Skip to content

The page reader interleaves two copies of a heading printed at an offset #239

Description

@lemarier

A Victron product catalogue prints some of its headings twice, one copy about ten points to the right of the other and a third of a point off its baseline. Both copies are text, so the converter reads both, and because they stand on nearly the same baseline they are put on one line, where their letters interleave.

6e51ab40f8eab8bce02afc2d9775b4c7a7f6dd4e64430db762559f20a9573503 p5 reads BBaattteteryryPPrrootteecctt ssyysstetemm for "BatteryProtect system", and p20 SoSlaolra sr ysysstetemm wwiitthh BBaattteteryryPPrortoetcetct for "Solar system with BatteryProtect". The two copies of "Battery" on p5 are set at x 220.56 and x 230.18, on baselines 763.657 and 763.370.

This predates #238 and is unchanged by it: the same pages read the same way on main, which put the copies together through their boxes. A fix has to notice that one run repeats another rather than to separate them by where they are set, since a tenth of a line is well inside what holds a raised figure on its line.

Whatever a reader makes of such a heading is wrong: the page names the product, and the name it prints is not a word.

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