Class RegexSplitConfig

entity/configs/dynamic_base.py:69–185

Configuration for regex-based splitting. Split content by regex pattern matches. Each match becomes one execution unit. Attributes: pattern: Python regular expression used to split content. group: Capture group name or index. Defaults to the entire match (group 0).

@dataclass
class RegexSplitConfig(SplitTypeConfig):
    """Configuration for regex-based splitting.
    
    Split content by regex pattern matches. Each match becomes one execution unit.
    
    Attributes:
        pattern: Python regular expression used to split content.
        group: Capture group name or index. Defaults to the entire match (group 0).
        case_sensitive: Whether the regex should be case sensitive.
        multiline: Enable multiline mode (re.MULTILINE).
        dotall: Enable dotall mode (re.DOTALL).
        on_no_match: Behavior when no match is found.
    """

    pattern: str = ""
    group: str | int | None = None
    case_sensitive: bool = True
    multiline: bool = False
    dotall: bool = False
    on_no_match: str = "pass"

    FIELD_SPECS = {
        "pattern": ConfigFieldSpec(
            name="pattern",
            display_name="Regex Pattern",
            type_hint="str",
            required=True,
            description="Python regular expression used to split content.",
        ),
        "group": ConfigFieldSpec(
            name="group",
            display_name="Capture Group",
            type_hint="str",
            required=False,
            description="Capture group name or index. Defaults to the entire match (group 0).",
        ),
        "case_sensitive": ConfigFieldSpec(
            name="case_sensitive",
            display_name="Case Sensitive",
            type_hint="bool",
            required=False,
            default=True,
            description="Whether the regex should be case sensitive.",
        ),
        "multiline": ConfigFieldSpec(
            name="multiline",
            display_name="Multiline Flag",
            type_hint="bool",
            required=False,
            default=False,
            description="Enable multiline mode (re.MULTILINE).",
            advance=True,
        ),
        "dotall": ConfigFieldSpec(
            name="dotall",
            display_name="Dotall Flag",
            type_hint="bool",
            required=False,