Skip to content

tree construction test assumes inefficient reprocessing #172

Open
@not-my-profile

Description

@not-my-profile

The test in question is the following of tests19.dat:

#data
<!doctype html><math></html>
#errors
(1,28): unexpected-end-tag
(1,28): expected-one-end-tag-but-got-another
(1,28): unexpected-end-tag
#document
| <!DOCTYPE html>
| <html>
|   <head>
|   <body>
|     <math math>

The "in body" insertion mode contains the following instruction:

An end tag whose tag name is "html"
[...]
Reprocess the token.

If this token reprocessing is implemented efficiently by reprocessing according to the current insertion mode (as opposed to unnecessarily stepping through the rules for parsing tokens in foreign content again), then the second error is never reported, which causes this test to fail due to there only being 2 instead of 3 errors.

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