Support to search portuguese caracters

Question

Support to search portuguese caracters

ruifrazao

Old Hand

Points: 336
More actions
August 19, 2009 at 5:47 am

#213572

Hi,
Have you used the Irony library (Irony.dll) like it is in http://irony.codeplex.com/, or have you made same changes?
I am trying to look in Irony code to incorporate portuguese caracters (not suported until now), but when I try to run you code with Irony.dll from
http://irony.codeplex.com gives some error.
Thanks & Regards,
Rui

Viewing 6 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic. Login to reply

Gift Peddie SSC Guru Points: 73570 More actions · Answer 1

Portuguese is standard Latin alphabet western European language that you can support by using System.Text.Encoding in .NET and using either Portuguese Windows collation so you can separate Portugal Portuguese from either Brazilian or Angolan. That is going to VS2005/8 Advanced save as and save you code as Portuguese.

It helps if you start with translated text so it is just simple encoding and collation, in .NET 3.5 most known languages and text encoding formats are supported by default. Post again if you still need help.

Kind regards,
Gift Peddie

ruifrazao Old Hand Points: 336 More actions · Answer 2

Gift Peddie (8/19/2009)
Portuguese is standard Latin alphabet western European language that you can support by using System.Text.Encoding in .NET and using either Portuguese Windows collation so you can separate Portugal Portuguese from either Brazilian or Angolan. That is going to VS2005/8 Advanced save as and save you code as Portuguese.
It helps if you start with translated text so it is just simple encoding and collation, in .NET 3.5 most known languages and text encoding formats are supported by default. Post again if you still need help.

Hi Gift,

I am using the Windows-1252 code page that includes Portuguese characters. But when I try to search a word with an Portuguese character appears a error message saying that the character are not valid.

Regards,

Rui

Gift Peddie SSC Guru Points: 73570 More actions · Answer 3

If you are storing more than Portuguese in one column then you may need Nvarchar definition of your text columns data type and then using Portuguese collation on either table level or column level. You could also add it in your T-SQL code. Here is the System.Text.Encoding page and the SQL Server Windows collation page.

http://msdn.microsoft.com/en-us/library/system.text.encoding.aspx

http://msdn.microsoft.com/en-us/library/ms143508.aspx

Kind regards,
Gift Peddie

ruifrazao Old Hand Points: 336 More actions · Answer 4

Gift Peddie (8/19/2009)
If you are storing more than Portuguese in one column then you may need Nvarchar definition of your text columns data type and then using Portuguese collation on either table level or column level. You could also add it in your T-SQL code. Here is the System.Text.Encoding page and the SQL Server Windows collation page.
http://msdn.microsoft.com/en-us/library/system.text.encoding.aspx
http://msdn.microsoft.com/en-us/library/ms143508.aspx

The problem is not in SQL database. In database all text are saved correctly.

My question is about using the code from article “A Google-like Full Text Search”, and message error about the Portuguese characters appears when I try to convert a word with a Portuguese character.

Thanks & Regards,

Rui

Gift Peddie SSC Guru Points: 73570 More actions · Answer 5

That article is calling a .NET dll that is parsing text on the binary level, you really don't need to parse most Western European text on the binary level. So you could just use a combination or saving .NET code as Portuguese per my first post or do Text encoding using Portuguese encoding before using the Google like search. The reason the article added Case statements to ANSI SQL predefined and Microsoft implemented Full Text predicates.

Kind regards,
Gift Peddie