mirror of
https://github.com/denizsafak/abogen.git
synced 2026-07-18 13:40:27 +02:00
refactor: replace hardcoded backend ID sets with registry checks
Add TTSBackendRegistry.is_registered() and module-level is_registered_backend() to validate backend IDs dynamically. Replace all Category A hardcoded sets (validation-only) in voice_profiles, api routes, conversion_runner, and form utils.
This commit is contained in:
@@ -30,6 +30,10 @@ class TTSBackendRegistry:
|
|||||||
self._backends[metadata.id] = metadata
|
self._backends[metadata.id] = metadata
|
||||||
self._factories[metadata.id] = factory
|
self._factories[metadata.id] = factory
|
||||||
|
|
||||||
|
def is_registered(self, backend_id: str) -> bool:
|
||||||
|
"""Return True if a backend with the given id is registered."""
|
||||||
|
return backend_id in self._backends
|
||||||
|
|
||||||
def list_backends(self) -> list[TTSBackendMetadata]:
|
def list_backends(self) -> list[TTSBackendMetadata]:
|
||||||
"""Return metadata for all registered backends."""
|
"""Return metadata for all registered backends."""
|
||||||
return list(self._backends.values())
|
return list(self._backends.values())
|
||||||
@@ -88,3 +92,9 @@ def get_default_voice(backend_id: str, fallback: str = "") -> str:
|
|||||||
def create_backend(backend_id: str, **kwargs: Any) -> TTSBackend:
|
def create_backend(backend_id: str, **kwargs: Any) -> TTSBackend:
|
||||||
"""Create a TTS backend instance by provider id."""
|
"""Create a TTS backend instance by provider id."""
|
||||||
return _registry.create_backend(backend_id, **kwargs)
|
return _registry.create_backend(backend_id, **kwargs)
|
||||||
|
|
||||||
|
|
||||||
|
def is_registered_backend(backend_id: str) -> bool:
|
||||||
|
"""Return True if *backend_id* is a registered TTS backend."""
|
||||||
|
import abogen.tts_backends # noqa: F401 — triggers backend registration
|
||||||
|
return _registry.is_registered(backend_id)
|
||||||
|
|||||||
+230
-230
@@ -1,230 +1,230 @@
|
|||||||
import json
|
import json
|
||||||
import os
|
import os
|
||||||
from typing import Any, Dict, Iterable, List, Tuple
|
from typing import Any, Dict, Iterable, List, Tuple
|
||||||
|
|
||||||
from abogen.tts_backend_registry import get_metadata
|
from abogen.tts_backend_registry import get_metadata, is_registered_backend
|
||||||
from abogen.utils import get_user_config_path
|
from abogen.utils import get_user_config_path
|
||||||
|
|
||||||
|
|
||||||
def _get_profiles_path():
|
def _get_profiles_path():
|
||||||
config_path = get_user_config_path()
|
config_path = get_user_config_path()
|
||||||
config_dir = os.path.dirname(config_path)
|
config_dir = os.path.dirname(config_path)
|
||||||
return os.path.join(config_dir, "voice_profiles.json")
|
return os.path.join(config_dir, "voice_profiles.json")
|
||||||
|
|
||||||
|
|
||||||
def load_profiles():
|
def load_profiles():
|
||||||
"""Load all voice profiles from JSON file."""
|
"""Load all voice profiles from JSON file."""
|
||||||
path = _get_profiles_path()
|
path = _get_profiles_path()
|
||||||
if os.path.exists(path):
|
if os.path.exists(path):
|
||||||
try:
|
try:
|
||||||
with open(path, "r", encoding="utf-8") as f:
|
with open(path, "r", encoding="utf-8") as f:
|
||||||
data = json.load(f)
|
data = json.load(f)
|
||||||
# always expect abogen_voice_profiles wrapper
|
# always expect abogen_voice_profiles wrapper
|
||||||
if isinstance(data, dict) and "abogen_voice_profiles" in data:
|
if isinstance(data, dict) and "abogen_voice_profiles" in data:
|
||||||
return data["abogen_voice_profiles"]
|
return data["abogen_voice_profiles"]
|
||||||
# fallback: treat as profiles dict
|
# fallback: treat as profiles dict
|
||||||
if isinstance(data, dict):
|
if isinstance(data, dict):
|
||||||
return data
|
return data
|
||||||
except Exception:
|
except Exception:
|
||||||
return {}
|
return {}
|
||||||
return {}
|
return {}
|
||||||
|
|
||||||
|
|
||||||
def save_profiles(profiles):
|
def save_profiles(profiles):
|
||||||
"""Save all voice profiles to JSON file."""
|
"""Save all voice profiles to JSON file."""
|
||||||
path = _get_profiles_path()
|
path = _get_profiles_path()
|
||||||
os.makedirs(os.path.dirname(path), exist_ok=True)
|
os.makedirs(os.path.dirname(path), exist_ok=True)
|
||||||
with open(path, "w", encoding="utf-8") as f:
|
with open(path, "w", encoding="utf-8") as f:
|
||||||
# always save with abogen_voice_profiles wrapper
|
# always save with abogen_voice_profiles wrapper
|
||||||
json.dump({"abogen_voice_profiles": profiles}, f, indent=2)
|
json.dump({"abogen_voice_profiles": profiles}, f, indent=2)
|
||||||
|
|
||||||
|
|
||||||
def delete_profile(name):
|
def delete_profile(name):
|
||||||
"""Remove a profile by name."""
|
"""Remove a profile by name."""
|
||||||
profiles = load_profiles()
|
profiles = load_profiles()
|
||||||
if name in profiles:
|
if name in profiles:
|
||||||
del profiles[name]
|
del profiles[name]
|
||||||
save_profiles(profiles)
|
save_profiles(profiles)
|
||||||
|
|
||||||
|
|
||||||
def duplicate_profile(src, dest):
|
def duplicate_profile(src, dest):
|
||||||
"""Duplicate an existing profile."""
|
"""Duplicate an existing profile."""
|
||||||
profiles = load_profiles()
|
profiles = load_profiles()
|
||||||
if src in profiles and dest:
|
if src in profiles and dest:
|
||||||
profiles[dest] = profiles[src]
|
profiles[dest] = profiles[src]
|
||||||
save_profiles(profiles)
|
save_profiles(profiles)
|
||||||
|
|
||||||
|
|
||||||
def export_profiles(export_path):
|
def export_profiles(export_path):
|
||||||
"""Export all profiles to specified JSON file."""
|
"""Export all profiles to specified JSON file."""
|
||||||
profiles = load_profiles()
|
profiles = load_profiles()
|
||||||
with open(export_path, "w", encoding="utf-8") as f:
|
with open(export_path, "w", encoding="utf-8") as f:
|
||||||
json.dump({"abogen_voice_profiles": profiles}, f, indent=2)
|
json.dump({"abogen_voice_profiles": profiles}, f, indent=2)
|
||||||
|
|
||||||
|
|
||||||
def serialize_profiles() -> Dict[str, Dict[str, Iterable[Tuple[str, float]]]]:
|
def serialize_profiles() -> Dict[str, Dict[str, Iterable[Tuple[str, float]]]]:
|
||||||
"""Return profiles in canonical dictionary form."""
|
"""Return profiles in canonical dictionary form."""
|
||||||
return load_profiles()
|
return load_profiles()
|
||||||
|
|
||||||
|
|
||||||
def _normalize_supertonic_voice(value: Any) -> str:
|
def _normalize_supertonic_voice(value: Any) -> str:
|
||||||
raw = str(value or "").strip().upper()
|
raw = str(value or "").strip().upper()
|
||||||
supertonic_voices = get_metadata("supertonic").voices
|
supertonic_voices = get_metadata("supertonic").voices
|
||||||
return raw if raw in supertonic_voices else "M1"
|
return raw if raw in supertonic_voices else "M1"
|
||||||
|
|
||||||
|
|
||||||
def _coerce_supertonic_steps(value: Any) -> int:
|
def _coerce_supertonic_steps(value: Any) -> int:
|
||||||
try:
|
try:
|
||||||
steps = int(value)
|
steps = int(value)
|
||||||
except (TypeError, ValueError):
|
except (TypeError, ValueError):
|
||||||
return 5
|
return 5
|
||||||
return max(2, min(15, steps))
|
return max(2, min(15, steps))
|
||||||
|
|
||||||
|
|
||||||
def _coerce_supertonic_speed(value: Any) -> float:
|
def _coerce_supertonic_speed(value: Any) -> float:
|
||||||
try:
|
try:
|
||||||
speed = float(value)
|
speed = float(value)
|
||||||
except (TypeError, ValueError):
|
except (TypeError, ValueError):
|
||||||
return 1.0
|
return 1.0
|
||||||
return max(0.7, min(2.0, speed))
|
return max(0.7, min(2.0, speed))
|
||||||
|
|
||||||
|
|
||||||
def normalize_profile_entry(entry: Any) -> Dict[str, Any]:
|
def normalize_profile_entry(entry: Any) -> Dict[str, Any]:
|
||||||
"""Normalize a stored profile entry.
|
"""Normalize a stored profile entry.
|
||||||
|
|
||||||
Backwards compatible:
|
Backwards compatible:
|
||||||
- Legacy Kokoro-only entries: {language, voices}
|
- Legacy Kokoro-only entries: {language, voices}
|
||||||
- New entries: include provider.
|
- New entries: include provider.
|
||||||
"""
|
"""
|
||||||
|
|
||||||
if not isinstance(entry, dict):
|
if not isinstance(entry, dict):
|
||||||
return {}
|
return {}
|
||||||
|
|
||||||
provider = str(entry.get("provider") or "kokoro").strip().lower()
|
provider = str(entry.get("provider") or "kokoro").strip().lower()
|
||||||
if provider not in {"kokoro", "supertonic"}:
|
if not is_registered_backend(provider):
|
||||||
provider = "kokoro"
|
provider = "kokoro"
|
||||||
|
|
||||||
language = str(entry.get("language") or "a").strip().lower() or "a"
|
language = str(entry.get("language") or "a").strip().lower() or "a"
|
||||||
|
|
||||||
if provider == "supertonic":
|
if provider == "supertonic":
|
||||||
return {
|
return {
|
||||||
"provider": "supertonic",
|
"provider": "supertonic",
|
||||||
"language": language,
|
"language": language,
|
||||||
"voice": _normalize_supertonic_voice(
|
"voice": _normalize_supertonic_voice(
|
||||||
entry.get("voice") or entry.get("voice_name") or entry.get("name")
|
entry.get("voice") or entry.get("voice_name") or entry.get("name")
|
||||||
),
|
),
|
||||||
"total_steps": _coerce_supertonic_steps(
|
"total_steps": _coerce_supertonic_steps(
|
||||||
entry.get("total_steps")
|
entry.get("total_steps")
|
||||||
or entry.get("supertonic_total_steps")
|
or entry.get("supertonic_total_steps")
|
||||||
or entry.get("quality")
|
or entry.get("quality")
|
||||||
),
|
),
|
||||||
"speed": _coerce_supertonic_speed(
|
"speed": _coerce_supertonic_speed(
|
||||||
entry.get("speed") or entry.get("supertonic_speed")
|
entry.get("speed") or entry.get("supertonic_speed")
|
||||||
),
|
),
|
||||||
}
|
}
|
||||||
|
|
||||||
voices = _normalize_voice_entries(entry.get("voices", []))
|
voices = _normalize_voice_entries(entry.get("voices", []))
|
||||||
if not voices:
|
if not voices:
|
||||||
return {}
|
return {}
|
||||||
return {
|
return {
|
||||||
"provider": "kokoro",
|
"provider": "kokoro",
|
||||||
"language": language,
|
"language": language,
|
||||||
"voices": voices,
|
"voices": voices,
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
||||||
def _normalize_voice_entries(entries: Iterable) -> List[Tuple[str, float]]:
|
def _normalize_voice_entries(entries: Iterable) -> List[Tuple[str, float]]:
|
||||||
normalized: List[Tuple[str, float]] = []
|
normalized: List[Tuple[str, float]] = []
|
||||||
kokoro_voices = get_metadata("kokoro").voices
|
kokoro_voices = get_metadata("kokoro").voices
|
||||||
for item in entries or []:
|
for item in entries or []:
|
||||||
if isinstance(item, dict):
|
if isinstance(item, dict):
|
||||||
voice = item.get("id") or item.get("voice")
|
voice = item.get("id") or item.get("voice")
|
||||||
weight = item.get("weight")
|
weight = item.get("weight")
|
||||||
elif isinstance(item, (list, tuple)) and len(item) >= 2:
|
elif isinstance(item, (list, tuple)) and len(item) >= 2:
|
||||||
voice, weight = item[0], item[1]
|
voice, weight = item[0], item[1]
|
||||||
else:
|
else:
|
||||||
continue
|
continue
|
||||||
if voice not in kokoro_voices:
|
if voice not in kokoro_voices:
|
||||||
continue
|
continue
|
||||||
if weight is None:
|
if weight is None:
|
||||||
continue
|
continue
|
||||||
try:
|
try:
|
||||||
weight_val = float(weight)
|
weight_val = float(weight)
|
||||||
except (TypeError, ValueError):
|
except (TypeError, ValueError):
|
||||||
continue
|
continue
|
||||||
if weight_val <= 0:
|
if weight_val <= 0:
|
||||||
continue
|
continue
|
||||||
normalized.append((voice, weight_val))
|
normalized.append((voice, weight_val))
|
||||||
return normalized
|
return normalized
|
||||||
|
|
||||||
|
|
||||||
def normalize_voice_entries(entries: Iterable) -> List[Tuple[str, float]]:
|
def normalize_voice_entries(entries: Iterable) -> List[Tuple[str, float]]:
|
||||||
"""Public helper to normalize voice-weight pairs from arbitrary payloads."""
|
"""Public helper to normalize voice-weight pairs from arbitrary payloads."""
|
||||||
|
|
||||||
return _normalize_voice_entries(entries)
|
return _normalize_voice_entries(entries)
|
||||||
|
|
||||||
|
|
||||||
def save_profile(name: str, *, language: str, voices: Iterable) -> None:
|
def save_profile(name: str, *, language: str, voices: Iterable) -> None:
|
||||||
"""Persist a single profile after validating its data."""
|
"""Persist a single profile after validating its data."""
|
||||||
|
|
||||||
name = (name or "").strip()
|
name = (name or "").strip()
|
||||||
if not name:
|
if not name:
|
||||||
raise ValueError("Profile name is required")
|
raise ValueError("Profile name is required")
|
||||||
|
|
||||||
normalized = _normalize_voice_entries(voices)
|
normalized = _normalize_voice_entries(voices)
|
||||||
if not normalized:
|
if not normalized:
|
||||||
raise ValueError("At least one voice with a weight above zero is required")
|
raise ValueError("At least one voice with a weight above zero is required")
|
||||||
|
|
||||||
if not language:
|
if not language:
|
||||||
language = "a"
|
language = "a"
|
||||||
|
|
||||||
profiles = load_profiles()
|
profiles = load_profiles()
|
||||||
profiles[name] = {"provider": "kokoro", "language": language, "voices": normalized}
|
profiles[name] = {"provider": "kokoro", "language": language, "voices": normalized}
|
||||||
save_profiles(profiles)
|
save_profiles(profiles)
|
||||||
|
|
||||||
|
|
||||||
def remove_profile(name: str) -> None:
|
def remove_profile(name: str) -> None:
|
||||||
delete_profile(name)
|
delete_profile(name)
|
||||||
|
|
||||||
|
|
||||||
def import_profiles_data(data: Dict, *, replace_existing: bool = False) -> List[str]:
|
def import_profiles_data(data: Dict, *, replace_existing: bool = False) -> List[str]:
|
||||||
"""Merge profiles from a dictionary structure and persist them.
|
"""Merge profiles from a dictionary structure and persist them.
|
||||||
|
|
||||||
Returns the list of profile names that were added or updated.
|
Returns the list of profile names that were added or updated.
|
||||||
"""
|
"""
|
||||||
|
|
||||||
if not isinstance(data, dict):
|
if not isinstance(data, dict):
|
||||||
raise ValueError("Invalid profile payload")
|
raise ValueError("Invalid profile payload")
|
||||||
|
|
||||||
if "abogen_voice_profiles" in data:
|
if "abogen_voice_profiles" in data:
|
||||||
data = data["abogen_voice_profiles"]
|
data = data["abogen_voice_profiles"]
|
||||||
|
|
||||||
if not isinstance(data, dict):
|
if not isinstance(data, dict):
|
||||||
raise ValueError("Invalid profile payload")
|
raise ValueError("Invalid profile payload")
|
||||||
|
|
||||||
current = load_profiles()
|
current = load_profiles()
|
||||||
updated: List[str] = []
|
updated: List[str] = []
|
||||||
for name, entry in data.items():
|
for name, entry in data.items():
|
||||||
normalized = normalize_profile_entry(entry)
|
normalized = normalize_profile_entry(entry)
|
||||||
if not normalized:
|
if not normalized:
|
||||||
continue
|
continue
|
||||||
if name in current and not replace_existing:
|
if name in current and not replace_existing:
|
||||||
# skip duplicates unless explicit replacement is requested
|
# skip duplicates unless explicit replacement is requested
|
||||||
continue
|
continue
|
||||||
current[name] = normalized
|
current[name] = normalized
|
||||||
updated.append(name)
|
updated.append(name)
|
||||||
|
|
||||||
if updated:
|
if updated:
|
||||||
save_profiles(current)
|
save_profiles(current)
|
||||||
return updated
|
return updated
|
||||||
|
|
||||||
|
|
||||||
def export_profiles_payload(names: Iterable[str] | None = None) -> Dict[str, Dict]:
|
def export_profiles_payload(names: Iterable[str] | None = None) -> Dict[str, Dict]:
|
||||||
"""Return profiles limited to the provided names for download/export."""
|
"""Return profiles limited to the provided names for download/export."""
|
||||||
|
|
||||||
profiles = load_profiles()
|
profiles = load_profiles()
|
||||||
if names is None:
|
if names is None:
|
||||||
subset = profiles
|
subset = profiles
|
||||||
else:
|
else:
|
||||||
subset = {name: profiles[name] for name in names if name in profiles}
|
subset = {name: profiles[name] for name in names if name in profiles}
|
||||||
return {"abogen_voice_profiles": subset}
|
return {"abogen_voice_profiles": subset}
|
||||||
|
|||||||
@@ -20,7 +20,7 @@ import numpy as np
|
|||||||
import soundfile as sf
|
import soundfile as sf
|
||||||
import static_ffmpeg
|
import static_ffmpeg
|
||||||
|
|
||||||
from abogen.tts_backend_registry import get_metadata
|
from abogen.tts_backend_registry import get_metadata, is_registered_backend
|
||||||
from abogen.epub3.exporter import build_epub3_package
|
from abogen.epub3.exporter import build_epub3_package
|
||||||
from abogen.kokoro_text_normalization import ApostropheConfig, normalize_for_pipeline, HAS_NUM2WORDS
|
from abogen.kokoro_text_normalization import ApostropheConfig, normalize_for_pipeline, HAS_NUM2WORDS
|
||||||
from abogen.normalization_settings import (
|
from abogen.normalization_settings import (
|
||||||
@@ -1574,7 +1574,7 @@ def run_conversion_job(job: Job) -> None:
|
|||||||
def get_pipeline(provider: str) -> Any:
|
def get_pipeline(provider: str) -> Any:
|
||||||
nonlocal kokoro_cache_ready
|
nonlocal kokoro_cache_ready
|
||||||
provider_norm = str(provider or "kokoro").strip().lower() or "kokoro"
|
provider_norm = str(provider or "kokoro").strip().lower() or "kokoro"
|
||||||
if provider_norm not in {"kokoro", "supertonic"}:
|
if not is_registered_backend(provider_norm):
|
||||||
provider_norm = "kokoro"
|
provider_norm = "kokoro"
|
||||||
|
|
||||||
existing = pipelines.get(provider_norm)
|
existing = pipelines.get(provider_norm)
|
||||||
|
|||||||
@@ -34,6 +34,7 @@ from abogen.normalization_settings import (
|
|||||||
)
|
)
|
||||||
from abogen.llm_client import list_models, LLMClientError
|
from abogen.llm_client import list_models, LLMClientError
|
||||||
from abogen.kokoro_text_normalization import normalize_for_pipeline
|
from abogen.kokoro_text_normalization import normalize_for_pipeline
|
||||||
|
from abogen.tts_backend_registry import is_registered_backend
|
||||||
from abogen.integrations.audiobookshelf import AudiobookshelfClient, AudiobookshelfConfig
|
from abogen.integrations.audiobookshelf import AudiobookshelfClient, AudiobookshelfConfig
|
||||||
from abogen.integrations.calibre_opds import (
|
from abogen.integrations.calibre_opds import (
|
||||||
CalibreOPDSClient,
|
CalibreOPDSClient,
|
||||||
@@ -63,7 +64,7 @@ def api_save_voice_profile() -> ResponseReturnValue:
|
|||||||
if profile is None:
|
if profile is None:
|
||||||
# Speaker Studio payload format
|
# Speaker Studio payload format
|
||||||
provider = str(payload.get("provider") or "kokoro").strip().lower()
|
provider = str(payload.get("provider") or "kokoro").strip().lower()
|
||||||
if provider not in {"kokoro", "supertonic"}:
|
if not is_registered_backend(provider):
|
||||||
provider = "kokoro"
|
provider = "kokoro"
|
||||||
if provider == "supertonic":
|
if provider == "supertonic":
|
||||||
profile = {
|
profile = {
|
||||||
@@ -230,7 +231,7 @@ def api_speaker_preview() -> ResponseReturnValue:
|
|||||||
use_gpu = settings.get("use_gpu", False)
|
use_gpu = settings.get("use_gpu", False)
|
||||||
|
|
||||||
base_spec, speaker_name = split_profile_spec(voice)
|
base_spec, speaker_name = split_profile_spec(voice)
|
||||||
resolved_provider = tts_provider if tts_provider in {"kokoro", "supertonic"} else ""
|
resolved_provider = tts_provider if is_registered_backend(tts_provider) else ""
|
||||||
|
|
||||||
if speaker_name:
|
if speaker_name:
|
||||||
entry = normalize_profile_entry(load_profiles().get(speaker_name))
|
entry = normalize_profile_entry(load_profiles().get(speaker_name))
|
||||||
|
|||||||
@@ -7,6 +7,7 @@ from flask.typing import ResponseReturnValue
|
|||||||
|
|
||||||
from abogen.webui.service import PendingJob, JobStatus
|
from abogen.webui.service import PendingJob, JobStatus
|
||||||
from abogen.webui.routes.utils.service import get_service
|
from abogen.webui.routes.utils.service import get_service
|
||||||
|
from abogen.tts_backend_registry import is_registered_backend
|
||||||
from abogen.webui.routes.utils.settings import (
|
from abogen.webui.routes.utils.settings import (
|
||||||
load_settings,
|
load_settings,
|
||||||
coerce_bool,
|
coerce_bool,
|
||||||
@@ -579,7 +580,7 @@ def apply_book_step_form(
|
|||||||
# spec (e.g. "speaker:Name" for saved speakers, or a Kokoro mix formula).
|
# spec (e.g. "speaker:Name" for saved speakers, or a Kokoro mix formula).
|
||||||
# This enables mixed-provider conversions (e.g. narrator=SuperTonic, characters=Kokoro).
|
# This enables mixed-provider conversions (e.g. narrator=SuperTonic, characters=Kokoro).
|
||||||
provider_value = str(form.get("tts_provider") or "").strip().lower()
|
provider_value = str(form.get("tts_provider") or "").strip().lower()
|
||||||
if provider_value in {"kokoro", "supertonic"}:
|
if is_registered_backend(provider_value):
|
||||||
pending.tts_provider = provider_value
|
pending.tts_provider = provider_value
|
||||||
|
|
||||||
# Determine the base speaker selection (saved speaker ref or raw voice).
|
# Determine the base speaker selection (saved speaker ref or raw voice).
|
||||||
|
|||||||
Reference in New Issue
Block a user