Possible BUG: af:query with several LOVs throws bitIndex exception

Hi all

I use the af:query for the search component and I get constantly "bitIndex < 0:-1 error" when searching on a LOV (by typing the value in the part of the LOV not using inputtext does not popup LOV).

Installation program:

I use Jdev 11.1.2.1 with JDEVADF_11.1.2.1.0_GENERIC_110907.2314.6081.

Using the schema HR, I created the entity departments and a Department display object based on the entity.

I created 2 objects of basis for the LOVs SQL view (not from entities not):

1. view LOV departments - select * from departments
2 view of the LOV locations - select * from places

The two views of Lov were created using the wizard with all the default values.

In my opinion main departments (based on the entity), I created 2 LOVs for LocationId and for DepartmentId.

The LOV for DepartmentId returns 2 attributes, DepartmentId and DepartmentName both are mapped to the corresponding attributes in the main view of Department. (I did using the popup "Edit the list of values", in the section 'List return values').

Finally, I dragged the "all searchable attributes" from the control panel of data to a test page, creating a group of ADF query with table

The error:

I type a valid service id in the LOV DepartmentId within the research section and I get the "bitIndex < 0:-1 error.» No results are returned.

When the server running with - Djbo.debugoutput = console I see the following error messages:

CtrlAttrs < JUCtrlValueBinding > < setInputValue > [18415]: exception caching: oracle.jbo.AttrValException: Houston-27019: the get method for attribute "DepartmentName" in the PREMISESof VIEW_USAGE_internal_vci_def_values is not resolved.
< DCBindingContainer > < cacheException > [18416] * updated BindingContainer cache EXCEPTION: oracle.jbo.AttrValException
< DCBindingContainer > < cacheException > [18417] java.lang.IndexOutOfBoundsException: bitIndex < 0:-1
at java.util.BitSet.get(BitSet.java:441)
at oracle.jbo.server.ViewRowStorage.getAttributeInternal(ViewRowStorage.java:1823)
at oracle.jbo.server.ViewRowImpl.getAttributeValue(ViewRowImpl.java:1962)
at oracle.jbo.server.ViewRowImpl.getAttributeInternal(ViewRowImpl.java:824)
at oracle.adf.model.bean.DCDataRow.getAttributeInternal(DCDataRow.java:366)
at oracle.adf.model.bean.DCCriteriaValueRowImpl.getAttributeInternal(DCCriteriaValueRowImpl.java:241)
at oracle.jbo.server.ViewRowImpl.getAttrInvokeAccessor(ViewRowImpl.java:906)
at oracle.jbo.server.ViewRowImpl.getAttribute(ViewRowImpl.java:854)
at oracle.jbo.uicli.binding.JUCtrlValueBinding.internalGetAttributeValueFromRow(JUCtrlValueBinding.java:1213)
at oracle.jbo.uicli.binding.JUCtrlListBinding.setTargetAttrsFromLovRow(JUCtrlListBinding.java:2799)
at oracle.jbo.uicli.binding.JUCtrlListBinding.setTargetAttrsFromLovRowAndUpdateMRU(JUCtrlListBinding.java:2712)
at oracle.adfinternal.view.faces.model.binding.FacesCtrlLOVBinding.setInputValueInRow(FacesCtrlLOVBinding.java:1163)
at oracle.jbo.uicli.binding.JUCtrlValueBinding.setInputValue(JUCtrlValueBinding.java:2926)
at oracle.jbo.uicli.binding.JUCtrlValueBinding.setInputValue(JUCtrlValueBinding.java:2889)
to oracle.adfinternal.view.faces.model.binding.FacesCtrlSearchBinding$ AdfCriterionValues.set (FacesCtrlSearchBinding.java:3589)
at javax.el.ListELResolver.setValue(ListELResolver.java:240)
at oracle.adfinternal.view.faces.model.AdfELResolver.setValue(AdfELResolver.java:162)
at oracle.adfinternal.view.faces.model.AdfELResolverProxy.setValue(AdfELResolverProxy.java:71)
at com.sun.faces.el.DemuxCompositeELResolver._setValue(DemuxCompositeELResolver.java:255)


Note:
The search works as expected if I only have a LOV attached to the main view. (if I don't have the Department LOV attached as described above everything works fine)
If both LOVs of DepartmentLOV and LocationsLOV return a single value, the departmnetId and the locationId, without mapping any additional return attributes research as planned as well.

This is test case the simplest I could create to reproduce the error I see in our code.

Here is a link to my test project: [http://www.2shared.com/file/QxZJDfax/bitIndexTest.html]

Help, please

I'm looking for workarounds / fixes.
Any advice would be greatly appreciated.

Thank you
Sasha

Hi Sasha,

which seems related to the bug 12851501 - ARRAYINDEXOUTOFBOUNDSEXCEPTION LIFTING OF VIEWROWSTORAGE CLASS
expected to be fixed in JDeveloper 11.1.2.2.0

If you have access to My Oracle Support, there are here available patch (patch 12851501).

Kind regards

Didier.

Tags: Java

Similar Questions

  • Possible bug in Anyconnect with smart card on linux

    Hello

    I got the authentication of smart card Anyconnect connecting Linux using NetId customers.

    My problem is that this only works at first starting a client anyconnect.

    I can do connect/reconnect whenever I want to, but if I left the anyconnect client and restart smart card authentication not working anymore.

    I've nailed sort of problem until beeing associated with the user profile for anyconnect beeing created (including the seams to read at the start of the customer).

    ~/. AnyConnect

    Still further the specific problem with the item seams

    If I delete this specific element of the profile or completely delete the profile, and then restart the client, smart card authentication will work.

    Newspapers anyconnect not sewing to shed light on the problem.

    Print written in the profile is always the same.

    Hope that is understandable and someone could give an explanation to this.

    Do not hesitate to ask if something is not clear or you would like more information.

    Best regards

    / Mattias

    Mattias,

    Please understand that these issues are all new to us.  We had not seen before two cases I know has opened in the same week as yours.  If you wish to pursue a fix, beyond the solution of effective workaround for this add-in, open that a TAC is necessary so that we can collect the details and file another bug.  Please be sure to include in your case, notes of the opening so that the TAC engineer who gets it may be noted that workaround.

    -Craig

  • Error exporting in the level of query with several conditions

    When I run the following query to export with the result.



    C:\Documents and Settings\ITL > exp scott/tiger@orcl tables = (emp) QUERY = "" WHERE deptno > 10 and sal! 2850 = "' LOG = log011.log FILE = exp.dmp"

    Export: Release 10.2.0.1.0 - Production on Wed Jan 30 10:01:27 2013

    Copyright (c) 1982, 2005, Oracle. All rights reserved.


    Connected to: Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - Production
    With partitioning, OLAP and Data Mining options
    Export performed WE8MSWIN1252 and AL16UTF16 NCHAR character set

    About to export specified tables by conventional means...
    . . export the rows in the table EMP 10 exported
    EXP-00091: exporting more than questionable statistics.
    EXP-00091: exporting more than questionable statistics.
    Export completed successfully with warnings.



    But when I run it with the following condition, then it displays the following error...




    C:\Documents and Settings\ITL > exp scott/tiger@orcl tables = (emp) QUERY ='"WHERE deptno > 10 and work!" = 'CLERK' ' ' LOG = log011.log FILE = Exp01.dmp.
    LRM-00111: no closing quote for value 'LOG = log01'

    EXP-00019: failure of the treatment of parameters, type 'HELP EXP = Y' help
    EXP-00000: export completed unsuccessfully


    C:\Documents and Settings\ITL > exp scott/tiger@orcl tables = (emp) QUERY = "WHERE deptno > 10 and work! = 'CLERK' ' LOG = log5.log FILE = exp01.dmp
    LRM-00112: multiple values not allowed for the parameter "query".

    EXP-00019: failure of the treatment of parameters, type 'HELP EXP = Y' help
    EXP-00000: export completed unsuccessfully


    Please suggest a solution for this.

    Hello

    Single quotes used for CLERKS are very probably the origin of the problem.
    C:\Documents and Settings\ITL > exp scott/tiger@orcl tables = (emp) QUERY ='"WHERE deptno 10 > and job!" = 'CLERK' ' ' LOG = log011.log FILE = Exp01.dmp.

    try to replace this
    C:\Documents and Settings\ITL > exp scott/tiger@orcl tables = (emp) QUERY ='"WHERE deptno 10 > and job!" = "REGISTRAR" "' LOG = log011.log FILE = Exp01.dmp.

    Two single quotes (no quotes) to place an apostrophe for CLERK.

    Thank you
    Padma...

  • Is it possible to use Lightroom with several computers that are not networked?

    I'm new to Lightroom, even though I am a professional photographer for over 15 years. Currently, I use the software that came with my camera to process my RAW files, Professional Digital Canon's Pro for a 5 d Mark II.

    I've been evaluating the lasest version of Lightroom to consider buying it. So far, I like the capacity expanded to setting the image and improvements. However, in my current workflow, I need to be able to save the settings to be able to transfer the files to another computer. Is this possible with Lightroom? With Digital Pro, I can edit a RAW file and it will save the adjustments any computer, I opened the image file on. I have to be able to replicate this process in order to convey to our production department how I want an adjusted and cropped image.

    I downloaded the version track on two computers, then the images imported into Lightroom. Adjustments made, then copied the images to a flash drive. Images downloaded from the flash drive to my computer laptop, then re-imported the images in Lightroom and adjustments were not saved.

    Thanks for any help.

    After some research, my answer lies in the XMP file. Whenever I make changes to the image, I can save my settings by creating an XMP (CTRL S) file. He rewrote it in the original image folder so when I burn the file by hand to my production team, the xmp files parleyed my settings of the image. In turn, they can make adjustments and send them to me, if necessary. Just took some time to find the information, because even if I am kicking of tires on LR4, it was a video LR1 who gave me the information.

  • Possible bug in LabVIEW with help chart

    The attached VI reads a group of bytes as U8 from a file and writes the each other in a new file.  It does this by using the function array decimate.  It seems that there is a bug in LabVIEW.  If the '+ 0' is deleted from the code nothing but 0 is written to the file.  If you explore the table that will be written in the file it shows no null values, but the resulting file is always all zeros.  Addition of '+ 0' was a work around the issue.

    I'm under LabVIEW version 8.5.

    This looks similar to the LabVIEW 8.5 suspiciosly bug, we discussed quite earlier.

    Try placing a primitive 'always copy' (from the palette ' application control... control memory') until the file write and see if things improve. Good luck!

  • Hierarchical query with several parents and children without parents

    Hi all,

    I have a problem of follow-up:

    1, parent_1,.

    2, child_1, 1

    3, child_2, 1

    4, grandchild_1, 2

    5, parent_2,.

    6, child_3, 5

    7, child_0,.

    In this case, I can have multiple parents, and I also have children who do not have parents. Can someone help me solve this problem

    Concerning

    Hello

    So, if a node has itself or 0 as its parent_id, it's the same as with the NULL value?

    In this case, include all these possibilities in the START WITH clause:

    SELECT DLI

    , LPAD (' ' ')

    , (LEVEL - 1) * 3

    , ' '

    ) || descr AS descr

    parent_id

    T

    START WITH NVL (parent_id, 0) (0, DLI)

    CONNECT BY parent_id = PRIOR DLI

    AND idd <> parent_id

    ;

  • possible bug: ListCollectionView.addAll with filter function

    Hello

    I wrote this post in the "General Discussion" Forum "LFex sdk", I also put it there.

    I think I found a bug in the flex SDK. I checked in JIRA and it seems it is not reported.
    This simple code raise an 'out of bouds exception' (flex SDK 4.1):

    var list: collection ArrayCollection = new ArrayCollection();

    list.filterFunction = filter;

    List.Refresh ();

    var items: ArrayList = new ArrayList ([1,2,3,4,5]);

    list.addAll (items);

    Function filter(obj:Object): Boolean {}

    var n: Number is obj in numbers;.

    return n < 2;

    }

    I checked the code in mx.collections.ListCollectionView and it's obvious: addAll called addAllAt. "addAllAt" launches a call loop on the list to add and calling the addItemAt for each element, but it increments the index arguments, even if the previous item has not really added (because of a restriction of the filter), and the "out of bounds exception occurs. In my opinion, addAllAt must be changed as as follows:

    current version:

    public function addAllAt (addList:IList, index: int):void {}

    var length: int = addList.length;

    for (var i: int = 0; i < length; i ++)

    {

    this.addItemAt (addList.getItemAt (i), I + index);

    }

    }

    proposed version:

    public function addAllAt (addList:IList, index: int):void {}

    var length: int = addList.length;

    for (var i: int = 0; i < length; i ++)

    {

    var to: int = I + index;

    if(at>Length) {}

    a = length;

    }

    this.addItemAt (addList.getItemAt (i), to);

    }

    }

    Should I open a bug? Or am I missing something?

    M1kal

    Fill out a bug report.

  • Are you aware of a possible bug in the version 3.6.23? In the browser, for example, "two /" appears with the slash through the 'o'. Is this a known bug? Thank you

    Are you aware of a possible bug in the version 3.6.23? In the browser, for example, "two /" appears with the slash through the 'o'. Is this a known bug?

    When the same text is seen in another browser, it is displayed correctly as "two."

    The same question is displayed on another computer.

    Thank you.

    Sounds that you use a font that supports ligatures (Palatino?).

  • Possible bug: save the table with double and extended precision to the worksheet

    If one concatenates an array of double-precision and an array of precision extended with the 'build' vi table, then recorded using 'Write in a spreadsheet file' vi any digits to the right of the decimal are reset to zero in the saved file. Regardless of the entry of signifier of format (for example %.10f) to the vi 'Write in a spreadsheet file'.

    I'm on Vista Ultimate 32 bit and labview 9.0

    This is a possible bug that is easily circumvented by the conversion of a type before you incorporate arrar in a worksheet. Nevertheless, it's a bug and it cost me some time.

    Hi JL,.

    No, this is not a bug - it's a feature

    Well, if you'd look closer you would recognize the 'save to spreadsheet' as polymorphic VI. As this polymorphic VI does not support the EXTENSION numbers internally (it only supports DBL, I64, and String) LabVIEW selects the instance with more precision: I64 (I64 a 64 bits of precision, DBL that 53...). Your options are:

    -the value of the instance to use as the DBL (by right click and "Select type... »)

    -make a copy of this VI, save it under a different name and make support number of POST (not rework the polymorphic VI like you would break compatibility with other facilities of LV or future revisions)

  • Is it possible to 'Load Up' the cursor with several elements of 'Place '?

    It happens quite often that I have a folder with several items that I want to place. I can go back individually and D command to the folder window to choose the order of the day and then place in the document. When there is page after page of items, it would be useful to be able to select multiple items that will go on a single page and then "spit their" so to speak, of the cursor by repeated clicks, machine gun style. Is this possible to do?


    TIA,

    Ken

    Altogether. Select all. If they are not in the right order when you go to the place, use the arrows to scroll and the ESC key to cancel an object.

  • Possible bug in implementing caching query ColdFusion 10

    I have stumbled upon a bug in caching of queries in ColdFusion 10.  I wanted to see if anyone else had seen this before as I've submitted a bug report.  It will be difficult to set up a full test case, so I hope that a description will be enough to confirm if this is reproducible.

    First, I create an array of structures.  Each structure is a set of key/value pairs of the arguments that will be passed to a query.  The first two keys contain numeric values, the third a string value.  I asked a Microsoft SQL Server database.  My query is inside a loop of table.  The query contains the CachedWithin attribute to 1 hour. In the query, I pass values in the structure in the clause WHERE using CFQueryParam.

    If two sets of numeric values are quite similar, but not identical, ColdFusion will jumble the values and hidden bad back the query result set.  For example, values in my table:

    [{BookNo = 97, NUMPAGE = 1, match = 'Card of the evaluator'}, {BookNo = 212, NUMPAGE = 50, match = 'Card of the evaluator'}, {BookNo = 250, NUMPAGE = 12, match = 'Card of the evaluator'}, {BookNo = 97, NUMPAGE = 1, match = 'Card of the evaluator'}]

    When I run the page for the first time with an empty query cache, this is what I get:

    1. Search results for BookNo 97, NUMPAGE 1. (non-cached)
    2. Search results for BookNo 212, NUMPAGE 50. (non-cached)
    3. Search results for BookNo 212, NUMPAGE 50. (cached)
    4. Search results for BookNo 97, NUMPAGE 1. (cached)

    What should I get:

    1. Search results for BookNo 97, NUMPAGE 1. (non-cached)
    2. Search results for BookNo 212, NUMPAGE 50. (non-cached)
    3. Search results for the BookNo 250, NUMPAGE 12. (non-cached)
    4. Search results for BookNo 97, NUMPAGE 1. (cached)

    See the problem with the third iteration, it's pulling the same results of the second iteration.  If I remove the CachedWithin, achieve the desired results.  If I create my own CacheID, I get the expected results.  It's only when I let ColdFusion handle the cache on its own I get incorrect results.

    Someone at - he seen anthing like that?  This look like a bug legitimate which must be filed?

    -Carl V.

    @BKBK,

    It's too complex for me to say. I have these questions, for example:

    (1) when the loop is processed, ColdFusion treats 4 queries consecutively?

    (2) a comparison of SQL statements is one of the controls that ColdFusion is running to decide whether to perform a given query or use cached data. Data using cfqueryparam binding has an effect on the SQL statement?

    (3) place a cfdump of the request as the last statement in the loop. A query dump tells you if a query is cached. You see what queries are cached?

    1. Yes.
    2. Oooh!  I've been using cfqueryparam.  I removed the cfqueryparams, and now he behaves as expected.  Seems that there is a bug when combining cfqueryparam and caching.
    3. Already had this and it was the results cached for the second query processing allegedly the third query (see the original post for the expected and actual results).

    -Carl V.

    Update: and it seems that it is a known bug, after all, with the layout of the query cache and cfqueryparam: https://bugbase.adobe.com/index.cfm?event=bug&id=3369530.  I'll add my vote, and perhaps others can also.

    Update 2: and looking more closely, it looks like it has been fixed in an everything-to-be-released version.  I hope that we will get it soon.  A suggested workaround solution was to place the values of cfqueryparam in a SQL comment, which seems to help CF recognize the single request instructions and cache correctly.

    Post edited by: Carl Von Stetten

  • Slide with several digital poster Down filter event Bug button?

    A key event filter on a slide with several digital screens down does not work as expected when not waived. If you for example tab of this slide with two digital screens digital display, the cursor remains on his position, rather than get KeyFocus to the next control. Same behavior at fault if you want to move the cursor on the left/right keyboard, nothing happens. Anyone an idea why?

    A simulation of that VI is attached. All other controls, a Slide with a digital display or a standard digital control, the key Down event filter behaves as expected... I tested in versions 8.5.1 LV (attachment), LV2012SP1 and LV2013.

    A teammate now found a solution or workaround. We do not know if this is the way it should be, if yes, then it's a bit hidden, but it works!

    It seems that the cursor needs to know on what digital display the FocusObj is when you use the event notifier KeyDown. Then, when to get the reference of the digital display active and the wiring of this reference to the terminal FocusObj filter event everything works as expected, numbers of tabs, typed, cursor etc and have to filter and notify events structure to the event.

    A vi and a printscreen are attached.

    The question is probably due to the specialty of slides and the class NumericWithScale in general, as its digital displays are a little under References and during the implementation of the features of event KeyDown for the class NumericWithScale filter, LV simply does not know on which side of the main action control part must be executed.

    Or someone has a better solution and explanation?

  • Is it possible to update a query with another query?

    I'm trying to update a query with another query (see the attachment of the code). Here is my setup: I have a table in an Access database in which I enter a string in a form and an update. This string is a single record in another table in the same data source. The first table has one record to provide the second, which has many and will have more. Basically what I was asking is: is this a valid thing to do in coldfusion? If this isn't the case, please help with a method to alter. I'm still new to coldfusion.

    The overall effect, I'm going to do is to display a folder as a profile of truck recommended on the website: www.truckerstoystore.net. Currently, I get an error when I try to view the page with the current configuration of the query.

    Check out this page to see the error: www.truckerstoystore.net/currentTOW2.cfm

    Assistance in this matter is greatly appreciated.
    ------------------------------------------------------------------------------------------ -----------------------------------------------------------------------

    The owner of the column has a data type of character and thus the value must be in single quotes

    WHERE owner = "#Owner #

    But as your learning, I suggest using cfqueryparam

    WHERE owner =

    Ken

  • I'm stuck with several GB of music files on my phone and I can't get down them!

    The other day I tried to install an application on my phone, but I didn't have enough memory.  My phone is full of music via iTunes synchronisation dock. So I connected it to my computer to try to make some more room. But there is a problem.

    (1) Apple does not play nice with Palm. Synchronization is no longer permissible or possible. I can't use iTunes to manage/delete music.

    (2) in USB Mode, there is no way to access/view the library. It's just not there.

    I tried Palm support, and they confirmed that the USB Mode is not working. I guess that the library is located in a hidden folder or directory. They suggested that I downgrade my iTunes. Goal that is unacceptable to me because one) it forces me to break the license Apple, b) it is not taken in charge, and c) if there is still a remote chance it will mess up my iTunes library, I won't.

    I'm ready to get off at Palm headquarters and throw my phone to their building.

    Can someone tell me how to delete my library without hacking iTunes?

    I thought about it. Support, put this in your KB. Everyone who was sync via iTunes on a Mac is currently STUCK with several GB of music on their phone that they cannot manage or remove, which makes the Palm Pre a lot less useful than announced.

    USB mode is a waste of time for people on a Mac. These hidden files are never going to be visible, and the library cannot be removed. The solution: go through a PC. Anywhere, any PC will do. Connect your phone to the PC, go to the USB drive Mode and show hidden files, then delete them from there.

    I got an old laptop lying around at the bottom, and the problem was solved lickety split.

  • Classic report with column LOV

    I have simple classic report (Apex 4.2.4) with two columns. One is to main table and shows the ID; another column is LOV that contains detail table names, which must FK main table.

    Table structure:

    master_table (id NUMBER);

    detail_table (detail_name VARCHAR2 (50), master_id NUMBER);

    Source report looks like:

    Select

    ID,

    lov_column null

    of master_table;

    lov_column is defined as a query in function LOV and query looks like:

    Select detail_name from detail_table

    where master_id = #ID #.

    But the application does not recognize the values #ID #? Can I use it without the WHERE statement, but in this case in LOV, I got all the records, and that I want in a folder have values only for this ID.

    What can I do?

    Thanks in advance.

    Kiran I thank you for your care, I finally found the solution.

    Apex.Oracle.com cloud works with Apex5, it is not possible to obtain a classical column LOV relation, but it is possible to do this by using APEX_ITEM. SELECT_LIST_FROM_QUERY. And so I did in my application.

    in the statement I wrote:

    Select

    Mt.ID,

    apex_item.select_list_from_query (1, null, "select dt.detail_code d, dt.detail_code r detail_table dt where dt.master_id = ' |") Mt.ID) retail

    of master_table TM;

    And it works!

Maybe you are looking for