Skip to content

Perf: replace regex splitters with Guava Splitter in VcfParser.java #9

Description

@felix1717-r

VcfParser.java currently calls Pattern.split per line, which is quite expensive for larger files.

Can I send a pull request to replace the four Pattern splitters with Google's Guava Splitter to speed up parsing? It's twice as fast on a 700k line VCF file and passes all of PharmCat's tests.

Note: Guava preserves trailing delimiters. While this wouldn't happen in a regular .vcf file, if a line had a trailing delimiter the code would need extra checks.

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