studioX for non-unicode windows application

[mountie]

unicode is default encoding at windows.
but windows support additional encodings for non-unicode applications.
default alternative locale is english.
but local sales windows version of Korea, Japan and China. default alternative locale is set to each country’s historical encodings.

when I try to use “extract table data” activity for non-unicode app,
studioX or the attended Bot does not follow the windows’s alternative locale setting.
the extracted data have broken character. only alphanumeric is visible.

other activity is working fine such as get text or anchors.

I think this is possibly bug.

[Nguyen_Van_Luong1 (Nguyen Van Luong)]

Extracting languages with special characters such as Chinese, Japanese, Korean…
In Vietnam I also have unicode for Vietnamese, however this will have an impact, usually I will turn off unicode to ensure accuracy.
Can you please share photos of your activities?
regards,

[mountie]

currently I removed the table data extracting activity.
the problem was because of dual encodings environment.
before unicode was accepted internationally, each countries developed their own encoding standards. for example, EUC-KR, Shift-JIS, Big5 for traditional chinese.

I was able to support unicode and non-unicode application both by setting followings

setup

time and landuage

choose language

adding language pack

choose date, time and local language

choose additional date, time and language

choose country or locale

choose admin option

change system locale
for non-unicode program

it was default english for non-unicode application.
with this default configuration, non-unicode CJK programs will result broken character.

[Luiza_Surdu-Bob (Luiza Bob)]

Hi @mountie ,

The Extract Table Data activity supports only Unicode characters.
This is because when extracting table data, we save the extracted data in an XML format. As XML documents consist of characters from the Unicode repertoire, we exclude the characters that are not valid in XML (Valid characters in XML - Wikipedia).

The Get Text activity needs no such filtering and therefore works for non-Unicode applications.

Best regards,
Luiza

[mountie]

using XML is your choice.
I think UiPath can fix it.
in Here(at least Korea), too much non-unicode applications are still running well where it can be automated only by RPA

I will test with other RPA solutions aswell.

[Luiza_Surdu-Bob (Luiza Bob)]

Thanks for your feedback @mountie . I’ll take your suggestion into the backlog for assessment and prioritization.

Thank you,
Luiza