Merge pull request #53 from OpenVoiceOS/dev

OpenVoiceOS · Jun 1, 2023 · e985034 · e985034
2 parents 5f3f3b3 + 89d5a6c
commit e985034
Show file tree

Hide file tree

Showing 6 changed files with 162 additions and 75 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,63 +1,32 @@
 # Changelog
 
-## [V0.0.9a6](https://github.com/OpenVoiceOS/ovos-config/tree/V0.0.9a6) (2023-05-24)
+## [V0.0.10a3](https://github.com/OpenVoiceOS/ovos-config/tree/V0.0.10a3) (2023-05-31)
 
-[Full Changelog](https://github.com/OpenVoiceOS/ovos-config/compare/V0.0.9a5...V0.0.9a6)
+[Full Changelog](https://github.com/OpenVoiceOS/ovos-config/compare/V0.0.10a2...V0.0.10a3)
 
-**Implemented enhancements:**
+**Fixed bugs:**
 
-- Port `init_module_config` from neon\_utils with unit tests [\#42](https://github.com/OpenVoiceOS/ovos-config/pull/42) ([NeonDaniel](https://github.com/NeonDaniel))
+- Update config hash to sort keys to ensure hashes compare [\#52](https://github.com/OpenVoiceOS/ovos-config/pull/52) ([NeonDaniel](https://github.com/NeonDaniel))
 
-**Closed issues:**
-
-- Add method to init configuration [\#22](https://github.com/OpenVoiceOS/ovos-config/issues/22)
-
-## [V0.0.9a5](https://github.com/OpenVoiceOS/ovos-config/tree/V0.0.9a5) (2023-05-24)
-
-[Full Changelog](https://github.com/OpenVoiceOS/ovos-config/compare/V0.0.9a4...V0.0.9a5)
-
-**Implemented enhancements:**
+## [V0.0.10a2](https://github.com/OpenVoiceOS/ovos-config/tree/V0.0.10a2) (2023-05-31)
 
-- os.env [\#46](https://github.com/OpenVoiceOS/ovos-config/issues/46)
-- Support envvar default configuration overrides [\#47](https://github.com/OpenVoiceOS/ovos-config/pull/47) ([NeonDaniel](https://github.com/NeonDaniel))
-
-## [V0.0.9a4](https://github.com/OpenVoiceOS/ovos-config/tree/V0.0.9a4) (2023-05-23)
-
-[Full Changelog](https://github.com/OpenVoiceOS/ovos-config/compare/V0.0.9a3...V0.0.9a4)
-
-**Merged pull requests:**
-
-- Add configuration update unit tests and update documentation [\#45](https://github.com/OpenVoiceOS/ovos-config/pull/45) ([NeonDaniel](https://github.com/NeonDaniel))
-
-## [V0.0.9a3](https://github.com/OpenVoiceOS/ovos-config/tree/V0.0.9a3) (2023-05-19)
-
-[Full Changelog](https://github.com/OpenVoiceOS/ovos-config/compare/V0.0.9a2...V0.0.9a3)
+[Full Changelog](https://github.com/OpenVoiceOS/ovos-config/compare/V0.0.10a1...V0.0.10a2)
 
 **Closed issues:**
 
-- 0.0.9a2 - Page not found [\#41](https://github.com/OpenVoiceOS/ovos-config/issues/41)
-- 0.0.9a2 - Page not found [\#40](https://github.com/OpenVoiceOS/ovos-config/issues/40)
-
-**Merged pull requests:**
-
-- Fix setup.py bug [\#44](https://github.com/OpenVoiceOS/ovos-config/pull/44) ([NeonDaniel](https://github.com/NeonDaniel))
-- Update build\_tests to use shared action [\#43](https://github.com/OpenVoiceOS/ovos-config/pull/43) ([NeonDaniel](https://github.com/NeonDaniel))
-
-## [V0.0.9a2](https://github.com/OpenVoiceOS/ovos-config/tree/V0.0.9a2) (2023-05-07)
-
-[Full Changelog](https://github.com/OpenVoiceOS/ovos-config/compare/V0.0.9a1...V0.0.9a2)
+- Clean install failed on missing ovos-tts-plugin-mimic [\#49](https://github.com/OpenVoiceOS/ovos-config/issues/49)
 
 **Merged pull requests:**
 
-- fix setup [\#39](https://github.com/OpenVoiceOS/ovos-config/pull/39) ([emphasize](https://github.com/emphasize))
+- Skip config file reload when a non-config file is modified [\#51](https://github.com/OpenVoiceOS/ovos-config/pull/51) ([NeonDaniel](https://github.com/NeonDaniel))
 
-## [V0.0.9a1](https://github.com/OpenVoiceOS/ovos-config/tree/V0.0.9a1) (2023-05-01)
+## [V0.0.10a1](https://github.com/OpenVoiceOS/ovos-config/tree/V0.0.10a1) (2023-05-29)
 
-[Full Changelog](https://github.com/OpenVoiceOS/ovos-config/compare/V0.0.8...V0.0.9a1)
+[Full Changelog](https://github.com/OpenVoiceOS/ovos-config/compare/V0.0.9...V0.0.10a1)
 
 **Merged pull requests:**
 
-- Bump PyYAML to 6.0 [\#38](https://github.com/OpenVoiceOS/ovos-config/pull/38) ([goldyfruit](https://github.com/goldyfruit))
+- refactor/better\_defaults [\#50](https://github.com/OpenVoiceOS/ovos-config/pull/50) ([JarbasAl](https://github.com/JarbasAl))
 
 
 

diff --git a/ovos_config/config.py b/ovos_config/config.py
@@ -278,7 +278,7 @@ def set_config_watcher(callback: Optional[callable] = None):
         """
         paths = [Configuration.system.path] + \
                 [c.path for c in Configuration.xdg_configs]
-        if callback:
+        if callback and callback not in Configuration._callbacks:
             Configuration._callbacks.append(callback)
         if not Configuration._watchdog:
             Configuration._watchdog = FileWatcher(
@@ -292,36 +292,36 @@ def _on_file_change(path: str):
         Callback method for FileWatcher
         @param path: Configuration file path reporting a change
         """
-        LOG.info(f'{path} changed on disk, reloading!')
         # reload updated config
-        for cfg in Configuration.xdg_configs + [Configuration.system]:
+        for cfg in Configuration.xdg_configs + [Configuration.system,
+                                                Configuration.remote]:
             if cfg.path == path:
+                old_cfg = hash(cfg)
                 try:
                     cfg.reload()
-                    break
+                    reloaded = True
                 except:
                     # got the file changed signal before write finished
                     sleep(0.5)
+                    reloaded = False
                 try:
-                    cfg.reload()
+                    if not reloaded:
+                        LOG.warning(f"Reload failed, retrying")
+                        cfg.reload()
                 except:
                     LOG.exception("Failed to load configuration, "
                                   "syntax seems invalid!")
+                new_cfg = hash(cfg)
+                if old_cfg == new_cfg:
+                    LOG.info(f"{path} unchanged")
+                    return
                 break
         else:
-            LOG.info(f"Reloading all configuration files, got: {path}")
-            # reload all configs
-            try:
-                Configuration.reload()
-            except:
-                # got the file changed signal before write finished
-                sleep(0.5)
-                try:
-                    Configuration.reload()
-                except:
-                    LOG.exception("Failed to load configuration, "
-                                  "syntax seems invalid!")
+            LOG.info(f"Ignoring non-config file change: {path}")
+            return
 
+        LOG.info(f'{path} changed on disk')
+        LOG.debug(f"Calling {len(Configuration._callbacks)} callbacks")
         for handler in Configuration._callbacks:
             try:
                 handler()

diff --git a/ovos_config/models.py b/ovos_config/models.py
@@ -64,6 +64,9 @@ def __init__(self, path):
         if path:
             self.load_local(path)
 
+    def __hash__(self):
+        return hash(json.dumps(dict(self), sort_keys=True))
+
     def _get_file_format(self, path=None):
         """The config file format
         supported file extensions:

diff --git a/ovos_config/mycroft.conf b/ovos_config/mycroft.conf
@@ -22,6 +22,10 @@
   // TODO: save unmodified, lowercase upon demand
   "lang": "en-us",
 
+  // Secondary languages will also have their resource files loaded into memory
+  // but intents will only be considered if that lang is tagged with the utterance at STT step
+  "secondary_langs": [],
+
   // Measurement units, either 'metric' or 'imperial'
   // Override: REMOTE
   "system_unit": "metric",
@@ -50,16 +54,19 @@
   },
 
   // Mechanism used to play WAV audio files
+  // by default ovos-utils will try to detect best player
   // Override: SYSTEM
-  "play_wav_cmdline": "paplay %1 --stream-name=mycroft-voice",
+  //"play_wav_cmdline": "paplay %1 --stream-name=mycroft-voice",
 
   // Mechanism used to play MP3 audio files
+  // by default ovos-utils will try to detect best player
   // Override: SYSTEM
-  "play_mp3_cmdline": "mpg123 %1",
+  //"play_mp3_cmdline": "mpg123 %1",
 
   // Mechanism used to play OGG audio files
+  // by default ovos-utils will try to detect best player
   // Override: SYSTEM
-  "play_ogg_cmdline": "ogg123 -q %1",
+  //"play_ogg_cmdline": "ogg123 -q %1",
 
   // Location where the system resides
   // NOTE: Although this is set here, an Enclosure can override the value.
@@ -140,6 +147,8 @@
     "sync2way": true,
     // values in skill settings missing in settingsmeta wont show up in selene
     // this flag auto generates settingsmeta on the fly
+    // TODO - settings this to true may run into permission issues until settingsmeta.json gets XDG support
+    // it writes to skill base folder
     "autogen_meta": false,
 
     // blacklisted skills to not load
@@ -272,11 +281,10 @@
   // Needs to be explicitly enabled, also see "protected_keys" under "system" above
   // Possible backends
   // - https://github.com/OpenVoiceOS/OVOS-local-backend
-  // - https://mycroft-ai.gitbook.io/docs/using-mycroft-ai/pairing-your-device
   "server": {
-    // Valid types: offline, selene, personal, neon, ovos
+    // Valid types: offline, personal
     "backend_type": "offline",
-    // url for selene -> https://api.mycroft.ai
+    // url for self hosted personal backend
     "url": "",
     "version": "v1",
     "update": false,
@@ -346,7 +354,7 @@
     // if enabled the noise level is saved to a ipc file, useful for
     // debuging if microphone is working but writes a lot to disk,
     // recommended that you set "ipc_path" to a tmpfs
-    "mic_meter_ipc": true,
+    "mic_meter_ipc": false,
 
     // Set 'save_path' to configure the location of files stored if
     // 'record_wake_words' and/or 'save_utterances' are set to 'true'.
@@ -368,6 +376,13 @@
       "url": ""
     },
 
+    // Microphone plugin to be read audio
+    // NOTE: only in ovos-dinkum-listener
+    "microphone": {
+      // use "ovos-microphone-plugin-sounddevice" for MacOS support
+      "module": "ovos-microphone-plugin-alsa"
+    },
+
     // Voice Activity Detection is used to determine when speech ended
     "VAD": {
         // silence method defined the main vad strategy
@@ -413,7 +428,7 @@
     // Stop listing to the microphone during playback to prevent accidental triggering
     // This is enabled by default, but instances with good microphone noise cancellation
     // can disable this to listen all the time, allowing 'barge in' functionality.
-    "mute_during_output" : true,
+    "mute_during_output" : false,
 
     // How much (if at all) to 'duck' the speaker output during listening.  A
     // setting of 0.0 will not duck at all.  A 1.0 will completely mute output
@@ -440,11 +455,11 @@
     // instant listen is an experimental setting, it removes the need for
     // the pause between "hey mycroft" and starting to speak the utterance,
     // this setting might slightly downgrade STT accuracy depending on engine used
-    "instant_listen": false,
+    "instant_listen": true,
 
     // continuous listen is an experimental setting, it removes the need for
     // wake words and uses VAD only, a streaming STT is strongly recommended
-    // this setting might downgrade STT accuracy depending on engine used
+    // this setting might downgrade STT accuracy depending on plugins used
     "continuous_listen": false,
 
     // hybrid listen is an experimental setting,
@@ -510,7 +525,7 @@
   },
 
    // DEPRECATED: the concept of enclosure will no longer exist in ovos-core
-   // this has been replaced with PHAL
+   // this has been replaced with PHAL, provides backwards compat for mycroft.client.enclosure module
   "enclosure": {
     // Platform name
     // Override: SYSTEM (set by specific enclosures)
@@ -556,7 +571,7 @@
   // Overrride: none
   "session": {
     // Time To Live, in seconds
-    "ttl": 180
+    "ttl": -1
   },
 
   // Speech to Text parameters
@@ -575,7 +590,7 @@
   "tts": {
     "pulse_duck": false,
     "module": "ovos-tts-plugin-mimic3-server",
-    "fallback_module": "ovos-tts-plugin-mimic",
+    "fallback_module": "",
     "ovos-tts-plugin-mimic": {
         "voice": "ap"
     },
@@ -584,15 +599,62 @@
     }
   },
 
+  // NLP plugins
+  // split utterances into words
+  // used by downstream tasks, handles lang specific nuances
+  "tokenization": {
+      // default plugin comes bundled with ovos-plugin-manager
+      "module": "ovos-tokenization-plugin-quebrafrases"
+  },
+  // split utterances into sub commands
+  // used to extract multiple intents from a single utterance
+  "segmentation": {
+      // default plugin comes bundled with ovos-plugin-manager
+      "module": "ovos-segmentation-plugin-quebrafrases"
+  },
+  // given an utterance extract keywords from it
+  // can be used to get search terms
+  "keyword_extract": {
+     // default plugin comes bundled with ovos-classifiers
+     "module": "ovos-keyword-extractor-heuristic"
+  },
+  // tag pronouns with the word they refer to
+  // used in normalization step
+  "coref": {
+     // default plugin comes bundled with ovos-classifiers
+     "module": "ovos-coref-solver-heuristic"
+  },
+  "postag": {
+     // default plugin comes bundled with ovos-classifiers
+     "module": "ovos-classifiers-postag-plugin"
+  },
+  // turn utterances into phoneme sequences, used to generate mouth movements
+  "g2p": {
+     "module": "ovos-g2p-plugin-dummy"
+  },
+
+  // DEPRECATED: this section is in the process of being fully removed,
+  // only providing default values for clean migration
   "padatious": {
     "intent_cache": "~/.local/share/mycroft/intent_cache",
     "train_delay": 4,
     "single_thread": false,
     // fallback settings for padacioso (pure regex)
-    "regex_only": false,
+    // set regex_only to False to re-enable padatious (ovos-core < 0.0.8)
+    "regex_only": true,
     "fuzz": true
   },
 
+  // Translation plugins
+  "language": {
+    // default plugin comes bundled with ovos-classifiers
+    "detection_module": "ovos-lang-detect-ngram-lm",
+    // default is hosted by Neon but can be self hosted
+    // https://github.com/NeonGeckoCom/neon-lang-plugin-libretranslate
+    "translation_module": "libretranslate_plug"
+  },
+
+  // Media playback
   "Audio": {
     // message.context may contains a source and destination
     // native audio (playback / TTS) will only be played if a

diff --git a/ovos_config/version.py b/ovos_config/version.py
@@ -1,6 +1,6 @@
 # START_VERSION_BLOCK
 VERSION_MAJOR = 0
 VERSION_MINOR = 0
-VERSION_BUILD = 9
+VERSION_BUILD = 10
 VERSION_ALPHA = 0
 # END_VERSION_BLOCK