Metadata-Version: 2.1
Name: ocrd-models
Version: 2.22.0b2
Summary: OCR-D framework - file format APIs and schemas
Home-page: https://github.com/OCR-D/core
Author: Konstantin Baierer
Author-email: unixprog@gmail.com
License: Apache License 2.0
Description: # ocrd_models
        
        > OCR-D framework - file format APIs and schemas
        
        See https://github.com/OCR-D/core
        
        ## Adding user methods to the generated PAGE API
        
        Let's say you want to add a method `get_FirstTextRegion` on the pc:Page element:
        
        1. Create a file `ocrd_models/ocrd_page_user_methods/get_FirstTextRegion.py`
        
        ```python
        def get_FirstTextRegion(self):
          return self.get_TextRegion[0]
        ```
        
        2. Edit `ocrd_models/ocrd_page_user_methods.py` and append to the `METHOD_SPECS` list:
        
        ```python
        METHOD_SPECS = (
          # ...
          _add_method(r'^PageType$', 'get_FirstTextRegion')
          # ...
        )
        ```
        
        If the filename (sans the `.py` extension) does not match the method_name, you
        can provide an additional `file_name` attribute to `_add_method`:
        
        ```python
        METHOD_SPECS = (
          # ...
          _add_method(r'^PageType$', 'exportChildren', 'exportChildren_PageType')
          # ...
        )
        ```
        
        Would add the method `exportChildren` from a file `exportChildren_PageType.py`.
        
        **NOTE** The method name in the file must match the method name passed to
        `_add_method`. This is *not* checked automatically, so double-check manually!
        
        
        3. Regenerate the PAGE API:
        
        ```sh
        make generate-page
        ```
        
Keywords: OCR,OCR-D
Platform: UNKNOWN
Description-Content-Type: text/markdown
