Method python_link

tensorflow/tools/docs/parser.py:303–331

Resolve a "@{python symbol}" reference to a Markdown link. This will pick the canonical location for duplicate symbols. The input to this function should already be stripped of the '@' and '{}'. This function returns a Markdown link. If `code_ref` is true, it is assumed that this

(self, link_text, ref_full_name, relative_path_to_root,
                  code_ref=True)

    return string

  def python_link(self, link_text, ref_full_name, relative_path_to_root,
                  code_ref=True):
    """Resolve a "@{python symbol}" reference to a Markdown link.

    This will pick the canonical location for duplicate symbols.  The
    input to this function should already be stripped of the '@' and
    '{}'.  This function returns a Markdown link. If `code_ref` is
    true, it is assumed that this is a code reference, so the link
    text will be rendered as code (using backticks).
    `link_text` should refer to a library symbol, starting with 'tf.'.

    Args:
      link_text: The text of the Markdown link.
      ref_full_name: The fully qualified name of the symbol to link to.
      relative_path_to_root: The relative path from the location of the current
        document to the root of the API documentation.
      code_ref: If true (the default), put `link_text` in `...`.

    Returns:
      A markdown link to the documentation page of `ref_full_name`.
    """
    url = self.reference_to_url(ref_full_name, relative_path_to_root)

    if code_ref:
      link_text = link_text.join(['<code>', '</code>'])
    else:
      link_text = self._link_text_to_html(link_text)

    return '<a href="{}">{}</a>'.format(url, link_text)

  @staticmethod
  def _link_text_to_html(link_text):