load 'data/odbc' Unicode not showing in terminal

[Michael Christodoulides]

Hi,
I am a newbie to J,
I am considering using it to more easily interact to SQL Server.

Why is this not showing the output as unicode in terminal?

load 'data/odbc'
[ch=. ddcon 'Driver={ODBC Driver 17 for SQL
Server};Server=(localdb)\db1;Trusted_Connection=Yes'
ddfet ('select N''w こんにちは ''' ddsel ch), 1000
┌────────────┐
│w ?????
└────────────┘

Many thanks,

Michael

[chris burke]

In J, text is assumed to be in utf8 format. Try:

load 'unicode'
toutf8 each ddfet ('select N''w こんにちは ''' ddsel ch), 1000

[Mike Chris]

|value error: toutf8

[Mike Chris]

toutf8 each ddfet ('select N''w こんにちは ''' ddsel ch), 1000

|index error in toutf8, executing dyad u:

| 8 u:y

Press ENTER to inspect

On Tuesday, 20 May 2025 at 15:14:24 UTC+3 chris burke wrote:

[bill lam]

I should be utf8

BTW J script is already utf8 encoded. and J doesn't understand
National string format.
try instead
ddfet ('select ''こんにちは'''ddsel ch), 1000

> To unsubscribe from this group and stop receiving emails from it, send an email to forum+un...@jsoftware.com.

[Mike Chris]

Hi,

try instead
ddfet ('select N''こんにちは'''ddsel ch), 1000

That is what I tried from the beginning.
Outputs ????
Maybe J cannot handle this.

Thanks, Michael

[bill lam]

I suggested

ddfet ('select ''こんにちは'''ddsel ch), 1000

NOT

ddfet ('select N''こんにちは'''ddsel ch), 1000

why there is a w in the output?

[Mike Chris]

Hi,
The N is needed before the single quote, to inform SQL Server that the
following is unicode string.
Select N'こ'
not
Select 'こ'

I have tried

ddfet ('select  ''こんにちは'''ddsel ch), 1000

returns ????

Thanks, Michael

[bill lam]

The N is not needed in J. If there are characters above 127, the whole sentence will be converted to utf16 before sending to SQL server.  I don't have access to SQL Server to test for the time being.

[Mike Chris]

from within MS SQL Server
select WithN=N'こんにちは'
こんにちは
select WithoutN='こんにちは'
?????