Skip to content

DECISIONS.md

Latest commit

 

History

History
77 lines (52 loc) · 3.55 KB

DECISIONS.md

File metadata and controls

77 lines (52 loc) · 3.55 KB

Parsing Decisions

This document outlines the key decisions and parsing rules applied to extract the maximum amount of usable data from .mxgrp files.

The initial focus was on Battery since it exclusively supports sample files. However, extracting sample paths from Battery proved more difficult than from Maschine.
For this reason, the Maschine Parser was implemented first, prioritizing sample extraction while discarding internal sound generators and VST references.

Maschine vs. Battery

Maschine can use internal sound generators (e.g., for a Kick), while Battery relies entirely on sample files:

Maschine vs Battery Kit Comparison

Because of the complexity of exporting anything else than samples, all non-sample properties are ignored during extraction.

Maschine .mxgrp Parsing Rules

Maschine can have single samples or multiple samples assigned to a pad, samples can be assigned to regions. To simplify, we call pads with single samples sample and pads with multiple samples multisample.

The goal is to get a 16-sound pack for each Maschine group, so in the final export the tool will select the best sample from each multisample.

  • Expansion Name
    Extracted from the two lines following the keyword serialization::archive.

    • If these two lines differ, the parent folder name is used as a fallback.

  • Kit Name
    Always derived from the .mxgrp file name, trimmed to remove spaces.

  • Pad Assignment
    Pads are assigned sequentially starting from 1 for each sample or multisample.

    • No wraparound occurs (pad numbers can exceed 16).
    • A warning is printed if more than 16 pads are detected.

  • Sample Path Handling

    • Sample paths split across multiple lines are merged.
      (e.g., if a line does not end with .wav but the next line does, and the path begins with Samples/, //Samples/, or *Samples/).
    • Only paths starting with Samples/, //Samples/, or *Samples/ are considered valid.
    • All other paths (e.g., C:/, /Users/, VST references) are ignored.

  • Multisample Detection

    • A multisample is identified when a sample’s preceding line matches the expansion name
      (or appears within the previous 4 lines, skipping garbage lines).
    • Consecutive samples with this property are grouped into a multisample.
    • If a multisample contains only one sample, it is downgraded to a regular sample.

  • Multisample Export

    When exporting multisamples, the goal is to select a single representative sample per multisample. The selection follows a priority order based on note designation in the filename:

    1. First, select the sample ending with _C4 (case-insensitive).
    2. If _C4 is not found, select _C3.
    3. If neither _C4 nor _C3 exists, select the sample ending with _C*, where * is any single digit (e.g., _C1, _C5).
    4. If no samples match the above, select the first sample in the multisample.

    This ensures consistency in exported kits, prioritizing central octave samples (C3, C4) while maintaining a fallback for any single-note multisample.

  • Cleanup Rules

    • Consecutive duplicate sample paths are removed.
    • Empty multisamples are deleted.
    • Kits with no valid samples are skipped, with a warning printed.
    • Garbage lines (length = 4 or containing disallowed characters) are ignored during multisample grouping.

  • Output Format
    Each parsed kit outputs:

    • Expansion name
    • Kit name
    • Base path
    • List of samples, including:
      • Type (sample or multisample)
      • Name
      • Pad number
      • File paths