Executing multiple DQL-s or UTF-8 in IDQL

Hi Experts,

Is there any way to execute multiple dql queries? I know that we can create a script file and execute using IDQL but this stuff did not properly set UTF-8 characters. Today I had to execute a dql script for updating some data and it was urgent, but I could solve it only by writing a custom DFC program.

Please let me know if there is a tool for doing this.

Thanks in advance.

Find more posts tagged with

Documentum

BPM

Comments

Imran_Yusubov

Any ides folks ?

kcrkarthik

Can you explain this "did not properly set UTF-8 characters" with an example? What issue did you face?

-Karthik

Imran_Yusubov

Hello Karthik ,

We have some special utf-8 characters such as 'ə','ü','ş','ö' and etc. After executing the dql script file I had found out that the utf-8 characters displayed as a set of ugly characters for example

"YazÄ±ÅŸmanÄ±n tarixÃ§É™si" . Then I executed the same dql-s with a dfc program, and everything worked as expected. Does IDQL support utf-8 or not? If yes are there any steps that I have to enable it?

Thanks in advance

kcrkarthik

I tried the following queries and they are working fine in IDQL:

create dm_document object set object_name='Test';

update dm_document object set object_name='Teösüt' where r_object_id='090447bd8050055a';

select object_name from dm_document where r_object_id='090447bd8050055a';

<Name returned properly>

Can you share the query you used?

-karthik

Imran_Yusubov

I see, that means IDQL supports utf-8, so the problem is the host or something, would you let me know how you did that.

Now I trying on a different host and I am facing another problem.

When I save the dql file as utf-8 then I am getting this error.

1> @C:\update.dql1> 2> [DM_QUERY_E_SYNTAX]error:  "A Parser Error (syntax error) has occurred inthe vicinity of:  ∩╗┐update"1> 1>

When I save it as ANSI it runs but it is clear that the UTF-8 characters can't be saved in ansi charset.

Here is what I have in my dql file.

update dm_document object set object_name='ƏƏƏƏƏƏ' where r_object_id='0900000c8001616a';go

Please give me some suggestions.

thanks in advance

kcrkarthik

what text editor are you using to create this DQL script file? Try using Notepad.

Also, try opening your existing script in Notepad and see if the chars are displayed fine. If not, correct them and execute.

-karthik

Imran_Yusubov

Yes, this is notepad, and when I open the file there is no any problems with the characters.

Imran_Yusubov

No luck,

this is my dql file, can you please try to see if it can be run.?

kcrkarthik

do you have any other characters giving trouble? or just this one "Ə"?

i think this particaular character is one of a kind which is not supported. Not sure even if this will work as a SQL query (https://forums.oracle.com/forums/thread.jspa?threadID=719181).

Not sure about this. I am not even able to copy this into VI editor in unix environment.

From Wiki: When using ə in Unicode, the Azerbaijani language has problems with the ISO/IEC Turkish encoding, so sometimes ä has been used instead, as in the Tatar language and Turkmen language.

-karthik

Message was edited by: Karthik Chokkaraman

Imran_Yusubov

Hello,

Ok.Not only this character others also are not properly displayed. But this works fine when I update through DA or DFC..

kcrkarthik

that could possibly be that DA (application sever) or DFC (java editor) can recognize this character. Whereas idql is a command-line utility which does not seem to recognize it.

-karthik

kcrkarthik

Can you take a look at this: https://solutions.emc.com/emcsolutionview.asp?id=esg113498

Symptoms:

After running a dump with iapi method all arabic letters are changed to '????' letters. This is already done in the dump file.

The arabic letters are displayed and handled correctly in DA

Also: if logging to docbase via idql arabic document listed as (e.g. Â¿Â¿Â¿ Â¿Â¿Â¿Â¿Â¿.doc) - select * from ... - it returns me ????.doc.

Cause:

java in dfc, which is used by idql and iapi can't handle correctly arabic characters.

Resolution:

Update dfc.properties file with the following:

client_codepage= ISO_8859-1 --> client_codepage= UTF8

Note: All the 5.3.x uses the DMCL. From D6 it uses the JAVA DMCL (DFC).

Imran_Yusubov

Hello,

Thank you very much for the information.Possibly this is the case for the other host.

On this host I added the entry to dfc.properties restarted docbase also changed Content Server's os codpage.

But I am still getting this error while trying to run a utf-8 query file.

1> 1> @C:\update.dql

1> 2> [DM_QUERY_E_SYNTAX]error: "A Parser Error (syntax error) has occurred in

the vicinity of: ´╗┐update"

kcrkarthik

this error is because there are some special characters in the file prior to the word UPDATE. You might not see that when you open it in Notepad or other windows based text editors. Open it through command prompt or move the file to any unix env and open it in VI editor, you will see the special chars.

But still, the special chars in "object_name" field will give trouble. I am also checking if there is any alternative way to get this working; Will keep you posted.

It is possible that idql is not supported for UTF-8 (need some experts to answer this; you can check with emc also). The reason i am saying this is, i ran the update query through DA.

update dm_document object set object_name='ƏƏƏƏƏƏ' where r_object_id='0900000c8001616a';

When i run the select query in DA, the special chars appear fine. But if i run the same query thru idql, the special chars appear as question marks.

-karthik

Message was edited by: Karthik Chokkaraman