Closed tiantun closed 2 years ago
tiantun
commented
3 years ago The code seems to be auto-generated based on /stable/2019-05-06/searchservice.json . The most recent preview version of this is preview/2020-06-30/searchservice.json .
Created a branch with what seem to be the required changes: Add "unk" value for "OcrSkillLanguage"
jsquire
commented
3 years ago Thank you for your feedback. Tagging and routing to the team member best able to assist.
Mohit-Chakraborty
commented
3 years ago Thanks @tiantun for reporting the issue and suggesting a change to fix the issue.
Mohit-Chakraborty
commented
3 years ago I see a PR has been created against the target branch - https://github.com/Azure/azure-rest-api-specs/pull/16279
tiantun
commented
3 years ago Thank you @Mohit-Chakraborty . Looks like the changes were already merged into the API specs.
Can we keep this one open until the generated code in the SDK is updated as well? That would simplify tracking of the "unk" value's availability in the SDK.
Mohit-Chakraborty
commented
3 years ago Sure. Since this issue is in the .NET SDK repo, we will keep it open till we consume the change in the SDK.
salilponde
commented
2 years ago Do we know when this will be available in the SDK?
Mohit-Chakraborty
commented
2 years ago It was added in '11.4.0-beta.6'. The Changelog has a mention.
I will close the issue. Let me know if there are issues.
Library or service name. Microsoft.Azure.Search Azure.Search.Documents
Is your feature request related to a problem? Please describe. Enabling auto-detection on the OCR Skill requires setting the "defaultLanguageCode" to "unk". See OCR cognitive skill: "If the language code is unspecified or null, the language will be set to English. If the language is explicitly set to "unk", the language will be auto-detected." This seems to be the behaviour since at least early 2019: Set to "unk" to have system auto-detect language. However, this is currently not possible in the Search SDKs.