Skip to content

IFT table bug - Incorrect lastEntryId initiation and missing implementation when entry Id could be a byte string in addition to integer #4134

Description

@yChenMonotype

I'm trying to read the IFT table from the font generated using IFT encoder and discovered there is inconsistencies between the entry's patch uri and the patch file names, they do not match up.

The culprit:

IFT patch format 2 specs states:

If the entryIdStringData offset is 0 then initialize last entry id to 0. Otherwise initialize it to an empty byte string. Set current byte to 0, and current id string byte to 0.

The lastEntryId should initialize as 0 if entryIdStringData is empty else it should initialize to b"" and use entry's entryIdStringLength to derive the actual entry ID.

Activity

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