SQL Loader (how to cut data header)

Similar Questions

• SQL * Loader does not import data

  Hi all -

  I have a very basic package which should load data from a file delimited by tabs.  My problem is that when I run my package, no data is loaded. Although the correct number of records is created (based on relaxation of the table).  All records contain no data.

  OPTIONS (skip = 1, errors = 10, lines = 10000, direct = True)

  DOWNLOAD THE DATA

  INFILE "C:\ECOMMERCE\VFT\Marin\inbound\DSGSiteCatalystPassbackKeywords.csv" "str"\n"

  BADFILE "C:\ECOMMERCE\MFT\NxtGen_Catalog_Int\ErrorFiles\MARIN_DSG_KEYWORDS.bad."

  DISCARDFILE 'C:\ECOMMERCE\MFT\NxtGen_Catalog_Int\ErrorFiles\MARIN_DSG_KEYWORDS.dsc '.

  IN THE TABLE "ETL_STAGE". "" MARIN_KEYWORD ".

  ADD

  EVALUATE CHECK_CONSTRAINTS

  "FIELDS TERMINATED BY ' 09 X."

  TRAILING NULLCOLS

  (

  MARIN_KEYWORD_ID,

  KEYWORD,

  BUSINESS_DATE,

  EDITOR,

  ACCOUNT,

  CAMPAIGN,

  AD_GROUP,

  TYPE CHAR (100000),

  DESTINATION_URL,

  UNIQUE_ID,

  PUB_ID,

  PRINT,

  CLICKS,

  PUB_COST,

  ATTRIBUTED_CONVERSIONS_CONV,

  CLICK_PATH_CONV,

  LAST_CLICK_CONV,

  EMAIL_SIGNUPS_CONV,

  SCORECARD_SIGNUPS_CONV,

  STORE_LOCATOR_PAGE_CONV

  )

  My table creation script is:

  CREATE THE TABLE ETL_STAGE. MARIN_KEYWORD

  (

  MARIN_KEYWORD_ID VARCHAR2 (1000 BYTE),

  KEYWORD VARCHAR2 (1000 BYTE).

  BUSINESS_DATE VARCHAR2 (200 BYTE),

  EDITOR VARCHAR2 (1000 BYTE),

  ACCOUNT VARCHAR2 (1000 BYTE),

  CAMPAIGN VARCHAR2 (1000 BYTE),

  AD_GROUP VARCHAR2 (1000 BYTE),

  TYPE VARCHAR2 (1000 BYTE),

  DESTINATION_URL VARCHAR2 (1000 BYTE),

  UNIQUE_ID VARCHAR2 (1000 BYTE),

  PUB_ID VARCHAR2 (1000 BYTE),

  VARCHAR2 (1000 BYTE) PRINT,

  VARCHAR2 (1000 BYTE) CLICKS,

  PUB_COST VARCHAR2 (1000 BYTE),

  ATTRIBUTED_CONVERSIONS_CONV VARCHAR2 (1000 BYTE),

  CLICK_PATH_CONV VARCHAR2 (1000 BYTE),

  LAST_CLICK_CONV VARCHAR2 (1000 BYTE),

  EMAIL_SIGNUPS_CONV VARCHAR2 (1000 BYTE),

  SCORECARD_SIGNUPS_CONV VARCHAR2 (1000 BYTE),

  STORE_LOCATOR_PAGE_CONV VARCHAR2 (1000 BYTE),

  IMEX_LOG_REFERENCE_ID VARCHAR2 (1000 BYTE),

  DATE_ADDED DATE DEFAULT SYSDATE NOT NULL,.

  ADDED_BY VARCHAR2 (50 BYTES) BY DEFAULT USER NOT NULL,.

  DATE_LAST_MODIFIED DATE DEFAULT SYSDATE NOT NULL,.

  MODIFIED_BY VARCHAR2 (50 BYTES) BY DEFAULT USER NOT NULL,.

  RECORD_STATUS VARCHAR2 (1 BYTE) DEFAULT 'A' NOT NULL

  )

  TABLESPACE ECOM_DATA

  PCTUSED 0

  PCTFREE 10

  INITRANS 1

  MAXTRANS 255

  STORAGE)

  64K INITIALS

  ACCORDING TO 1 M

  MINEXTENTS 1

  MAXEXTENTS UNLIMITED

  PCTINCREASE 0

  DEFAULT USER_TABLES

  )

  LOGGING

  NOCOMPRESS

  NOCACHE

  NOPARALLEL

  MONITORING;

  The log displays the following text:

  SQL * Loader: release 11.2.0.2.0 - Production on Thu Jun 25 14:31:35 2015

  Copyright (c) 1982, 2009, Oracle and/or its affiliates.  All rights reserved.

  Control file: C:\ECOMMERCE\MFT\NxtGen_Catalog_Int\SQLLoaderScripts\MARIN_DSG_CREATIVES.ctl

  Data file: C:\ECOMMERCE\VFT\Marin\inbound\DSGSiteCatalystPassbackCreatives.csv

  Bad leadership: C:\\ECOMMERCE\MFT\NxtGen_Catalog_Int\ErrorFiles\MARIN_DSG_CREATIVES.bad

  Delete the file: C:\ECOMMERCE\MFT\NxtGen_Catalog_Int\ErrorFiles\MARIN_DSG_CREATIVES.dsc

  (Allow all releases)

  Number of loading: ALL

  Number of jump: 10

  Authorized errors: 50

  Link table: 64 lines, maximum of 256000 bytes

  Continuation of the debate: none is specified

  Path used: classics

  Table 'ETL_STAGE '. "" MARIN_CREATIVE ", loaded from every logical record.

  Insert the option in effect for this table: APPEND

  TRAILING NULLCOLS option in effect

  Column Position Len term Encl. Datatype name

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

  CREATIVE_ID PRIME * WHT O(") CHARACTER

  The maximum field length is 100,000

  HEADLINE THEN * CHARACTER O(") WHT

  BUSINESS_DATE NEXT * CHARACTER O(") WHT

  The SQL string for the column: 'TRIM(:BUSINESS_DATE) '.

  DESCRIPTION_LINE_1 NEXT * CHARACTER O(") WHT

  DESCRIPTION_LINE_2 NEXT * CHARACTER O(") WHT

  DISPLAY_URL NEXT * CHARACTER O(") WHT

  DESTINATION_URL NEXT * WHT O(") CHARACTER

  The maximum field length is 100,000

  EDITOR NEXT * CHARACTER O(") WHT

  THEN THE CAMPAIGN * CHARACTER O(") WHT

  AD_GROUP NEXT * CHARACTER O(") WHT

  UNIQUE_ID NEXT * CHARACTER O(") WHT

  PUB_ID NEXT * CHARACTER O(") WHT

  PRINT NEXT * CHARACTER O(") WHT

  The SQL string for the column: "replace (: print, ',',") "."

  CLICK ON NEXT * CHARACTER O(") WHT

  The SQL string for the column: "replace (: CLICKS, ',',") "."

  ATTRIBUTED_CONVERSIONS_CONV NEXT * CHARACTER O(") WHT

  CLICK_PATH_CONV NEXT * CHARACTER O(") WHT

  LAST_CLICK_CONV NEXT * CHARACTER O(") WHT

  EMAIL_SIGNUPS_CONV NEXT * CHARACTER O(") WHT

  SCORECARD_SIGNUPS_CONV NEXT * CHARACTER O(") WHT

  STORE_LOCATOR_PAGE_HITS_CONV NEXT * CHARACTER O(") WHT

  IMEX_LOG_REFERENCE_ID NEXT * CHARACTER O(") WHT

  Why WHT, showing when I use tabs?

  In addition, why is not display all the data in the files?

  NLS_CHARACTERSET

  WE8ISO8859P1

• How to cut the header in three lines in OBIEE10g?

  Hello
  
  Can someone help by dividing header in three lines...
  
  I can able to divide into two rows by using the table and column names custom, but unable to divide it into 3 lines.
  
  My requirement is
  Row1) Table Name (customised Name)
  row2) column name (custom)
  name of column row3)
  
  Please tell me if it is possible in OBIEE10g or not. I use as oracle 10g database.
  
  
  Thank you
  Pavan
  
  Published by: 911761 on February 17, 2012 04:21

  Hi Pavan,

  You can try to display your name subcolumn within the column header section by adding a line break between your column name and column name. that is in your column heading specify the text as - column name < br / > the name of the column. The br tag will force the column displayed in another line. In addition, to make this work you must add this entry in your instanceconfig.xml file.

  < HardenXSS > false < / HardenXSS >

  See this link for more on the above tag:
  http://gerardnico.com/wiki/dat/OBIEE/hardenxss

  Thank you

• How to cut data in pieces, have been collected by DAQ assistant?

  like the image below, the data were won by a random number generator in a loop for a vector that contains 10 points each time

  scrip MATLAB to process will be. now, I want to get the data by the DAQ assistant

  but I can't my goal, can someone help me?

  

  You could say the DAQ Assistant, to return only the number of samples to be processed.  I'm still lost as to what exactly you are looking for.

• SQL Loader, CLOB, delimited fields

  Hello.
  
  I have to load using SQL Loader of csv file data into a table, a field is of type CLOB.
  
  Here's how ctl file looked like at the beginning:
  
  UNRECOVERABLE
  DOWNLOAD THE DATA
  INFILE «.\csv_files\TSH_DGRA.csv '.
  BADFILE «.\bad_files\TSH_DGRA.bad '.
  DISCARDFILE «.\dsc_files\TSH_DGRA.dsc '.
  ADD
  IN THE TSH_DGRA TABLE
  FIELDS ENDED BY ',' POSSIBLY FRAMED BY "" "
  TRAILING NULLCOLS
  (
  ID_OBJ_TSHD,
  PR_ZOOM_TSHD,
  PR_GRID_TSHD,
  PR_ELMGR_TSHD TANK (4000) OPTIONALLY ENCLOSED BY '< clob >' AND ' < / clob > ',
  PR_ALRMGR_TSHD TANK (4000) OPTIONALLY ENCLOSED BY '< clob >' AND ' < / clob > '
  )
  
  
  The problems are the fields PR_ELMGR_TSHD and PR_ALRMGR_TSHD (CLOB in the table TSH_DGRA). Until which data should be loaded into CLOB fields are at least 4000 characters, it works very well, but what do I do if I want to load data that is older of 4000 characters?
  
  If found on the link: [http://download.oracle.com/docs/cd/B14117_01/server.101/b10825/ldr_loading.htm#i1006803] which a sentence says:
  
  "SQL * Loader by default 255 bytes when moving CLOB data, but you can specify a value of up to 2 gigabytes." For a field bounded, if a length is specified, this length is used as a maximum. If no maximum is specified, it uses default 255 bytes. For a CHAR field that is bounded and is also greater than 255 bytes, you must specify a maximum length. (See TANK for more information on the CHAR data type".
  
  So, my question is, how specify "up to 2 GB" so that text says? I can't use the CHAR data type because it is limited to 4,000 characters. And I have to load about 60000 characters. I also can't use technique where all each CLOB field data in separate files.

  Simply specify the expected maximum size:

  ....
  PR_ELMGR_TSHD TANK (100000), POSSIBLY ENCLOSED BY ""AND"."
  PR_ALRMGR_TSHD TANK (1000000), OPTIONALLY ENCLOSED BY '... « ET ». »
  )

  The CHARIOT (1000000) will allow SQLLDR to manage up to 1000000 bytes of input text.

• How to load xml using sql loader data?

  Is there any load script of XML data in a table through sql loader?

  Help me out here...

  Thank you

  You can post your code here? while I can help you... your code format should be like this...

  Example code:

  DOWNLOAD THE DATA

  INFILE *.

  IN THE TABLE po_tab

  ADD

  XMLTYPE (xmldata)

  FIELDS

  (xmldata TANK (2000))

  BEGINDATA

  "" http://www.Oracle.com/po/"xmlns: xsi ="http://www.w3.org/2001/XMLSchema-instance"

  "" xsi: schemaLocation = "http://www.oracle.com/PO http://www.oracle.com/scha0/po1.xsd"

  orderDate = "1999-10-20" >

  Alice Smith

  123 maple Street

  Mill Valley

  CA

  90952

  Robert Smith

  8 oak Avenue

  Old town

  PA

  95819

  Hurry, my lawn is going wild!

  Lawn mower

  1

  148.95

  Confirm this is electric

  Baby monitor

  1

  39.98

  1999-05-21

• How to handle multiple formats of date to the same date field in SQL * Loader

  Dear all,
  
  I have a requirement where I need to get data from a text file and even insert into the oracle table.
  
  I'm using SQL * Loader to fill data from the text file in my table.
  
  The file has an area where I expect to date data date in different formats, such as MON/dd/yyyy mon/dd/yyyy mon/dd/yyyy, dd/mm/yyyy, DD/month/yyyy.
  
  When you use SQL * Loader, I can see loading fails for records where we have formats such as LUN/jj/aaaa, Mon/dd/yyyy, DD/month/yyyy.
  
  Is it possible in SQL * Loader where we can mention all these date formats so that data from that date should go smoothly in the underlying date column in the table.
  
  Receive your answer on that.
  
  Thank you
  Madhu K.

  I think following could be the solution to your problem:

  Suppose if you have data from four different date formats, you need to load the data into four different tables by running SQL * Loader four times and each time specify other date format in the control file whose data to load, and then at the end, merge data from these four different tables using the UNION clause.

  Elya

• Associating a header for his record in detail record in SQL Loader

  I have a SQL loader script that loads the data in 2 separate tables, a table containing the header and one giving details. Once the charge is completed there is no way of knowing what element is associated with his header.
  Is there a way to associate a header record with its corresponding details?
  
  Published by: 934858 on May 16, 2012 16:49

  >
  2. the header and detail have the item number in the column
  >
  So I expect that 'Article number' provides the association.

  If this isn't the case, then show the header and detail DDL table and some sample data and remember where is the problem.

  This is why I am asking about (and you do not answer) how data loading; a single file or two, flat or hierarchical, files etc. to see how your sql * control of loading script dealt with data.

• Import data... Wizard creates the file SQL Loader ctl with columns out of order

  4.1.1.19 SQL Developer version. Connected to Oracle XE to test this.

  I'm trying to understand what the problem with the data in my import file when finally, I realize that the Import Wizard of data did not care how I traced the columns at all. The ctl SQL Loader file generated by the Wizard expects that the columns of data in my file to match the order that they appear in the definition of the table and not how they have been mapped in the wizard. Manually edit the ctl file is a workaround. Has anyone else seen elsewhere?

  I see that this is a bug.

• SQL Loader loading data into two Tables using a single CSV file

  Dear all,

  I have a requirement where in I need to load the data into 2 tables using a simple csv file.

  So I wrote the following control file. But it loads only the first table and also there nothing in the debug log file.

  Please suggest how to achieve this.

  Examples of data

  Source_system_code,Record_type,Source_System_Vendor_number,$vendor_name,Vendor_site_code,Address_line1,Address_line2,Address_line3

  Victor, New, Ven001, Vinay, Vin001, abc, def, xyz

  Control file script

  ================

  OPTIONS (errors = 0, skip = 1)
  load data
  replace
  in the table1 table:
  fields ended by ',' optionally surrounded "" "
  (
  Char Source_system_code (1) POSITION "ltrim (rtrim (:Source_system_code)),"
  Record_type tank "ltrim (rtrim (:Record_type)),"
  Source_System_Vendor_number tank "ltrim (rtrim (:Source_System_Vendor_number)),"
  $vendor_name tank "ltrim (rtrim (:Vendor_name)),"
  )
  in the Table2 table
  1 = 1
  fields ended by ',' optionally surrounded "" "
  (
  $vendor_name tank "ltrim (rtrim (:Vendor_name)),"
  Vendor_site_code tank "ltrim (rtrim (:Vendor_site_code)),"
  Address_line1 tank "ltrim (rtrim (:Address_line1)),"
  Address_line2 tank "ltrim (rtrim (:Address_line2)),"
  Address_line3 tank "ltrim (rtrim (:Address_line3)).
  )

  the problem here is loading into a table, only the first. (Table 1)

  Please guide me.

  Thank you

  Kumar

  When you do not provide a starting position for the first field in table2, it starts with the following after a last referenced in table1 field, then it starts with vendor_site_code, instead of $vendor_name.  So what you need to do instead, is specify position (1) to the first field in table2 and use the fields to fill.  In addition, he dislikes when 1 = 1, and he didn't need anyway.  See the example including the corrected below control file.

  Scott@orcl12c > test.dat TYPE of HOST

  Source_system_code, Record_type, Source_System_Vendor_number, $vendor_name, Vendor_site_code, Address_line1, Address_line2, Address_line3

  Victor, New, Ven001, Vinay, Vin001, abc, def, xyz

  Scott@orcl12c > test.ctl TYPE of HOST

  OPTIONS (errors = 0, skip = 1)

  load data

  replace

  in the table1 table:

  fields ended by ',' optionally surrounded "" "

  (

  Char Source_system_code (1) POSITION "ltrim (rtrim (:Source_system_code)),"

  Record_type tank "ltrim (rtrim (:Record_type)),"

  Source_System_Vendor_number tank "ltrim (rtrim (:Source_System_Vendor_number)),"

  $vendor_name tank "ltrim (rtrim (:Vendor_name)).

  )

  in the Table2 table

  fields ended by ',' optionally surrounded "" "

  (

  source_system_code FILL (1) POSITION.

  record_type FILLING,

  source_system_vendor_number FILLING,

  $vendor_name tank "ltrim (rtrim (:Vendor_name)),"

  Vendor_site_code tank "ltrim (rtrim (:Vendor_site_code)),"

  Address_line1 tank "ltrim (rtrim (:Address_line1)),"

  Address_line2 tank "ltrim (rtrim (:Address_line2)),"

  Address_line3 tank "ltrim (rtrim (:Address_line3)).

  )

  Scott@orcl12c > CREATE TABLE table1:

  2 (Source_system_code VARCHAR2 (13),)

  3 Record_type VARCHAR2 (11),

  4 Source_System_Vendor_number VARCHAR2 (27),

  5 $vendor_name VARCHAR2 (11))

  6.

  Table created.

  Scott@orcl12c > CREATE TABLE table2

  2 ($vendor_name VARCHAR2 (11),)

  3 Vendor_site_code VARCHAR2 (16).

  4 Address_line1 VARCHAR2 (13),

  5 Address_line2 VARCHAR2 (13),

  Address_line3 6 VARCHAR2 (13))

  7.

  Table created.

  Scott@orcl12c > HOST SQLLDR scott/tiger CONTROL = test.ctl DATA = test.dat LOG = test.log

  SQL * Loader: release 12.1.0.1.0 - Production on Thu Mar 26 01:43:30 2015

  Copyright (c) 1982, 2013, Oracle and/or its affiliates.  All rights reserved.

  Path used: classics

  Commit the point reached - the number of logical records 1

  TABLE1 table:

  1 row loaded successfully.

  Table TABLE2:

  1 row loaded successfully.

  Check the log file:

  test.log

  For more information on the charge.

  Scott@orcl12c > SELECT * FROM table1

  2.

  RECORD_TYPE SOURCE_SYSTEM_VENDOR_NUMBER $VENDOR_NAME SOURCE_SYSTEM

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

  Victor Ven001 new Vinay

  1 selected line.

  Scott@orcl12c > SELECT * FROM table2

  2.

  $VENDOR_NAME VENDOR_SITE_CODE ADDRESS_LINE1 ADDRESS_LINE2 ADDRESS_LINE3

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

  Vinay Vin001 abc def xyz

  1 selected line.

  Scott@orcl12c >

• Date fromats loading using SQL * Loader

  Hello

  I have the data to load into an Oracle DB using sqlldr below

  333. 789. 6 ||| 01-08-2013 | 2014-08-01 |

  334. 789. 6 ||| 01-08-2013 | 2014-08-01 |

  335. 789. 6 ||| 01-08-2013 | 2014-08-01 |

  It fails while the failed to load on date formats. How can I fix everything by using load data.

  Thank you

  Sylvie

  Works for me. You must validate the CTL file.

  Here is an example:

  Control line: -.

  DOWNLOAD THE DATA

  INFILE *.

  IN THE TABLE test_table

  REPLACE

  FIELDS TERMINATED BY ', '.

  (

  Col DATE 'yyyy-mm-dd ".

  )

  BEGINDATA

  2013-08-01

  2014-09-01

  2015-10-01

  2016-11-01

  2017-12-01

  -Create table

  SQL > create table test_table (date of the pass);

  Table created.

  -SQL * Loader

  C:\Users\43729434>sqlldr user/password@db_alias name control=C:\fakepath\test_ctl.ctl

  SQL * Loader: release 11.2.0.1.0 - Production Tue Oct 24 10:26:40 2013

  Copyright (c) 1982, 2009, Oracle and/or its affiliates.  All rights reserved.

  Commit the point reached - the number of logical records 4
  Commit the point reached - the number of logical records 5

  -Check the data in the database

  SQL > alter session set nls_date_format = 'DD-Mon-YYYY ';

  Modified session.

  SQL > select col from test_table;

  COL
  -----------
  August 1, 2013
  01-sep-2014
  October 1, 2015
  November 1, 2016
  December 1, 2017

  SQL >

• SQL * Loader: load data in the format MM/DD/YYYY HH: mi: SS PM

  Please advice how to load data in the format MM/DD/YYYY HH: mi: SS PM in an Oracle Table using SQL * Loader.
  
  -Should what format I give in the control file?
  -What would be the type of column to create the table to load data.
  
  Examples of data below;
  
  MM/DD/YYYY HH: MI: SS PM
  
  14:40:20 09/12/2012
  29/11/2011 11:23:12
  
  Thanks in advance
  Avinash

  Astr0 wrote:
  Hello Srini,

  I tried with the date of creation as DATE data type, but I had an error like

  ORA-01830: date format picture ends before converting entire input string
  CREATION_DATE               "TO_CHAR(TO_DATE(:CREATION_DATE,'MM/DD/YYYY HH:MI:SS AM'),'DD-MON-YYYY HH:MI:SS AM')",
  

  Try

  CREATION_DATE "to_Date(:CREATION_DATE,'yyyy-mm-dd hh:mi:ss AM')",
  

  Does that help? It is charging correctly in the timestamp column

  REDA

• SQL Loader failed to load the same source sometimes data file

  I meet different types of data, loading errors when you try to load data. What makes it really annoying, is that I'm not able to identify the real culprit since the success of load depends on the amount of lines in the source data file but not its content. I use dataset Toad export feature to create delimited data set. When I decided to take the first 50 lines then succeeds data to load into the target table. When I decide to take the first 150 lines then the data to load into the target table fails indicating that:

  Record 13: Rejected - Error on table ISIKUD, column SYNLAPSI.
  ORA-01722: invalid number

  I can see .bad file that the same line has been loaded successfully when the data file that contains 50 rows was used. The content has no column for this particular row is NULL (no chain).
  I suspect that the toad generates faulty delimited text file when taking 150 lines. File data for reasons of confidentiality, I can't show. What can I do? How can we all further investigate this problem? Size of the array that must be loaded by SQL Loader is almost 600 MB. I use Windows XP.
  
  Published by: totalnewby on June 30, 2012 10:17

  I do not believe that customer sqlloader 11g allows you to load a 10g database. You can run sqlldr on the server of database 10g itself? PL also post the rest I asked information

  HTH
  Srini

• Ctl SQL Loader Date Format file

  Hi, I have a source file with the format of date as * [23:02:2012] [04:04:50:579] *, how can I remove ' [' and ']' and load it as 23:02:2012 04:04:50:579.
  
  I tried like
  
  CallSetupTime DATE completed by ',' ENCLOSED BY '[' AND '] ' HH24MISSFFFFF ' DD/MM/YYYY "
  
  BUT did not work. Please help me. Thank you very much..
  
  Kind regards
  Sam

  I suggest to use the correct data type. See here: sql loader timestamp

  for example CallSetupTime TIMESSTAMP "[JJ] [HH24:MI:SS:FFF].

  the format must exactly match the format of your data. Maybe it isa best format for the second millis, see the mentioned thread and formats the docs on the stamp

  concerning

• Encapsulate data problems when loading with sql loader

  Hi all
  
  I use sql loader to load data into a flat file HP UNIX.
  
  I find the fi the NUMBER data type or date type get wraped to the new line, the control file triggered errors.
  
  The data looks like to (field dilimiter is |, record dilimter is ~):
  
  1 A87CCH | 1 A87CCH | PLAN_ACCOUNT_PROMOTION | HIR6A-1 | 20100706 06:06:24 | 1 DNE1 | DC?
  2010.7 FY1011 Promoiton | 1 A87AW0 | 1 HJEZE | Private | 20100730 00:00:00 | 00 20100710
  : 00:00 | 0 | Completed | 20100730 00:00:00 | 20100710 00:00:00 | 0 | 1 4A6PKP | TFMAAI | N
  | 0 | 0 | 0 | 0 | 0 | 0 | 1 4A6PKP | Approved | 1 8U4E-163 | 00:00:20110630 00 |
  20100708 01:45:35 | 20100707 00:00:00 | -1||| 0 | 9000 | 0 | 0 ||| 100. N | 0 | 0 | 0 | 0
  | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | N | 20110426 04:01:34 | 1 8U4E-163 | 0|||||||||
  ||||| ~
  
  
  The control file looks like:
  
  OPTIONS (ERRORS = 1000, DIRECT = TRUE, PARALLEL = TRUE, DATE_CACHE = 5000, discardmax = 50)
  UNRECOVERABLE
  load data
  INFILE ' / home/bs7822/leon/leon.dat' "str" ~ "»
  BADFILE ' / home/bs7822/leon/leon.bad'
  DISCARDFILE ' / home/bs7822/leon/leon.discard'
  ADD THE LEON_123456 TABLE
  FIELDS TERMINATED BY ' | '. SURROUNDED OF POSSIBLY "" "
  TRAILING NULLCOLS
  (
  X_INTERFACED_DT EXPRESSION "to_date (replace (replace (: X_INTERFACED_DT_BF, chr (10),"), Chr (13), "), 'YYYYMMDD hh24:mi:ss')", "
  X_INTERFACED_DT_BF boundfiller,
  EXTERNAL DECIMAL X_ACCRUAL_AMT,
  X_PLAN_SHIPMENT_PATTERN_ID TANK (90)
  )
  
  I think that replace it can treat the wrapped date. But I want to know if we can find a faster or easier way to conquer this topic since the beginning.
  
  Best regards
  
  Leon

  user12064076 wrote:
  Thank you for your response. But how to ensure that a record is in a single line? For example, to unload data with coil?

  The table has more than 100 columns.

  Best regards
  Leon

  UH... which guarantee is implemented by anyone or anything that generates the data in the first place.
  for example if I am extracting data to CSV for a customer or another application, making sure that it is 1 card per line with delimiters known etc.

  What is your own code that produces the data in the first place? If so, how are you to produce? With large amounts of data I wouldn't do it with SQL * Plus spool command, but I do with UTL_FILE within PL/SQL, or generating data in a file CLOB and spell the CLOB in a go using one of the CLOB methods available writing files.