Oracle 11: Register XSDs in different ways

Hello,
i have two XSD-Files. One includes the other.
Is it possible that i register the two xsd-files with different options?
I mean, that i register one in the following way:
BEGIN
DBMS_XMLSchema.registerSchema(schemaurl=>'http://190.1.1.148:15000/home/user/myFirst.xsd'xsd',schemadoc=>
sys.UriFactory.getUri('/home/user/myFirst.xsd'));
END;
And the other one (as BinaryXml):
BEGIN
DBMS_XMLSCHEMA.registerSchema
(SCHEMAURL => 'http://190.1.1.148:15000/home/user/mySecond.xsd',
SCHEMADOC => sys.UriFactory.getUri('/home/user/mySecond.xsd'),
LOCAL => TRUE, -- local
GENTYPES => FALSE, -- generate object types
GENBEAN => FALSE, -- no java beans
GENTABLES => TRUE, -- generate object tables
OPTIONS => DBMS_XMLSCHEMA.REGISTER_BINARYXML,
OWNER => USER);
END;
Or should ´t i do this? If not, why. Or can i do this?
Thank you very much
Best regards
Heidi

Currently all schemas that will be used with Binary XML must be registered with the FOR BINARY XML option. The Schema compliation process is different for Object relational storage and Binary XML Storage.

Similar Messages

Database duplication oracle 10g different ways

Hi everyone,
I need to know the different ways of duplicating the oracle 10g database.Reply will be appreciated.If U know any document ID,kindly post it too.
Thanks
Jaya prataab J

you could:
- shutdown the database and simply copy all files
- use rman duplicate to make a clone
- use transportable tablespaces
- datapump export/import
- conventional export/import

XML validation against registered XSD

I have registered XSD into oracle database, while validating xml against this xsd I got stuck with non-mandatory tags.
Below is the tag-XSD structure:
Status field as String:
<xs:element name="Status" nillable="false"> <xs:annotation> <xs:documentation>Values are 'N' or 'E'</xs:documentation>
</xs:annotation>
<xs:simpleType>
<xs:restriction base="xs:string">
<xs:maxLength value="1"/>
</xs:restriction>
</xs:simpleType>
</xs:element>
idnumber tag field as Integer:
<xs:element name="idnumber" type="xs:int" nillable="false">
<xs:annotation>
<xs:documentation>id value </xs:documentation>
</xs:annotation>
</xs:element>
As status tag is of string type and I have mentioned it as nillable = "False", but even though when I am passing this tag as <status/>, then also its validating xml.... but as i mentioned nillable property as false then it should not validate it ..... please correct me if i m wrong..
In case of idnumber tag it is of int type. In this case <idnumber/> will not work in either case even it is nillable = true or false. Means we cannt pass null values in integer field..... right?
Please suggest me how to make tag of string datatype as not nillable.
Is there any significance of using nillable = "false" in int datatype tag. Means if we want to make it as optional then need to mention it as minoccurs = 0..... is it correct?
Requirement is simple:
Just want to make some tag as optional and some as mandatory, tag can be of integer or string type.
Thanks in advance.....
Edited by: 959352 on 15-Oct-2012 00:45

As status tag is of string type and I have mentioned it as nillable = "False", but even though when I am passing this tag as <status/>, then also its validating xml.... but as i mentioned nillable property as false then it should not validate it ..... please correct me if i m wrong.. The nillable attribute doesn't work that way. It's easy to look that up in any tutorial.
nillable = "true" means you allow the element to have a NULL content (different from empty string) by specifying an explicit null attribute :
<Status xsi:nil="true" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"/>nillable = "false" is the default behaviour.
Please suggest me how to make tag of string datatype as not nillable.
Is there any significance of using nillable = "false" in int datatype tag. Means if we want to make it as optional then need to mention it as minoccurs = 0..... is it correct?
Requirement is simple:
Just want to make some tag as optional and some as mandatory, tag can be of integer or string type. You have to be careful about the vocabulary.
In my mind, an "optional" element means you can omit it entirely in the instance document (minOccurs="0"), whereas an empty element means the start tag and end tag are there but with no content.
There are multiple ways to allow for empty or non-empty elements.
e.g. for a non-empty string :
<xs:minLength value="1"/>or, as in your example, if the content is actually constrained to a list of values, use enumerations.
xs:int datatype doesn't allow empty values (unless you use xsi:nil).
A possible solution is to build a new type by unioning (xs:union) an xs:int and an empty string type.

Register XSD-Files which includes Other XSD-Files

Hallo
i have two XML-Schemata. One is included by the other.
Here is an example:
This is the XSD, which is included by the other (metadata.xsd):
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema" elementFormDefault="qualified">
     
     <xs:element name="metadata" type="metadataType">
          <xs:annotation>
               <xs:documentation>ROOT ELEMENT von Metadata</xs:documentation>
          </xs:annotation>
     </xs:element>
     
     <xs:complexType name="metadataType">
          <xs:sequence>
               <xs:element ref="identifier"/>
               <xs:element ref="category"/>
               <xs:element ref="creator"/>
               <xs:element ref="creation_date"/>
               <xs:element ref="publisher"/>
               <xs:element ref="language"/>
               <xs:element ref="title"/>
               <xs:element ref="basic_year"/>
               <xs:element ref="statistic"/>
This is the XSD, which include the other XSD (lexikon.xsd):
<?xml version="1.0" encoding="UTF-8"?>
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xdb="http://xmlns.oracle.com/xdb" elementFormDefault="unqualified" attributeFormDefault="unqualified">
     
     <xs:include schemaLocation="metadata.xsd"/>
     
     <xs:element name="lexikon" type="lexikonType" xdb:defaultTable="LEXIKON">
          <xs:annotation>
               <xs:documentation>ROOT ELEMENT</xs:documentation>
          </xs:annotation>
     </xs:element>
     <xs:complexType name="lexikonType">
          <xs:sequence>
               <xs:element ref="metadata"/>
               <xs:element ref="lexikonEntries"/>
          </xs:sequence>
          <xs:attribute name="lexikonKat" type="xs:integer">
               <xs:annotation>
                    <xs:documentation>für spezielle UnterkategorienTypen zu den Lexikas</xs:documentation>
               </xs:annotation>
          </xs:attribute>
     </xs:complexType>
     
     
     
     
     <xs:element name="explicationTxt" type="explicationType"/>
     <xs:complexType name="explicationType">
          <xs:choice>
               <xs:element ref="verweis"/>
               <xs:element ref="txt"/>
               <xs:element ref="para"/>
Both documents have i copied to a webdav-folder. Now i want register the documents.
BEGIN
DBMS_XMLSchema.registerSchema(
schemaurl=>'http://190.1.1.148:8080/home/DEV/xsd/lexikon.xsd',
schemadoc=>sys.UriFactory.getUri('/home/DEV/xsd/lexikon.xsd'));
END;
The following error-message is displayed:
ERROR at line 1:
ORA-31000: Resource 'metadata.xsd' is not an XDB schema document
ORA-06512: at "XDB.DBMS_XMLSCHEMA_INT", line 0
ORA-06512: at "XDB.DBMS_XMLSCHEMA", line 166
ORA-06512: at line 2
I made the XSD-Files by XMLSpy. The Documents are well-formated and valid.
When i make a new XML-File based ONLY (of course) on lexikon.xsd everything works fine with XMLSpy.
How must i register these documents? Is it possible? Or can´t i register XSD-Files which includes others?
Thank you very much
cu
Heidi

Heidi,
register the metadata.xsd first inthe same way and then register the lixikon.xsd.
it should work.

I have an iPod of rock and pop tracks which i generally use on shuffle mode. Now I have a second iPod to store classical music. If I plug it in to my pc will it duplicate my rock and pop or will it register as a different device?

I have an iPod Classic with rock and pop tracks on which I generally play in shuffle mode. I have now bought a second iPod Classic to store and play classical music as I don't want to shuffle rock with classical. What will happen when I attach the new iPod to my pc? Will it register as a new device with an iTunes file of its own, or will it duplicate what I already have on the original iPod?

You can attach the second iPod and configure it to sync whatever content you like without affecting the first iPod. I would take a look at this article for different ways to manage multiple iPods on one computer.
How to use multiple iPods, iPads, or iPhones with one computer
B-rock

Different ways to referencing Session State variables

Hi,
According to APEX documentation there's 4 different ways to reference session state variables: http://download-west.oracle.com/docs/cd/B32472_01/doc/appdev.300/b32471/concept.htm#BEICHBBG
In an inline PL/SQL statement, what's the difference when using the different methods? I remember reading something that the bind and static text have a size restrictions. What's the difference between the V() and NV() functions?
Thank you.
Martin

Martin,
In PL/SQL, the preferable method is to use bind variable notation, e.g., :P1_ITEM. In HTML contexts, you must use &ITEM. notation. In stored procedures, you can use v or nv, the latter function being identical to the former with the additional characteristic that it raises an exception if the retrieved value is non-numeric.
Scott

Different ways of passing procedure's paramters

Hi,
In the following example, I am calling the procedure p_test in different ways.
I understood from reading examples, that we can pass parameters using the syntaxe : " Name of the parameters => Value".
But I did not find anything official about it.
I'm just looking for Oracle Documentation where it is explained the syntaxe " Name of the parameters => Value".
CREATE OR REPLACE procedure test_call
is
procedure p_test(I_FIRST IN NUMBER DEFAULT 1, I_SECOND IN NUMBER DEFAULT 2 )
is
begin
dbms_output.put_line ('-----------------');
dbms_output.put_line ('FIRST:'||I_FIRST);
dbms_output.put_line ('SECOND:'||I_SECOND);
end;
begin
p_test(10,11);
p_test(10);
p_test();
p_test(I_FIRST => 100);
p_test(I_SECOND => 200);
p_test(I_SECOND => 2000, I_FIRST => 1000);
exception
when others then
ks_Error.log();
end;
/

note that the => is illegal in sql in 10g and before
select f(x=>1) from dual;
ERROR at line 1:
ORA-00907: missing right parenthesisbut it will probably be possible in 11g

Different ways to shutdown and startup instance in 10g

I have just installed 10g on my win2003 server. There seems to be 3 different ways to shutdown / startup an instance in 10g.
1. via SQLPLUS (how its always been) with shutdown and startup commands
2. via the new OEM Web interface
3. Start / Stop the NT Service for the instance
questions:
- Did I miss anything other methods?
- how come OEM ask for O/S login as well as DB login, when shutdown or startup? but via SQLPLUS I only have to login using as DBA? different level of security?
- When shutdown via either SQLPLUS or OEM, the NT service is still shown as "started"? so does this mean that the NT service is more than the oracle instance???
Thanks
Clo

how come OEM ask for O/S login as well as DB loginWith SQL*Plus, you are already logged into the Operating System. There is a OS session in progress. With OEM web interface, it has to first start a OS session and then connect to Oracle using Oracle credentials supplied to actually shutdown the DB.
the NT service is still shown as "started"? The difference between NT service started or not started (given that the database was already shutdown) will be:
When NT service is started but the DB is shutdown, your users will get this error message:
Enter user-name: scott
Enter password:
ERROR:
ORA-01034: ORACLE not available
ORA-27101: shared memory realm does not exist
Enter user-name:When NT service is stopped, your users will get this error message:
Enter user-name: scott
Enter password:
ERROR:
ORA-12560: TNS:protocol adapter error
Enter user-name:The NT service needs to be running for you to be able to do a STARTUP or SHUTDOWN on the database. You will NOT be able to STARTUP if the NT service was not running (STARTUP will not start your NT service automatically).

What are the different ways of retrieving data from Oracle8i

What are the different ways of retrieving data from Oracle8i
into my HTML page ?
Is it JDBC and ODBC ?
Is there any other way ?
null

Methods I tried,
Applet using SQLJ/JDBC with JDBC drivers thin or Oci8,
Oracle Web Publishing Assistant,
HTP/HTF PL/SQL packages (if you have OAS 4.0)
Webserver Generator of Designer 2000 (if you have OAS 4.0)
Arun (guest) wrote:
: What are the different ways of retrieving data from Oracle8i
: into my HTML page ?
: Is it JDBC and ODBC ?
: Is there any other way ?
null

Different ways for inter-application communcations

Hi,
Can anyone please tell me how many different ways by which an application created in java can communicate with:
1. Application developed in Java and
2. Application Developed in other platform
Thanks and Regards,
Faisal

The most common ones might be;
- TCP sockets
- UDP datagrams.
- HTTP/webservices.
- JMS/Messaging
- FTP (not so common because its a often a bad idea)
- RMI/RPC/IIOP
- Distributed data solutions, like JGroups.
- Distributed JVM solutions, like Terracotta.
- Network relational databases, like MySQL, Oracle, Sybase, SQL Server.
- Directory services, e.g. LDAP servers.
- local/NFS/SAN/NAS/SMB file systems.
- Shared memory/Other shared resources like named pipes.
- serial communications (e.g. to a J2ME device)
There are probably a few more I've forgotten.

Different ways of enabling cache

Hi
Doubt:
what are the different ways of enabling cache in obiee,other than "enable=yes" in Nqsconfig.ini file.
Regards
mervin

Hi Mervin,
You have to specify the cache settings on the Physical Table;
http://download.oracle.com/docs/cd/E12096_01/books/admintool/admintool_QueryCaching5.html
Daan Bakboord
http://obibb.wordpress.com

BI server generating query in a different way between two instances

Hi All,
We have executed a report in dev,test instances,BI server generated query in a different way in two instances where as dev BI server is on AIX operating system(recently we migrated from windows),test instance is on still on Windows environment.
For a report below are the queries
DEV(AIX)
WITH
SAWITH0 AS (select sum(T316025.SALES_QUOTA) as c1,
T329697.DIVISION_DESC as c2,
T329697.AREA_DESC as c3,
T329697.TERRITORY_DESC as c4,
case when T329697.ACCOUNT_NUM is null then T329697.BILL_TO_PARTY_NAME else concat(concat(concat(T329697.BILL_TO_PARTY_NAME, '('), T329697.ACCOUNT_NUM), ')') end as c5,
T150993.X_CONS_MAJOR_GROUP as c6,
T66755.PER_NAME_ENT_YEAR as c7
from
W_DAY_D T66755 /* Dim_W_DAY_D_Common */ ,
W_PRODUCT_D T67704 /* Dim_W_PRODUCT_D */ ,
WC_SLX_DATA_F T316025 /* Fact_WC_SLX_DATA_F */ ,
WC_CUSTOMER_HIERARCHY_D T329697 /* Dim_WC_Customer_Hierarchy_D_With_Error */ ,
OBIEE_SECURITY_LOCATION_SALES T339204,
W_PROD_CAT_DH T150993 /* Dim_W_PROD_CAT_DH_General */
where ( T66755.ROW_WID = T316025.DAY_WID and T316025.CUSTOMER_HIERARCHY_WID = T329697.ROW_WID and T67704.ROW_WID = T316025.PRODUCT_WID and T67704.PROD_CAT2_WID = T150993.ROW_WID and T329697.TERRITORY_CODE = nvl(T339204.LOCATION , T329697.TERRITORY_CODE) and T329697.AREA_DESC = 'GROCERY AREA - EAST' and T329697.DIVISION_DESC = 'DOMESTIC SALES DIVISION' and T339204.USER_NAME = upper('Administrator') and case when T329697.ACCOUNT_NUM is null then T329697.BILL_TO_PARTY_NAME else concat(concat(concat(T329697.BILL_TO_PARTY_NAME, '('), T329697.ACCOUNT_NUM), ')') end = 'JETRO CASH AND CARRY ENTERPRISES INC(10313)' and (T66755.PER_NAME_ENT_YEAR in ('2011', '2012')) and (T329697.TERRITORY_DESC in ('BOSTON', 'CHARLOTTE', 'FLORIDA', 'GREAT LAKES', 'MID-SOUTH', 'NEW YORK', 'WHITE ROSE')) )
group by T66755.PER_NAME_ENT_YEAR, T150993.X_CONS_MAJOR_GROUP, T329697.TERRITORY_DESC, T329697.AREA_DESC, T329697.DIVISION_DESC, case when T329697.ACCOUNT_NUM is null then T329697.BILL_TO_PARTY_NAME else concat(concat(concat(T329697.BILL_TO_PARTY_NAME, '('), T329697.ACCOUNT_NUM), ')') end )
select distinct SAWITH0.c2 as c1,
SAWITH0.c3 as c2,
SAWITH0.c4 as c3,
SAWITH0.c5 as c4,
SAWITH0.c5 as c5,
SAWITH0.c6 as c6,
SAWITH0.c7 as c7,
SAWITH0.c1 as c8
from
SAWITH0
order by c1, c6
Test(Windows)
select distinct D1.c2 as c1,
D1.c3 as c2,
D1.c4 as c3,
D1.c5 as c4,
D1.c5 as c5,
D1.c6 as c6,
D1.c7 as c7,
D1.c1 as c8
from
(select sum(T316025.SALES_QUOTA) as c1,
T329697.DIVISION_DESC as c2,
T329697.AREA_DESC as c3,
T329697.TERRITORY_DESC as c4,
case when T329697.ACCOUNT_NUM is null then T329697.BILL_TO_PARTY_NAME else concat(concat(concat(T329697.BILL_TO_PARTY_NAME, '('), T329697.ACCOUNT_NUM), ')') end as c5,
T150993.X_CONS_MAJOR_GROUP as c6,
T66755.PER_NAME_ENT_YEAR as c7
from
W_DAY_D T66755 /* Dim_W_DAY_D_Common */ ,
W_PRODUCT_D T67704 /* Dim_W_PRODUCT_D */ ,
WC_SLX_DATA_F T316025 /* Fact_WC_SLX_DATA_F */ ,
WC_CUSTOMER_HIERARCHY_D T329697 /* Dim_WC_Customer_Hierarchy_D_With_Error */ ,
OBIEE_SECURITY_LOCATION_SALES T339204,
W_PROD_CAT_DH T150993 /* Dim_W_PROD_CAT_DH_General */
where ( T66755.ROW_WID = T316025.DAY_WID and T316025.CUSTOMER_HIERARCHY_WID = T329697.ROW_WID and T67704.ROW_WID = T316025.PRODUCT_WID and T67704.PROD_CAT2_WID = T150993.ROW_WID and T329697.TERRITORY_CODE = nvl(T339204.LOCATION , T329697.TERRITORY_CODE) and T329697.AREA_DESC = 'GROCERY AREA - EAST' and T329697.DIVISION_DESC = 'DOMESTIC SALES DIVISION' and T339204.USER_NAME = upper('Administrator') and case when T329697.ACCOUNT_NUM is null then T329697.BILL_TO_PARTY_NAME else concat(concat(concat(T329697.BILL_TO_PARTY_NAME, '('), T329697.ACCOUNT_NUM), ')') end = 'JETRO CASH AND CARRY ENTERPRISES INC(10313)' and (T66755.PER_NAME_ENT_YEAR in ('2011', '2012')) and (T329697.TERRITORY_DESC in ('BOSTON', 'CHARLOTTE', 'FLORIDA', 'GREAT LAKES', 'MID-SOUTH', 'NEW YORK', 'WHITE ROSE')) )
group by T66755.PER_NAME_ENT_YEAR, T150993.X_CONS_MAJOR_GROUP, T329697.TERRITORY_DESC, T329697.AREA_DESC, T329697.DIVISION_DESC, case when T329697.ACCOUNT_NUM is null then T329697.BILL_TO_PARTY_NAME else concat(concat(concat(T329697.BILL_TO_PARTY_NAME, '('), T329697.ACCOUNT_NUM), ')') end
) D1
order by c1, c6
If we observe test query is very simple and easy to back track but in Dev it is appending like sawwith0,sawwith1 etc..looking difficult.
Is there in any configuration to change to generate query like Test(with out SAWWITH0).
NOTE:Any how results are same in both instances.
Please help me to resolve this issue.
Thank You,
Anil Kumar.

Anil,
Are your database settings the same in both of the rpds dev and test?
Check wether you didn't change anything to the default. Open your RDP double click on the Database in the physical layer and go to Feature you can check all the settings there
Adil

Different ways to setup applets... what's best?

Alright. I need some help deciding how to setup an applet. I'm a high school student and I'm going to be in a competition on Saturday that will be using Java applets. I've been trying to get my hands on as much info about it as I can, but I've seen so many different ways of displaying things on an applet that I'm not sure what to do anymore.
So here's the deal. In a book I have, they set up applets like this:
import javax.swing.*;
public class Example extends JApplet
   public void init()
      getContentPane().add(new ExamplePanel ());
      set size (300, 300)
}And the other class:
import java.awt.*;
import java.awt.event.*;
import javax.swing.*;
public class ExamplePanel extends JPanel
   private vars
   public ExamplePanel ()
      set up listeners and set background
   public void paintComponent (Graphics page)
      super.paintComponent (page);
      paint stuff
   event methods
}I'm pretty sure this works fine, but while on the Java website, looking for examples, I noticed that sun sets up applets like this:
import java.awt.*;
import java.awt.event.*;
import javax.swing.*;
import javax.swing.event.*;
public class Example extends JPanel implements listener interfaces
   protected and private vars
   public Example ()
      set up panels, labels, basically any JComponent stuff
      create listeners
      use add(...) to put stuff on applet
   event methods
   private static void createAndShowGUI()
        JFrame frame = new JFrame ("Example");
        frame.setDefaultCloseOperation (JFrame.EXIT_ON_CLOSE);
        JComponent newContentPane = new Example();
        newContentPane.setOpaque (true);
        frame.setContentPane (newContentPane);
        frame.pack();
        frame.setVisible(true);
    public static void main()
        javax.swing.SwingUtilities.invokeLater(new Runnable()
            public void run()
                createAndShowGUI();
}I don't know which way is better, but we don't have much time so I need something that's easy. The problem with the second one is that I don't know how to draw shapes on the second one and I might need to do that. I tried using paintComponent in it and things just got screwy. If anyone knows of a better way, please let me know. And since that one was developed directly from Sun, I'm guessing it's more efficient, so I would like to use that one if I can just learn how to use it as easily as the first.
Also, if anyone knows a completely different way to do this that's better, please let me know.

The second Example class shows an important issue: swing is not thread safe, and thus all GUI operations, including creating the GUI itself, must be done in the event dispatch thread. Read more about it at http://java.sun.com/docs/books/tutorial/uiswing/misc/threads.html#EDT.
The second class Example does also not seem to be an applet, that is why main() is used instead of init(). But even in init() you are not guaranteed that the current thread is the EDT - it is choosen by the browser you are using - and you should thus use invokeLater() there as well.
Appart from that, a component should not care if it is displayed in an applet or in an application.

A different way to deal with "unknown" user after upgrade from Tiger.

It's a known issue that doing upgrade and install or archive and install from Tiger to Leopard, produces a lot of folders with "unknown" group in their "get info" panel. This has to do with a different group structure in Tiger and Leopard.
In Tiger, every user has their own private group with the same name as the username. That group is set to be user's primary group.
In leopard, every new user with an account is a member of the group staff (gid=20) which is not present in Tiger. Leopard doesn't fully understand Tiger's private group structure, hence the "unknown" group. This causes various problems such as Finder crashes when trying to change the permissions and ownership in the get info window.
Apple's proposed way of dealing with this is described [here|http://docs.info.apple.com/article.html?artnum=307128].
However, this method doesn't change the existing group structure and doesn't resolve all the problems.
I'd like to suggest a different way of dealing with it.
1. create a new admin user (if you don't have one already), log out of your primary account and log into the new one.
2. enter the following in terminal:
*sudo dscl . delete /users/"username"*
where username is the short name of the user on the main account (not the one you are currently logged into).
This will delete that user.
3. Log out/in.
4. go to system preferences->accounts and create a new user with the name and short name exactly as the one you've deleted in step 2.
You'll be get a popup saying that a home directory by this name already exists and asking if you want to use it. Say "Yes".
That's it. Your main user will be recreated using the native group structure of Leopard with gid=20(staff). Permissions on your home directory will be reset with correct group and ownership.
I tested the process on my powerbook after an upgrade and install from Tiger to leopard and it worked without a hitch.
I like this method better than Apple's because it completely gets rid of Tiger's group structure which was creating all the confusion. The only downside as far as I can see is that you might have to manually change the group ownership on some files belonging to the main user that sit outside your home directory.
I would appreciate any comments on this.

hmm, I confess, I did this before installing 10.5.2 so i can't say for sure. However, I've dealt with a few people (see e.g. this [thread|http://discussions.apple.com/thread.jspa?messageID=7012957]) who are running 10.5.2 and still have this issue. I don't know what if anything 10.5.2 did about this but I'm quite sure that it didn't change the group structure inherited from Tiger.
Message was edited by: V.K.

Two separate versions of the same file corrupted in different ways...

I was working on a Photoshop CS4 file last night that was created in CS4 and has only been opened on this machine in CS4. I saved the file and closed Photoshop. When I open the file today, all my layers are gone, flattened into one Background layer, and the image looks like colored TV static. On top of that, the separate, older version of this file with a different name (created earlier in the week on the same machine in the same CS4) will not open at all - I get the error message, "Could not complete your request because the file is empty" and the file size has mysteriously shrunk to 76 KB from 20+ MB. What happened?! I'm guessing these files are corrupted beyond recovery, but why and how did this happen? Has anyone had this or a similar problem? I want to know how to prevent this from happening in the future. Please help! Thank you. =)

Thanks for the suggestion, Chris.
Same issue, new question - don't ask me why I did this, but while both Photoshop and the now corrupted file were open, I changed the name of the folder containing the file, then continued to save the open version of the file without first closing it and reopening it from the newly renamed folder. Could this have caused both files in that folder to become corrupted in different ways? The other file that gives me the "file is empty" error was not open at the time, but it was in that folder. I'm thinking maybe by changing the path to the files while Photoshop was running, that could have triggered this. ?? I have 3 1TB internal drives raided to read as 1 drive, so I'm really hoping that one of these disks is not dying on me as you suggested. I was also backing up an external drive using BackBlaze when this all occurred - any possible connection? How can I find out if one of my three raided drives (and which one) might be going bad?
Thanks!!

Oracle 11: Register XSDs in different ways

Similar Messages

Maybe you are looking for