Hi guys,
Matthew, are you interested in joining the team? You could commit your
changes on your own.
Pablo, Matthew, regarding the 0/1 start index, that's a good point, and we
have two options:
1. give to each potentially problematic SQL generation method both values,
the one where index starts at 0, the other where index starts at 1.
2. add some characteristics to the ISqlProvider, telling if string indexes
start at 0 or 1, for example.
What's you favorite flavor?
On Sun, Sep 14, 2008 at 11:56, Pablo Iñigo Blasco <[EMAIL PROTECTED]> wrote:
>
>
> On Sun, Sep 14, 2008 at 6:32 AM, Ardekantur <[EMAIL PROTECTED]>wrote:
>
>>
>> Hello all -
>
>
> Hi Matthew!
>
>
>>
>>
>> Puttering through some of the tests for Oracle and caught a fairly
>> simple fix for a couple of tests. Where the default SQL provider users
>> `STRPOS` for getting the right behavior from #IndexOf LINQ queries,
>> Oracle has INSTR and SUBSTR.
>>
>> Thus:
>>
>> Index: src/DbLinq.Oracle/OracleSqlProvider.cs
>> ===================================================================
>> --- src/DbLinq.Oracle/OracleSqlProvider.cs (revision 876)
>> +++ src/DbLinq.Oracle/OracleSqlProvider.cs (working copy)
>> @@ -92,5 +92,28 @@
>> @"SELECT * FROM ({3}{0}{3}) WHERE {2} > {1}",
>> GetLiteralLimit(select, offsetAndLimit), offset,
>> LimitedRownum, NewLine);
>> }
>> +
>> + protected override string GetLiteralStringIndexOf(string
>> baseString, string searchString, string startIndex, string count)
>> + {
>> + // SUBSTR(baseString, StartIndex)
>> + string substring = GetLiteralSubString(baseString,
>> startIndex, count);
>> +
>> + // INSTR(SUBSTR(baseString, StartIndex), searchString) ---
>> > range 1:n , 0 => doesn't exist
>> + return string.Format("INSTR({0},{1})", substring,
>> searchString);
>> + }
>> +
>> + protected override string GetLiteralStringIndexOf(string
>> baseString, string searchString, string startIndex)
>> + {
>> + // SUBSTR(baseString,StartIndex)
>> + string substring = GetLiteralSubString(baseString,
>> startIndex);
>> +
>> + // INSTR(SUBSTR(baseString, StartIndex), searchString) ---
>> > range 1:n , 0 => doesn't exist
>> + return string.Format("INSTR({0},{1})", substring,
>> searchString);
>> + }
>> +
>> + protected override string GetLiteralStringIndexOf(string
>> baseString, string searchString)
>> + {
>> + return GetLiteralSubtract(string.Format("INSTR({0},{1})",
>> baseString, searchString), "1");
>> + }
>> }
>> }
>>
>
> It seems to be fine. The main potential problem you should consider here is
> the 'indexes start offset' difference between pl-sql and clr. Most of
> sql-engine has as first array index = 1. In the other hand, clr and most of
> programming languages has as first array index = 0. If you have considered
> it, it shoud be ok.
>
>
>> This knocks out 5 more tests, I think.
>
>
> Great :-D
>
>
>> I may write some more just to
>> make sure we're hitting as many cases as we can.
>
>
> That would be fantastic, they would be reused by others vendors too.
>
> I have tried apply the patch over the Oracle folder, but it crashes, wich
> revision are you using? could you send an attached .patch file?
>
> Regards.
>
>
> >
>
--
Pascal.
jabber/gtalk: [EMAIL PROTECTED]
msn: [EMAIL PROTECTED]
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"DbLinq" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/dblinq?hl=en
-~----------~----~----~----~------~----~------~--~---
