Multiselect begins with gives inconsistent fetch results

Hi All,
We have a Multiselect item in our Dashboard Tab Page. We are making the values coming from database as UPPER to avoid case sensitive search.
The problem now comes when the user keys in different partial values of same word, the results are different.
To illustrate,
When we search by OPI and option is "begins with", it fetches all the comapny names starting with OPI incluidng OPITZ SOLUTIONS which we are looking for.
But when we give OPITZ or OPIT, it does not fetch any results.
When I give OPITZ i.e. with space, again the results are fetched.
Is there any explanation for this behaviour?

Thanks to both of you for your inputs.
-> I checked the SQL generated after applying the filters - it is same for all columns. Moreover this is an issue that I see on the using the "begins with" option to restrict the number of available choices - so at that level I am not sure how the corresponding SQL can be tracked.
-> Also, there is no data level security implemented for any column/table in the repository.
I am still working to resolve this and would get back if there is any luck. Please do let me know if you have additional inputs on this issue.
Thanks,
Piyush

Similar Messages

'Begins with' not working in multiselect

Hi All,
Our project is using a number of multiselect dashboard prompts.
In version 10.1.3.3 we find there is a provision of using operators (match). However, the 'begins with' operator doesnot seem to be working.
I have checked this in multiple environments (Windows and Linux) but the problem is consistant everywhere.
Anyone else encountered this problem? Any way to byepass this?
Regards,

I understand and I don't understand... So you have a dashboard prompt set to multi-select, yes? You have changed "Show" to "SQL Results" and added your "active products" filter, yes? Now with the prompt and report in the dashboard, you click on the ellipsis of the prompt and for example, you change "Match" to "begins with," type "A" and click on "Go" which then displays all the values that begin with "A," yes?
Now you select those "A" products and click on the left arrow to bring them to the left pane, yes? Then you click "Ok" which closes the window and click on "OK" one more time to run the prompt, yes?
Now you are saying that the report filters "all the products that start with 'A' both active and inactive, yes? And this is what I don't understand. The report should be filtered by the specific "A" starting products you selected in the prompts, not "all 'A' starting values" regardless of whether they are active or inactive. If you have "Is Prompted" on the appropriate column, you should get only those values from the prompt, nothing more.
How are you getting "all values that start with 'A'"? Please explain.

ORA-20001:Error fetching column value:Query must begin with SELECT or WITH

Hi all,
I have implemented an application with a master-detail form using Apex 3.1.2 and I imported that to Apex 3.2 on different server. Everything else seems to be fine except when I add detail record (clicking on Add Rows), I am getting following error.
=================
report error:
ORA-20001: Error fetching column value: ORA-20001: Query must begin with SELECT or WITH
====================
Both the databases have same version 10gR2. I didn't make any change after importing it as well. Can someone let me know what could be the issue? Thanks in advance.
SHY

May we know your first name?
Can you put the application in your workspace on apex.oracle.com so we can take a look?
Scott

Spotlight search cannot begin with "a"

If I type a search in the Spotlight field which begins with "a" (or is simply that single letter), the progress icon (rotating lines) never stops spinning, and the mds process eats up all available physical memory, as well as writing gigabytes worth of bogus data to disk as virtual memory.
Pressing "esc" to clear the text entered causes the mds process to return to normal.
For example, if I type "activity" (looking for activity monitor) into the Spotlight search field, the lines will spin indefinitely and mds will do what I describe above. *However, typing "bactivity" and then using arrow keys to delete the starting 'b', leaving "activity" in the search field, gives a NORMAL search where the rotating lines disappear and correct results are given, with no mds problems.*
Words starting with any letter other than 'a' return normal search results and do not cause mds to spiral out of control.
What on earth might be causing this? (I have rebuilt the Spotlight index twice already, to no avail.)

Glad to know I'm not the only one!
Now I am only having the 'a' problem intermittently.
I tried doing two things... (bad science, I know, as now I'm not sure which one actually helped).
1.) I changed the permissions of the Applications folder (/Applications). For some reason my account didn't have read+write permissions on that folder. I added myself as having read/write permissions. Check using cmd-i on the folder in finder, at the bottom -- "sharing and permissions".
2.) I reindexed my Applications folder by adding it to the Spotlight exclude list, then removing it from that list (and then rebooted).

Hi - When I go into Contacts and type a letter in the search bar to get to last names that begin with that letter I get a list of random names that don't correlate to the letter. My settings are fine. Help ?

Hi. When I go into my contacts and type a letter in the saerch bar to get to a list of last names that begin with that letter I get a list of random names that don't begin or end with the letter. My Settings appear fine. Any insights would be greatly appreciated. Cheers Bill

The search bar is not designed to take you to names beginning with the letter you put in. If you want to do that, use the letters down the side. In the search bar, the device will return results for all contacts that have that have that letter in them. If you type a couple of letters of the name, you'll find what you want faster. Once you get used to how it works, it's quite efficient.
Best of luck.

How to declare internal table in BADI with variable name beginning with 0..

Gurus,
How to declare an internal table within a BADI. I have to implement a BADI UC_TASK_CUSTOM for BCS to BW load and there i have to declare an internal table like:
TYPES:          BEGIN OF t_cube_data,
                0cs_version       TYPE /bi0/oics_version,
                0sem_cgcomp       TYPE /bi0/oisem_cgcomp,
                bcs_vers          TYPE /bic/oibcs_vers,
                bcs_lcus          TYPE /bic/oibcs_lcus,
                bcs_ldch          TYPE /bic/oibcs_ldch,
                bcs_invcom       TYPE /bi0/oibcs_invcom,
                bcs_litem         TYPE /bic/oibcs_litem,
                bcs_llob          TYPE /bic/oibcs_llob,
                bcs_lmay          TYPE /bic/oibcs_lmay,
                0move_type        TYPE /bi0/oimove_type,
                pcompany         TYPE /bi0/oipcompany,
                bcs_lprg          TYPE /bic/oibcs_lprg,
                figlxref3         TYPE /bic/oifiglxref3,
                fiscyear         TYPE /bi0/oifiscyear,
                fiscper3         TYPE /bi0/oifiscper3,
                fiscvarnt        TYPE /bi0/oifiscvarnt,
                curkey_gc        TYPE /bi0/oicurkey_gc,
                unit             TYPE /bi0/oiunit,
                cs_trn_gc        TYPE /bi0/oics_trn_gc,
                cs_trn_qty       TYPE /bi0/oics_trn_qty,
          END OF t_cube_data
But with this declaration it gives a error saying that u cannot have a variable beginning with 0...like 0cs_version....
but i have to do it for my functionality to wrk...
Please help me do it....
how can i declare a internal table that allows me to have variable names that start with 0....
Please help....Its URGENT....
Thanks
Sam

Murali,
I need to have 0 before the variable name in the declaration of the internal table....how can i attain that....
Please suggest...
Thanks
Sam

What is a file that begins with ._ ?

A client just sent me some files that begin with ._ in the extension.
(Client claimed they are vector files for a video project)
I tried to download and Mac told me they are system files and will be hidden.
sure enough they didn't show up my desktop. I download again adding a " to the front of the extension. I got the file on my desktop- but it would not open.
I trashed it.
I then did a spotlight system search for the ._ files I think I download and I got no result.
I'm now worried.
Are these ._ somewhere on my system?
Can they do harm to the system?

In Unix the initial period in a filename causes the file to become invisible. Spotlight's default configuration will not locate invisible files. If the files are downloaded to your Downloads folder or to your Desktop then you can make them visible in a number of different ways.
1. Enable Finder to Show Invisible Files and Folders
Open the Terminal application in your Utilities folder. At the prompt enter or paste the following command line then press RETURN.
defaults write com.apple.finder AppleShowAllFiles TRUE
To turn off the display of invisible files and folders enter or paste the following command line and press RETURN.
defaults write com.apple.finder AppleShowAllFiles FALSE
Alternatively you can use one of the numerous third-party utilities such as TinkerTool or ShowHideInvisibleFiles - VersionTracker or MacUpdate.
2. Open the Terminal application in your Utilities folder. At the prompt enter the following commands pressing RETURN after each. I will assume for the purpose of illustration that you have downloaded the files to your Desktop.
cd /Users/yourusername/Desktop
mv ._filename.ext filename.ext
This will rename each file to the same filename but without the initial "._". Repeat the "mv" command for each file substituting the actual filename.ext for the "filename.ext" in the command line. If you have downloaded the files to a different location, then substitute the full path to that location in the "cd" command line.
3. Download a utility such as Batch Rename'Em from CNet Downloads or MacUpdate and use it to mass rename the files by removing the initial "._".
4. Have your client rename the files before sending them to you by explaining that they are invisible on the Mac with the initial period in the filename.
5. Configure Spotlight to search in system folders and to display invisible files in search results.
6. Download Easy Find - CNet Downloads or MacUpdate - and use it instead of Spotlight.
Message was edited by: Kappy

Error: File does not begin with '

After changing two javascript "else if" statements to "if" in an Adobe LiveCycle v8 form, I received errors upon saving. Now the file seems to be corrupt because when I attempt to open it I get an error stating "File does not begin with '". Does anyone know what could be causing this AND MORE IMPORTANTLY if there is a way to salvage my form? Please help!

Hi VetSrini,
Thanks for the update. I was able to figure out what was wrong.
Now there is another problem. When i try "View Output", the excel file is shown and the following options are also shown:
1. As an XML List
2. As a read-only workbook
3. Use the XML Source Task Pane
When I try the first option, I get the following error:
An Invalid Character was found in text content
When I try the third option, I get the following error:
Cannot load the specified XML or schema source.
when i press the details button, it gives me the following detailed error:
An Invalid Character was found in text content
When I try the second option, i get the same error, but with some xml tags
Can anyone please help me and tell me why i am getting these errors.
Thanks in Advance,
Arun

Renaming startup disk to begin with a dot

I realized that 10.6 prevents you from renaming a startup disk to begin with a dot (a period), so that you can hide the volume on the desktop/Finder. I've tried the diskutil rename command in terminal, but it gives me an error saying some like "not suitable name for drive type." It worked in 10.5, but not Snow Leopard. Well, for me at least. Anyone else have this problem? Thanks.
P.S. Not sure exactly where I should have posted this. I guess it's related to Finder.

DO NOT start the name of the drive with a dot. if you want to make it hidden from finder (not sure why you'd want to do it with the startup drive) you can add a hidden flag to the drive.
run this in terminal
sudo chflags hidden /
and it will hide the startup drive from finder. to unhide it change hidden to nohidden in the command. and of course you can just go to finder preferences->general and uncheck the box to show internal drives on the desktop. that will remove all internal drives from teh desktop.

Same query takes different time to fetch result from Database

Hi all,
I am having a scenario in which a query is taking different time keeping other environmental variables constant.
When that query runs for 1 user it takes just 2 minutes to fetch result and the DB connection becomes inactive after fetching the result.
But if I run the same query after some time in similar environment it takes 25 minutes. sometimes 40 minutes to execute and give the result to the app server from the same database.
I am not able to understand this behavior from DB. Can anybody try to explain this behavior?
The details of the DB are,
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
PL/SQL Release 11.2.0.1.0 - Production
CORE 11.2.0.1.0 Production
TNS for 32-bit Windows: Version 11.2.0.1.0 - Production
NLSRTL Version 11.2.0.1.0 - Production
OS Details:
Windows 2008 server Enterprise Edition.
I tried analyzing that query in oracle, it recommended there are lots of hard parsing in that query.
Regards,

user10915512 wrote:
Hi all,
I am having a scenario in which a query is taking different time keeping other environmental variables constant.
When that query runs for 1 user it takes just 2 minutes to fetch result and the DB connection becomes inactive after fetching the result.
But if I run the same query after some time in similar environmentBut not exactly the same environment. So what is different that it is only "similar"?
it takes 25 minutes. sometimes 40 minutes to execute and give the result to the app server from the same database.
I am not able to understand this behavior from DB. Can anybody try to explain this behavior?Run a statspack on the 'well behaved' query and on the 'not well behaved' query. Compare and contrast the results.
>
The details of the DB are,
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
PL/SQL Release 11.2.0.1.0 - Production
CORE 11.2.0.1.0 Production
TNS for 32-bit Windows: Version 11.2.0.1.0 - Production
NLSRTL Version 11.2.0.1.0 - Production
OS Details:
Windows 2008 server Enterprise Edition. To paraphrase Forest Gump, "My momma always said Windows was like a box of chocolates ...."
>
>
I tried analyzing that query in oracle, it recommended there are lots of hard parsing in that query.
Regards,

Query not fetching results and shows executing

All,
When I run a query from SQL Plus I am getting query results immediately table with more than 50,000 rows), but when I run from TOAD or PL/SQL Developer am not getting any result and all I see is query executing no error message also.
Surprising part is I am able to see results if the table is less than 100 rows in SQL Plus, TOAD and PL/SQL Developer.
Please tell me is there anything I have to do in the database configuration, I am facing this issue due to database move to a new server.
Thanks in advance

Anand,
I don't see this as an issue with TOAD or PL/SQL Developer, the reason is using my TOAD or PL/SQL Developer I am able to connect to the copy of this database (DEV) and able to fetch results. Only in the database I migrated to the new server I am not able to fetch results from connecting through any application, the only place I am able to run and get results is SQL PLUS.
So I believe this is something which I am missing in my database configuration or in my new server.
Please help me
Thanks in Advance

CFUPDATE problem with field name beginning with a number

I have a form (actually several) that is gathering data for a survey. I tried to use cfupdate to update the data elements and got a string index out of range error. I tested again without the form fields that began with numbers and it worked just fine. I then tried to enclose the offending field names in () and []. Both times I got same error. Does anyone have a work around for this? The code and error are below.
Changing the field names is not really an option. This survey has almost 400 data points and most of the fields begin with the question number.
Thanks
DW
<cfupdate datasource="#application.fdp_datasource#" tablename="arra_survey1" dbtype="ODBC" formfields = "entry_no, 2_a_fy09, '2_a_fy10', '2_a_fy11', '2_a_fy12', '2_b_fy09', '2_b_fy10', '2_b_fy11', '2_b_fy12', '2_c_fy09', '2_c_fy10', '2_c_fy11', '2_c_fy12', '2_d_fy09', '2_d_fy10', '2_d_fy11', '2_d_fy12', '2_e_fy09', '2_e_fy10', '2_e_fy11', '2_e_fy12', '2_f_fy09', '2_f_fy10', '2_f_fy11', '2_f_fy12', '2_g_fy09', '2_g_fy10', '2_g_fy11', '2_g_fy12', '2_h_fy09', '2_h_fy10', '2_h_fy11', '2_h_fy12', '2_i_fy09', '2_i_fy10', '2_i_fy11', '2_i_fy12', '2_j_fy09', '2_j_fy10', '2_j_fy11', '2_j_fy12', '2_k_fy09', '2_k_fy10', '2_k_fy11', '2_k_fy12', '2_l_fy09', '2_l_fy10', '2_l_fy11', '2_l_fy12', central_office, dept_office, incl_other, 3_a_fy09, 3_a_fy10, 3_a_fy11, 3_a_fy12, 3_b_fy09, 3_b_fy10, 3_b_fy11, 3_b_fy12, 3_c_fy09, 3_c_fy10, 3_c_fy11, 3_c_fy12, 3_d_fy09, 3_d_fy10 3_d_fy11, 3_d_fy12, 3_e_fy09, 3_e_fy10, 3_e_fy11, 3_e_fy12, 3_f_fy09, 3_f_fy10, 3_f_fy11, 3_f_fy12, 3_g_fy09, 3_g_fy10, 3_g_fy11, 3_g_fy12, 3_h_fy09, 3_h_fy10, 3_h_fy11, 3_h_fy12, 3_i_fy09, 3_i_fy10, 3_i_fy11, 3_i_fy12, 3_j_fy09, 3_j_fy10, 3_j_fy11, 3_j_fy12, 3_k_fy09, 3_k_fy10, 3_k_fy11, 3_k_fy12, 3_l_fy09, 3_l_fy10, 3_l_fy11, 3_l_fy12">
String index out of range: -2
The error occurred in C:\Inetpub\wwwroot\directory\process_arra_suvey_2.cfm: line 12
12 : <cfupdate datasource="#application.fdp_datasource#" tablename="arra_survey1" dbtype="ODBC" formfields = "entry_no, 2_a_fy09, '2_a_fy10', '2_a_fy11', ...

I think you have run one of CFUPDATE's limitations. I am not 100% certain, but I do not think cfupdate provides a way to escape invalid column names like yours. In which case you may need to do a regular UPDATE within a cfquery instead... or rename your columns.
fields that began with numbers
Most databases discourage creating column names that start with a number (or other restricted characters). While many databases will allow you to escape invalid names, it is simpler all around to avoid them altogether. Otherwise, you may have to escape the names in every single query. The recommendations vary, but usually object names that begin with a letter, and contain only letters, numbers and underscores are considered safe.
This survey has almost 400 data points and most of the fields begin with the question number.
Four-hundred (400) is a lot of columns for a single table. The fact that they all them seem to contain the same type of information (ie response to a question) is an indication that data should probably be stored in a separate table, as rows, not columns. For example questions could be stored in one table, possible answers in another, and the results of an individual test/survey stored in a third table. A structure like that is much easier to query, can easily accommodate changes (add/remove questions, responses, etcetera) without having modify the table every time.

Change Data Source not showing tables (in the taget drop down) that begin with "SYSTEM"

I am trying to change the data source to a database table that is named someschema.SYSTEM_XXX_XXX but none of my tables that are named SYSTEM_....... are in the combo box list of the taget tables (Change Data Source) dialog. I have tried this on several database with the same results. Are you filtering out tables that begin with system ?

This issue has been resolved in the latest hot fix for CR4E. You can download the latest version of the hot fix by following these steps: <ol><li>Go under the Help menu</li><li>Select the option Software Updates -> Manage Configuration</li><li>Expand the tree until you see Crystal Reports for Eclipse 1.0.1.v671 (or similar)</li><li>Highlight this entry and select the Scan for Updates option in the window on the right.</li></ol> This will allow you to update to the latest version which is currently 1.0.2.v795 Please let me know if this fix does not resolve the issue outlined in this thread. Regards, Sean Johnson (CR4E Product Manager) <a href="http://www.eclipseplugincentral.com/Web_Links-index-req-ratelink-lid-639.html">Rate this plugin @ Eclipse Plugin Central</a>

NOTICE: Keys beginning with // are reserved for internal use

If you're currently developing an application using NoSQL DB, or designing your schema for a NoSQL DB application, please take notice.
In an upcoming release of NoSQL DB, there will be a new restriction on keys: no key may have an empty (zero length string) first component. An IllegalArgumentException will be thrown when attempting to use a key with a KVStore method, if the first component of the major path is empty.
Using the standard Key.toString format, this means that any key beginning with // will be prohibited and the empty key / will also be rejected. For example, the following keys will be prohibited:
//a
//-/a
//a/-/bThe reason for this restriction is to allow the NoSQL DB implementation to store internal information or metadata using key-value pairs. These internal key-value pairs will begin with an empty component and will not be accessible via the API. For example, these internal keys will not be returned by the KVStore.storeIterator method. This capability will be used internally in an upcoming release.
If your application currently uses keys with an empty first component, please reply to this post and we will suggest alternative approaches.
Thanks,
--mark

Greg,
Oracle does not follow the road for abstraction using views for it's E-Business suite.
In general it's better to use the tables. It's faster and in case a table changes or becomes obsolete a view is provided anyway (i.e. RA_CUSTOMERS used to be a table prior to 11i, now it's a view).
There are some exceptions (of course ;-):
1. Multi-Org tables (_ALL suffix).
If possible you should use the views if you run your reports from Oracle Applications. If you want to extract data for your datamart then the using the tables would be
your preferred option.
2. Multi-language tables (_TL suffix)
For tables that have a "_TL" peer table the view with the _VL suffix should be used.
If you are using Discoverer with 11i it may be good to have a look at the BIS end-user-layer that Oracle provides with 11i. This will give you the data abstraction you were seeking.
Regards,
Diederik

Secure Global Desktop and RDPBroker doesn't like passwords that begin with

That begin with a #
RDP kicks you out and SGD gives you a nice white X window for a short while.
Any suggestions besides removing the # key from all my keyboards?

I have raised a support call with Sun and they have been able to reproduce the problem!
It is my intension to post any resolution information here.

Multiselect begins with gives inconsistent fetch results

Similar Messages

Maybe you are looking for