Problem with LONG column in 8.1.7

Hello All,
Our db version is 8.1.7.0.0. ( Unix OS ).
We have a table with a LONG column, which stores BINARY data(Billing Details). Is there any way to manipulate it, other than converting it to LOB column?
I dont know where to start.Any help or hint is appreciated..
Thanks..

I did the following test in 10g but not with 8i:
dev001> set serveroutput on
dev001>
dev001> declare
2 v_b blob;
3 v_c varchar2(10);
4 begin
5 v_b := utl_raw.cast_to_raw('OK');
6 v_c := utl_raw.cast_to_varchar2(v_b);
7 dbms_output.put_line(v_c);
8 end;
9 /
OK
PL/SQL procedure successfully completed.However it is not a good idea to use BLOB data type to store character data: you should use CLOB for character data and BLOB for binary data.

Similar Messages

Problem with long column names

I'm trying to fetch rows via pl/sql from an Access db table that has long(over 30 characters) column names. I've tried using pass-through sql but it still returns generic ODBC error. Is there a way to get this to work without changing column names in table?

Andy,
Oracle does have a restriction in column name size but if you are using pass-through this should not come into play.
It may be a restriction of the ODBC driver or a bug. Please take a look at the ODBC driver trace file.
Vira

Problem with checkbox column in matrix

Hello.
I have a little problem with checkbox column in matrix.
Column is binded to the UserData.
It has ValOn="Y", ValOff="N".
I use C++. It is wird problem. In matrix I have 10 columns - scrollbar role and if You want see checkbox column, You must role to the right. If this column is on the screen, and I use:
checkcell->PutChecked(VARIANT_TRUE);
then the checkbox is cheched, and if the checkbox isn`t on the screen and I use this comment - it nothing happening.
I tried to use ValOn="Y", PutChecked...
The problem i solved if the column is on the screen - if the column is first in matrix or second, but if it`s last I have a big problem.
My column with checkbox is not editable, but I tried to make it editable, check it, and then make it uneditable - the same efect.
How can I solve it ?
Sorry for my english.
Kamil Wydra

Hello Kamil,
I am not sure about your problem, but here is an example of how to use checkbox in UI API.
First, create the matrix with checkbox column in Screen painter, and the output is an xml file, like this. Type as 121 indicates that it is a check box.
- <column AffectsFormMode="0" backcolor="-1" description="" disp_desc="0" editable="0" right_just="0" title="Rented" type="121" uid="Rented" val_off="N" val_on="Y" visible="1" width="41">
<databind alias="U_RENTED" databound="1" table="@VIDS" />
<ExtendedObject />
Second, bind the column to table from DB. This is a bug of 2004 Screen Painter, so if you are using 2005 Screen Painter, there is no problem.
Third, when you open the form, you can check and uncheck the cell.
BTW, please set the editable of the column to true.
Hope this helps,
Nick

Error -43 when try playback on win7 64bits over the network QT that recide MacPro osx 10.5 however i can play with VLC player, This happend when the QT is inside a Folder with name longer 8 chars other files has no problem with long names just the QT

error -43 when try playback on win7 64bits over the network QT that recide MacPro osx 10.5 however i can play with VLC player, This happend when the QT is inside a Folder with name longer 8 chars other files has no problem with long names just the QT nfs sharing

Never mind, I already found the solution myself.
What I did wrong was:
- not copying the master image to the nbi folder
- selecting the netinstall-restore.dmg image as source to copy to my HD.
The thing is, when you create a netinstall image for 10.5, the image itself is already included in the netinstall image so you don't have to do anything else.
With the 10.4 image however, you also have to copy the master image to the NetBootSP0 directory.
In the *.nbi folder contains an netinstall-restore.dmg file. But that is only to boot you to netrestore, it's not the image itself.
Other alternative is to copy the images to another folder that you share with AFP and adjust the configuration of netrestore like described in this manual:
http://www.shellharbourd.det.nsw.edu.au/pdf/misc/osxrestoringnet.pdf
This manual was also how I figured out that I forgot to copy the image to the NetBootSP0 folder.

WebRowSet Problem with database columns defined as TEXT??

Hello,
Can somebody help me on this subject. (http://forum.java.sun.com/thread.jspa?forumID=31&threadID=778586)
I have the same problem with TEXT column when I try to populate the WebRowset.
Thanks,
Stephane

OK,
I change my postgresql driver for the lastest version (postgresql-8.2-506.jdbc3.jar) for JVM 1.5 and it's find...
This driver support the javax.sql.
St�phane
Edited by: Borealis on Oct 15, 2007 12:43 PM

Has anyone else had problems with long wakeup times on Cinema displays?

We're having problems with long wakeup times on our Cinema Display. Noticed it after the Maverick update.
Need help please. This thing is nothing more than a boat anchor to my developers... frustrating, expensive boat anchor.
Thoughts?
Thanks,
Jeff Stutzman

Hey Jeff,
Thanks for the question. I understand you are experiencing issues with an external display and your MacBook Air. The following resource may provide a solution:
Apple computers: Troubleshooting issues with video on internal or external displays
http://support.apple.com/kb/ht1573
While you should give all of the troubleshooting steps a whirl, definitely make sure to follow the steps in this section:
Reset the system
You can reset the Mac's parameter RAM and SMC.
Thanks,
Matt M.

Change Tablespace of a table with LONG column

I have a 9i database that I have just gotten control of. At this point there is just one big dictionary managed tablespace for everything created by users. I am trying to move to multiple locally managed tablespaces with fixed extent sizes but I have run into a problem.
I have one table with one LONG datatype column. Apparently there is a huge amount of work involved to change the code if I make it a BLOB so that is out.
At this point I would like to change the tablespace of this table but I can't move it the normal way because of the LONG column. I have found mention of being able to do this with "COPY" but I can't find any documentation on the "COPY" command in the 9i Docs.
Any help would be appreciated,
Chris S.

Chris-
Can't you create your new table ahead of time in your new tablespace?
You could then use a statement like:
COPY FROM old/your_password@olddb TO new/your_password@newdb -
REPLACE NEWTABLE -
USING SELECT * FROM OLDTABLE;

Selecting from a SQL Server 2005 with long column names ( 30 chars)

Hi,
I was able to set up a db link from Oracle 11.2.0.1 to SQL Server 2005 using DG4ODBC.
My problem is that some column names in the Sql Server are longer than 30 chars and trying to select them gives me the ORA-00972: identifier is too long error.
If I omit these columns the select succeeds.
I know I can create a view in the sql server and query it instead of the original table, but I was wondering if there's a way to overcome it with sql.
My select looks like this:
select "good_column_name" from sometable@sqlserver_dblink -- this works
select "good_column_name","very_long_column_name>30 chars" from sometable@sqlserver_dblink -- ORA-00972Thanks

I tried creating a view with shorter column names but selecting from the view still returns an error.
create view v_Boards as (select [9650_BoardId] as BoardId, [9651_BoardType] as BoardType, [9652_HardwareVendor] as
HardwareVendor, [9653_BoardVersion] as BoardVersion, [9654_BoardName] as BoardName, [9655_BoardDescription] as BoardDescription,
[9656_SlotNumber] as SlotNumber, [9670_SegmentId] as SegmentId, [MasterID] as MasterID, [9657_BoardHostName] as BoardHostName,
[9658_BoardManagementUsername] as BoardManagementUsername, [9659_BoardManagementPassword] as BoardManagementPassword,
[9660_BoardManagementVirtualAddress] as BoardManagementVirtualAddress, [9661_BoardManagementTelnetLoginPrompt] as
MANAGEMENTTELNETLOGINPROMPT, [9662_BoardManagementTelnetPasswordPrompt] as MANAGEMENTTELNETPASSPROMPT,
[9663_BoardManagementTelnetCommandPrompt] as MANAGEMENTTELNETCOMMANDPROMPT FROM Boards)performing a select * from this view in sqlserver works and show the short column names
this is the error i'm getting for performing a select * from v_boards@sqlserver_dblink
ORA-28500: connection from ORACLE to a non-Oracle system returned this message:
[Microsoft][SQL Native Client][SQL Server]Invalid column name 'BoardManagementTelnetLoginProm'. {42S22,NativeErr = 207}[Microsoft]
[SQL Native Client][SQL Server]Invalid column name 'BoardManagementTelnetPasswordP'. {42S22,NativeErr = 207}[Microsoft][SQL Native
Client][SQL Server]Invalid column name 'BoardManagementTelnetCommandPr'. {42S22,NativeErr = 207}[Microsoft][SQL Native Client][SQL
Server]Statement(s) could not be prepared. {42000,NativeErr = 8180}
ORA-02063: preceding 2 lines from sqlserver_dblinkI also tried replacing the * with specific column names but it fails on the columns that have a long name (it doesn't recognize the short names from the view)
what am I doing wrong?
Edited by: Pyrocks on Dec 22, 2010 3:58 PM

Problem with LONG FIELD

Hello,
In HTML DB, I would like to insert a line in a table with a database link (in a PL/SQL Block).
Description of my table :
Nom NULL ? Type
LCOMMENTID NOT NULL NUMBER(10)
DTLASTMODIF DATE
MEMCOMMENT LONG
TABLENAME VARCHAR2(18)
My command :
insert into amcomment@B_ASSET.WORLD values (1,'01/01/2005','toto','AMTICKET');
Each time I try to insert a value in the long field, I have this error :
"ORA-06550: Ligne 8, colonne 13 : PL/SQL: ORA-00600: code erreur interne, arguments : [qctstc2o1], [1], [0], [31], [8], [0], [0], [] ORA-06550: Ligne 8, colonne 1 : PL/SQL: SQL Statement ignored"
I try also with null value. Same problem ...
Have you a solution for this problem ?
Thanks.
FW

From the SQL Reference guide:
"LONG and LONG RAW columns cannot be used in distributed SQL statements"
It should also be noted that the LONG datatype has been deprecated and may not be supported in the future. It is suggested that LONG columns be replaced with CLOB columns. These are much easier to work with than LONGs and have fewer restrictions. You can easily use the TO_LOB function to convert a LONG to a CLOB (see http://download-west.oracle.com/docs/cd/B10501_01/server.920/a96540/functions139a.htm#79466)

Problem with long text

Hi folks,
iam uploading longtext in transaction KP06 in bdc programing using create_text.
now my problem is if i pass selection criteria
case 1:
Vertion : N0
fiscal year : 2007
planner profile: CCPLAN4
then the budeget plan data uploading successfully along with long text.
case2: Vertion : N1
in this case also everthing wrking fine
case3: Vertion : 001
in this case budgetplan data uploading sucessfully but longtext is not uploading,
i debbug the proragm , create_text also returning sy-subrc = 0. but if i check in KP06 the long text is not there.
data declaration
vertion : VERSN type C(3)
can u tell me the problem is in logic or with FM ?
note : i checked with save_text also.
regards
neeru

HI Naimesh,
the FM Create_text internally contains both FMs save_text and commit_text,
even i debug the FM in case of vertion N1, N0 and 001,
all the cases sy-subrc = 0.
code:
CLEAR G_OBJNR.
LOOP AT RECORD.
COST = RECORD-BDC01.
CALL FUNCTION 'CONVERSION_EXIT_ALPHA_INPUT'
    EXPORTING
      input = COST
    IMPORTING
      output = COST.
    CONCATENATE 'KS' p_kokrs record-bdc00 INTO g_objnr.
        SELECT SINGLE * FROM COSP WHERE OBJNR = g_objnr
                          AND GJAHR = P_GJAHR
                          AND VERSN = p_ver
                          AND KSTAR = COST.
IF SY-SUBRC = 0.
      LEDNR = COSP-LEDNR.
      OBJNR = COSP-OBJNR.
      GJAHR = COSP-GJAHR.
      WRTTP = COSP-WRTTP.
      VERSN = COSP-VERSN.
      KSTAR = COSP-KSTAR.
      HRKFT = COSP-HRKFT.
      VRGNG = COSP-VRGNG.
      VBUND = COSP-VBUND.
      PARGB = COSP-PARGB.
      BEKNZ = COSP-BEKNZ.
      TWAER = COSP-TWAER.
CONCATENATE LEDNR OBJNR INTO OBJNR.
CONCATENATE OBJNR GJAHR INTO TNAME1 separated BY SPACE10.
CONCATENATE TNAME1 WRTTP VERSN INTO TNAME2.
CONCATENATE TNAME2 KSTAR INTO TNAME3 separated BY SPACE.
CONCATENATE TNAME3 VRGNG INTO TNAME4 separated BY SPACE4.
CONCATENATE TNAME4 BEKNZ INTO TNAME5 separated BY SPACE10.
CONCATENATE TNAME5 TWAER INTO TNAME.
I_THEAD-TDOBJECT = 'CCSS'.
I_THEAD-TDID = 'COSP'.
I_THEAD-TDNAME = TNAME.
I_THEAD-TDSPRAS = 'E'.
APPEND I_THEAD.
i_tline-tdformat = '*'.
I_TLINE-TDLINE = RECORD-LONGTEXT.
APPEND I_TLINE.
loop at i_thead.
CALL FUNCTION 'CREATE_TEXT'
EXPORTING
    FID               = I_THEAD-TDID
    FLANGUAGE         = I_THEAD-TDSPRAS
    FNAME             = I_THEAD-TDNAME
    FOBJECT           = I_THEAD-TDOBJECT
SAVE_DIRECT       = 'X'
FFORMAT           = '*'
TABLES
    FLINES            = i_tline
EXCEPTIONS
NO_INIT           = 1
NO_SAVE           = 2
OTHERS            = 3
IF SY-SUBRC <> 0.
MESSAGE ID SY-MSGID TYPE SY-MSGTY NUMBER SY-MSGNO
         WITH SY-MSGV1 SY-MSGV2 SY-MSGV3 SY-MSGV4.
ENDIF.
ENDLOOP.
CLEAR I_TLINE.
REFRESH I_TLINE.
CLEAR I_THEAD.
REFRESH I_THEAD.
SELECT SINGLE * from cokp into itab WHERE OBJNR = g_objnr
                          AND GJAHR = P_GJAHR
                          AND VERSN = p_ver
                          AND KSTAR = COST.
if sy-subrc = 0.
   itab-TXFLG = 'X'.
   append itab.
   write:/ 'Text saved for', record-bdc00 , record-bdc01.
   else.
      write:/ 'Text saved for', record-bdc00 , record-bdc01.
      ENDIF.
ENDIF.
my program wrking fine but the long text is not visible for vertion 001.
regards
niru

Problem with Stacked Columns in Business Graphics

Hi,
Has anyone used the stacked columns chart type in business graphics? Am facing a problem with this.
I am using the SimpleGraphics example demo with 2 dataseries, the values ranging from 0-10. It works fine if the chart type is Columns (this is default). When I change the chart type to stacked_columns, the graph goes for a toss. The y-axis range automatically changes to 0-1 (values as 0, 0.2, 0.4 ... 1.0) and it does not display stacked columns as expected.
Is there any specific property(s) that need to be set for this chart type?
Appreciate any quick response.
cheers,
Vittal

Hi friend,
See the link below it is having the solution of hiding the columns in smart forms
Hide table columns in smart form?
Create a table to display your values with 12 col and hide the columns based on the idea provided in the link above.
I think this will solve your issue if you still have queries please revert back to me i will help you.
Thanks,
Sri Hari

Problems with Long Interactive PDFs with Extensive Interactivity via Buttons and Hyperlinks

I have inherited a project which requires the creation of large interactive pdfs with lots of buttons and hyperlinks. I am having continuing problems with buttons not working and long periods of wasted time as attempted exports fail.
Background:
The largest books range from 264 MB to 411 MB – (764 to 1024 pages). Per book, the maximum number of buttons on each page is 16, linking separate files (subsections) in the book via text anchors. The buttons are created using Master Pages which are set for each file (subsection) in the book. 50% to 60% of the pages in each file (subsection) will also be referenced as 2 hyperlinks. One is the source page hyperlink on each workflow and the second is a listing on the source page as a possible navigation path. I have created hyperlink destinations for each page which is used to generate the source and navigation hyperlinks. With 2 hyperlinks representing 50%-60% of the pages - that means the largest book contains a maximum of 1200 hyperlinks. Each file (subsection) also has it’s own separate TOC and the bookmarks that are created (optionally) during the TOC creation process.   As currenlty formatted, the book is primarily images of workflows which mean 50%-60% of the pages contain pdf images.
Several questions:
1. How do I get my buttons to work consistently? There is nothing worse than working hard and then ending up with faulty product. I have been thinking about combining all the files into 1 document per book. Then I could change the buttons to objects, add a hyperlink over each one and set them for a specific hyperlink destination. Can I do that if the buttons are on Master Pages? Can InDesign handle a document with 1024 pages?    I can't fix the problems in Acrobat later because it would mean 1024 fixes for just 1 button correction.
2. How do I stop InDesign from freezing when exporting these large pdfs? It seems to wait until the final steps to crash which sometimes means a 4-6 hour waste of time per export.
3. Are we making a mistake using pdfs as our workflow images? I'm not pleased with the quality of the images we have and wonder if we switched to another image format if this would improve the results.   Would it be worth the effort and relinking? I have been reading different information on what the preferred source image format is for InDesign. One article I found yesterday says using pdf images is a problem. Another site said the Photoshop, TIF and JPG are preferred but PDFs were also acceptable.
4. Do I need to propose scrapping InDesign as the source of the project? If so, what application can better meet my needs?

Those are the limitations you have to put up with.
Apple doesn't follow the PDF spec. It chooses which PDF features it will support.
As to the version number (v5 or v6 or the new multimedia features in v9), there is different support in different versions of Reader/Acrobat as the product has developed. You just need to post a message to downloaders that they need to view the file in Adobe Reader version X or above to get the best experience.
Other industry standards face the same problem: How can you add new feature like multimedia? You can't always be backwards compatible.

Problem with dynamic columns in smartforms.

Dear SDN Experts,
I have a requirement in smartforms for dynamic columns.
i have used template with 10 columns, So from these 10 columns,Columns may vary monthly MIN 2 to MAX 10 depending on
readings with them for that month.
i cannot fix column headings also,Because headings also changes dynamically.
So Problem is if there is no data in columns,Columns is displayng empty.
For EX: In this month i have 2 columns data remaining all columns is displaying empty boxes.
Please suggest me a solution is this posible in smartforms if i use table also.
<removed by moderator>
Regrds,
MNR
Edited by: Thomas Zloch on Sep 11, 2011 3:50 PM

Hi friend,
See the link below it is having the solution of hiding the columns in smart forms
Hide table columns in smart form?
Create a table to display your values with 12 col and hide the columns based on the idea provided in the link above.
I think this will solve your issue if you still have queries please revert back to me i will help you.
Thanks,
Sri Hari

Mail problem with long titles using utl_smtp / mixed content / attachements

Hello,
I'm having a problem with sending emails. The title as well as the body needs to be able to contain accented letters (é ç ...).
For the body i got it working.
And the title works fine too, untill the encoding passes a certain length, then goes all wrong and my mail header starts showing up in the mail body.
Could anyone help me in finding a solution?
Thanks in advance
Below my code:
DECLARE
   v_From       VARCHAR2(80) := 'test@localhost';
   v_Recipient VARCHAR2(80) := 'test@localhost';
   v_Cc         VARCHAR2(80);
-- long subject causing problems
   v_Subject    VARCHAR2(32000) := '123456789111111111111111111111118888888888888111111111111111991234567891111111111111111111111188888888888881111111111111119912345678911111111111111111111111888888888888811111111111111199';
   v_body       VARCHAR2(32000) := '<a href="#">tester</a><br><br>en nog wat éékes en ççkes';
   v_Mail_Host VARCHAR2(30) := 'localhost';
   v_Mail_Conn utl_smtp.Connection;
   crlf         VARCHAR2(2) := chr(13)||chr(10);
BEGIN
v_Mail_Conn := utl_smtp.Open_Connection(v_Mail_Host, 25);
utl_smtp.Helo(v_Mail_Conn, v_Mail_Host);
utl_smtp.Mail(v_Mail_Conn, v_From);
utl_smtp.Rcpt(v_Mail_Conn, v_Recipient);
utl_smtp.Data(v_Mail_Conn,
-- write header
    'Date: '   || to_char(sysdate, 'Dy, DD Mon YYYY hh24:mi:ss') || crlf ||
    'From: '   || v_From || crlf ||
    'Subject: ' || utl_encode.MIMEHEADER_ENCODE(v_Subject) || crlf ||
    'To: '     || v_Recipient || crlf ||
    'Cc: '     || v_Cc || crlf ||
    'X-Mailer: Mailer by Oracle UTL_SMTP' || crlf ||
    'X-Priority: 1' || crlf ||
    'MIME-Version: 1.0'|| crlf ||       -- Use MIME mail standard
    'Content-Type: multipart/mixed;'|| crlf ||
    ' boundary="-----SECBOUND"'|| crlf ||
    crlf ||
-- add message body
    '-------SECBOUND'|| crlf ||
    'Content-type: text/html; charset=UTF-8'|| crlf ||
    'Content-Transfer-Encoding: base64'|| crlf ||
    crlf ||
    utl_raw.cast_to_varchar2(utl_encode.base64_encode(utl_raw.cast_to_raw(v_body)))||
    crlf ||
-- add csv attachment
    '-------SECBOUND'|| crlf ||
    'Content-Type: text/plain;'|| crlf ||
    ' name="excel.csv"'|| crlf ||
    'Content-Transfer_Encoding: 8bit'|| crlf ||
    'Content-Disposition: attachment;'|| crlf ||
    ' filename="excel.csv"'|| crlf ||
    crlf ||
    'CSV,file,attachement'|| crlf ||   -- Content of attachment
    crlf ||
    '-------SECBOUND--'                            -- End MIME mail
utl_smtp.Quit(v_mail_conn);
EXCEPTION
WHEN utl_smtp.Transient_Error OR utl_smtp.Permanent_Error then
    raise_application_error(-20000, 'Unable to send mail: '||sqlerrm);
END;
/

Hi,
I did not think the characterset and database would matter for this problem, but here is the info you requested.
The database is an Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - 64bi
The characterset : NLS_CHARACTERSET     AL32UTF8
And i don't have a clue about the characterset of the operating system (not even sure what the operating system is), but i think that will be utf-8 aswell.
I'm not sure what i did, to come over as rude, but i didn't mean to and i apologize for it.

Problem with shared column mapping

I have a problem with mapping different fields onto the same column. In most cases Kodo handles it very well but one tough case it does not
Modification references contract N-1 using contract_id field
Contract back references two modifications out of all modification owned by THIS contract (last_modification and executed_modifcation)
when both modifications are not null it works fine since I make sure both fields point to modification owned by the same contract
hovewer when I set one of the fields (executed_modifcation) to null (after it was not null) Kodo attempts to set not only executed_modifcation field to null but contract_id as well. Which obviously not desired
I understand that it is a pretty hard corner case. Do you think you can handle it or it is a dead end?
Attached are my files

Abe,
I fully agree with your analysis. In my case we can use precedence of direct
mappings over reference mapping in null situation.
However as you noted, there are cases when we have only references and so
you can not say which mapping is more important. However since we are
resolving NULL issue here, I am not sure we need to know precedence of
references - not null case should always take precedence. In other words:
priority 1: direct mapping
priority 2: not null reference
priority 3: null reference
Of course all three cases are subject rule of not changing value of shared
column
"Abe White" <[email protected]> wrote in message
news:[email protected]...
I admit that I didn't go through all your metadata, but I do believe that
3.0 final will handle this case. My plan is to implement updates such
that direct field mappings always take precedence over foreign key
mappings.
Let's say column CONTRACT_ID is used by int field contractId and is
also used by field lastModification as part of an FK to a Modification
instance (which I believe is close to your mapping). Because contractId
is a simple field rather than a foreign key, Kodo will give it the final
say on nulling/defaulting the CONTRACT_ID column. So if you null the
lasetModification field, CONTRACT_ID will retain the value of the
contractId field. Any non-null value of the lastModification field will
stilly have to jive with the contractId field, though, or else we'll
throw an exception like we do currently when you try to set 1 column to
multiple values.
I believe this solves the common case of shared
columns between direct mappings and FK mappings. There are other cases
in which multiple FK mappings overlap on certain columns, and this does
not solve nulling problems in those instances. However, I think those
cases are very rare and not worth the effort. Correct me if I'm wrong
and your situation actually is one of sharing columns between multiple
FKs and no primitives after all.

Problem with LONG column in 8.1.7

Similar Messages

Maybe you are looking for