Skip to content

Naming with a truncated version of the ancestor's name. #314

@ruck94301

Description

@ruck94301

User reports that in some operations in LabGym, new files or directories are named with a truncated version of
their ancestor.

When automatically naming some data containers produced by LabGym (files and dirs), if the naming is truncated/restricted in length, then can the outputs overlap and lose info that's encoded in the naming?

On 2025-10-18, Henry says this would be a problem, but he hasn't encountered it, so needs details.

All the information encoded in the file name is critical for users to know their settings if they forget what they set. This is useful if a user wants to go back to a dataset generated years ago. So yes, if the naming is truncated/restricted in length, such information will be lost. By the way, which system restricts the length of file names? I have never seen this on my systems.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions