!

  SELECT * FROM emp_exadata;

  no selected line
  

  When I look at the db server in the directory /export/home/oracle/archives I see no archived file employees_exp.csv. Here is the result of the first three lines:

  
  

  bash - $3.2-3 employees_exp.csv head

  198, Donald, Ollivier, DOCONNEL, 650.507.9833, 21/06/2007 00:00:00, SH_CLERK, 2600, 124, 50,.

  199, Douglas, grant, DGRANT, 650.507.9844, 2008-01-13 00:00:00, SH_CLERK, 2600, 124, 50,.

  200 Jennifer Whalen, JWHALEN 515.123.4444, 17/09/2003 00:00:00, AD_ASST, 4400, 101, 10.

  The end of the lines in the file line is LF (unix style). The encoding is in ANSI format.

  I tried to experiment around, but cannot view records when I select in the external table. Please help me to solve it.

  I also register the generated log file:

  LOG file opened at 01/06/15 16:40:11

  For table EMP_EXADATA field definitions

  Record format DELIMITED BY newline

  Data in file have same "endianness" as platform

  Rows with all null fields are accepted

  Fields of the Data Source:

  EMPLOYEE_ID CHAR (255)

  Completed by «,»

  Trim whitespace same as SQL Loader

  FIRST NAME CHAR (255)

  Completed by «,»

  Trim whitespace same as SQL Loader

  LAST_NAME CHAR (255)

  Completed by «,»

  Trim whitespace same as SQL Loader

  EMAIL CHAR (255)

  Completed by «,»

  Trim whitespace same as SQL Loader

  PHONE_NUMBER CHAR (255)

  Completed by «,»

  Trim whitespace same as SQL Loader

  HIRE_DATE TANK (19)

  Day DATE data type, date mask dd-mm-yyyy hh24:mi:ss

  Completed by «,»

  Trim whitespace same as SQL Loader

  JOB_ID CHAR (255)

  Completed by «,»

  Trim whitespace same as SQL Loader

  SALARY CHAR (255)

  Completed by «,»

  Trim whitespace same as SQL Loader

  COMMISSION_PCT CHAR (255)

  Completed by «,»

  Trim whitespace same as SQL Loader

  MANAGER_ID CHAR (255)

  Completed by «,»

  Trim whitespace same as SQL Loader

  DEPARTMENT_ID CHAR (255)

  Completed by «,»

  Trim whitespace same as SQL Loader

  You need to carefully examine the description of the PREPROCESSOR option in the chapter of the manual external Tables utility.

  The first point that applies to your question is that the preprocessor must write its data transformed to stdout. To do this with gunzip, you use the - c command line parameter. The second point that applies to your case, in view of the answer to the first point, is that you must write a script shell if your preprocessor requires command line settings.

  Kind regards

  Bob

• How to select the csv data stored in a BLOB column as if it were an external table?

  Hi all

  (Happy to be back after a while! )

  Currently I am working on a site where users should be able to load the data of csv (comma is the separator) of their client machines (APEX 3.2 application) in the Oracle 11.2.0.4.0 EE database.

  My problem is:

  I can't use an external table (for the first time in my life ) so I'm a little clueless what to do as the csv data is stored by the application of the APEX in a BLOB column, and I'm looking for an elegant way (at least SQL PL/SQL/maximization) to insert the data into the destination table (run validations by a MERGER would be the most effective way to do the job).

  I found a few examples, but I think they are too heavy and there could be a more elegant way in Oracle DB 11.2.

  Simple unit test:

  drop table CBC purge;

  drop table dst serving;

  create table src

  (myblob blob

  );

  create table dst

  (num number

  , varchar2 (6) str

  );

  Insert in src

  Select utl_raw.cast_to_raw (1; AAAAAA ;'|| Chr (10) |

  2; BATH; »

  )

  Double;

  Desired (of course) output based on the data in table SRC:

  SQL > select * DST;

  NUM STR

  ---------- ------

  1 ABDELKRIM

  2 BATH

  Does anybody know a solution for this?

  Any ideas/pointers/links/examples are welcome!

  / * WARNING: I was 'off' for about 3 months, then the Oracle - the part of my brain has become a bit "rusty and I feel it should not be so complicated as the examples I've found sofar ' * /"

  Haha, wonder about regexp is like the blind leading the blind!

  However, it's my mistake: I forgot to put the starting position setting (so 1, 2, 3,... was in fact the starting position, not the nth occurrence. duh!)

  So, it should actually be:

  select x.*
  ,      regexp_substr(x.col1, '[^;]+', 1, 1)
  ,      regexp_substr(x.col1, '[^;]+', 1, 2)
  ,      regexp_substr(x.col1, '[^;]+', 1, 3)
  ,      regexp_substr(x.col1, '[^;]+', 1, 4)
  ,      regexp_substr(x.col1, '[^;]+', 1, 5)
  ,      regexp_substr(x.col1, '[^;]+', 1, 6)
  from  src
  ,      xmltable('/a/b'
                passing xmltype(''||replace(conv_to_clob(src.myblob), chr(10), '')||'')
                columns
                  col1 varchar2(100) path '.') x;
  

  Note: that's assuming that all the "columns" passed in the string won't be lame.

  If one of them might be null, then:

  select x.*
  ,      regexp_substr(ltrim(x.col1, ';'), '[^;]+', 1, 1)
  ,      regexp_substr(ltrim(x.col1, ';'), '[^;]+', 1, 2)
  ,      regexp_substr(ltrim(x.col1, ';'), '[^;]+', 1, 3)
  ,      regexp_substr(ltrim(x.col1, ';'), '[^;]+', 1, 4)
  ,      regexp_substr(ltrim(x.col1, ';'), '[^;]+', 1, 5)
  ,      regexp_substr(ltrim(x.col1, ';'), '[^;]+', 1, 6)
  from   src
  ,      xmltable('/a/b'
                passing xmltype(replace(';'||replace(conv_to_clob(src.myblob), chr(10), ';')||'', ';;', '; ;'))
                columns
                  col1 varchar2(100) path '.') x;
  

• How to load the file tabs with field-delimited a comma in an external table

  I am trying to create an external table in oracle 11 g r2. the script is as below. It fails if the field contains commas.

  for example, the following data will not be loaded.

  ERT 123, poipoipoi, yutio 567

  Please suggest how to solve this problem. Thank you!

  CREATE TABLE external_interaction
  (
  TAX_ID NUMBER (8).
  pubmed_id_list varchar2 (36),
  interaction_id_type varchar2 (36)
  )
  EXTERNAL ORGANIZATION
  (TYPE ORACLE_LOADER
  THE DEFAULT DIRECTORY ET_NCBI_DIR
  ACCESS SETTINGS
  (records delimited by 0 x '0A'
  jump 1
  BADFILE et_ncbi_log_dir: 'interactions.bad'
  Et_ncbi_log_dir LOG file: 'interactions.log'
  fields terminated by '\t '.
  missing field values are null
  REJECT ROWS WITH ALL FIELDS ARE NULL
  (
  TAX_ID,

  pubmed_id_list,
  interaction_id_type
  )
  )
  LOCATION (ET_NCBI_DIR: "interactions")
  )
  REJECT LIMIT UNLIMITED
  NOPARALLEL
  NOMONITORING;

  Thanks for the reply.

  Should I use

  collumn_name tank (2000) in the table creation script because it is larger than the default char (255)

• In the external table files forming themselves!

  Hi all

  I am facing a weird situation.

  Few days back, DB has been improved to 11g 10g R2 and we are facing the situation of the training of logfiles, throw files have joined with external tables in the directory of the OS!

  Without job is calling them and the only procedure in which these paintings have been used in the SELECT statement, are not running, even while these files form themselves and fill space in the BONE.

  Here is the version information:

  Oracle Database 11 g Enterprise Edition Release 11.2.0.3.0 - 64 bit Production

  PL/SQL Release 11.2.0.3.0 - Production

  "CORE 11.2.0.3.0 Production."

  AMT for IBM/AIX RISC System/6000: Version 11.2.0.3.0 - Production

  NLSRTL Version 11.2.0.3.0 - Production

  Please help in this regard.

  Thank you.

  Hi all

  Problem has been identified.

  11 g, dbms_stats.gather_stats_job runs all day until 22:00 and after every 4 hours on weekends.

  He also tried gathering statistics for external tables given through the GATHER_TABLE_STATS procedure.

  In the process, he joint/written in the log files.

  Now, I think of locking statistics from this table, so that no stats will be gathered for what is just an EXTERNAL table which is used/populated only once a month.

  Suggestions or comments?

  Thank you.

• External table with the empty file

  Hello

  My version of db: Oracle 11 g

  I have a csv file that is empty.

  I created an external table the empty csv file.

  When I run:

  Select count (*) in the externaltblname;

  It returns 1.

  It should return 0 to the right.

  In the definition, I specified "SKIP 1.

  But he always returns 1.

  When I use this external table to load into a table target. It loads a single row with null values.

  How to solve this problem. Please advice.

  What works for me is the following (t_ext points to a blank csv):

  SQL > select count (field) in the t_ext;

  COUNT (FIELD)

  ------------

  1

  1 selected line.

  SQL > select ascii (field) in the t_ext;

  ASCII (FIELD)

  ------------

  13

  1 selected line.

  SQL > select count (replace (field, chr (13))) of t_ext;

  COUNT (REPLACE (FIELD, CHR (13)))

  -----------------------------

  0

  1 selected line.