diff --git a/CODEOWNERS b/CODEOWNERS index 66ea80f8d6..a0812c9cd6 100644 --- a/CODEOWNERS +++ b/CODEOWNERS @@ -490,7 +490,7 @@ esphome/components/vbus/* @ssieb esphome/components/veml3235/* @kbx81 esphome/components/veml7700/* @latonita esphome/components/version/* @esphome/core -esphome/components/voice_assistant/* @jesserockz +esphome/components/voice_assistant/* @jesserockz @kahrendt esphome/components/wake_on_lan/* @clydebarrow @willwill2will54 esphome/components/watchdog/* @oarcher esphome/components/waveshare_epaper/* @clydebarrow diff --git a/Doxyfile b/Doxyfile index a4dd65bf45..edbb5b1206 100644 --- a/Doxyfile +++ b/Doxyfile @@ -48,7 +48,7 @@ PROJECT_NAME = ESPHome # could be handy for archiving the generated documentation or if some version # control system is used. -PROJECT_NUMBER = 2025.6.1 +PROJECT_NUMBER = 2025.6.2 # Using the PROJECT_BRIEF tag one can provide an optional one line description # for a project that appears at the top of each page and should give viewer a diff --git a/esphome/components/audio/audio_reader.cpp b/esphome/components/audio/audio_reader.cpp index b82c6db9ee..6966c95db7 100644 --- a/esphome/components/audio/audio_reader.cpp +++ b/esphome/components/audio/audio_reader.cpp @@ -5,6 +5,7 @@ #include "esphome/core/defines.h" #include "esphome/core/hal.h" #include "esphome/core/helpers.h" +#include "esphome/core/log.h" #if CONFIG_MBEDTLS_CERTIFICATE_BUNDLE #include "esp_crt_bundle.h" @@ -16,13 +17,13 @@ namespace audio { static const uint32_t READ_WRITE_TIMEOUT_MS = 20; static const uint32_t CONNECTION_TIMEOUT_MS = 5000; - -// The number of times the http read times out with no data before throwing an error -static const uint32_t ERROR_COUNT_NO_DATA_READ_TIMEOUT = 100; +static const uint8_t MAX_FETCHING_HEADER_ATTEMPTS = 6; static const size_t HTTP_STREAM_BUFFER_SIZE = 2048; -static const uint8_t MAX_REDIRECTION = 5; +static const uint8_t MAX_REDIRECTIONS = 5; + +static const char *const TAG = "audio_reader"; // Some common HTTP status codes - borrowed from http_request component accessed 20241224 enum HttpStatus { @@ -94,7 +95,7 @@ esp_err_t AudioReader::start(const std::string &uri, AudioFileType &file_type) { client_config.url = uri.c_str(); client_config.cert_pem = nullptr; client_config.disable_auto_redirect = false; - client_config.max_redirection_count = 10; + client_config.max_redirection_count = MAX_REDIRECTIONS; client_config.event_handler = http_event_handler; client_config.user_data = this; client_config.buffer_size = HTTP_STREAM_BUFFER_SIZE; @@ -116,12 +117,29 @@ esp_err_t AudioReader::start(const std::string &uri, AudioFileType &file_type) { esp_err_t err = esp_http_client_open(this->client_, 0); if (err != ESP_OK) { + ESP_LOGE(TAG, "Failed to open URL"); this->cleanup_connection_(); return err; } int64_t header_length = esp_http_client_fetch_headers(this->client_); + uint8_t reattempt_count = 0; + while ((header_length < 0) && (reattempt_count < MAX_FETCHING_HEADER_ATTEMPTS)) { + this->cleanup_connection_(); + if (header_length != -ESP_ERR_HTTP_EAGAIN) { + // Serious error, no recovery + return ESP_FAIL; + } else { + // Reconnect from a fresh state to avoid a bug where it never reads the headers even if made available + this->client_ = esp_http_client_init(&client_config); + esp_http_client_open(this->client_, 0); + header_length = esp_http_client_fetch_headers(this->client_); + ++reattempt_count; + } + } + if (header_length < 0) { + ESP_LOGE(TAG, "Failed to fetch headers"); this->cleanup_connection_(); return ESP_FAIL; } @@ -135,7 +153,7 @@ esp_err_t AudioReader::start(const std::string &uri, AudioFileType &file_type) { ssize_t redirect_count = 0; - while ((esp_http_client_set_redirection(this->client_) == ESP_OK) && (redirect_count < MAX_REDIRECTION)) { + while ((esp_http_client_set_redirection(this->client_) == ESP_OK) && (redirect_count < MAX_REDIRECTIONS)) { err = esp_http_client_open(this->client_, 0); if (err != ESP_OK) { this->cleanup_connection_(); @@ -267,27 +285,29 @@ AudioReaderState AudioReader::http_read_() { return AudioReaderState::FINISHED; } } else if (this->output_transfer_buffer_->free() > 0) { - size_t bytes_to_read = this->output_transfer_buffer_->free(); - int received_len = - esp_http_client_read(this->client_, (char *) this->output_transfer_buffer_->get_buffer_end(), bytes_to_read); + int received_len = esp_http_client_read(this->client_, (char *) this->output_transfer_buffer_->get_buffer_end(), + this->output_transfer_buffer_->free()); if (received_len > 0) { this->output_transfer_buffer_->increase_buffer_length(received_len); this->last_data_read_ms_ = millis(); - } else if (received_len < 0) { + return AudioReaderState::READING; + } else if (received_len <= 0) { // HTTP read error - this->cleanup_connection_(); - return AudioReaderState::FAILED; - } else { - if (bytes_to_read > 0) { - // Read timed out - if ((millis() - this->last_data_read_ms_) > CONNECTION_TIMEOUT_MS) { - this->cleanup_connection_(); - return AudioReaderState::FAILED; - } - - delay(READ_WRITE_TIMEOUT_MS); + if (received_len == -1) { + // A true connection error occured, no chance at recovery + this->cleanup_connection_(); + return AudioReaderState::FAILED; } + + // Read timed out, manually verify if it has been too long since the last successful read + if ((millis() - this->last_data_read_ms_) > MAX_FETCHING_HEADER_ATTEMPTS * CONNECTION_TIMEOUT_MS) { + ESP_LOGE(TAG, "Timed out"); + this->cleanup_connection_(); + return AudioReaderState::FAILED; + } + + delay(READ_WRITE_TIMEOUT_MS); } } diff --git a/esphome/components/esp32/__init__.py b/esphome/components/esp32/__init__.py index 157fd9db11..ba9a4894c2 100644 --- a/esphome/components/esp32/__init__.py +++ b/esphome/components/esp32/__init__.py @@ -605,7 +605,7 @@ ESP_IDF_FRAMEWORK_SCHEMA = cv.All( CONF_ENABLE_LWIP_DHCP_SERVER, "wifi", default=False ): cv.boolean, cv.Optional( - CONF_ENABLE_LWIP_MDNS_QUERIES, default=False + CONF_ENABLE_LWIP_MDNS_QUERIES, default=True ): cv.boolean, cv.Optional( CONF_ENABLE_LWIP_BRIDGE_INTERFACE, default=False @@ -760,7 +760,7 @@ async def to_code(config): and not advanced[CONF_ENABLE_LWIP_DHCP_SERVER] ): add_idf_sdkconfig_option("CONFIG_LWIP_DHCPS", False) - if not advanced.get(CONF_ENABLE_LWIP_MDNS_QUERIES, False): + if not advanced.get(CONF_ENABLE_LWIP_MDNS_QUERIES, True): add_idf_sdkconfig_option("CONFIG_LWIP_DNS_SUPPORT_MDNS_QUERIES", False) if not advanced.get(CONF_ENABLE_LWIP_BRIDGE_INTERFACE, False): add_idf_sdkconfig_option("CONFIG_LWIP_BRIDGEIF_MAX_PORTS", 0) diff --git a/esphome/components/i2c/__init__.py b/esphome/components/i2c/__init__.py index e47dec650d..e20b7febeb 100644 --- a/esphome/components/i2c/__init__.py +++ b/esphome/components/i2c/__init__.py @@ -1,5 +1,8 @@ +import logging + from esphome import pins import esphome.codegen as cg +from esphome.components import esp32 import esphome.config_validation as cv from esphome.const import ( CONF_ADDRESS, @@ -12,6 +15,8 @@ from esphome.const import ( CONF_SCL, CONF_SDA, CONF_TIMEOUT, + KEY_CORE, + KEY_FRAMEWORK_VERSION, PLATFORM_ESP32, PLATFORM_ESP8266, PLATFORM_RP2040, @@ -19,6 +24,7 @@ from esphome.const import ( from esphome.core import CORE, coroutine_with_priority import esphome.final_validate as fv +LOGGER = logging.getLogger(__name__) CODEOWNERS = ["@esphome/core"] i2c_ns = cg.esphome_ns.namespace("i2c") I2CBus = i2c_ns.class_("I2CBus") @@ -40,6 +46,32 @@ def _bus_declare_type(value): raise NotImplementedError +def validate_config(config): + if ( + config[CONF_SCAN] + and CORE.is_esp32 + and CORE.using_esp_idf + and esp32.get_esp32_variant() + in [ + esp32.const.VARIANT_ESP32C5, + esp32.const.VARIANT_ESP32C6, + esp32.const.VARIANT_ESP32P4, + ] + ): + version: cv.Version = CORE.data[KEY_CORE][KEY_FRAMEWORK_VERSION] + if version.major == 5 and ( + (version.minor == 3 and version.patch <= 3) + or (version.minor == 4 and version.patch <= 1) + ): + LOGGER.warning( + "There is a bug in esp-idf version %s that breaks I2C scan, I2C scan " + "has been disabled, see https://github.com/esphome/issues/issues/7128", + str(version), + ) + config[CONF_SCAN] = False + return config + + pin_with_input_and_output_support = pins.internal_gpio_pin_number( {CONF_OUTPUT: True, CONF_INPUT: True} ) @@ -65,6 +97,7 @@ CONFIG_SCHEMA = cv.All( } ).extend(cv.COMPONENT_SCHEMA), cv.only_on([PLATFORM_ESP32, PLATFORM_ESP8266, PLATFORM_RP2040]), + validate_config, ) diff --git a/esphome/components/lvgl/widgets/qrcode.py b/esphome/components/lvgl/widgets/qrcode.py index 742b538938..7d8d13d8c4 100644 --- a/esphome/components/lvgl/widgets/qrcode.py +++ b/esphome/components/lvgl/widgets/qrcode.py @@ -3,7 +3,7 @@ import esphome.config_validation as cv from esphome.const import CONF_SIZE, CONF_TEXT from esphome.cpp_generator import MockObjClass -from ..defines import CONF_MAIN, literal +from ..defines import CONF_MAIN from ..lv_validation import color, color_retmapper, lv_text from ..lvcode import LocalVariable, lv, lv_expr from ..schemas import TEXT_SCHEMA @@ -34,7 +34,7 @@ class QrCodeType(WidgetType): ) def get_uses(self): - return ("canvas", "img") + return ("canvas", "img", "label") def obj_creator(self, parent: MockObjClass, config: dict): dark_color = color_retmapper(config[CONF_DARK_COLOR]) @@ -45,10 +45,8 @@ class QrCodeType(WidgetType): async def to_code(self, w: Widget, config): if (value := config.get(CONF_TEXT)) is not None: value = await lv_text.process(value) - with LocalVariable( - "qr_text", cg.const_char_ptr, value, modifier="" - ) as str_obj: - lv.qrcode_update(w.obj, str_obj, literal(f"strlen({str_obj})")) + with LocalVariable("qr_text", cg.std_string, value, modifier="") as str_obj: + lv.qrcode_update(w.obj, str_obj.c_str(), str_obj.size()) qr_code_spec = QrCodeType() diff --git a/esphome/components/mcp23xxx_base/mcp23xxx_base.cpp b/esphome/components/mcp23xxx_base/mcp23xxx_base.cpp index 14a703fb9f..fc49f216ee 100644 --- a/esphome/components/mcp23xxx_base/mcp23xxx_base.cpp +++ b/esphome/components/mcp23xxx_base/mcp23xxx_base.cpp @@ -6,7 +6,11 @@ namespace mcp23xxx_base { float MCP23XXXBase::get_setup_priority() const { return setup_priority::IO; } -void MCP23XXXGPIOPin::setup() { pin_mode(flags_); } +void MCP23XXXGPIOPin::setup() { + pin_mode(flags_); + this->parent_->pin_interrupt_mode(this->pin_, this->interrupt_mode_); +} + void MCP23XXXGPIOPin::pin_mode(gpio::Flags flags) { this->parent_->pin_mode(this->pin_, flags); } bool MCP23XXXGPIOPin::digital_read() { return this->parent_->digital_read(this->pin_) != this->inverted_; } void MCP23XXXGPIOPin::digital_write(bool value) { this->parent_->digital_write(this->pin_, value != this->inverted_); } diff --git a/esphome/components/speaker/media_player/audio_pipeline.cpp b/esphome/components/speaker/media_player/audio_pipeline.cpp index ac122b6e0c..333a076bec 100644 --- a/esphome/components/speaker/media_player/audio_pipeline.cpp +++ b/esphome/components/speaker/media_player/audio_pipeline.cpp @@ -343,13 +343,12 @@ void AudioPipeline::read_task(void *params) { xEventGroupSetBits(this_pipeline->event_group_, EventGroupBits::READER_MESSAGE_FINISHED); // Wait until the pipeline notifies us the source of the media file - EventBits_t event_bits = - xEventGroupWaitBits(this_pipeline->event_group_, - EventGroupBits::READER_COMMAND_INIT_FILE | EventGroupBits::READER_COMMAND_INIT_HTTP | - EventGroupBits::PIPELINE_COMMAND_STOP, // Bit message to read - pdFALSE, // Clear the bit on exit - pdFALSE, // Wait for all the bits, - portMAX_DELAY); // Block indefinitely until bit is set + EventBits_t event_bits = xEventGroupWaitBits( + this_pipeline->event_group_, + EventGroupBits::READER_COMMAND_INIT_FILE | EventGroupBits::READER_COMMAND_INIT_HTTP, // Bit message to read + pdFALSE, // Clear the bit on exit + pdFALSE, // Wait for all the bits, + portMAX_DELAY); // Block indefinitely until bit is set if (!(event_bits & EventGroupBits::PIPELINE_COMMAND_STOP)) { xEventGroupClearBits(this_pipeline->event_group_, EventGroupBits::READER_MESSAGE_FINISHED | @@ -434,12 +433,12 @@ void AudioPipeline::decode_task(void *params) { xEventGroupSetBits(this_pipeline->event_group_, EventGroupBits::DECODER_MESSAGE_FINISHED); // Wait until the reader notifies us that the media type is available - EventBits_t event_bits = xEventGroupWaitBits(this_pipeline->event_group_, - EventGroupBits::READER_MESSAGE_LOADED_MEDIA_TYPE | - EventGroupBits::PIPELINE_COMMAND_STOP, // Bit message to read - pdFALSE, // Clear the bit on exit - pdFALSE, // Wait for all the bits, - portMAX_DELAY); // Block indefinitely until bit is set + EventBits_t event_bits = + xEventGroupWaitBits(this_pipeline->event_group_, + EventGroupBits::READER_MESSAGE_LOADED_MEDIA_TYPE, // Bit message to read + pdFALSE, // Clear the bit on exit + pdFALSE, // Wait for all the bits, + portMAX_DELAY); // Block indefinitely until bit is set xEventGroupClearBits(this_pipeline->event_group_, EventGroupBits::DECODER_MESSAGE_FINISHED | EventGroupBits::READER_MESSAGE_LOADED_MEDIA_TYPE); diff --git a/esphome/components/voice_assistant/__init__.py b/esphome/components/voice_assistant/__init__.py index b9309ab422..59c7ec8383 100644 --- a/esphome/components/voice_assistant/__init__.py +++ b/esphome/components/voice_assistant/__init__.py @@ -17,10 +17,11 @@ from esphome.const import ( AUTO_LOAD = ["socket"] DEPENDENCIES = ["api", "microphone"] -CODEOWNERS = ["@jesserockz"] +CODEOWNERS = ["@jesserockz", "@kahrendt"] CONF_ON_END = "on_end" CONF_ON_INTENT_END = "on_intent_end" +CONF_ON_INTENT_PROGRESS = "on_intent_progress" CONF_ON_INTENT_START = "on_intent_start" CONF_ON_LISTENING = "on_listening" CONF_ON_START = "on_start" @@ -136,6 +137,9 @@ CONFIG_SCHEMA = cv.All( cv.Optional(CONF_ON_INTENT_START): automation.validate_automation( single=True ), + cv.Optional(CONF_ON_INTENT_PROGRESS): automation.validate_automation( + single=True + ), cv.Optional(CONF_ON_INTENT_END): automation.validate_automation( single=True ), @@ -282,6 +286,13 @@ async def to_code(config): config[CONF_ON_INTENT_START], ) + if CONF_ON_INTENT_PROGRESS in config: + await automation.build_automation( + var.get_intent_progress_trigger(), + [(cg.std_string, "x")], + config[CONF_ON_INTENT_PROGRESS], + ) + if CONF_ON_INTENT_END in config: await automation.build_automation( var.get_intent_end_trigger(), diff --git a/esphome/components/voice_assistant/voice_assistant.cpp b/esphome/components/voice_assistant/voice_assistant.cpp index a692a7556e..879d9492f0 100644 --- a/esphome/components/voice_assistant/voice_assistant.cpp +++ b/esphome/components/voice_assistant/voice_assistant.cpp @@ -555,7 +555,7 @@ void VoiceAssistant::request_stop() { break; case State::AWAITING_RESPONSE: this->signal_stop_(); - break; + // Fallthrough intended to stop a streaming TTS announcement that has potentially started case State::STREAMING_RESPONSE: #ifdef USE_MEDIA_PLAYER // Stop any ongoing media player announcement @@ -599,6 +599,14 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) { switch (msg.event_type) { case api::enums::VOICE_ASSISTANT_RUN_START: ESP_LOGD(TAG, "Assist Pipeline running"); +#ifdef USE_MEDIA_PLAYER + this->started_streaming_tts_ = false; + for (auto arg : msg.data) { + if (arg.name == "url") { + this->tts_response_url_ = std::move(arg.value); + } + } +#endif this->defer([this]() { this->start_trigger_->trigger(); }); break; case api::enums::VOICE_ASSISTANT_WAKE_WORD_START: @@ -622,6 +630,8 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) { if (text.empty()) { ESP_LOGW(TAG, "No text in STT_END event"); return; + } else if (text.length() > 500) { + text = text.substr(0, 497) + "..."; } ESP_LOGD(TAG, "Speech recognised as: \"%s\"", text.c_str()); this->defer([this, text]() { this->stt_end_trigger_->trigger(text); }); @@ -631,6 +641,27 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) { ESP_LOGD(TAG, "Intent started"); this->defer([this]() { this->intent_start_trigger_->trigger(); }); break; + case api::enums::VOICE_ASSISTANT_INTENT_PROGRESS: { + ESP_LOGD(TAG, "Intent progress"); + std::string tts_url_for_trigger = ""; +#ifdef USE_MEDIA_PLAYER + if (this->media_player_ != nullptr) { + for (const auto &arg : msg.data) { + if ((arg.name == "tts_start_streaming") && (arg.value == "1") && !this->tts_response_url_.empty()) { + this->media_player_->make_call().set_media_url(this->tts_response_url_).set_announcement(true).perform(); + + this->media_player_wait_for_announcement_start_ = true; + this->media_player_wait_for_announcement_end_ = false; + this->started_streaming_tts_ = true; + tts_url_for_trigger = this->tts_response_url_; + this->tts_response_url_.clear(); // Reset streaming URL + } + } + } +#endif + this->defer([this, tts_url_for_trigger]() { this->intent_progress_trigger_->trigger(tts_url_for_trigger); }); + break; + } case api::enums::VOICE_ASSISTANT_INTENT_END: { for (auto arg : msg.data) { if (arg.name == "conversation_id") { @@ -653,6 +684,9 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) { ESP_LOGW(TAG, "No text in TTS_START event"); return; } + if (text.length() > 500) { + text = text.substr(0, 497) + "..."; + } ESP_LOGD(TAG, "Response: \"%s\"", text.c_str()); this->defer([this, text]() { this->tts_start_trigger_->trigger(text); @@ -678,7 +712,7 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) { ESP_LOGD(TAG, "Response URL: \"%s\"", url.c_str()); this->defer([this, url]() { #ifdef USE_MEDIA_PLAYER - if (this->media_player_ != nullptr) { + if ((this->media_player_ != nullptr) && (!this->started_streaming_tts_)) { this->media_player_->make_call().set_media_url(url).set_announcement(true).perform(); this->media_player_wait_for_announcement_start_ = true; diff --git a/esphome/components/voice_assistant/voice_assistant.h b/esphome/components/voice_assistant/voice_assistant.h index 865731522f..2424ea6052 100644 --- a/esphome/components/voice_assistant/voice_assistant.h +++ b/esphome/components/voice_assistant/voice_assistant.h @@ -177,6 +177,7 @@ class VoiceAssistant : public Component { Trigger<> *get_intent_end_trigger() const { return this->intent_end_trigger_; } Trigger<> *get_intent_start_trigger() const { return this->intent_start_trigger_; } + Trigger *get_intent_progress_trigger() const { return this->intent_progress_trigger_; } Trigger<> *get_listening_trigger() const { return this->listening_trigger_; } Trigger<> *get_end_trigger() const { return this->end_trigger_; } Trigger<> *get_start_trigger() const { return this->start_trigger_; } @@ -233,6 +234,7 @@ class VoiceAssistant : public Component { Trigger<> *tts_stream_start_trigger_ = new Trigger<>(); Trigger<> *tts_stream_end_trigger_ = new Trigger<>(); #endif + Trigger *intent_progress_trigger_ = new Trigger(); Trigger<> *wake_word_detected_trigger_ = new Trigger<>(); Trigger *stt_end_trigger_ = new Trigger(); Trigger *tts_end_trigger_ = new Trigger(); @@ -268,6 +270,8 @@ class VoiceAssistant : public Component { #endif #ifdef USE_MEDIA_PLAYER media_player::MediaPlayer *media_player_{nullptr}; + std::string tts_response_url_{""}; + bool started_streaming_tts_{false}; bool media_player_wait_for_announcement_start_{false}; bool media_player_wait_for_announcement_end_{false}; #endif diff --git a/esphome/const.py b/esphome/const.py index 3ef82aff5b..956edfd6be 100644 --- a/esphome/const.py +++ b/esphome/const.py @@ -1,6 +1,6 @@ """Constants used by esphome.""" -__version__ = "2025.6.1" +__version__ = "2025.6.2" ALLOWED_NAME_CHARS = "abcdefghijklmnopqrstuvwxyz0123456789-_" VALID_SUBSTITUTIONS_CHARACTERS = ( diff --git a/tests/components/lvgl/lvgl-package.yaml b/tests/components/lvgl/lvgl-package.yaml index d8452bdd2a..e3391a5bac 100644 --- a/tests/components/lvgl/lvgl-package.yaml +++ b/tests/components/lvgl/lvgl-package.yaml @@ -646,7 +646,9 @@ lvgl: on_click: lvgl.qrcode.update: id: lv_qr - text: homeassistant.io + text: + format: "A string with a number %d" + args: ['(int)(random_uint32() % 1000)'] - slider: min_value: 0