Wildcard search

Similar Questions

• Short: activate the wildcard search

  Hello world

  I have an Oracle Linux 6.6 3 components installed:

  -Oracle short Server 7.6.1

  -Oracle short discovered Information Integrator 3.1.1

  -Oracle short Information discovered Studio 3.1.0

  I'm sorry if this is too newbie...

  I want to activate the on an attribute wildcard search to locate text in the middle of words (for example, * in * would correspond to 'find').

  I found a page with a few tracks: http://docs.oracle.com/cd/E40518_01/server.761/es_dev/toc.htm#Configuring%20wildcard%20search%20in%20record%20search

  However, I don't know how to proceed. This page shows an example of use of a web service request (' soap: envelope "). You could point me please in the directions? Do I have to create a graph, adding components "Universal Data Reader" and "Web Service Client"?

  I'm sorry if the question seems vague, but I am a bit lost...

  Thank you

  Fausto

  Dear Fausto,

  You have found the necessary information and are very near doing this task. You are also right in the doc, while we provide information on the level of WEb Services, it not there no procedure end-to-end in the Integrator user's Guide. The reason for this is because this procedure relies on the use of the standard Integrator component (not a component EID created). The component that you can put in the Integrator to send in the XML graph is called Web Services Client.

  Below, I describe the steps.

  Context:

  To activate the search by wildcard characters, you must set the schema property for the configuration of indexing to true. It's the mdex-property_TextSearchAllowsWildcards. Must be true for the wildcard search is activated.

  This property is one of several properties on the GCR (record of the Global Configuration), which is used by the indexes of the Dgraph.

  To change this property, there are several ways: using SOAPUI, or graphic Integrator with the Web Services Client component. The two methods are described below.

  1. with the help of SOAP UI for the Web Service of Config for the loading of the DGC (Configuration global registry, which is an XML document)

  You can use any Web Service, such as SOAP UI user interface tool and that it points to the Dgraph running. In this document, you can access the Configuration Web Service with this URL:
  : /short-server/ws/config / ? wsdl (this is documented on the "On the Web Service Setup").

  Then, you can send the RCMS of together again to the Dgraph, using the putProperties of the Config Web Service operation. See this topic for info; It uses the same operation for the addition of configuration XML documents: load the schema of attribute

  Then, in the SOAP UI, you can issue a query similar to the following:

  <>

  "xmlns:config ="http://www.endeca.com/MDEX/config/services/types/3/0"

  xmlns:MDEX ="http://www.endeca.com/MDEX/config/XQuery/2009/09" >. "

  true

  2 using the Integrator and its graph based on XML, for loading XML documents. For info, see this topic and some subjects beneath him.

  The Integrator user's Guide > data domains configuration > Global Configuration record, you will see that this implies essentially to create a chart that uses a Web Services Client (it is a component of standard clover you can use), to load the same RCM document that you free of charge if you use SoapUI.

  For the WebServiceClient component, specify a string similar to the following in the name of the component operation field:

  {http://www.endeca.com/MDEX/config/services/config/3}Config#ConfigPort#DoConfigTransaction
  

  Notice that this string includes a namespace target for the Web service that faithfully reflects the major Web service as version / 3 in this example (you must have this web service as version 3.0 too, but please check what version of the WS Config you use.) In the Installation Guide for server short 7.6.1 you can see how to check it: list of Web Services of short and their versions

  I hope this helps!

  Julia

• Wildcard search query ADF

  Hi Experts,
  I have a requirement in the ADF search panel should allow wildcard search.
  I would like to know is there OOTB or how can achieve us?
  appreciate your help
  
  Thank you
  KT

  A criterion to display that contains all the attributes of the caisse and you set the search as 'start with' criteria which generate search by wildcard characters. Set each attribute "ignore null values" allow research where the attribute that is not defined.
  Then you use this criterion of the view and drag it onto the page and drop it on the page as a query.

  Timo

• How to do a wildcard search in El Capitan

  Why Apple did not come to the top with the possibility to search and sort the files with a wildcard character. My Sony videos each contain an .xml file that I didn't need. I want to compile a list of these research files and sorting using a wildcard. i.e. *.xml. I am amazed that computers Apple do not have a generic search capability.

  Try this link (list of all my files)

  https://support.Apple.com/en-us/HT201732

• Default behavior of short wildcard search

  
  Hi friends,

  Is there a way to control the short wild-card as default search behavior when we search and if we "searchterm *", we get results for ' * searchterm * "but it is possible to restrict the result set so that it contains only" searchterm * ", we understand that it can be filtered at the code level, but we are looking for a way to do it in short itself.

  Inputs/suggestions are appreciated.

  Kind regards

  Shreyas

  Sarah,

  I'm a bit confused by your first sentence and would like to clarify.

  When you say a search on ' term * ' returns ' * term * "-you don't want to say something like ABCtermDEF being returned but rather ABC termDEF.»»

  Without the wildcard character in front of 'term', short does not return all the words that begin with anything else of "term".  Thus, it is the case, then you could check the reqlogs to see that something else is not inject the leading asterisk.

  So, if your use case so more like "I want to find all the records that have, as their first word, a word that begins with"term", then it is not supported without a little manipulator gyration.  In this case, I create a separate property called say 'AllFirstWords' and have a manipulator put all the first words in this property.  Then your search for these can go against that property.

  For example, if the original property was 'Draped Velvet' and you want to find "Vel * ' but not ' Dra *", and then with your new property which has just "Velvet" you are set.

• The wildcard search

  Blocks the im trying to a lookup field in a database of a block no DB and if I type in what is exactly the value of the field, nothing returns. I have a field that can hold up to 2000 characters and I want to be able to search by keyword...% did not work for me...

  You would still have the % in the database-user field no it is entered. You could even encapsulate field '%' when you set your v_where clause:

  v_where: = 'like % | : nondatabaseblock.field | '%';

  I did similar things to this, but you should keep in mind when you use AS you end up doing full table scans which can slow down the Application if the table is large enough.

  -mike

• How to restrict the wildcard character in the search form?

  Hello
  I am new to ADF and learning from Steve's blog. I was going to for example 158th in which he restricted the wildcard search. I could not
  understand where he defined the restriction of wildcard. I have checked the VO.xml Emp but found no clue. Can someone help me please?

  It is in test.model.CustomViewObjectImpl.java in the model project

• Search for wildcard characters

  Hello
  
  Can we use the wildcard search in Berkeley Java Edition. The joker that interests me is to search a text field that begin with the specific prefix. For example, in SQL query can find all text that start with the prefix "XML", by specifying the lookup key to be "XML %". Can it be done in Berkeley I if I use a Direct persistence layer?
  Thank you
  
  Samir

  Samir,

  If you search the forum for 'DPL query', you will find the wire this query is possible using DPL? which is the same kind of question. In particular, if you read the white paper recommended (http://www.oracle.com/technology/products/berkeley-db/pdf/performing%20queries%20in%20oracle%20berkeley%20db%20java%20edition.pdf) you can see an example of the query, that you ask the subject.

  Kind regards

  Linda

• How to disable 'Quick search' at the University Complutense of MADRID 10g for users in bulk

  Hi friends,

  We use the AAU 10 g and there are 1,000 users to use it. There are millions of records in the server and users tend to use the wildcard searches in the 'Quick search' option on the homepage.

  There is only one way, we can disable the quick search.

  (1) first, the layout SHOULD be classic. To make a layout by default for all users, the config.cfg goto file and edit it with entry below.

  DisableAmberLayouts = true - this will change all users to the classic layout.

  (2) goto user profile and click on "design portal for XXX (username).

  (3) uncheck the quick search option.

  -My question is, how to realize for all whithout users asking them to perform steps 2 and 3.

  Y at - it no service call, we can do?

  Looking forward to your response.

  Thank you

  Srikanth

  controlledQuery = 1 turns off quick search on 11g...

  try to add the URL parameter and see if it works in 10g more

• Wild search in the name column

  Hello
  
  IAM using oracle 11.2
  
  IAM having a query that finds a particular part of the word name
  for example
  
  CREATE TABLE TEST123
  (type1, varchar2 (10))
  LIST1 VARCHAR2 (10),
  NAME VARCHAR2 (302).
  NAME2 VARCHAR2 (500),
  ANAG_NAME VARCHAR2 (2000))
  
  
  
  SELECT *.
  OF test123
  WHERE ='s TYPE1' AND
  (LIST1 = "CL"
  AND (NAME2 = "AIILN" AND)
  (instr (NAME, 'IAN') > 0 AND instr (NAME, 'LI') > 0))
  OR LIST1 = 'TT '.
  AND (NAME2 = "AIILN" AND)
  (instr (ANAG_NAME, ' un ') > 0 AND instr (ANAG_NAME, "L") > 0))
  )
  
  the table can contain 2 million records.
  
  I want to replace it with the index ctxsys.
  because research is a wildcard search
  If use to create indexes ctxsys.context I have to use the CONTAINS clause to find records that I have to use %
  and if I create index ctxsys.ctxcat I use CATSEARCH to search for record, I used ' *'
  
  
  but both takes more time...
  However, an error occurs.
  
  Can someone give some suggestion...
  
  Thank you
  user
  
  Published by: user9080289 on June 22, 2012 12:10

  In general, you can create a CONTEXT index on a column, then use CONTAINS instead of the Instr. You can also create a MULTI_COLUMN_DATASTORE on several columns and use a sort of GROUP of SECTIONS, so that you can use a CONTAINS clause and with an index affected research in several columns. You can also use FILTER BY in your index creation, so that you can include columns structured in your index, so that you can use SURLABASEDESDONNEESDUFABRICANTDUBALLAST in your CONTAINS clause, so that your entire query uses one index. You can also use SUBSTRING_INDEX to increase lookup speed when you use wildcard characters. You may also need to increase the WILDCARD_MAXTERMS. If you are looking for a string that is separated by spaces, punctuation or some other delimiter, you don't need wildcards. However, if you search for a string in a Word, then you need wildcards. If you are looking for short strings, then you must be sure to only use a list of empty words that does not include them. If your channels are too short, then you are likely to exceed the WILDCARD_MAXTERMS, unless set you it to 0, while there is no limit, but then this type of query can be very slow and consume a lot of resources. Please see the demo below. I added a column to create the index on, but you can create index on a column of text. The columns that are searched are determined by the columns in the store of data and filtering, not the column in the index in created the.

  -test environment consisting of table, data and similar to the original query:

  SCOTT@orcl_11gR2> -- table:
  SCOTT@orcl_11gR2> CREATE TABLE TEST123
    2    (type1        varchar2(10),
    3       LIST1        VARCHAR2(10),
    4       NAME        VARCHAR2(302),
    5       NAME2        VARCHAR2(500),
    6       ANAG_NAME  VARCHAR2(2000),
    7       names        VARCHAR2(1))
    8  /
  
  Table created.
  
  SCOTT@orcl_11gR2> -- data:
  SCOTT@orcl_11gR2> INSERT ALL
    2  INTO test123 VALUES ('s', 'CL', 'IAN LI', 'AIILN', NULL, NULL)
    3  INTO test123 VALUES ('s', 'TT', NULL, 'AIILN', 'AN L', NULL)
    4  INTO test123 VALUES ('s', 'CL', 'IANLI', 'AIILN', NULL, NULL)
    5  INTO test123 VALUES ('s', 'TT', NULL, 'AIILN', 'ANL', NULL)
    6  INTO test123 VALUES ('s', 'CL', 'LIIAN', 'AIILN', NULL, NULL)
    7  INTO test123 VALUES ('s', 'TT', NULL, 'AIILN', 'LAN', NULL)
    8  SELECT * FROM DUAL
    9  /
  
  6 rows created.
  
  SCOTT@orcl_11gR2> -- query similar to original:
  SCOTT@orcl_11gR2> COLUMN name        FORMAT A10
  SCOTT@orcl_11gR2> COLUMN name2        FORMAT A10
  SCOTT@orcl_11gR2> COLUMN anag_name FORMAT A10
  SCOTT@orcl_11gR2> SELECT *
    2  FROM   test123
    3  WHERE  type1 ='s'
    4  AND    ((list1 = 'CL'
    5             AND name2 = 'AIILN'
    6             AND INSTR (name, 'IAN') > 0
    7             AND INSTR (name, 'LI') > 0)
    8            OR
    9            (list1 = 'TT'
   10             AND name2 = 'AIILN'
   11             AND INSTR (anag_name, 'AN') > 0
   12             AND INSTR (anag_name, 'L') > 0))
   13  /
  
  TYPE1      LIST1      NAME       NAME2      ANAG_NAME  N
  ---------- ---------- ---------- ---------- ---------- -
  s          CL         IAN LI     AIILN
  s          TT                    AIILN      AN L
  s          CL         IANLI      AIILN
  s          TT                    AIILN      ANL
  s          CL         LIIAN      AIILN
  s          TT                    AIILN      LAN
  
  6 rows selected.
  

  -index multi_column_datastore, list of words and text:

  SCOTT@orcl_11gR2> -- multi_coumn_datastore and wordlist:
  SCOTT@orcl_11gR2> BEGIN
    2    -- multi_column_datastore:
    3    CTX_DDL.CREATE_PREFERENCE ('test123_mds', 'MULTI_COLUMN_DATASTORE');
    4    CTX_DDL.SET_ATTRIBUTE ('test123_mds', 'COLUMNS', 'name, name2, anag_name');
    5    -- worlist:
    6    CTX_DDL.CREATE_PREFERENCE ('test123_wl', 'BASIC_WORDLIST');
    7    CTX_DDL.SET_ATTRIBUTE ('test123_wl', 'SUBSTRING_INDEX', 'TRUE');
    8    CTX_DDL.SET_ATTRIBUTE ('test123_wl', 'WILDCARD_MAXTERMS', 50000);
    9  END;
   10  /
  
  PL/SQL procedure successfully completed.
  
  SCOTT@orcl_11gR2> -- text index using FILTER BY, multi_column_datastore, section group, wordlist:
  SCOTT@orcl_11gR2> CREATE INDEX test123_idx
    2  ON test123 (names)
    3  INDEXTYPE IS CTXSYS.CONTEXT
    4  FILTER BY type1, list1
    5  PARAMETERS
    6    ('DATASTORE     test123_mds
    7        SECTION GROUP     CTXSYS.AUTO_SECTION_GROUP
    8        WORDLIST     test123_wl
    9        STOPLIST     CTXSYS.EMPTY_STOPLIST')
   10  /
  
  Index created.
  

  -query using CONTAINS and SURLABASEDESDONNEESDUFABRICANTDUBALLAST with a single hit index:

  SCOTT@orcl_11gR2> SET AUTOTRACE ON EXPLAIN
  SCOTT@orcl_11gR2> SELECT *
    2  FROM   test123
    3  WHERE  CONTAINS
    4             (names,
    5              'SDATA (type1 = ''s'')
    6            AND AIILN WITHIN name2
    7            AND ((SDATA (list1 = ''CL'')
    8                  AND (%IAN% AND %LI% WITHIN name))
    9                 OR
   10                 (SDATA (list1 = ''TT'')
   11                  AND (%AN% AND %L% WITHIN anag_name)))') > 0
   12  /
  
  TYPE1      LIST1      NAME       NAME2      ANAG_NAME  N
  ---------- ---------- ---------- ---------- ---------- -
  s          CL         IAN LI     AIILN
  s          TT                    AIILN      AN L
  s          CL         IANLI      AIILN
  s          TT                    AIILN      ANL
  s          CL         LIIAN      AIILN
  s          TT                    AIILN      LAN
  
  6 rows selected.
  
  Execution Plan
  ----------------------------------------------------------
  Plan hash value: 3044757981
  
  -------------------------------------------------------------------------------------------
  | Id  | Operation                   | Name        | Rows  | Bytes | Cost (%CPU)| Time     |
  -------------------------------------------------------------------------------------------
  |   0 | SELECT STATEMENT            |             |     1 |  1435 |     4   (0)| 00:00:01 |
  |   1 |  TABLE ACCESS BY INDEX ROWID| TEST123     |     1 |  1435 |     4   (0)| 00:00:01 |
  |*  2 |   DOMAIN INDEX              | TEST123_IDX |       |       |     4   (0)| 00:00:01 |
  -------------------------------------------------------------------------------------------
  
  Predicate Information (identified by operation id):
  ---------------------------------------------------
  
     2 - access("CTXSYS"."CONTAINS"("NAMES",'SDATA (type1 = ''s'')            AND
                AIILN WITHIN name2            AND ((SDATA (list1 = ''CL'')                  AND
                (%IAN% AND %LI% WITHIN name))                 OR                 (SDATA (list1 =
                ''TT'')                  AND (%AN% AND %L% WITHIN anag_name)))')>0)
  
  Note
  -----
     - dynamic sampling used for this statement (level=2)
  
  SCOTT@orcl_11gR2>
  

• Logical search strategies

  With the help of Oracle 9.2.0.8 on AIX, Oracle text. Front end .net
  
  I have a very demanding client, requiring precise results (high relevance). Want it provide a simple search for end users (ala Google) interface, but take advantage of more advanced in the back-end logic. We deal with only the French and English content... for most of indexing CLOB and varchar2s.
  
  The dilemma I face to, is the end user is not familiar with Oracle Advanced search syntax, but still want accurate and relevant results... "The advanced search screen offers 3 text boxes, 'All words', exact phrase" and "words". There is also a 'part of the word' checkbox, when enabled, tells stored proc to make a joker on each keyword match.
  
  Based on what text box is selected, the stored procedure injects the appropriate Boolean operators and encapsulates each term with a BLUR (70, weighted score) operator...
  
  The current implementation of research, according to them, sucks.
  
  I was responsible to improve the results of the research... so far I have set up the following:
  
  -conversion of base_letter
  -arising (index stems)
  -fuzzy matching enabled
  -skipjoins for hyphens (treat words as post-secondary education)
  -index_themes, prove_themes (on request)
  -index of the substring (for wildcard matching)
  
  I also have a user_datastore to combine the 3 columns (a varchar2 and two CLOB) that were previously indexed separately and controlled on the varchar2 and column scores a clob.
  
  I also replaced it with FUZZY query operator searches using 'a WAY' integrated knowledge base.
  
  With these changes, they find the results improved, but I think that you can make more.
  
  The stored procedure strips on special characters (used by Oracle Text in the control unit) in the query terms and injects its own operators and Boolean logic based on what text box was used to submit the query.
  
  Is there a better way to do it? There seems to be a delicate balance between maintaining the simple search interface, while still developing advanced search on the back-end functions.
  
  Some scenarios I played wiith:
  
  -Wrap each with a CONNECTION query term, GOLD ' ing or AND'ing of each term, depending on whether the user entered terms in 'words' or 'all words '. Problem here is WHAT ignores special characters... even a hyphen which is part of a Word, even if I escape by using "-" or "{"... base don feedback: query, words like "post-secondary" will be searched as:
  
  Search for ALL words, 1, 0, CRT ACCUMULATE, 1
  Search for all words, 2, 1 CRT, on, posting, 1
  CRT search for all words, 3, 1, SUBJECT, secondariness, 2
  
  How a handful of Oracle research? Asktom?
  
  Thanks for the tips and advice
  
  Stéphane

  Any chance of move you to a supported version, I guess?

  The best way to improve the relevance without losing callback is to use the Progressive Relaxation - see
  http://www.Oracle.com/technology/products/text/htdocs/prog_relax.html
  but unfortunately this requires 10g.

  Otherwise, you might try something like this:

  Application of single word for dog:

  (dog) * 4, subject (dog) * 3% * 2 dog,? dog * 1

  The ',' is the ACCUM operator and the "*" multiplies the score. So we say that a match exact higher scores, followed by a generic, followed by a fuzzy character followed by a theme.

  Several Word searches, looking for searches of words to match highest, followed by and, followed by or. So for a search for 'black dog '...

  (black dog) * 4 (black and dog) * 3, (about (black) and about (dog)) * 2.5, (black or dog) * 2, (% black and dog%)*1.5, (black or dog %) * 1... etc.)

  There are some combinations missing here - and it will get exponentially more complex with terms more, but you get the picture. In addition, you could be smart when dealing with sentences...

  "camping or hiking in Western Europe"-there are three groups of Word it: camping, hiking, and western europe. Looking for empty words will help you separate sentences, then you can search for 'western europe', which appears in the same place but worry not "camping hiking" or "hiking western."

  Moreover, I do not recommend using SUBSTRING_INDEX. It is useful when you perform wildcard searches, and it is rarely necessary to search for % black or dog %. SUBSTRING_INDEX will greatly increase the size and timing of the generation of your index.

  Published by: Roger Ford on February 12, 2010 09:28

• Generic without using % in the search field

  Hello
  
  I have a small database + app .asp, containing addresses to about 500 firms.
  Users are able to search for any company name address, street, etc.
  And using %, it is possible to wildcard search the DB. But I don't want the user to have to use the % character.
  Lets assume that the company is called "Big apples Inc.". Apples using %% the company would appear.
  But is it possible to search by wildcard characters automatically, which means that the company would show if fair "apples" are typed without the % sign. I searched the forum, but I can't really find anything on it.
  
  Best regards / / Pat!
  
  
  

  Try to use % as your default.

• Awesome Bar is not working properly

  The Awesome Bar used to be great and to suggest previous websites, we were that we typed. Now he's doing a wildcard search. You should know the exact address approaching what we are looking for. Even if a complete with the ".com" address or the end is successfully added, the awesome bar responds as if we just did a search and shows several suggestions of Web sites. We uninstalled and reinstalled firefox - no change. Help, please.

  Hello

  The reset Firefox feature can solve a lot of problems in restaurant Firefox to its factory default condition while saving your vital information.
  Note: This will make you lose all the Extensions, open Web sites and preferences.

  To reset Firefox, perform the following steps:

  1. Go to Firefox > help > troubleshooting information.
  2. Click on the button 'Reset Firefox'.
  3. Firefox will close and reset. After Firefox is finished, it will display a window with the imported information. Click Finish.
  4. Firefox opens with all the default settings applied.

  Information can be found in the article Firefox Refresh - reset the settings and Add-ons .

  This solve your problems? Please report to us!

  Thank you.

• Yoga 11 s - accidents

  Recently bought the 11s Yoga. After some (windows) updates and application installations, he started planting. Last week, it crashed regularly (less than 5 to 10 minutes walk). I found an answer that solves the problem on a wildcard search. It goes, once again, although not as often as before. I had not thought to copy the instructions down or to bookmark the page.

  The answer was a set of instructions on how to enter the system and have it perform a check to fix or reset the windows updates that may have been incomplete or corrupt. It worked like a charm, and I expect to find him. I tried to do research in the same way, but can't locate.

  Can anyone help? Thank you in advance.

  Hi Vic450,

  Welcome to the Community Forums of Lenovo!

  This could be the guide you're looking for

  • Open Control Panel.
  
  
  • In the search box, type troubleshooting, and then click Troubleshooting.
  
  
  • Under system and security, click Fix problems with Windows Update

  This launches a convenience store of WIndows 8, the analysis and the difficulty of the Windows updates.

  If you still have problems after that, post the question shows on how and how often.

  Hope this helps,

  See you soon!

• Find the text and apply the format paragraph tag in Framemaker 11

  Help!

  I am trying to automate the search text in a file FM11 and CE marking with a specific paragraph tag. Once it is, I want to be able to search for the same string again and delete it. The research will include wildcard characters. I'm moving Word files on and bring with manual numbering of paragraphs. Thus, when they arrive, they "2.3.4.5" before paragraphs for example. I want to search all of the numbers at the beginning of paragraphs, tag all those with the appropriate para format (starting a catalog) and then remove the manual numbering.

  At this time do us it manually, using find/replace and by using wildcard search strings to find (such as '\P[1-9]*.*.*.*[0-9]', which will find all the para numbering at the beginning of a paragraph 4-digit), then using the change option to paste to paste in a paragraph of the catalog tag. Then we repeat the same search and delete the text, leaving only the tag para to the number format. It works very well, but errors can occur if the search string in set incorrectly or if the numbers are not deleted (in this case, it is found on the next search and re-labeled incorrectly). For example

  I work for the Air Force collection of entries from the field and creating publications out of them, so there is no mobile profit here, just trying to be efficient. Any help is greatly appreciated!

  Thanks in advance

  b

  Hi Brian,.

  I guess right here because I've never done this before, but... In my view, there is definitely something wrong with your mission to find the property values. The way you wrote, you are trying to assign two properties to a single table space. That's why you need to remove these two lines to make him find work, because they are replacing the FindText property apparently. Here is my proposal for how it should be:

  findParams = AllocatePropVals (2);

  findParams [0].propIdent.num = Constants.FS_FindText;

  findParams [0].propVal.valType = Constants.FT_String;

  findParams [0].propVal.sval = searchString.

  findParams [1].propIdent.num = Constants.FS_FindCustomizationFlags;

  findParams [1].propVal.valType = Constants.FT_Integer;

  findParams [1].propVal.ival = Constants.FF_FIND_USE_WILDCARDS;

  In any case, let me point out again once it is a hypothesis and I did not test this. I am sure, however, that the solution is somewhere along this path.

  Russ