Function _prepend_source_to_frontmatter

openkb/agent/compiler.py:1099–1137

Prepend ``source_file`` to the inline ``sources:`` list in YAML frontmatter. Creates the frontmatter or the ``sources:`` line if missing. Returns the text unchanged if ``source_file`` is already present in the list, or if the frontmatter is malformed (no closing ``---``).

(text: str, source_file: str)

def _prepend_source_to_frontmatter(text: str, source_file: str) -> str:
    """Prepend ``source_file`` to the inline ``sources:`` list in YAML frontmatter.

    Creates the frontmatter or the ``sources:`` line if missing. Returns the
    text unchanged if ``source_file`` is already present in the list, or if
    the frontmatter is malformed (no closing ``---``).
    """
    if not text.startswith("---"):
        return f"---\n{_yaml_list_line('sources', [source_file])}\n---\n\n" + text

    parts = frontmatter.split(text)
    if parts is None:
        return text

    fm_block, body = parts
    # Strip the trailing closing delimiter to get the prefix lines (opening
    # "---" + content lines), then re-append it. `frontmatter.split` leaves the
    # closing at the end of fm_block as either "\n---\n" or a bare "\n---" (when
    # the page ends at the delimiter with no trailing newline). Assuming only
    # "\n---\n" would, for the bare form, make the strip below collapse the
    # whole block and drop every existing frontmatter key.
    closing = "\n---\n" if fm_block.endswith("\n---\n") else "\n---"
    fm_prefix = fm_block[: -len(closing)]
    fm_lines = fm_prefix.split("\n")

    for i, line in enumerate(fm_lines):
        if not line.lstrip().startswith("sources:"):
            continue
        items = _parse_yaml_list_value(line)
        if items is None:
            return text
        if source_file in items:
            return text
        items.insert(0, source_file)
        fm_lines[i] = _yaml_list_line("sources", items)
        return "\n".join(fm_lines) + closing + body

    fm_lines.insert(1, _yaml_list_line("sources", [source_file]))
    return "\n".join(fm_lines) + closing + body


def _remove_source_from_frontmatter(text: str, source_file: str) -> tuple[str, bool]: