CLOB variable: concatenate or append?

Similar Questions

• Write a CLOB variable in a file in the log file

  Hi, I am currently building a dynamic sql statement in a plsql procedure that stores the instruction in a CLOB variable (the statement is very long), which has been declared as:
  
  vc_sql_statement CLOB.
  Now, I want to write this statement to a file using the UTL FILE function.
  
  Because the variable is greater than 32K, can anyone please suggest another way of how I can file this statement in a log file.
  
  I use version oracle 11g.
  
  Thank you very much.

  Three ways to do it.

  http://www.morganslibrary.org/howcani.html

  Look at #4 on the page linked above.

• Store > 32K in a CLOB variable

  Oracle 11.1

  CREATE OR REPLACE PROCEDURE foo(p_input IN clob,p_output OUT INTEGER)
  IS
  BEGIN
      WITH data AS (
          SELECT xmltype(p_input) xml
          FROM dual
      )
      SELECT
          length(extractValue(d.xml,'//value'))
      INTO p_output
      FROM data d;
  END;
  /
  
  

  The incoming XML content passed in FOO can be greater than 32K so I just want to make sure that Oracle takes care of everything, but I am not able to set up a simple test case. I Googled, looked up the old sons of RTO and AskTom, but no matter that I do either ORA-01706: user function result value was too big or ORA-06502: PL/SQL: digital or value error: character string buffer too small

  OK so extractValue returns a VARCHAR2 so that it cannot handle more than 4000, but so I tried to use extract(d.dml,'//value').getclobval , but would not compile with PL/SQL: ORA-22806: not an object or REF.  , I'm missing a few concepts here.

  DECLARE
      l_clob clob;
      l_open clob := '<node><value>';
      l_close clob := '</value></node>';
      l_output INTEGER;
  BEGIN
       DBMS_LOB.createtemporary(l_clob,TRUE);
       dbms_lob.writeappend(l_clob,length(l_open),l_open);
          for i in 1..10 loop
            dbms_lob.writeAppend( l_clob, 32000, rpad('*',32000,'*') );
          end loop;
       dbms_lob.writeappend(l_clob,length(l_CLOSE),l_CLOSE);
  
      foo(l_clob,l_output);
      dbms_output.put_line('Output='||l_output);
  END;
  /
  
  

  Or simply l_clob: = l_open | RPAD('*',32000,'*') | l_close

  What Miss me? I want to just prove to me that I can successfully in an XML document that is greater than 32K and XMLTYPE methods allow to process the data in a relational format.

  Thank you

  Quick test...

  SQL > create or replace procedure process_xml (xml in xmltype) is
  2 cursor cur_xml is
  3. Select x.*
  xmltable 4 ('/ / employee ')
  5 passage xml
  RN 6 columns for ordinalite
  7 road of varchar2 (10) ename '. / ename'
  8 road of varchar2 (10) work '. / job'
  9 road number sal '. / sal'
  10                     ) x;
  11 start
  12 x in cur_xml
  loop 13
  14 dbms_output.put_line ('[' | x.RN |']) '|| x.Ename |' -' || x.Job |' -' || x.SAL);
  15 end loop;
  16 end;
  17.

  Created procedure.

  SQL > declare
  2 xml xmltype.
  3. start
  4. Select xmlelement ("AllDepartments",
  () 5 xmlagg
  6 xmlelement ("Department", xmlattributes (x.l as copy, d.deptno as deptno, dname as d.dname),)
  () 7 xmlagg
  8 xmlelement ("employee", xmlattributes (e.empno as empno, e.mgr as Bishop),)
  9 xmlelement ("ename", ename)
  10 xmlelement ("employment", work),
  11 xmlelement ("sal", sal)
  12                             )
  13                       )
  14                         )
  15                   )
  16                     )
  17 in xml
  18 Department d e outer join emp left (e.deptno = d.deptno)
  19 cross join (select l dual connect by level level<= 1024)="">
  Group of 20 by x.l, d.deptno, d.dname;
  21 dbms_output.put_line ('XMLSize: ' |) Length (XML.getClobVal ()));
  22 xml is so 1 715 925 about 1.63 Mb in size
  23 process_xml (xml);
  24 end;
  25.
  XMLSize: 1715925
  [1] CLARK - MANAGER - 2450
  [2] MILLER - CLERK - 1300
  [3] THE KING - PRESIDENT - 5000
  [4] SMITH - CLERK - 800
  [5] FORD - ANALYST - 3000
  [6] ADAMS - CLERK - 1100
  [7] SCOTT - ANALYST - 3000
  [8] JONES - MANAGER - 2975
  [9] ALLEN - SELLER - 1600

  
  [10] JAMES - CLERK - 950
  [11] TURNER - SELLER - 1500
  [12] BLAKE - MANAGER - 2850
  [13] MARTIN - SELLER - 1250
  [14] WARD - SELLER - 1250
  [15]  -  -
  [16] CLARK - MANAGER - 2450
  [17] MILLER - CLERK - 1300
  [18] THE KING - PRESIDENT - 5000
  [19] SMITH - CLERK - 800
  [20] FORD - ANALYST - 3000
  [21] ADAMS - CLERK - 1100
  [22] SCOTT - ANALYST - 3000
  [23] JONES - MANAGER - 2975
  [24] ALLEN - SELLER - 1600
  [25] JAMES - CLERK - 950
  [26] TURNER - SELLER - 1500
  [27] BLAKE - MANAGER - 2850
  [28] MARTIN - SELLER - 1250
  [29] WARD - SELLER - 1250
  [30]  -  -
  [31] CLARK - MANAGER - 2450
  [32] MILLER - CLERK - 1300
  [33] KING - PRESIDENT - 5000
  [34] SMITH - CLERK - 800
  [35] FORD - ANALYST - 3000
  [36] ADAMS - CLERK - 1100
  [37] SCOTT - ANALYST - 3000
  [38] JONES - MANAGER - 2975

  [15356] TURNER - SELLER - 1500
  [15357] BLAKE - MANAGER - 2850
  MARTIN [15358] - SELLER - 1250
  WARD [15359] - SELLER - 1250
  [15360]-

  PL/SQL procedure successfully completed.

  Don't know why I have some empty employees in this test, but not to worry, it was just for show...

  So, there we have a XML being passed in a procedure that is being processed, and that XML is more than 32 K (in my case it was > 1.6 Mb but could easily be larger).

  If you are passing in a CLOB, there is no real difference as you just convert to XMLTYPE as you did already.

• Problem with the CLOB data type.

  Greetings,
  
  I am facing a problem with the CLOB data type. I know that the CLOB data type is 4 GB (I use Oracle 9i). But in the Pl Sql procedure, I can only store 34305 size for a CLOB variable character data.
  
  This is the test script that I am trying to run.
  
  DECLARE
  -Local variables here*.
  I have INTEGER;
  C_1 CLOB.
  BEGIN
  FOR Rec IN (SELECT
  *
  ACCORDING TO THE TABLE)
  LOOP
  C_1: = c_1 | Rec.Clo_1;
  END LOOP;
  EXCEPTION
  WHILE OTHERS THEN
  Dbms_Output.put_line (SQLERRM);
  END;
  
  
  Here variable C_1 range value 34305 good character regardless of type CLOB. Now the above script fails if query my buckle - huge number of return values. It is throwing the exception "error during transfer of files ORA-06502: PL/SQL: digital error or value."
  
  He would be grateful if someone can help me on this.
  
  Thank you.

  You are probably better off using DBMS_LOB.append, instead of the concatenation of varchar2 (|).

  And... take off your when-other Manager exceptions, please...

• XQuery, Clob and Ora: view used... Please suggest

  Our approach is as:
  
  -We do all the /business processing logic in Xquery.
  In Oracle Store us our code of entire xquery as a clob.
  
  
  The code looks like:
  
  Create function nomfonction (ExecDate as)
  Return XMLTYPE AS
  
  v_result XMLTYPE.
  
  v_xquerycode1 CLOB: = to_clob ('declare... all of the code in xquery');
  v_xquerycode2 CLOB: = to_clob ("remaining code v_xquerycode1...'");
  v_xquerycode3 CLOB: = to_clob ("let $p1: = ora: view("T1")/ROW / / Root: MessageStart")
  "let $p2: = ora: view("T2")/ROW / / Root: MessageStart"
  Func(P1,_P2,_asOfDate) ');
  -Then we club the CLOB as:
  DBMS_LOB. APPEND (v_xquerycode1, v_xquerycode2);
  DBMS_LOB. APPEND (v_xquerycode1, v_xquerycode3);
  
  -Call the xquerycode as
  Select "COLUMN_VALUE.
  IN v_result
  from XMLTable (v_xquerycode1
  By PASSING ExecDate AS "reportDate".
  ) Name;
  
  Return v_result;
  
  End nomfonction;
  
  -In the comic book, we have retained the xml files to be transformed into tables T1 and T2.
  Structure of T1:
  Number XMLID
  XMLSource varchar2 (20)
  XMLContent XMLType
  .............
  
  Structure of T1:
  Number XMLID
  XMLSource varchar2 (20)
  XMLContent XMLType
  .............
  
  XmlContent column stores XML files.
  
  Then when I query as:
  Select XMLType.getCLOBVal (FunctionName (to_date (Sysdate, 'YYYYMMDD'))) of the double
  
  
  It takes hours without running
  If I reduce the trades to 100 rather than 1000 s, then it is executed in a minute.
  
  Please suggest a better way here.

  -We do all the /business processing logic in Xquery.
  In Oracle Store us our code of entire xquery as a clob.

  What is the version of database?
  I guess probably 11.x since you can use a dynamic XQuery query string...

  Which is certainly not in my list of top 10 of the best drawings about XML in the comic book.

  Dynamic XQuery: security issues, no optimization...

  Please suggest a better way here.

  Why must you dynamic XQuery?
  What is the underlying storage of XMLType columns?

  Could you give a test case? Examples of data + expected result?

• CLob update fails with PLS-00172: string literal too long

  I have a clob column that needs updating with a message about message_id
  
  Description of the table
  
  SQL > desc cl_message1
  Name Null? Type
  ----------------------------------------- -------- ----------------------------
  MSG_ID(2) NOT NULL NUMBER (10)
  DATA NOT NULL CLOB
  
  UPDATE statement:
  CLOB variable data1;
  Start
  : data1: ='TESTHDR ^ add ^ 200812310374 ^ 15:53:22.0 2012-08-19 ^.
  TESTHDR ^ add ^ 200812310374 ^ 5652 ^ C0000029 ^ 15:46:49.0 2012-08-19 ^.
  TESTDWN ^ ADD ^ 200812310374 ^ 5652 ^ G ^ 03 ^ 00000999990000722444 ^.
  .....
  ... ----------------------------------------------------------------------------------------- 2376 lines of text-------------------------
  TESTDWN ^ ADD ^ 200812310374 ^ 5674 ^ G ^ 10 ^ 00000999990000743289 ^.
  TESTEND ^ 5674 ^';
  UPDATE cl_message1
  SET DATA =: data1
  WHERE
  msg_id(2) = '989';
  END;
  /
  
  When I run this statement it fails with
  
  
  SQL > @test_clob_insert.sql
  : data1: ='WAVEHDR ^ add ^ 200812310374 ^ 15:53:22.0 2012-08-19 ^.
  *
  ERROR on line 2:
  ORA-06550: line 2, column 10:
  PLS-00172: string literal too long
  
  According to my understanding of a bind variable can store a maximum of clob. Not sure why it is throwing this error. Help, please
  
  Concerning
  
  This column is stored in_row.
  
  
  SQL > r
  1 * select table_name, column_name, owner, in_row from dba_lobs where table_name = 'CL_MESSAGE1 '.
  
  OWNER, TABLE_NAME COLUMN_NAME IN2
  ------------------------------ ------------------------------ ---------------------------------------- ---
  VEERA CL_MESSAGE1 DATA YES
  
  Published by: 953621 on August 19, 2012 22:34
  
  Published by: 953621 on August 19, 2012 23:44

  953621 wrote:
  According to my understanding of a bind variable can store a maximum of clob. Not sure why it is throwing this error. Help, please

  Yes, if the binding variable is declared as a clob, so it can store the maximum size of the CLOB, but this isn't your problem.
  Your problem is that you set the clob data using a string literal that is a single quoted string. There is a limit to the size of the string literals, so you will have to accumulate the clob in adding a whole lot string literals instead.

• How to read XML data stored as a Clob

  Hello
  I'm new on clob & XML types...
  Data XML has been inserted into a Clob field in a Table in Oracle (9.2.0.8.0)
  -----------------------------------------------------------
  CREATE THE TEMP TABLE
  (
  SNO NUMBER (5),
  STR_VAL LONG,
  CREATED_DT DATE default sysdate,
  COL2 CLOB,
  COL3 SYS. XMLTYPE
  )
  -----------------------------------------------------------
  SELECT dbms_lob.getlength (col2) of temp
  ---
  -> 24754
  -----------------------------------------------------------
  SQL > select temp col2;
  
  COL2
  ----------------------------------------------
  & lt; DataSet1 & gt;
  & lt; TAGSDATATABLE & gt;
  & lt; TAG_NAME & gt; KST20001 & lt; / TAG_NA
  --------------------------------------------------------------------------
  If I use the stmt above it shows only a pice of data
  
  
  How can I get the data in this column.
  
  could someone help to obtain the data.
  
  Concerning
  Prakash
  
  Published by: user12957183 on August 25, 2010 12:25 AM

  Insert data in to table XMLTYPE clob variable:

  SQL> ed
  Wrote file afiedt.buf
  
    1  declare
    2  my_clob CLOB := 'KST20001
    3  KST20002
    4  KST20003
    5  
    6  KST20004
    7  
    8  
    9  KST20005
   10  
   11  
   12  KST20006
   13  
   14  
   15  KST20007
   16  
   17  
   18  KST20008
   19  
   20  
   21  KST20009
   22  
   23  
   24  KST20010
   25  
   26  
   27  KST20009
   28  
   29  ';
   31  BEGIN
   33  INSERT INTO my_tab1 VALUES(XMLTYPE(my_clob));
   34* end;
  SQL> /
  
  PL/SQL procedure successfully completed.
  
  SQL> desc my_tab1;
   Name
   ---------------------------------------------------------------------------------------------------
   TABLE of XMLTYPE
  
  SQL> 
  
  -- For larger data:
  
  SQL> ed
  Wrote file afiedt.buf
  
    1  declare
    2  my_clob CLOB := '
    3  
    4  KST20001
    5  
    6  
    7  KST20002
    8  
    9  
   10  KST20003
   11  
   12  
   13  KST20004
   14  
   15  
   16  KST20005
   17  
   18  
   19  KST20006
   20  
   21  
   22  KST20007
   23  
   24  
   25  KST20008
   26  
   27  
   28  KST20009
   29  
   30  
   31  KST20010
   32  
   33  
   34  KST20009
   35  
   36  
   37  KST20009
   38  
   39  
   40  KST20009
   41  
   42  
   43  KST20009
   44  
   45  
   46  KST20010
   47  
   48  
   49  KST20009
   50  
   51  
   52  KST20009
   53  
   54  
   55  KST20009
   56  
   57  
   58  KST20009
   59  
   60  
   61  KST20010
   62  
   63  
   64  KST20009
   65  
   66  
   67  KST20009
   68  
   69  
   70  KST20009
   71  
   72  
   73  KST20009
   74  
   75  
   76  KST20010
   77  
   78  
   79  KST20009
   80  
   81  
   82  KST20009
   83  
   84  
   85  KST20009
   86  
   87  
   88  KST20009
   89  
   90  
   91  KST20010
   92  
   93  
   94  KST20009
   95  
   96  
   97  KST20009
   98  
   99  
  100  KST20009
  101  
  102  
  103  KST20009
  104  
  105  
  106  KST20010
  107  
  108  
  109  KST20009
  110  
  111  
  112  KST20009
  113  
  114  
  115  KST20009
  116  
  117  
  118  KST20009
  119  
  120  
  121  KST20010
  122  
  123  
  124  KST20009
  125  
  126  
  127  KST20009
  128  
  129  
  130  KST20009
  131  
  132  
  133  KST20009
  134  
  135  
  136  KST20010
  137  
  138  
  139  KST20009
  140  
  141  
  142  KST20009
  143  
  144  
  145  KST20009
  146  
  147  ';
  148  l_xmltype xmltype;
  149  BEGIN
  150  --l_xmltype := my_clob;
  151  INSERT INTO my_tab1 VALUES(XMLTYPE(my_clob));
  152* end;
  SQL> /
  
  PL/SQL procedure successfully completed.
  
  SQL> 
  

  Published by: AP on August 25, 2010 04:46

• "EXECUTE IMMEDIATE" of CLOB

  I have plsql-variable with CLOB datatype. It contains the source code of a package body. The size of the variable does not fit into 'varchar2' data type, it is larger.
  Now, I want to run a "EXECUTE IMMEDIATE" statement to create my package body defined in my CLOB variable.
  How can I use the "EXECUTE IMMEDIATE"?
  Note that "EXECUTE IMMEDIATE" does not allow for CLOB-variable as an argument.
  Note that "EXECUTE IMMEDIATE" allows "varchar2" - variables as arguments but "varchar2" contains maximum only some 32 KB of data.
  
  
  --
  
  Maybe it's possible to run something like
  "EXECUTE IMMEDIATE v1 | v2 | v3»
  where v1, v2, v3 are of type varchar2 and altogether they contain ca 100K of data and execute immediately would succeed?
  I wil outside, a s test...

  I wil outside, a s test...

  You may have searched for as well ;)

  Re: immediate execution of a CLOB field.

  Note: In 11 GR 2 run immediately accepts a clob as well...

• How to make a base64_decode in UTF - 8?

  Hello

  my version of the database is 11.2.0.2.0 and set to WE8ISO8859P15 NLS_CHARACTERSET and NLS_NCHAR_CHARACTERSET is set to AL16UTF16.

  I get a UTF-8 stream that is base64 encoded via a CLOB variable in PL/SQL. When I now decode (see my decodeClobBase642Clob function) this flow in CLOB, it works fine except the special characters because I think that the characters in the database is set to WE8ISO8859P15.

  How can I tell the utl_encode.base64_decode that his is decoded to UTF - 8?

  Here's the function with I use to decode the base64:

  -- -----------------------------------------------------------------------------
  -- Decode Base64 CLOB --> CLOB.
  -- -----------------------------------------------------------------------------
  PROCEDURE decodeClobBase642Clob(pBase64 IN CLOB, pClob OUT CLOB,
    pReturn OUT NUMBER, pMsg OUT VARCHAR2)
  IS
    mName        CONSTANT VARCHAR2(64) := 'decodeClobBase642Clob';
    vAmount      PLS_INTEGER := 192;
    vBase64Len   PLS_INTEGER;
    vBufferChar  VARCHAR2(200);
    vBufferRaw   RAW(200);
    vClob        CLOB;
    vClobLen     PLS_INTEGER;
    vOffset      PLS_INTEGER := 1;
  BEGIN
    pMsg := NULL;
    pReturn := 0;
    dbms_output.put_line('Start base64 decoding ...');
    vBase64Len := dbms_lob.getlength(lob_loc => pBase64);
    IF vBase64Len IS NULL THEN
      pMsg := 'Base64 Stream ist leer.';
      pReturn := 10;
    END IF;
    IF pReturn = 0 THEN
      dbms_output.put_line('<Base64Length>' || vBase64Len);
      dbms_lob.createtemporary(lob_loc => vClob, cache => TRUE);
      LOOP
        IF vOffset >= vBase64Len THEN
          EXIT;
        END IF;
        dbms_lob.read(lob_loc => pBase64, amount => vAmount, offset => vOffset,
          buffer => vBufferChar);
        vBufferRaw := utl_raw.cast_to_raw(vBufferChar);
        BEGIN
          dbms_lob.append(
            dest_lob => vClob,
            src_lob => utl_raw.cast_to_varchar2(utl_encode.base64_decode(vBufferRaw)));
        EXCEPTION
          WHEN OTHERS THEN
            dbms_output.put_line('<vAmount>' || vAmount || '<vOffset>' || vOffset ||
              '<vBufferChar>' ||  vBufferChar);
            dbms_output.put_line('Error in dbms_lob.append <SQLERRM>' || SQLERRM);
            RAISE;
        END;
        vOffset := vOffset + vAmount;
      END LOOP;
      vClobLen := dbms_lob.getlength(lob_loc => vClob);
      dbms_output.put_line('<ClobLength>' || vClobLen);
      dbms_output.put_line('... finshed base64 decoding.' || CHR(10));
      pClob := vClob;
    END IF;
  EXCEPTION
    WHEN OTHERS THEN
      RAISE;
  END decodeClobBase642Clob;
  

  Thanks in advance.

  Kind regards

  Martin

  How can I tell the utl_encode.base64_decode that his is decoded to UTF - 8?

  This isn't the concern of UTL_ENCODE, which processes only bytes, not characters.

  Assuming that your "special characters", including equivalent encoding in your database character (ISO-8859-15), then replace this:

  utl_raw.cast_to_varchar2 (utl_encode.base64_decode (vBufferRaw))

  with

  utl_i18n.raw_to_char (utl_encode.base64_decode (vBufferRaw), "AL32UTF8")

  It will convert the stream of bytes to characters, considering that the source bytes represent characters encoded in UTF - 8.

  If you want to manage the source characters that do not exist in ISO-8859-15, then you are out of luck with a CLOB release.

  You can consider using a NCLOB output.

• UTL_SMTP line wraps in csv. files after 990 characters.

  I have a pkg and it does some validation. My requirement is to capture this validation result and send it to the user by e-mail.

  For this I use UTL_SMTP.

  In the package, I use a CLOB variable to get all the data required by the separation with commas.

  Like this

  IF g_error_tab. COUNT > 0

  THEN

  BECAUSE me in 1... g_error_tab. COUNTY

  LOOP

  l_msg_clob: = l_msg_clob | TO_CHAR (g_error_tab (i));

  END LOOP;

  END IF;

  and then I call my UTL_SMTP package.

  SEND_MAIL. SEND_MAIL (p_to = > ' [email protected]', )

                       p_from          => ' [email protected] ',

  p_subject = > 'Message of Test. "

  p_text_msg = > "This is a test message."

  p_attach_name = > 'test.csv ';

  p_attach_mime = > ' text/plain; charset = us-ascii',.

  p_attach_clob = > l_msg_clob,

  p_smtp_host = > 'localhost');

  I am able to send the email with the data you want, BUT I see a few lines get shortened with exclamation point!. Like this

  1759

  110

  0

  0

  19926

  0

  0

  00!

  I tried different MIME types, but did not work.

  Appreciate your help.

  Hello

  The problem here is the way the linesize treats smtp server and not the utl_smtp package.

  According to RFC 822, the maximum total length of a line of text including the is 1000 characters.

  Some mail servers have their own limitation on the size of the line and insert a space or '!' as a newline after the maximum line length has been reached.

  The solution for this is to handle this by inserting the UTL_TCP newline character sequence. CRLF programmatically before reaching the length of the line max (authorized by the mail server).

  Note:

  It would normally be concatenate Chr (10) / as a newline separator in chains of plsql, but for new lines in smtp based email if some implementations may choose to use any line break to designate a new line, most of them need to have/ CHR (13) | Chr (10) and not only , it is advisable to use UTL_TCP. CRLF as a sequence of line break characters.

  Reference:

  Extra spaces and punctuation marks in the e-mail message when you send mail with utl_smtp (Doc ID 461659.1)

  Kind regards

  Ravi

• Get the ORA-01489 "result of concatenating string is too long."

  Hello world

  Im having problems with updating of a specific field. I got this error after I run my Update statement.

  * Cause: Result of concatenation of string exceeds the maximum size.

  * Action: Make sure that the result is less than the maximum size.

  Here is my script:

  declare
  cursor material is
      SELECT tt.column2,
             m.name, 
             m.MATERIAL_ID,
             tt.column3,
             tt.column4,
             tt.column5,
             tt.column6,
             tt.column7,
             tt.column8
      FROM material m, tmp_text tt
      WHERE m.NAME like tt.column2 || '%'
      AND tt.column9 = 'SI';
     
              begin
  
    for u in material loop
     
      update material
      set fields_xml =  '<?xml version="1.0" encoding="UTF-8"?>
                          <fields>
                            <field type="text" name="Size EU" maxlength="200"/>
                           <field type="text" name="Size UK" maxlength="200"/> 
                           <field type="text" name="Weight per size" maxlength="200"/>
                           <field type="text" name="Shaft height" maxlength="200"/>  
                           <field type="textarea" name="Product Description Norwegian" maxlength="2000">' || u.column8 || '</field>
                           <field type="textarea" name="Product Description Swedish" maxlength="2000">' || u.column7 || '</field>
                           <field type="textarea" name="Product Description Danish" maxlength="2000">' || u.column4 || '</field>
                           <field type="textarea" name="Product Description Finnish" maxlength="2000">' || u.column5 || '</field>
                           <field type="textarea" name="Product Description German" maxlength="2000">' || u.column6 || '</field>
                           <field type="textarea" name="Product Description English" maxlength="2000">' || u.column3 || '</field>
                           <field type="text" name="Additional Information" maxlength="200"/>
  
  
                           
                          </fields>'
      where material_id = u.material_id;
  
  
  
       end loop;
  
  end;
  
  
  
  
  
  
  
  
  
  
    
  

  That the script is that it will update the 'FIELD_XML' (CLOB) field in the table "material". I know that it exceeds the maximum size. Is there another way to do this?

  I would like to know your opinion... Thank you guys.

  Hello

  You can try with CLOB variable declaration and concatenate the value you want in it.

  For example:

  declare
  cursor material is
      SELECT tt.column2,
             m.name,
             m.MATERIAL_ID,
             tt.column3,
             tt.column4,
             tt.column5,
             tt.column6,
             tt.column7,
             tt.column8
      FROM material m, tmp_text tt
      WHERE m.NAME like tt.column2 || '%'
      AND tt.column9 = 'SI';
  
      v_clob CLOB;  
  
              begin  
  
    for u in material loop
      v_clob := '
                          
                            
                           
                           
                           
                           ';
      v_clob := v_clob || u.column8 || ' ';
      v_clob := v_clob || '' || u.column7 || ' ';
      v_clob := v_clob || ''  || u.column4 || ' ';
      v_clob := v_clob || '' || u.column5 || ' ';
      v_clob := v_clob || ''  || u.column6 || ' ';
      v_clob := v_clob || '' || u.column3 || ' ';
      v_clob := v_clob || ''
      update material
      set fields_xml =  v_clob
      where material_id = u.material_id;
       end loop;
  end;
  

  Cheerz,

  Stako

• Help required on DBMS_ADVISOR. TUNE_MVIEW

  Hello

  I'm trying to settle a materialized view.  He has the length and complicated SQL statement.

  When I spent all the materialized create sql procedure.  I'm SP2-0027: entry is too long (> 2499 characters) - ignored line.

  I did some research on the above question, I came across the asktom solution to create a table and place it

  So I modified my code as

  execute DBMS_ADVISOR.TUNE_MVIEW (:mv1,select field1||field2||field3||field4||field5||';' from work_tbl);
  

  Now, I get the following error

  ORA-06550: line 1, column 37:

  PLS-00103: encountered the symbol "SELECT" at the expected in the following way:

  (- + new case mod not null < an ID >)

  < between double quote delimited identifiers of > < a variable binding >

  continue the current avg County are min max sql stddev prior

  variance of the sum run forall fusion timestamp interval

  Date < a literal string with the specified set of characters >

  < a > < a SQL string between single quotes > hose

  < a literal character in string quoted also defined specifications >

  Can you please help me to solve this problem.

  Thanks in advance.

  Kind regards

  S. Subramanian

  DBMS_ADVISOR. TUNE_MVIEW second parameter is MV create statement. It can be VARCHAR2 or CLOB. In your case MV create statement exceeds 4000 bytes. Therefore, you have to move from the CLOB as a second parameter. To do that you need to split into pieces create statement MV, convert the first piece to CLOB and concatenate them into a CLOB. And use literals of Q, so you don't need to deal with quotes in the definition of MV. For example:

  DECLARE

  v_task_name VARCHAR2 (30): = ' tune this MV.

  BEGIN

  DBMS_ADVISOR. TUNE_MVIEW (v_task_name, TO_CLOB (Q '[CREATE MATERIALIZED VIEW EMP_VW]') |) ' Q'[QUICK UPDATE] | Q'["ON COMMIT"] | Q'[AS SELECT ENAME |] "works like" | (JOB - EMP]');

  END;

  /

  SY.

• How to use the referential cursor in this context

  Hello everyone I want to create a procedure and I am able to create it, but it does not work as I want it to be
  Please suggest the best way.

  CREATE OR REPLACE PROCEDURE UPDATE_COMMAND(V_TABLE_NAME VARCHAR2, V_COLUMN_NAME VARCHAR2)
  AS
  
  V_SQL CLOB;
  V_SQL_1 CLOB;
  TYPE V_SQL_2 IS REF CURSOR;
  V_SQL_22 V_SQL_2;
  V_SQL_23 VARCHAR2(2000);
  A clob;
  
  BEGIN
  
  V_SQL :='UPDATE '||V_TABLE_NAME ||'SET '||V_COLUMN_NAME ||' = ';
  
  V_SQL_23 := 'SELECT'|| V_COLUMN_NAME|| ' FROM ' ||V_TABLE_NAME;
  
  OPEN V_SQL_22 FOR V_SQL_23;
  
  loop
     FETCH V_SQL_22 INTO A;
   
    V_SQL_1 :=V_SQL | |V_COLUMN_NAME;
  
     dbms_output.put_line(v_sql_1);
     END LOOP;
     
     CLOSE V_SQL_22;
     
     END;

  first of all I don't know how to refer to this v_coluimn_name which I use just after the fetch command.
  
  When I run the procedure it gives error
  
  ORA-06550: line 1, column 22:
  PLS-00357: Table, view or reference sequence "QM_PRODUCT" not allowed in this context
  ORA-06550: line 1, column 7:
  PL/SQL: Statement ignored

  Prost wrote:
  I want the update command to be stored in a clob variable so that I can insert this variable in a column of a fictitious table, and then I'll get update of this table scripts manually.

  Well, what I gave just you out individual lines, so just adapt it to concatenate these lines in a CLOB variable for the loop process and then store you CLOB when finished. You should be able to handle that yourself.

  Still don't know why you want to do it right. Looks like a bad way to do something.

• What is the difference of the temporary lob in plsql

  Hi all
  
  What is the difference of a temporary lob for plsql, I mean for ex what is the difference for these codes:
  
  Declare
  MyClob Clob.
  X PLS_INTEGER;
  Start
  MyClob: = ' something clob operation ';
  x: = DBMS_Lob.InStr (MyClob, 'lob');
  end;
  /
  
  
  Declare
  MyClob Clob.
  X PLS_INTEGER;
  Start
  DBMS_Lob.CreateTemporary (MyClob, True);
  MyClob: = ' something clob operation ';
  x: = DBMS_Lob.InStr (MyClob, 'lob');
  DBMS_Lob.FreeTemporary (MyClob);
  end;
  /
  
  I know that temporary clob is stored in temporary tablespace, but in the first block of code above, what is the difference?
  Thank you

  What is the difference?

  In this case, there is no difference, because the temporary lob is created' in silence ': the Temporary LOB performance guidelines:

  »
  If you use the recently improved SQL semantics functionality in your applications, then there are many more temporary LOBs created silently in the SQL and PL/SQL than before.
  ...
  In particular, these temporary LOBs are created in silent mode when you use the following:

  On the LOBs SQL functions
  Functions of characters embedded PL/SQL on LOBs
  VARCHAR2/RAW for CLOB/BLOB variable assignments, respectively.
  Migrate from LONG-to-LOB
  «

  Things are different when the"silent" creation takes place, for example in.

  SQL> declare
    2     myclob   clob;
    3     myclob2 clob;
    4  begin
    5     myclob := 'something clob operation';
    6     dbms_lob.append(myclob2, myclob);
    7  end;
    8  /
  declare
  *
  ERROR at line 1:
  ORA-06502: PL/SQL: numeric or value error: invalid LOB locator specified:
  ORA-22275
  ORA-06512: at "SYS.DBMS_LOB", line 639
  ORA-06512: at line 6
  

  Then you can either use createtemporary procedure

  SQL> declare
    2     myclob   clob;
    3     myclob2 clob;
    4  begin
    5     dbms_lob.createtemporary(myclob2,true);
    6     myclob := 'something clob operation';
    7     dbms_lob.append(myclob2, myclob);
    8  end;
    9  /
  
  PL/SQL procedure successfully completed.
  

  Or to force another initialization:

  SQL> declare
    2     myclob   clob;
    3     myclob2 clob;
    4  begin
    5     myclob := 'something clob operation';
    6     myclob2 := trim(to_clob(' '));
    7     dbms_lob.append(myclob2, myclob);
    8  end;
    9  /
  
  PL/SQL procedure successfully completed.
  

• error of the oracle database to send alerts log

  Hi all
  
  actually I'm wondering about the status of database mail, and is the error produced in the last 24 hours in the critical mission of oracle 10.2.0.1 database every day... I set up SMTP and UTL_MAIL on the server but do not know how to send alert.log error and listener.log in the last 24 hours?
  
  I can handle the other requirements, but it will really be difficult to mail that a few lines and error codes occurred in the last 24 hours in the alert.log file.is there any table in the database that records all error messages written in the alert.log with description?
  
  I have my using RHEL5.2 ACE. so any help will be much appreciated...
  
  Thanks and greetings
  VD

  Hello

  How can I insert the latest 24 hours of UTL_FILE data? I have that in mind to create the file and send it and there is no parameter "attr_inline" I think it shows mesga ein mail itself...

  I do not have you here.

  so, how can I insert 24 hours of data in this file?

  For this you have to use some OS commands to read the data of alertlog last 24 hours being transferred to the tempfile.

  Try to play the file using UTL_FILE and type checking 'ORA' messages and if found none and then concatenates the string to clob variable and you can use it in utl_mail.

  I use this code to read the log file of the last until the (300) limit in size. It will return the contents of the logfile from lastline.

  UTL_FILE. FSEEK (F1, flength - 300);
  I'm in reverse 0... Flength-2
  loop
  Start
  -utl_file.fseek (F1, null,-2);
  UTL_FILE.get_line (F1, Err1);
  Exp_error: = Exp_error | LRC | Err1;
  exception
  When no_data_found then
  UTL_FILE. FCLOSE (F1);
  p_Exp_error: = Exp_error;
  "exit";
  -output when Err1 is null;
  end;
  end loop;

  Kind regards
  CGI