Coverage for src / quber / core / ocr / __init__.py: 100%

1"""Reading text off page images, with provenance. 

2 

3One module lives here today: `well`, the OCR grounding well. It exists for 

4content the parse cannot ground — scanned pages and image-bound figures, 

5where no native text layer sits under the region. The well reads the rendered 

6page with RapidOCR and keeps each fragment with its box and confidence, in 

7PDF points, so consumers can trace an extracted value back to the pixels that 

8support it. 

9 

10The split of responsibilities: this package only reads and locates. Deciding 

11which regions need a well (the zero-native-text discriminator), attributing 

12values to fragments, and correcting readings all belong to consumers. 

13""" 

14 

15from quber.core.ocr.well import DocumentWell, OcrFragment, PageWell, build_well, read_image 

16 

17__all__ = [ 

18 "DocumentWell", 

19 "OcrFragment", 

20 "PageWell", 

21 "build_well", 

22 "read_image", 

23]