![]() | ![]() Products ![]() ![]() ![]() ![]() |
Home » Technical Support » DBISAM Technical Support » Support Forums » DBISAM General » View Thread |
Messages 1 to 8 of 8 total |
![]() |
Thu, Apr 27 2006 7:35 AM | Permanent Link |
"Santy Concepción" | Hi!
I'm using DBISAM 4.23 I want to filter a table as follows: text := 'a'; table1.filter := 'UPPER(NAME) LIKE UPPER(''%' + text + '%'')'; I want to show every records that contains "a", "A", "á", "Á", "à", etc... but it only shows the ones without accents ('a' and 'A'). Any help? Should I change the table 'Locale' or index type? Thanks! |
Thu, Apr 27 2006 9:25 AM | Permanent Link |
Dan Rootham | Santy,
<<I want to filter a table as follows: text := 'a'; table1.filter := 'UPPER(NAME) LIKE UPPER(''%' + text + '%'')'; I want to show every records that contains "a", "A", "á", "Á", "à", etc... but it only shows the ones without accents ('a' and 'A'). >> For you this may not be the best way to handle it, but I can tell you how we solved exactly this problem for a multilingual dictionary. The queries were being sent to a DBISAM database through mobile phone networks, so we didn't even have any guarantee that accented characters would arrive intact! We created an extra column on each dictionary-language table. This new column contained a version of the search column, with these changes: - truncated from 100 characters to 20 characters - forced to uppercase - all accents stripped off This new column was then indexed and used as the lookup for all searches. Obviously this means that you need to "preprocess" any table to create the new accent-free index. And it also means that in real time you must "strip accents" from any search term and uppercase it before the term is submitted as a parameter to a SQL query. We just wrote a language-aware Delphi unit to do this. Why language-aware? Because some languages are different: for German, we wanted to change a-umlaut to "ae" rather than to "a". But the extra work is worth it, believe me. We found that this solution worked perfectly for many West European languages. We could do a round-trip translation from cellphone to DBISAM and back to the user's cellphone in around 5 seconds - even with the accent-stripping function included before the SQL query. HTH, Dan Lexicon Software Ltd |
Thu, Apr 27 2006 12:01 PM | Permanent Link |
Tim Young [Elevate Software] Elevate Software, Inc. ![]() | Santy,
<< I want to show every records that contains "a", "A", "á", "Á", "à", etc... but it only shows the ones without accents ('a' and 'A'). Any help? Should I change the table 'Locale' or index type? >> What is the current Locale for the table ? -- Tim Young Elevate Software www.elevatesoft.com |
Thu, Apr 27 2006 12:45 PM | Permanent Link |
"Santy Concepción" | > What is the current Locale for the table ?
"ANSI Standar" > > -- > Tim Young > Elevate Software > www.elevatesoft.com > > |
Thu, Apr 27 2006 12:47 PM | Permanent Link |
"Santy Concepción" | If I set the table Locale to "Spanish", then it works fine, BUT...
Will I have any problems changing the table Locale? Any problems migrating/converting from paradox? What are the advantages/disadvantages/differences of setting the Locale to Spanish or ANSI Standar? Thanks! "Tim Young [Elevate Software]" <timyoung@elevatesoft.com> escribió en el mensaje news:593EF802-BECD-4547-B2B6-499DD3983D9D@news.elevatesoft.com... > Santy, > > << I want to show every records that contains "a", "A", "á", "Á", "à", > etc... but it only shows the ones without accents ('a' and 'A'). > > Any help? > Should I change the table 'Locale' or index type? >> > > What is the current Locale for the table ? > > -- > Tim Young > Elevate Software > www.elevatesoft.com > > |
Fri, Apr 28 2006 3:57 PM | Permanent Link |
Tim Young [Elevate Software] Elevate Software, Inc. ![]() | Santy,
<< If I set the table Locale to "Spanish", then it works fine, BUT... Will I have any problems changing the table Locale? Any problems migrating/converting from paradox? >> Nope, the only difference is that the collation/sorting of indexes and ORDER BYs is done according to Spanish collation rules. << What are the advantages/disadvantages/differences of setting the Locale to Spanish or ANSI Standar? >> There are really no differences at all, other than the overhead of string comparison calls to the OS for Spanish collation/sorting. But, I doubt if you'll even notice the difference. -- Tim Young Elevate Software www.elevatesoft.com |
Sat, Apr 29 2006 6:34 AM | Permanent Link |
"Santy Concepción" | Hi, Tim...
Thanks for your answers. How can I get the Spanish 'LocaleID' form creating the tables? What about if one of my customers doesn't have the spanish LocaleID in his computer? Or how can I be sure that the "LocalID" number is always the same in every computers? Thanks! "Tim Young [Elevate Software]" <timyoung@elevatesoft.com> escribió en el mensaje news:E2C125D9-B960-4CEB-AC0A-1B95C6C850B1@news.elevatesoft.com... > Santy, > > << If I set the table Locale to "Spanish", then it works fine, BUT... > > Will I have any problems changing the table Locale? Any problems > migrating/converting from paradox? >> > > Nope, the only difference is that the collation/sorting of indexes and > ORDER BYs is done according to Spanish collation rules. > > << What are the advantages/disadvantages/differences of setting the Locale > to Spanish or ANSI Standar? >> > > There are really no differences at all, other than the overhead of string > comparison calls to the OS for Spanish collation/sorting. But, I doubt if > you'll even notice the difference. > > -- > Tim Young > Elevate Software > www.elevatesoft.com > > |
Mon, May 1 2006 2:37 PM | Permanent Link |
Tim Young [Elevate Software] Elevate Software, Inc. ![]() | Santy,
<< How can I get the Spanish 'LocaleID' form creating the tables? >> Are you asking how to specify the LocaleID while creating the tables ? If so, then see the docs for creating tables: http://www.elevatesoft.com/dbisam4d5_creating_altering_tables.htm << What about if one of my customers doesn't have the spanish LocaleID in his computer? >> As long as they're using Windows 2000 or higher, they should have Spanish language support already installed. Windows 9x might require some additional installation from the Windows CD. << Or how can I be sure that the "LocalID" number is always the same in every computers? >> Windows locale IDs are standard across all versions of Windows and are always the same. -- Tim Young Elevate Software www.elevatesoft.com |
This web page was last updated on Wednesday, August 14, 2024 at 03:12 PM | Privacy Policy![]() © 2024 Elevate Software, Inc. All Rights Reserved Questions or comments ? ![]() |