Skip to content

escaped characters should be recognized  #233

Description

@luzpaz

The term wasn\'t is being recognized as wasn by codespell. There should be logic to distinguish this better.

Activity

  1. peternewman commented on Feb 7, 2018

    @peternewman
    Collaborator

    #201

    This currently causes false positives due to not matching wasn't, but I guess that would be the case even without the PR above, it would just be suggesting was instead of wasn't.

  2. arm-in commented on Jan 7, 2019

    @arm-in
    Contributor

    +1

    "\nTo terminate ..."
    leads to
    "nto->not, disabled due to \n"

  3. yarikoptic commented on Feb 24, 2023

    @yarikoptic
    Contributor

    IMHO the original issue of escaped \' (where result would constitute just removal of \) is different from \n problem (where result should constitute of completely removing \n and indicating word boundary there), the oldest instance of which is mentioned in the oldest still open issue #17 where I just provided an extra example.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

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