Tom Kyte -v- Oracle Documentation - Character Semantics for AL32UTF8 ?

On a unicode database, what is best practice for varchar columns - character semantics or byte semantics ?
Tom Kyte's new book Expert Oracle Database Architecture says "When using a multibyte character set such as UTF8, you would be well advised to use the CHAR modifier in the VARCHAR2/CHAR definition - that is, use VARCHAR2(80 CHAR), not VARCHAR2(80), since your intention is likely to define a column that can in fact store 80 characters of data." (p499). This I would agree with.
Yet, when I read the 10gR2 documentation "Globalization Support Guide", it says "The BYTE and CHAR qualifiers shown in the VARCHAR2 definitions should be avoided when possible because they lead to mixed-semantics databases. Instead, set NLS_LENGTH_SEMANTICS in the initialization parameter file and define column datatypes to use the default semantics based on the value of NLS_LENGTH_SEMANTICS. Byte semantics is the default for the database character set."
I'd much rather explicitly state that I'm using character semantics, than relying on some default setting. So who's right ? Tom Kyte, or the Oracle documentation ? Or are they saying different things ?
If I use character semantics, is this going to come back and bite me later ?
I'm using Oracle 10gR2, and the database character set is the default of AL32UTF8, using only VARCHAR columns - not NVARCHAR.
Thanks,
Andy Mackie

I would say that both want to say the same thing in a different way.
Whether you always specify CHAR modifier or define NLS_LENGTH_SEMANTICS to CHAR is the same ... until you decide to change NLS_LENTGH_SEMANTICS to BYTE but this change will only apply to new columns created after NLS_LENGTH_SEMANTICS parameter change and I wonder why you would to do it.
You have to use character semantics and make a choice: if you don't need to change it, it should not bite you back.

Similar Messages

While generating reports in Oracle BI Publisher in pdf format, the generated pdf reports have hindi इ matra displaced by one character. For example, रिपोर्ट is printed as रपेिोरट. Word file generated of the same report have correct hindi इ matra positi

While generating reports in Oracle BI Publisher in pdf format, the generated pdf reports have hindi इ matra displaced by one character. For example, रिपोर्ट is printed as रपेिोरट. Word file generated of the same report have correct hindi इ matra position and also pdf generated from this word file also contains the same.

Documentation and Installer for "Oracle VM Management Pack" for EMGC?

Hi,
Can you please provide locations/links to download the Documentation and the Installer for "*Oracle VM Management Pack"* for EMGC? I need to install the virtualization pack for provisioning OVM templates on Virtual machines.
Thanks in advance,
Ranganath

>
Can you please provide locations/links to download the Documentation and the Installer for "*Oracle VM Management Pack"* for EMGC? I need to install the virtualization pack for provisioning OVM templates on Virtual machines.Hi Ranganath,
This feature is present in EM Grid Control 10.2.0.5 and you need to enable it..
Please refer to
How to Enable Oracle VM Management Pack in EM Grid Control 10.2.0.5 [ID 781879.1]
http://www.oracle.com/us/corporate/press/017955_EN
Regards
Rajesh

Character semantics when setting the length of NVARCHAR2 field (Oracle 10g)

SQL> CREATE TABLE viv_naseleni_mesta (
2 naseleno_mjasto_id INTEGER NOT NULL PRIMARY KEY,
3 ime NVARCHAR2(15 CHAR) NOT NULL,
4 oblast_id INTEGER NOT NULL REFERENCES viv_oblasti(oblast_id),
5 grad CHAR NOT NULL CHECK (grad IN (0,1)),
6 de_timestamp TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
7 de_update_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
8 de_operator INTEGER DEFAULT NULL,
9 de_versija NUMBER(5,2) DEFAULT NULL
10 );
ime NVARCHAR2(15 CHAR) NOT NULL,
ERROR at line 3:
ORA-00907: missing right parenthesis
It seems that the DB does not accept the length of a field to be given in character
semantics. I have several sources that say it is possible, but is that
only possible in 11g?

According to the docs, NCHAR and NVARCHAR2 types are always character-based. So specifying in CHAR doesn't make sense explicitly.
http://download.oracle.com/docs/cd/B19306_01/server.102/b14225/ch3globenv.htm#sthref385
Cheers
Sarma.

Download Oracle 10g Character set scanner for Solaris

http://www.oracle.com/technology/software/tech/globalization/index.html
the above link contains only the oracle 9i Character set scanner but I need the Oracle 10g Character set scanner for Solaris
please assist me.

Pl see MOS Doc 745809.1 (Installing and configuring Csscan in 10g and 11g (Database Character Set Scanner)
HTH
Srini

Support for NVARCHAR2 and/or CHARACTER semantics

Does streams support replication of NVARCHAR2 between databases whose char set is ISO8859-1?
Same question for char fields declared with CHARACTER semantics.
Same questions as above but for GoldenGate.
Does GoldenGate support replication of NVARCHAR2 between databases whose char set is ISO8859-1?
Same question for GoldenGate for char fields declared with CHARACTER semantics.

No, we have consolidated on the Apache XML/XSL stuff. You are free to use
whatever other parser you would like however. The Sun XML pkg should work
fine.
Thanks,
Michael
Michael Girdley
BEA Systems Inc
"Boris Tabenkin" <[email protected]> wrote in message
news:[email protected]..
Are you still including the sun XML package, and if so which version?
"Toby Allsopp" <[email protected]> wrote in message
news:[email protected]..
Hi, Craig.
Craig Macha wrote:
Our company typically does NOT use Open Source products. We can't
risk
not having support.
I think you're a little confused about what "Open Source" means. Itdoesn't mean "unsupported." If you want support then you need to pay forit,
just like any software. One of the advantages of open source software is
that you aren't tied to a single vendor for your support - anyone with the
appropriate skills can take the source code and
provide support, as BEA have chosen to do in this case.
Regards,
Toby Allsopp.

Oracle documentation for patch & update

I'm just learning to make a focused and rapid search.
Someone could help me on find in the oracle documentation, given an instal baseline where i find the patchset that should i apply on?
Or how i should set my search to find it.
thanks

Your question is not clear for me, but you can find all the last available patchset (except the last 10.2.0.3) :
http://www.oracle.com/technology/support/patches.htm
Nicolas.

Oracle 9iAs documentation and patches for aix 5.3

Hi,
I have Oracle 9i application server release 1.0.2.2.2. I work on aix 5.3 and i've installed oracle server release 3. In Oracle 9iAs i need to work with oracle forms 6i and reports 6i.
I need Oracle Oracle 9i application server release 1.0.2.2.2. documentation and patches for aix 5.3
best regards

What is Oracle server release 3? Application server 1.0.2.2.2 is desupported now, it worked in the past with server release 9.2. Certification matrix on metalink shows a lot of necessary additional steps (OS patches and software patches) to be done. Metalink is also the place for Oracle software patches, OS patches come from IBM.
OS dumps (but for pre-6i versions) are documented in metalink note
DEVELOPER TOOLS CORE DUMP OR OPERATING SYSTEM ERRORS RELATED TO MOTIF
Note:1077772.6
Werner

How to set or change character set for Oracle 10 XE

Installing via RPM on Linux.
I need to have my database set to use UTF8 and WE8ISO8859P15 as the character set and national character set. (Think those are in the right order. If not, it's the opposite.)
If I do a standard "yum localinstall rpm-file-name," it installs Oracle. I then run the "/etc/init.d/oracle-xe configure" command to set my ports.
Every time I do this, I end up with AL32/AL16 character sets.
I finally hardcoded ISO-8859-15 as the Linux 'locale' character set and set this in the various bash profile config files. Now, I end up with WE8MSWIN1252 as the character set and AL16UTF16 as the national character set.
I've tried editing the createdb.sh script to hard code the character set types and then copied that file over the original while the RPM is still installing. I've tried editing the nls_lang.sh script to hard code the settings there and copied over the original shell script while the RPM is still installing.
Doesn't matter.
HOW can I do this? If I wait until after the RPM is installed and try running the createdb.sh file, then it ends up creating a database but not doing everything properly. I end up missing pfiles or spfiles. Various errors crop up.
If I try to change them from the sql command line, I am told that the new character set must be a superset of the old one. It fails.
I'm new to Oracle, so I'm treading water that's uncharted. In short, I need community help. It's important to the app I'm running and attempting to migrate from to maintain these character sets.
Thanks.

I don't think you can change Oracle XE character set. When downloading Oracle XE you must choose to download:
- either the Universal Edition using AL32UTF8
- or the Western Euopean Edition using WE8MSWIN1252.
See http://download.oracle.com/docs/cd/B25329_01/doc/install.102/b25144/toc.htm#BABJACJJ
If you really need UTF8 instead of AL32UTF8 you need to use Oracle Standard Edition or Oracle Entreprise Edition:
these editions allow to select database character set at database creation time which is not really possible with Oracle XE
Note that changing environment variable NLS_LANG has nothing to do with changing database character set:
http://download.oracle.com/docs/cd/B25329_01/doc/install.102/b25144/toc.htm#BABBGFIC

Converting an old Tom Kyte function to straight SQL

I am on Oracle 10.2
There is an old tom kyte function (10+ years old) floating around that will remove constants so I can look for queries that don't use bind variables. It has to loop through each character i a sql so its slow. I was able to convert part of it to using a regular expression. There are alot of new SQL features, so I am trying to replace it with sql.
The old REMOVE_CONSTRAINTS function is at the bottom.
I have already converted the really slow part of the function which loops through every character in a query (first loop). I did it with a regular expression. See test case below.
I am having trouble with the second part that replaces numbers.
The translate is there because I was having trouble passing a single quote to a regular expression and got tired of looking at it. So I converted all single quotes to '@''
Basically the translate turns all single quotes to @
So anything surrounded by an @ and an @ gets turned into a single @
This is a bit flawed. If I have a literal that has a quote in it, this will get screwed up. It is not something I have to worry about right now though.
select mytest,regexp_replace(translate(to_char(mytest),chr(39),'@'),'@[^@]+@','@') from test
create table test (mytest clob);
insert into test values ('select 1 from mytable where x = '||chr(39)||'v'||chr(39)||' and y = '||chr(39)||'q'||chr(39));
insert into test values ('select 1 from mytabe where a = '||chr(39)||'xyz'||chr(39));
commit;
SQLt> select mytest,regexp_replace(translate(mytest,chr(39),'@'),'@[^@]+@','@') from test
2 /
MYTEST
REGEXP_REPLACE(TRANSLATE(MYTEST,CHR(39),'@'),'@[^@]+@','@')
select 1 from mytable where x = 'v' and y = 'q'
select 1 from mytable where x = @ and y = @
select 1 from mytabe where a = 'xyz'
select 1 from mytabe where a = @I can put this in a pipeline function and it will run fast enough. I think I can probably make this straight sql. anyone know how? also, I have some column names with numbers in it. anyone know a good way to ignore numbers on the left side of the where clause?
    l_query := translate( l_query, '0123456789', '@@@@@@@@@@' );
--   dbms_output.put_line(l_query);
    for i in 0 .. 8 loop
        l_query := replace( l_query, lpad('@',10-i,'@'), '@' );
        l_query := replace( l_query, lpad(' ',10-i,' '), ' ' );
    end loop;
function remove_constraints ( p_query in clob ) return clob AS
    l_query clob;
    l_char varchar2(1);
    l_in_quotes boolean default FALSE;
    l_length number := 0;
    vmodule varchar2(30);
    vaction varchar2(30);
begin
    dbms_application_info.read_module(vaction,vmodule);
    dbms_application_info.set_module(vaction,nvl(vmodule,0)+1);
    for i in 1 .. length( p_query )
    loop
        l_char := substr(p_query,i,1);
        if ( l_char = '''' and l_in_quotes )
        then
            l_in_quotes := FALSE;
        elsif ( l_char = '''' and NOT l_in_quotes )
        then
            l_in_quotes := TRUE;
            l_query := l_query || '''#';
            dbms_output.put_line('a '||l_char);
            dbms_output.put_line( '''#');
        end if;
        if ( NOT l_in_quotes ) then
            l_query := l_query || l_char;
        end if;
    end loop;
--dbms_output.put_line(l_query);
    l_query := translate( l_query, '0123456789', '@@@@@@@@@@' );
--   dbms_output.put_line(l_query);
    for i in 0 .. 8 loop
        l_query := replace( l_query, lpad('@',10-i,'@'), '@' );
        l_query := replace( l_query, lpad(' ',10-i,' '), ' ' );
    end loop;
    return upper(l_query);
end;


In 10g you can use
FORCE_MATCHING_SIGNATURE     The signature used when the CURSOR_SHARING parameter is set to FORCE
http://docs.oracle.com/cd/B19306_01/server.102/b14237/dynviews_2113.htm#REFRN30246

Need explanation about Oracle documentation (DBMS_LOB.LOADCLOBFROMFILE)

Hello World,
I try to understand Oracle documentation about DBMS_LOB.LOADCLOBFROMFILE function.
I am not shure the explanation is written in a very good english,
but i can't understand a word !!!
Is anybody could explain me in more simple sentences ??
DBMS_LOB.LOADCLOBFROMFILE (
dest_lob IN OUT NOCOPY BLOB,
src_bfile IN BFILE,
amount IN INTEGER,
dest_offset IN OUT INTEGER,
src_offset IN OUT INTEGER,
src_csid IN NUMBER,
lang_context IN OUT INTEGER,
warning OUT INTEGER);
Oracle documentation explanation :
src_csid Character set id of the source (BFILE) file.
lang_context (IN) Language context, such as shift status, for the current
load. (OUT) The language context at the time when the
current load stopped, and what the next load should be using if
continuing loading from the same source. This information is
returned to the user so that they can use it for the continuous
load without losing or misinterpreting any source data. For the
very first load or if do not care, simply use the default 0. The
details of this language context is hidden from the user. One
does not need to know what it is or whats in it in order to
make the call
warning (OUT) Warning message. This indicates something abnormal
happened during the loading. It may or may not be caused by the
users mistake. The loading is completed as required, and its up to the
user to check the warning message. Currently, the only possible
warning is the inconvertible character. This happens when the
character in the source cannot be properly converted to a character in
destination, and the default replacement character (e.g., ?) is used in
place. The message is defined as warn_inconvertable_char in
DBMSLOB.

I am not kidding. I posted the link but I did not check if it were the same source. Really I do not have time to realize it.
I think you can get a fast answer in SQL & PL/SQL forum.
Good Luck!
Joel Pérez

Expdp with network_link and different character sets for source and target?

DB versions 10.2 and 10.1
Is there any limitation for implementing expdp with network_link but target and source have different character sets?
I tried with many combinations, but only had success when target and source have the same character sets. In other combinations there was invalid character set conversion (loose of some characters in VARCHAR2 and CLOB fields).
I didn't find anything on the internet, forums or Oracle documentation. There was only limitation for transportable tablespace mode export (Database Utilities).
Thanks

Hi DBA-One
This link
http://download-east.oracle.com/docs/cd/B19306_01/server.102/b14215/dp_overview.htm#CEGFCFFI
is from Database Utilities 10g Release 2 (10.2) (I read it many times)
There is only one thing about different character sets.
I quote:
Data Pump supports character set conversion for both direct path and external tables. Most of the restrictions that exist for character set conversions in the original Import utility do not apply to Data Pump. The one case in which character set conversions are not supported under the Data Pump is when using transportable tablespaces.
Parameter VERSION also doesn't play role here, because the behaviour (character set) is the same if I use for target and source only 10.2 or (10.2 and 10.1) respectivly.

Oracle database character set

Hello all -
Please help..................
I will be exporting/importing 6/7 users/schemas/data from one database to the another database on solaris. Users are created.
I am confused about NLS_LANG variable and database characterset.
I have the following questions -
1. What is impact of NLS_LANG variable setting of user session while import/export the data ?
2. Why do we need to set this NLS_LANG user session variable before export/import ?
3.If NLS_LANG variable is not set (doesnot have any value) what would happen ?
4. If I have to set NLS_LANG varible, what should I set it to?
5. How can I see the characterset of my database?
6. Where can I get more info about database charaterset and What are the valid values for database characterset and NLS_LANG varibale ?
Any help would really appreciated...
Thanks a lot.....
RAMA

1. What is impact of NLS_LANG variable setting of user session while import/export the data ?
On export, the data will be converted from the database character set to the character set specified by NLS_LANG. In import, the database will assume that the data is in the character set specified by NLS_LANG and use that value to perform the conversion to the database character set if the two values to not match.
2. Why do we need to set this NLS_LANG user session variable before export/import ?
If your database character set is the same as your OS, you don't necessarily have to set NLS_LANG. For instance, if you have a US7ASCII db, and your OS locale is set to AMERICA_AMERICAN.US7ASCII, there won't be any problems. The only time it's really important to set this is when the db and OS settings don't match.
3.If NLS_LANG variable is not set (doesnot have any value) what would happen ?
If your database character set doesn't match your OS, the data could be garbled because the db will incorrectly transcode the data on import/export.
4. If I have to set NLS_LANG varible, what should I set it to?
Depends on what your database character set is set to (see below).
5. How can I see the characterset of my database?
select * from nls_database_parameters and look for the value set for the NLS_CHARACTERSET parameter. Don't get confused by the NLS_NCHAR_CHARACTERSET, that's for NCHAR datatypes.
So, for instance, if your NLS_CHARACTERSET value is set to UTF8, you would set NLS_LANG to .UTF8 (the dot is important because that's actually shorthand for territory_language.characterset, or language_territory.characterset, I can never remember which comes first. In any case, use the dot). For example:
setenv NLS_LANG .UTF8
6. Where can I get more info about database charaterset and What are the valid values for database characterset and NLS_LANG varibale ?
It's all in the Oracle documentation.
hope this helps.
Tarisa.

Where is the official Oracle documentation that describes Hot Backup?

I have used the woefully inadequate Metalink search utilities and Google but the nearest thing I have found is a description on "Ask Tom".
I am looking for the official Oracle documentation that describes what happens internally during a Hot Backup. There are a lot of web sites and blogs out there that describe it but I am looking for the description from Oracle themselves. The only things I have found in the Oracle documentation so far is how to start and stop Hot Backup but not what happens internally. Preferably 11gR2 if it makes any difference.
Thanks in advance.

Thanks mseberg. This is what I was looking for. I had read this manual before but apparently glossed over this paragraph:
"When performing a user-managed online backup, you must place your datafiles into
backup mode with the ALTER DATABASE or ALTER TABLESPACE statement with the
BEGIN BACKUP clause. When a tablespace is in backup mode, the database writes the
before image for an entire block to the redo stream before modifying a block. The
database also records changes to the block in the online redo log. Backup mode also
freezes the data file checkpoint until the file is removed from backup mode. Oracle
Database performs this safeguard because it cannot guarantee that a third-party
backup tool copies the file header before copying the data blocks."
Hot backup is still quite useful and Rman doesn't cover all cases. What we are doing with it is using storage-based replication (Shadow Image) to another set of LUNS which we then mount on another server and backup using Rman. This removes the overhead of Rman on the Production Database server and extends our backup window to almost 24 hours.

Strange happenings with examples from Tom Kyte's book. 10 XE

Hi all,
I was twiddling my thumbs and decided to brush up on some of Oracle's fundamentals.
Tom Kyte's book - Effective Oracle by Design. Examples on pp. 141 - 142 (to do with
bind variables). Typed in the example and obtained a very strange system reaction.
It consists of a package "DEMO_141_PKG" with 1 procedure "parse_bind_execute_close"
which is called by the procedure QUICKFIX142.. The system then proceeds to go mental.
If anybody could explain to me what's going on, I would be grateful. I'm running 10 XE on
Ubuntu Linux.
On running the code (below) I get this in the "Run" window of SQLDeveloper.
Connecting to the database demo.
ORA-01000: maximum open cursors exceeded
ORA-06512: at "SYS.DBMS_SYS_SQL", line 884
ORA-06512: at "SYS.DBMS_SQL", line 9
ORA-06512: at "DEMO.DEMO_141_PKG", line 17
ORA-06512: at "DEMO.QUICKFIX142", line 8
ORA-06512: at line 2And the logging window of SQLDevloper just shows a continuous
stream of the text below - I have to kill the sessions as SYS from
within Oracle - the Terminate process from within SQLDeveloper
doesn't manage to kill this runaway process.
Logging output
SEVERE     2474100     1     oracle.dbtools.raptor.runner.DBStarterFactory     logDbmsOutput: ORA-01000: maximum open cursors exceeded
SEVERE     2474099     1     oracle.dbtools.raptor.runner.DBStarterFactory     logDbmsOutput: ORA-01000: maximum open cursors exceeded
SEVERE     2474098     0     oracle.dbtools.raptor.runner.DBStarterFactory     logDbmsOutput: ORA-01000: maximum open cursors exceeded
SEVERE     2474097     0     oracle.dbtools.raptor.runner.DBStarterFactory     logDbmsOutput: ORA-01000: maximum open cursors exceeded
SEVERE     2474096     1     oracle.dbtools.raptor.runner.DBStarterFactory     logDbmsOutput: ORA-01000: maximum open cursors exceeded
<Millions of lines snipped - it just keeps going>Now, the code is here for those who would like to help me get to the bottom of this phenomenon.
create or replace
PACKAGE DEMO_141_PKG AS
procedure parse_bind_execute_close(p_input in varchar2);
END DEMO_141_PKG;
create or replace
PACKAGE BODY DEMO_141_PKG AS
g_first_time boolean := TRUE;
g_cursor number;
procedure parse_bind_execute_close(p_input in varchar2)
AS
l_cursor number;
l_output varchar2(4000);
l_status number;
BEGIN
l_cursor := dbms_sql.open_cursor;
dbms_sql.parse(l_cursor, 'SELECT * FROM Dual WHERE Dummy = :x', dbms_sql.native);
dbms_sql.bind_variable(l_cursor, ':x', p_input);
dbms_sql.define_column(l_cursor, 1, l_output, 4000);
l_status := dbms_sql.execute(l_cursor);
if(dbms_sql.fetch_rows(l_cursor) <= 0)
then
    l_output := null;
else
    dbms_sql.column_value(l_cursor, 1, l_output);
end if;
END parse_bind_execute_close;
END DEMO_141_PKG;and the above is called here
create or replace
PROCEDURE QUICKFIX142 AS
BEGIN
-- demo.runstats_pkg.rs_start; // Don't worry about runstats - it's a Tom Kyte package for
-- timings and measuring contention.
execute immediate 'alter session set session_cached_cursors = 0';
for i in 1..1000
loop
    DEMO_141_PKG.parse_bind_execute_close('Y');
end loop;
-- runstats_pkg.rs_middle;
execute immediate 'alter session set session_cached_cursors = 100'; // reduced this to 50, 20, 10 & 5 - no effect.
for i in 1..1000
loop
    DEMO_141_PKG.parse_bind_execute_close('Y');
end loop;
-- runstats_pkg.rs_stop;
END QUICKFIX142;Edited by: Paulie on Aug 12, 2011 2:18 PM

If you modify the called package by adding :
-- CLOSE THE CURSOR
dbms_sql.close_cursor(l_cursor);in
procedure parse_bind_execute_close(p_input in varchar2)
AS
l_cursor number;
l_output varchar2(4000);
l_status number;
BEGIN
l_cursor := dbms_sql.open_cursor;
dbms_sql.parse(l_cursor, 'SELECT * FROM Dual WHERE Dummy = :x', dbms_sql.native);
dbms_sql.bind_variable(l_cursor, ':x', p_input);
dbms_sql.define_column(l_cursor, 1, l_output, 4000);
l_status := dbms_sql.execute(l_cursor);
if(dbms_sql.fetch_rows(l_cursor) <= 0)
then
    l_output := null;
else
    dbms_sql.column_value(l_cursor, 1, l_output);
end if;
-- CLOSE THE CURSOR
dbms_sql.close_cursor(l_cursor);
END parse_bind_execute_close;
END DEMO_141_PKG;Then script runs with default OPEN_CURSORS setting:
SQL> connect / as sysdba
Connected.
SQL> show parameter open_c
NAME                                 TYPE        VALUE
open_cursors                         integer     300
SQL> connect test/test
Connected.
SQL> exec quickfix142;
PL/SQL procedure successfully completed.

Tom Kyte -v- Oracle Documentation - Character Semantics for AL32UTF8 ?

Similar Messages

Maybe you are looking for