migrate to mikroservices architecture, and add more features

core/config.py

@@ -0,0 +1,85 @@
+"""Pipeline configuration dataclasses with JSON save/load."""
+from __future__ import annotations
+
+import json
+from dataclasses import asdict, dataclass, field, fields
+from pathlib import Path
+from typing import Optional
+
+
+@dataclass
+class ConvertConfig:
+    enabled: bool = True
+    sample_rate: int = 44100
+    channels: int = 2
+    ffmpeg_path: str = "ffmpeg"
+
+
+@dataclass
+class SeparateConfig:
+    enabled: bool = True
+    model: str = "bs_roformer_anvuew"
+    isolate_stem: str = "vocals"
+    unblend_path: str = "unblend"
+    custom_kernels: bool = False
+
+
+@dataclass
+class TranscribeConfig:
+    enabled: bool = True
+    engine: str = "whisperx"           # "whisperx" | "gigaam"
+    model: str = "large-v3"
+    language: Optional[str] = "ru"     # None = autodetect
+    device: str = "auto"               # "cuda" | "cpu" | "auto"
+    compute_type: Optional[str] = None  # None = auto (int8_float16 / int8)
+    batch_size: int = 8
+    gigaam_provider: str = "auto"      # "auto" | "cuda" | "cpu"
+
+
+@dataclass
+class PipelineConfig:
+    input_dir: str = ""
+    output_dir: str = "transcribed"
+    recursive: bool = False
+    keep_intermediate: bool = True
+    overwrite: bool = True
+
+    convert: ConvertConfig = field(default_factory=ConvertConfig)
+    separate: SeparateConfig = field(default_factory=SeparateConfig)
+    transcribe: TranscribeConfig = field(default_factory=TranscribeConfig)
+
+    # --- serialization ---
+    def to_dict(self) -> dict:
+        return asdict(self)
+
+    def to_json(self) -> str:
+        return json.dumps(self.to_dict(), indent=2, ensure_ascii=False)
+
+    @classmethod
+    def from_dict(cls, data: dict) -> "PipelineConfig":
+        cfg = cls()
+        for top in fields(cls):
+            name = top.name
+            if name not in data:
+                continue
+            value = data[name]
+            current = getattr(cfg, name)
+            if hasattr(current, "__dataclass_fields__"):
+                sub_fields = {f.name for f in fields(current)}
+                for k, v in value.items():
+                    if k in sub_fields:
+                        setattr(current, k, v)
+            else:
+                setattr(cfg, name, value)
+        return cfg
+
+    @classmethod
+    def from_json(cls, text: str) -> "PipelineConfig":
+        return cls.from_dict(json.loads(text))
+
+    def save(self, path: Path) -> None:
+        Path(path).write_text(self.to_json(), encoding="utf-8")
+
+    @classmethod
+    def load(cls, path: Path) -> "PipelineConfig":
+        return cls.from_json(Path(path).read_text(encoding="utf-8"))