Merge pull request #2 from gbel/claude/heuristic-merkle

chore: release v0.4.0 — ADDAC112 support + PyPI auto-publish

Author: gbel

.github/workflows/release.yml

@@ -0,0 +1,41 @@
+name: Publish to PyPI
+
+on:
+  release:
+    types: [published]
+
+jobs:
+  build:
+    name: Build distribution
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v5
+
+      - name: Build package
+        run: uv build
+
+      - name: Upload dist artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+  publish:
+    name: Publish to PyPI
+    needs: build
+    runs-on: ubuntu-latest
+    environment: pypi
+    permissions:
+      id-token: write  # required for OIDC trusted publishing
+    steps:
+      - name: Download dist artifacts
+        uses: actions/download-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+      - name: Publish to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1

CHANGELOG.md

@@ -1,5 +1,14 @@
 # Changelog
 
+## v0.4.0 — 2026-03-23
+
+### Features
+
+- **ADDAC112 VC Looper & Granular Processor** — new driver; audio format (sample rate, bit depth, channels) is configured per-bank via the wizard. Produces `BANK0/{WAV/,DELETED/,SETTINGS.CFG,SCALES.CFG,0.CFG}` — the complete folder structure the module expects on first load. WAV files are fully stripped of metadata (`-map_metadata -1 -fflags +bitexact`). Warns when total bank size approaches the ~60 MB limit. Scale data (factory defaults + HARMONIC/WELL TUNED custom presets) and CFG formats derived from the ADDAC112 user manual (pp. 40–43). Developed without access to physical hardware.
+- **`wizard_settings` driver hook** — new extension point in `__main__.py` for drivers that need format configuration before source selection; consumed by the ADDAC112 driver and available for future devices.
+
+---
+
 ## v0.3.0 — 2026-03-21
 
 ### Features

README.md

@@ -29,6 +29,7 @@ Over the course of development, what started as a simple converter accumulated a
 | ALM/Busy Circuits Squid Salmple | WAV | 44100 Hz | 16-bit | Mono | USB stick |
 | Rossum Electro-Music Assimil8or | WAV | 48000 Hz | 16-bit | Mono | SD card |
 | WMD Clutch | WAV | 48000 Hz | 16-bit | Mono | SD card |
+| ADDAC112 VC Looper & Granular Processor | WAV (no metadata) | configurable | configurable | configurable | microSD |
 
 ---
 
@@ -267,6 +268,39 @@ The HIHAT.INI settings file at the SD card root is not touched by SamplerPrep.
 > etc.) and to validate your SD card layout.  This driver was developed without access
 > to physical hardware.
 
+### ADDAC112 VC Looper & Granular Processor
+
+```
+card_folders/addac112/my-bank/
+  BANK0/
+    WAV/
+      1_sample.wav   ← up to 99 files, 1-indexed
+      2_sample.wav
+      …
+    DELETED/         ← empty dir, required by firmware
+    SETTINGS.CFG     ← format, granular, and CV settings
+    SCALES.CFG       ← quantizer scale definitions (Default + Custom, 7 slots each)
+    0.CFG            ← seed granular preset
+```
+
+Unlike other supported devices, the ADDAC112 has no fixed audio format — sample rate, bit depth, and channel count are configured per-bank during the wizard. The driver prompts for:
+
+- **Channels** — Mono or Stereo
+- **Sample rate** — 8k / 11.025k / 16k / 22.05k / 36k / 44.1k / 48k / 96k Hz
+- **Bit depth** — 8-bit, 16-bit, or 24-bit (note: 24-bit halves available loop recording time)
+- **Dry volume position** — Pre-FX or Post-FX
+- **Pause mode** — Toggle or Momentary
+- **Grain pitch on loop change** — Keep or Retune
+- **Grain pan mode** — Fixed or Travel
+- **Grain deviation mode** — Random or Spread
+- **Scales** — Factory defaults only, or optionally add HARMONIC or WELL TUNED custom scale presets
+
+All metadata is stripped from output WAVs (`-map_metadata -1 -fflags +bitexact`). A size warning fires if the total WAV folder content exceeds the recommended ~60 MB bank limit.
+
+`SETTINGS.CFG` and a seed `0.CFG` preset are generated from your wizard choices, giving the module a valid configuration to load on first use. Scale data in `SCALES.CFG` is derived from the ADDAC112 user manual (pp. 40–43).
+
+> **Note:** `bit_depth` in `SETTINGS.CFG` uses a non-obvious inverted encoding: `0` = 24-bit, `1` = 8-bit, `2` = 16-bit. SamplerPrep handles this automatically.
+
 ---
 
 ## macOS and AppleDouble files
@@ -447,6 +481,7 @@ samplerprep/
     squid.py
     assimil8or.py
     clutch.py
+    addac112.py
 config.json       ← runtime config and Radio Music profiles
 data.json         ← downloadable sample pack catalogue
 empty_folder/     ← placeholder RAW files for Radio Music skeleton

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "sampler-prep"
-version = "0.3.0"
+version = "0.4.0"
 description = "Prepare audio sample packs for hardware samplers"
 requires-python = ">=3.11"
 dependencies = ["questionary"]