Merge branch 'dev' into api-sint32-short-varint

.ai/instructions.md

@@ -239,6 +239,123 @@ This document provides essential context for AI models interacting with this pro
               var = await switch.new_switch(config)
           ```
 
+*   **Automations (Triggers, Actions, Conditions):**
+
+    Automations have three building blocks: **Triggers** (fire when something happens), **Actions** (do something), and **Conditions** (check if something is true).
+
+    *   **Triggers -- Callback method (preferred):**
+
+        Use `build_callback_automation()` for simple triggers. This eliminates the need for a C++ Trigger class by using a lightweight pointer-sized forwarder struct registered directly as a callback. No `CONF_TRIGGER_ID` in the schema.
+
+        **Python:**
+        ```python
+        from esphome import automation
+
+        CONFIG_SCHEMA = cv.Schema({
+            cv.GenerateID(): cv.declare_id(MyComponent),
+            cv.Optional(CONF_ON_STATE): automation.validate_automation({}),
+        }).extend(cv.COMPONENT_SCHEMA)
+
+        async def to_code(config):
+            var = cg.new_Pvariable(config[CONF_ID])
+            await cg.register_component(var, config)
+            for conf in config.get(CONF_ON_STATE, []):
+                await automation.build_callback_automation(
+                    var, "add_on_state_callback", [(bool, "x")], conf
+                )
+        ```
+
+        `build_callback_automation` arguments: `parent`, `callback_method` (C++ method name), `args` (template args as `[(type, name)]` tuples), `config`, and optional `forwarder` (defaults to `TriggerForwarder<Ts...>`).
+
+        For boolean filtering (e.g. `on_press`/`on_release`), use built-in forwarders with `args=[]`:
+        ```python
+        for conf_key, forwarder in (
+            (CONF_ON_PRESS, automation.TriggerOnTrueForwarder),
+            (CONF_ON_RELEASE, automation.TriggerOnFalseForwarder),
+        ):
+            for conf in config.get(conf_key, []):
+                await automation.build_callback_automation(
+                    var, "add_on_state_callback", [], conf, forwarder=forwarder
+                )
+        ```
+
+        **C++ -- no trigger class needed.** The callback registration method must be templatized to accept both `std::function` and lightweight forwarder structs (which avoid heap allocation):
+        ```cpp
+        class MyComponent : public Component {
+         public:
+          // Must be a template -- accepts both std::function and pointer-sized forwarder structs
+          template<typename F> void add_on_state_callback(F &&callback) {
+            this->state_callback_.add(std::forward<F>(callback));
+          }
+         protected:
+          // Use CallbackManager when callbacks are always registered (e.g. core components)
+          CallbackManager<void(bool)> state_callback_;
+          // Use LazyCallbackManager when callbacks are often not registered -- saves 8 bytes
+          // (nullptr vs empty std::vector) per instance when no callbacks are added
+          // LazyCallbackManager<void(bool)> state_callback_;
+        };
+        ```
+
+    *   **Triggers -- Trigger class method:**
+
+        Use `build_automation()` with a `Trigger<Ts...>` subclass only when the forwarder needs **mutable state beyond a single `Automation*` pointer** (e.g. edge detection tracking previous state, timing logic).
+
+        **Python:**
+        ```python
+        TurnOnTrigger = my_ns.class_("TurnOnTrigger", automation.Trigger.template())
+
+        CONFIG_SCHEMA = cv.Schema({
+            cv.Optional(CONF_ON_TURN_ON): automation.validate_automation(
+                {cv.GenerateID(CONF_TRIGGER_ID): cv.declare_id(TurnOnTrigger)}
+            ),
+        })
+
+        async def to_code(config):
+            for conf in config.get(CONF_ON_TURN_ON, []):
+                trigger = cg.new_Pvariable(conf[CONF_TRIGGER_ID], var)
+                await automation.build_automation(trigger, [], conf)
+        ```
+
+        **C++:**
+        ```cpp
+        class TurnOnTrigger : public Trigger<> {
+         public:
+          explicit TurnOnTrigger(MyComponent *parent) : last_on_{false} {
+            parent->add_on_state_callback([this](bool state) {
+              if (state && !this->last_on_)
+                this->trigger();
+              this->last_on_ = state;
+            });
+          }
+         protected:
+          bool last_on_;
+        };
+        ```
+
+    *   **Actions:**
+        ```cpp
+        template<typename... Ts> class MyAction : public Action<Ts...> {
+         public:
+          explicit MyAction(MyComponent *parent) : parent_(parent) {}
+          void play(const Ts &...) override { this->parent_->do_something(); }
+         protected:
+          MyComponent *parent_;
+        };
+        ```
+        Register with `@automation.register_action("my_component.do_something", MyAction, schema, synchronous=True)`. Use `synchronous=True` for actions that run to completion inside `play()` without deferring. Use `synchronous=False` if the action may suspend/defer execution (e.g. `delay`, `wait_until`, `script.wait`) or store trigger arguments for later use.
+
+    *   **Conditions:**
+        ```cpp
+        template<typename... Ts> class MyCondition : public Condition<Ts...> {
+         public:
+          explicit MyCondition(MyComponent *parent) : parent_(parent) {}
+          bool check(const Ts &...) override { return this->parent_->is_active(); }
+         protected:
+          MyComponent *parent_;
+        };
+        ```
+        Register with `@automation.register_condition("my_component.is_active", MyCondition, schema)`.
+
 *   **Configuration Validation:**
     *   **Common Validators:** `cv.int_`, `cv.float_`, `cv.string`, `cv.boolean`, `cv.int_range(min=0, max=100)`, `cv.positive_int`, `cv.percentage`.
     *   **Complex Validation:** `cv.All(cv.string, cv.Length(min=1, max=50))`, `cv.Any(cv.int_, cv.string)`.
@@ -274,10 +391,39 @@ This document provides essential context for AI models interacting with this pro
     *   **Component Tests:** YAML-based compilation tests are located in `tests/`. The structure is as follows:
         ```
         tests/
-        ├── test_build_components/ # Base test configurations
-        └── components/[component]/ # Component-specific tests
+        ├── test_build_components/
+        │   └── common/          # Shared bus packages (uart, i2c, spi, etc.)
+        │       ├── uart/        # UART at default baud rate
+        │       ├── uart_115200/ # UART at 115200 baud
+        │       ├── i2c/         # I2C bus
+        │       └── spi/         # SPI bus
+        └── components/[component]/
+            ├── common.yaml      # Component-only config (no bus definitions)
+            ├── test.esp32-idf.yaml
+            ├── test.esp8266-ard.yaml
+            └── test.rp2040-ard.yaml
         ```
         Run them using `script/test_build_components`. Use `-c <component>` to test specific components and `-t <target>` for specific platforms.
+
+    *   **Test Grouping with Packages:** Components that use shared bus packages can be grouped together in CI to reduce build count. **Never define buses (uart, i2c, spi, modbus) directly in test YAML files** — always use packages from `test_build_components/common/`:
+        ```yaml
+        # test.esp32-idf.yaml — use packages for buses
+        packages:
+          uart: !include ../../test_build_components/common/uart_115200/esp32-idf.yaml
+
+        <<: !include common.yaml
+        ```
+        ```yaml
+        # common.yaml — component config only, NO bus definitions
+        my_component:
+          id: my_instance
+
+        sensor:
+          - platform: my_component
+            name: My Sensor
+        ```
+        Components that define buses directly are flagged as "NEEDS MIGRATION" and cannot be grouped, increasing CI build time.
+
     *   **Testing All Components Together:** To verify that all components can be tested together without ID conflicts or configuration issues, use:
         ```bash
         ./script/test_component_grouping.py -e config --all
@@ -417,6 +563,30 @@ This document provides essential context for AI models interacting with this pro
 
         Note: Avoiding heap allocation after `setup()` is always required regardless of component type. The prioritization above is about the effort spent on container optimization (e.g., migrating from `std::vector` to `StaticVector`).
 
+        **Callback Managers:**
+
+        ESPHome provides two callback manager types in `esphome/core/helpers.h` for the observer pattern. Both support `std::function`, lambdas, and lightweight forwarder structs via their templatized `add()` method.
+
+        | Type | Idle overhead (32-bit) | When to use |
+        |------|----------------------|-------------|
+        | `CallbackManager<void(Ts...)>` | 12 bytes (empty `std::vector`) | Callbacks are always or almost always registered |
+        | `LazyCallbackManager<void(Ts...)>` | 4 bytes (`nullptr`) | Callbacks are often not registered (common case) |
+
+        `LazyCallbackManager` is a drop-in replacement for `CallbackManager` that defers allocation until the first callback is added. Prefer it for entity-level callbacks where most instances have no subscribers.
+
+        **Important:** Registration methods that add to a callback manager **must always be templatized** to accept both `std::function` and pointer-sized forwarder structs (used by `build_callback_automation`). Never use `std::function` in the method signature:
+        ```cpp
+        // Bad -- forces heap allocation for forwarder structs
+        void add_on_state_callback(std::function<void(bool)> &&callback) {
+          this->state_callback_.add(std::move(callback));
+        }
+
+        // Good -- accepts any callable without forcing std::function wrapping
+        template<typename F> void add_on_state_callback(F &&callback) {
+          this->state_callback_.add(std::forward<F>(callback));
+        }
+        ```
+
     *   **State Management:** Use `CORE.data` for component state that needs to persist during configuration generation. Avoid module-level mutable globals.
 
         **Bad Pattern (Module-Level Globals):**

esphome/components/binary_sensor/binary_sensor.cpp

@@ -32,13 +32,6 @@ void BinarySensor::publish_initial_state(bool new_state) {
   this->invalidate_state();
   this->publish_state(new_state);
 }
-void BinarySensor::send_state_internal(bool new_state) {
-  // copy the new state to the visible property for backwards compatibility, before any callbacks
-  this->state = new_state;
-  // Note that set_new_state_ de-dups and will only trigger callbacks if the state has actually changed
-  this->set_new_state(new_state);
-}
-
 bool BinarySensor::set_new_state(const optional<bool> &new_state) {
   if (StatefulEntityBase::set_new_state(new_state)) {
     // weirdly, this file could be compiled even without USE_BINARY_SENSOR defined

esphome/components/binary_sensor/binary_sensor.h

@@ -32,7 +32,10 @@ void log_binary_sensor(const char *tag, const char *prefix, const char *type, Bi
  */
 class BinarySensor : public StatefulEntityBase<bool> {
  public:
-  explicit BinarySensor(){};
+  explicit BinarySensor() = default;
+
+  const bool &get_state() const override { return this->state; }
+  void set_trigger_on_initial_state(bool value) { this->trigger_on_initial_state_ = value; }
 
   /** Publish a new state to the front-end.
    *
@@ -54,16 +57,24 @@ class BinarySensor : public StatefulEntityBase<bool> {
 
   // ========== INTERNAL METHODS ==========
   // (In most use cases you won't need these)
-  void send_state_internal(bool new_state);
+  void send_state_internal(bool new_state) {
+    // Fast path: skip virtual dispatch when state hasn't changed
+    if (this->flags_.has_state && this->state == new_state)
+      return;
+    this->set_new_state(new_state);
+  }
 
   /// Return whether this binary sensor has outputted a state.
   virtual bool is_status_binary_sensor() const;
 
-  // For backward compatibility, provide an accessible property
-
+  /// The current state of this binary sensor. Also used as the backing storage for StatefulEntityBase.
   bool state{};
 
  protected:
+  bool get_trigger_on_initial_state() const override { return this->trigger_on_initial_state_; }
+  void set_state_value(const bool &value) override { this->state = value; }
+
+  bool trigger_on_initial_state_{true};
 #ifdef USE_BINARY_SENSOR_FILTER
   Filter *filter_list_{nullptr};
 #endif
@@ -73,7 +84,7 @@ class BinarySensor : public StatefulEntityBase<bool> {
 
 class BinarySensorInitiallyOff : public BinarySensor {
  public:
-  bool has_state() const override { return true; }
+  BinarySensorInitiallyOff() { this->set_has_state(true); }
 };
 
 }  // namespace esphome::binary_sensor

esphome/components/esp8266/__init__.py

@@ -1,5 +1,6 @@
 import logging
 from pathlib import Path
+import re
 
 import esphome.codegen as cg
 import esphome.config_validation as cv
@@ -18,8 +19,9 @@ from esphome.const import (
     PLATFORM_ESP8266,
     ThreadModel,
 )
-from esphome.core import CORE, CoroPriority, coroutine_with_priority
+from esphome.core import CORE, CoroPriority, Lambda, coroutine_with_priority
 from esphome.helpers import copy_file_if_changed
+from esphome.types import ConfigType
 
 from .boards import BOARDS, ESP8266_LD_SCRIPTS
 from .const import (
@@ -40,12 +42,42 @@ from .const import (
 )
 from .gpio import PinInitialState, add_pin_initial_states_array
 
+CONF_ENABLE_SCANF_FLOAT = "enable_scanf_float"
+# Heuristically matches scanf/sscanf calls with float format specifiers.
+# Standard scanf float conversions: %f %F %e %E %g %G %a %A
+# With optional modifiers: %*f (suppression), %8f (width), %lf %Lf (length)
+# Also matches non-standard patterns like %.2f as a heuristic — these are
+# invalid in scanf but users may write them by analogy with printf.
+# Uses [^;]*? to stay within a single statement, preventing false positives
+# from e.g. sscanf(buf, "%d", &x); printf("%f", val);
+_SCANF_FLOAT_RE = re.compile(r"scanf\s*\([^;]*?%[*\d.]*[hlL]*[feEgGaAF]")
+
 CODEOWNERS = ["@esphome/core"]
 _LOGGER = logging.getLogger(__name__)
 AUTO_LOAD = ["preferences"]
 IS_TARGET_PLATFORM = True
 
 
+def lambdas_use_scanf_float(config: ConfigType) -> bool:
+    """Check if any lambda in the config uses scanf with a float format specifier.
+
+    Comments are stripped before matching to avoid false positives from
+    commented-out code. The cost of a false positive is only ~8KB flash.
+    """
+    stack: list = [config]
+    while stack:
+        obj = stack.pop()
+        if isinstance(obj, Lambda):
+            src = obj.comment_remover(obj.value)
+            if _SCANF_FLOAT_RE.search(src):
+                return True
+        elif isinstance(obj, dict):
+            stack.extend(obj.values())
+        elif isinstance(obj, list):
+            stack.extend(obj)
+    return False
+
+
 def set_core_data(config):
     CORE.data[KEY_ESP8266] = {}
     CORE.data[KEY_CORE][KEY_TARGET_PLATFORM] = PLATFORM_ESP8266
@@ -181,6 +213,7 @@ CONFIG_SCHEMA = cv.All(
             cv.Optional(CONF_ENABLE_SERIAL): cv.boolean,
             cv.Optional(CONF_ENABLE_SERIAL1): cv.boolean,
             cv.Optional(CONF_ENABLE_FULL_PRINTF, default=False): cv.boolean,
+            cv.Optional(CONF_ENABLE_SCANF_FLOAT): cv.boolean,
         }
     ),
     set_core_data,
@@ -201,16 +234,23 @@ async def to_code(config):
     cg.add_define("ESPHOME_VARIANT", "ESP8266")
     cg.add_define(ThreadModel.SINGLE)
 
-    cg.add_platformio_option(
-        "extra_scripts",
-        [
-            "pre:testing_mode.py",
-            "pre:exclude_updater.py",
-            "pre:exclude_waveform.py",
-            "pre:remove_float_scanf.py",
-            "post:post_build.py",
-        ],
-    )
+    enable_scanf_float = config.get(CONF_ENABLE_SCANF_FLOAT)
+    if enable_scanf_float is None and lambdas_use_scanf_float(CORE.config):
+        enable_scanf_float = True
+        _LOGGER.warning(
+            "Lambda uses scanf with a float format specifier; "
+            "enabling scanf float support (~8KB flash)"
+        )
+
+    extra_scripts = [
+        "pre:testing_mode.py",
+        "pre:exclude_updater.py",
+        "pre:exclude_waveform.py",
+    ]
+    if not enable_scanf_float:
+        extra_scripts.append("pre:remove_float_scanf.py")
+    extra_scripts.append("post:post_build.py")
+    cg.add_platformio_option("extra_scripts", extra_scripts)
 
     conf = config[CONF_FRAMEWORK]
     cg.add_platformio_option("framework", "arduino")

esphome/components/sensor/__init__.py

@@ -381,7 +381,7 @@ async def filter_out_filter_to_code(config, filter_id):
     if not isinstance(config, list):
         config = [config]
     template_ = [await cg.templatable(x, [], float) for x in config]
-    return cg.new_Pvariable(filter_id, template_)
+    return cg.new_Pvariable(filter_id, cg.TemplateArguments(len(template_)), template_)
 
 
 QUANTILE_SCHEMA = cv.All(
@@ -650,7 +650,9 @@ async def throttle_with_priority_filter_to_code(config, filter_id):
     if not isinstance(config[CONF_VALUE], list):
         config[CONF_VALUE] = [config[CONF_VALUE]]
     template_ = [await cg.templatable(x, [], float) for x in config[CONF_VALUE]]
-    return cg.new_Pvariable(filter_id, config[CONF_TIMEOUT], template_)
+    return cg.new_Pvariable(
+        filter_id, cg.TemplateArguments(len(template_)), config[CONF_TIMEOUT], template_
+    )
 
 
 HEARTBEAT_SCHEMA = cv.Schema(

esphome/components/sensor/filter.cpp

@@ -222,16 +222,14 @@ MultiplyFilter::MultiplyFilter(TemplatableValue<float> multiplier) : multiplier_
 
 optional<float> MultiplyFilter::new_value(float value) { return value * this->multiplier_.value(); }
 
-// ValueListFilter (base class)
-ValueListFilter::ValueListFilter(std::initializer_list<TemplatableValue<float>> values) : values_(values) {}
-
-bool ValueListFilter::value_matches_any_(float sensor_value) {
-  int8_t accuracy = this->parent_->get_accuracy_decimals();
+// ValueListFilter helper (non-template, shared by all ValueListFilter<N> instantiations)
+bool value_list_matches_any(Sensor *parent, float sensor_value, const TemplatableValue<float> *values, size_t count) {
+  int8_t accuracy = parent->get_accuracy_decimals();
   float accuracy_mult = pow10_int(accuracy);
   float rounded_sensor = roundf(accuracy_mult * sensor_value);
 
-  for (auto &filter_value : this->values_) {
-    float fv = filter_value.value();
+  for (size_t i = 0; i < count; i++) {
+    float fv = values[i].value();
 
     // Handle NaN comparison
     if (std::isnan(fv)) {
@@ -248,16 +246,6 @@ bool ValueListFilter::value_matches_any_(float sensor_value) {
   return false;
 }
 
-// FilterOutValueFilter
-FilterOutValueFilter::FilterOutValueFilter(std::initializer_list<TemplatableValue<float>> values_to_filter_out)
-    : ValueListFilter(values_to_filter_out) {}
-
-optional<float> FilterOutValueFilter::new_value(float value) {
-  if (this->value_matches_any_(value))
-    return {};   // Filter out
-  return value;  // Pass through
-}
-
 // ThrottleFilter
 ThrottleFilter::ThrottleFilter(uint32_t min_time_between_inputs) : min_time_between_inputs_(min_time_between_inputs) {}
 optional<float> ThrottleFilter::new_value(float value) {
@@ -269,17 +257,13 @@ optional<float> ThrottleFilter::new_value(float value) {
   return {};
 }
 
-// ThrottleWithPriorityFilter
-ThrottleWithPriorityFilter::ThrottleWithPriorityFilter(
-    uint32_t min_time_between_inputs, std::initializer_list<TemplatableValue<float>> prioritized_values)
-    : ValueListFilter(prioritized_values), min_time_between_inputs_(min_time_between_inputs) {}
-
-optional<float> ThrottleWithPriorityFilter::new_value(float value) {
+// ThrottleWithPriorityFilter helper (non-template, keeps App access in .cpp)
+optional<float> throttle_with_priority_new_value(Sensor *parent, float value, const TemplatableValue<float> *values,
+                                                 size_t count, uint32_t &last_input, uint32_t min_time_between_inputs) {
   const uint32_t now = App.get_loop_component_start_time();
-  // Allow value through if: no previous input, time expired, or is prioritized
-  if (this->last_input_ == 0 || now - this->last_input_ >= min_time_between_inputs_ ||
-      this->value_matches_any_(value)) {
-    this->last_input_ = now;
+  if (last_input == 0 || now - last_input >= min_time_between_inputs ||
+      value_list_matches_any(parent, value, values, count)) {
+    last_input = now;
     return value;
   }
   return {};

esphome/components/sensor/filter.h

@@ -3,6 +3,7 @@
 #include "esphome/core/defines.h"
 #ifdef USE_SENSOR_FILTER
 
+#include <array>
 #include <utility>
 #include <vector>
 #include "esphome/core/automation.h"
@@ -328,28 +329,42 @@ class MultiplyFilter : public Filter {
   TemplatableValue<float> multiplier_;
 };
 
-/** Base class for filters that compare sensor values against a list of configured values.
+/// Non-template helper for value matching (implementation in filter.cpp)
+bool value_list_matches_any(Sensor *parent, float sensor_value, const TemplatableValue<float> *values, size_t count);
+
+/** Base class for filters that compare sensor values against a fixed list of configured values.
  *
- * This base class provides common functionality for filters that need to check if a sensor
- * value matches any value in a configured list, with proper handling of NaN values and
- * accuracy-based rounding for comparisons.
+ * Templated on N (the number of values) so the list is stored inline in a std::array,
+ * avoiding heap allocation and the overhead of FixedVector.
+ *
+ * @tparam N Number of values in the filter list, set by code generation to match
+ *           the exact number of values configured in YAML.
  */
-class ValueListFilter : public Filter {
+template<size_t N> class ValueListFilter : public Filter {
  protected:
-  explicit ValueListFilter(std::initializer_list<TemplatableValue<float>> values);
+  explicit ValueListFilter(std::initializer_list<TemplatableValue<float>> values) {
+    init_array_from(this->values_, values);
+  }
 
   /// Check if sensor value matches any configured value (with accuracy rounding)
-  bool value_matches_any_(float sensor_value);
+  bool value_matches_any_(float sensor_value) {
+    return value_list_matches_any(this->parent_, sensor_value, this->values_.data(), N);
+  }
 
-  FixedVector<TemplatableValue<float>> values_;
+  std::array<TemplatableValue<float>, N> values_{};
 };
 
 /// A simple filter that only forwards the filter chain if it doesn't receive `value_to_filter_out`.
-class FilterOutValueFilter : public ValueListFilter {
+template<size_t N> class FilterOutValueFilter : public ValueListFilter<N> {
  public:
-  explicit FilterOutValueFilter(std::initializer_list<TemplatableValue<float>> values_to_filter_out);
+  explicit FilterOutValueFilter(std::initializer_list<TemplatableValue<float>> values_to_filter_out)
+      : ValueListFilter<N>(values_to_filter_out) {}
 
-  optional<float> new_value(float value) override;
+  optional<float> new_value(float value) override {
+    if (this->value_matches_any_(value))
+      return {};   // Filter out
+    return value;  // Pass through
+  }
 };
 
 class ThrottleFilter : public Filter {
@@ -363,13 +378,21 @@ class ThrottleFilter : public Filter {
   uint32_t min_time_between_inputs_;
 };
 
+/// Non-template helper for ThrottleWithPriorityFilter (implementation in filter.cpp)
+optional<float> throttle_with_priority_new_value(Sensor *parent, float value, const TemplatableValue<float> *values,
+                                                 size_t count, uint32_t &last_input, uint32_t min_time_between_inputs);
+
 /// Same as 'throttle' but will immediately publish values contained in `value_to_prioritize`.
-class ThrottleWithPriorityFilter : public ValueListFilter {
+template<size_t N> class ThrottleWithPriorityFilter : public ValueListFilter<N> {
  public:
   explicit ThrottleWithPriorityFilter(uint32_t min_time_between_inputs,
-                                      std::initializer_list<TemplatableValue<float>> prioritized_values);
+                                      std::initializer_list<TemplatableValue<float>> prioritized_values)
+      : ValueListFilter<N>(prioritized_values), min_time_between_inputs_(min_time_between_inputs) {}
 
-  optional<float> new_value(float value) override;
+  optional<float> new_value(float value) override {
+    return throttle_with_priority_new_value(this->parent_, value, this->values_.data(), N, this->last_input_,
+                                            this->min_time_between_inputs_);
+  }
 
  protected:
   uint32_t last_input_{0};

esphome/components/version/version_text_sensor.cpp

@@ -1,6 +1,5 @@
 #include "version_text_sensor.h"
 #include "esphome/core/application.h"
-#include "esphome/core/build_info_data.h"
 #include "esphome/core/helpers.h"
 #include "esphome/core/log.h"
 #include "esphome/core/progmem.h"
@@ -36,7 +35,9 @@ void VersionTextSensor::setup() {
   if (!this->hide_timestamp_) {
     size_t len = strlen(version_str);
     ESPHOME_strncat_P(version_str, BUILT_STR, sizeof(version_str) - len - 1);
-    ESPHOME_strncat_P(version_str, ESPHOME_BUILD_TIME_STR, sizeof(version_str) - strlen(version_str) - 1);
+    char build_time_buf[Application::BUILD_TIME_STR_SIZE];
+    App.get_build_time_string(build_time_buf);
+    strncat(version_str, build_time_buf, sizeof(version_str) - strlen(version_str) - 1);
   }
 
   // The closing parenthesis is part of the config-hash suffix and must

esphome/components/web_server_base/web_server_base.h

@@ -135,7 +135,7 @@ class WebServerBase {
   uint16_t get_port() const { return port_; }
 
  protected:
-  int initialized_{0};
+  uint8_t initialized_{0};
   uint16_t port_{80};
   AsyncWebServer *server_{nullptr};
   std::vector<AsyncWebHandler *> handlers_;

esphome/core/component.cpp

@@ -84,6 +84,8 @@ void store_component_error_message(const Component *component, const char *messa
 
 static constexpr uint16_t WARN_IF_BLOCKING_INCREMENT_MS =
     10U;  ///< How long the blocking time must be larger to warn again
+// Threshold in ms (computed from centiseconds constant in component.h)
+static constexpr uint32_t WARN_IF_BLOCKING_OVER_MS = static_cast<uint32_t>(WARN_IF_BLOCKING_OVER_CS) * 10U;
 
 float Component::get_setup_priority() const { return setup_priority::DATA; }
 
@@ -268,15 +270,18 @@ void Component::call() {
       break;
   }
 }
+const LogString *Component::get_component_log_str() const {
+  return component_source_lookup(this->component_source_index_);
+}
 bool Component::should_warn_of_blocking(uint32_t blocking_time) {
-  if (blocking_time > this->warn_if_blocking_over_) {
-    // Prevent overflow when adding increment - if we're about to overflow, just max out
-    if (blocking_time + WARN_IF_BLOCKING_INCREMENT_MS < blocking_time ||
-        blocking_time + WARN_IF_BLOCKING_INCREMENT_MS > std::numeric_limits<uint16_t>::max()) {
-      this->warn_if_blocking_over_ = std::numeric_limits<uint16_t>::max();
-    } else {
-      this->warn_if_blocking_over_ = static_cast<uint16_t>(blocking_time + WARN_IF_BLOCKING_INCREMENT_MS);
-    }
+  // Convert centisecond threshold to milliseconds for comparison
+  uint32_t threshold_ms = static_cast<uint32_t>(this->warn_if_blocking_over_) * 10U;
+  if (blocking_time > threshold_ms) {
+    // Set new threshold: blocking_time + increment, converted back to centiseconds
+    uint32_t new_threshold_ms = blocking_time + WARN_IF_BLOCKING_INCREMENT_MS;
+    uint32_t new_cs = new_threshold_ms / 10U;
+    // Saturate at uint8_t max (255 = 2550ms)
+    this->warn_if_blocking_over_ = static_cast<uint8_t>(new_cs > 255U ? 255U : new_cs);
     return true;
   }
   return false;
@@ -537,4 +542,7 @@ void clear_setup_priority_overrides() {
 }
 #endif
 
+// Weak default for component_source_lookup - overridden by generated code
+__attribute__((weak)) const LogString *component_source_lookup(uint8_t) { return LOG_STR("<unknown>"); }
+
 }  // namespace esphome

esphome/core/component.h

@@ -11,6 +11,10 @@
 #include "esphome/core/log.h"
 #include "esphome/core/optional.h"
 
+// Forward declarations for friend access from codegen-generated setup()
+void setup();           // NOLINT(readability-redundant-declaration) - may be declared in Arduino.h
+void original_setup();  // NOLINT(readability-redundant-declaration)
+
 namespace esphome {
 
 // Forward declaration for LogString
@@ -79,11 +83,14 @@ inline constexpr uint8_t STATUS_LED_WARNING = 0x08;
 inline constexpr uint8_t STATUS_LED_ERROR = 0x10;
 // Component loop override flag uses bit 5 (set at registration time)
 inline constexpr uint8_t COMPONENT_HAS_LOOP = 0x20;
-
 // Remove before 2026.8.0
 enum class RetryResult { DONE, RETRY };
 
-inline constexpr uint16_t WARN_IF_BLOCKING_OVER_MS = 50U;
+inline constexpr uint8_t WARN_IF_BLOCKING_OVER_CS = 5U;  // 50ms in centiseconds (1cs = 10ms)
+
+/// Lookup component source name by index (1-based). Generated by Python codegen.
+/// Weak default returns "<unknown>" so builds without codegen still link.
+const LogString *component_source_lookup(uint8_t index);
 
 class Component {
  public:
@@ -275,23 +282,25 @@ class Component {
 
   bool has_overridden_loop() const { return (this->component_state_ & COMPONENT_HAS_LOOP) != 0; }
 
-  /** Set where this component was loaded from for some debug messages.
-   *
-   * This is set by the ESPHome core, and should not be called manually.
-   */
-  void set_component_source(const LogString *source) { component_source_ = source; }
   /** Get the integration where this component was declared as a LogString for logging.
    *
    * Returns LOG_STR("<unknown>") if source not set
    */
-  const LogString *get_component_log_str() const {
-    return this->component_source_ == nullptr ? LOG_STR("<unknown>") : this->component_source_;
-  }
+  const LogString *get_component_log_str() const;
 
   bool should_warn_of_blocking(uint32_t blocking_time);
 
  protected:
   friend class Application;
+  friend void ::setup();
+  friend void ::original_setup();
+
+  /** Set where this component was loaded from for some debug messages.
+   *
+   * This is set by the ESPHome core during setup, and should not be called manually.
+   * @param index 1-based index into the component source lookup table (0 = not set)
+   */
+  void set_component_source_(uint8_t index) { this->component_source_index_ = index; }
 
   virtual void call_setup();
   void call_dump_config_();
@@ -509,9 +518,9 @@ class Component {
   void status_clear_warning_slow_path_();
   void status_clear_error_slow_path_();
 
-  // Ordered for optimal packing on 32-bit systems
-  const LogString *component_source_{nullptr};
-  uint16_t warn_if_blocking_over_{WARN_IF_BLOCKING_OVER_MS};  ///< Warn if blocked for this many ms (max 65.5s)
+  // Ordered for optimal packing on 32-bit systems (8 bytes total with vtable)
+  uint8_t component_source_index_{0};  ///< Index into component source PROGMEM lookup table (0 = not set)
+  uint8_t warn_if_blocking_over_{WARN_IF_BLOCKING_OVER_CS};  ///< Warn threshold in centiseconds (max 2550ms)
   /// State of this component - each bit has a purpose:
   /// Bits 0-2: Component state (0x00=CONSTRUCTION, 0x01=SETUP, 0x02=LOOP, 0x03=FAILED, 0x04=LOOP_DONE)
   /// Bit 3: STATUS_LED_WARNING
@@ -588,6 +597,8 @@ class WarnIfComponentBlockingGuard {
     this->record_runtime_stats_();
 #endif
 #ifndef USE_BENCHMARK
+    // Fast path: compare against constant threshold in ms (computed at compile time from centiseconds)
+    static constexpr uint32_t WARN_IF_BLOCKING_OVER_MS = static_cast<uint32_t>(WARN_IF_BLOCKING_OVER_CS) * 10U;
     if (blocking_time > WARN_IF_BLOCKING_OVER_MS) [[unlikely]] {
       warn_blocking(this->component_, blocking_time);
     }

esphome/core/entity_base.h

@@ -296,15 +296,36 @@ void log_entity_device_class(const char *tag, const char *prefix, const EntityBa
 #define LOG_ENTITY_UNIT_OF_MEASUREMENT(tag, prefix, obj) log_entity_unit_of_measurement(tag, prefix, obj)
 void log_entity_unit_of_measurement(const char *tag, const char *prefix, const EntityBase &obj);
 
-/**
- * An entity that has a state.
- * @tparam T The type of the state
+/** Base class for entities that track a typed state value with change-detection and callbacks.
+ *
+ * This class does not store the state value — subclasses own their storage. Whether a state
+ * has been set is tracked by EntityBase::has_state().
+ *
+ * Subclasses must implement:
+ *   - get_state(): return a const reference to the current value
+ *   - set_state_value(): store a new value (called only when the state actually changes)
+ *   - get_trigger_on_initial_state(): return whether callbacks should fire on the first state
+ *
+ * Subclasses may override set_new_state() to add behavior (logging, notifications) after calling
+ * the base implementation. Since set_new_state() is virtual, callers like invalidate_state()
+ * dispatch through the vtable to the subclass override in the .cpp, avoiding template code
+ * bloat at inline call sites. Subclasses may also add a fast-path dedup check before calling
+ * set_new_state() to skip virtual dispatch entirely when the state hasn't changed.
+ *
+ * Callback behavior:
+ *   - full_state_callbacks_: fired on every change, receives optional<T> previous and current
+ *   - state_callbacks_: fired only when the new state has a value, and either this is not the
+ *     first state (had_state) or trigger_on_initial_state is set
+ *
+ * @tparam T The type of the state value
  */
 template<typename T> class StatefulEntityBase : public EntityBase {
  public:
-  virtual bool has_state() const { return this->state_.has_value(); }
-  virtual const T &get_state() const { return this->state_.value(); }  // NOLINT(bugprone-unchecked-optional-access)
-  virtual T get_state_default(T default_value) const { return this->state_.value_or(default_value); }
+  /// Return the current state value. Only valid when has_state() is true.
+  virtual const T &get_state() const = 0;
+  /// Return the current state if available, otherwise return the provided default.
+  T get_state_default(T default_value) const { return this->has_state() ? this->get_state() : default_value; }
+  /// Clear the state — sets has_state() to false and fires callbacks with nullopt.
   void invalidate_state() { this->set_new_state({}); }
 
   template<typename F> void add_full_state_callback(F &&callback) {
@@ -314,33 +335,46 @@ template<typename T> class StatefulEntityBase : public EntityBase {
     this->state_callbacks_.add(std::forward<F>(callback));
   }
 
-  void set_trigger_on_initial_state(bool trigger_on_initial_state) {
-    this->trigger_on_initial_state_ = trigger_on_initial_state;
-  }
-
  protected:
-  optional<T> state_{};
-  /**
-   * Set a new state for this entity. This will trigger callbacks only if the new state is different from the previous.
+  /// Subclasses return whether callbacks should fire on the very first state.
+  virtual bool get_trigger_on_initial_state() const = 0;
+
+  /** Apply a new state, de-duplicating and firing callbacks as needed.
    *
-   * @param new_state The new state.
-   * @return True if the state was changed, false if it was the same as before.
+   * Pass nullopt to invalidate (clear) the state. Pass a value to set it.
+   * Returns true if the state actually changed, false if it was the same.
+   * Subclasses may override to add logging/notifications after calling the base.
    */
   virtual bool set_new_state(const optional<T> &new_state) {
-    if (this->state_ != new_state) {
-      // call the full state callbacks with the previous and new state
-      this->full_state_callbacks_.call(this->state_, new_state);
-      // trigger legacy callbacks only if the new state is valid and either the trigger on initial state is enabled or
-      // the previous state was valid
-      auto had_state = this->has_state();
-      this->state_ = new_state;
-      if (new_state.has_value() && (this->trigger_on_initial_state_ || had_state))
-        this->state_callbacks_.call(new_state.value());
-      return true;
+    // Access flags_ directly to avoid function call overhead in this hot path
+    bool had_state = this->flags_.has_state;
+    // Use pointer to avoid requiring T to be default-constructible
+    const T *current = had_state ? &this->get_state() : nullptr;
+    if (new_state.has_value()) {
+      if (current != nullptr && *current == new_state.value())
+        return false;  // same value, no change
+    } else if (!had_state) {
+      return false;  // already invalidated, no change
     }
-    return false;
+    // Capture old_state before set_state_value — current pointer aliases subclass storage
+    bool has_full_cbs = !this->full_state_callbacks_.empty();
+    optional<T> old_state;
+    if (has_full_cbs)
+      old_state = current != nullptr ? optional<T>(*current) : nullopt;
+    // Update storage before firing callbacks so callback code can inspect current state
+    this->flags_.has_state = new_state.has_value();
+    if (new_state.has_value()) {
+      this->set_state_value(new_state.value());
+    }
+    if (has_full_cbs)
+      this->full_state_callbacks_.call(old_state, new_state);
+    // had_state first: on every change except the first, skips the virtual call
+    if (new_state.has_value() && (had_state || this->get_trigger_on_initial_state()))
+      this->state_callbacks_.call(new_state.value());
+    return true;
   }
-  bool trigger_on_initial_state_{true};
+  /// Subclasses implement this to store the actual value into their own storage.
+  virtual void set_state_value(const T &value) = 0;
   LazyCallbackManager<void(optional<T> previous, optional<T> current)> full_state_callbacks_;
   LazyCallbackManager<void(T)> state_callbacks_;
 };

esphome/core/helpers.h

@@ -2,6 +2,7 @@
 
 #include <algorithm>
 #include <array>
+#include <cassert>
 #include <cmath>
 #include <cstdarg>
 #include <cstdint>
@@ -497,6 +498,23 @@ template<typename T, size_t MAX_CAPACITY = std::numeric_limits<uint16_t>::max()>
   index_type capacity_{0};
 };
 
+/// Initialize a std::array from an initializer_list. Uses memcpy for trivially copyable types (optimal codegen),
+/// falls back to element-wise copy for non-trivially copyable types (e.g. TemplatableValue).
+/// N is set by code generation; assert catches mismatches in debug/integration tests.
+template<typename T, size_t N> inline void init_array_from(std::array<T, N> &dest, std::initializer_list<T> src) {
+#ifdef ESPHOME_DEBUG
+  assert(src.size() == N);
+#endif
+  if constexpr (std::is_trivially_copyable_v<T>) {
+    __builtin_memcpy(dest.data(), src.begin(), N * sizeof(T));
+  } else {
+    size_t i = 0;
+    for (const auto &v : src) {
+      dest[i++] = v;
+    }
+  }
+}
+
 /// Fixed-capacity vector - allocates once at runtime, never reallocates
 /// This avoids std::vector template overhead (_M_realloc_insert, _M_default_append)
 /// when size is known at initialization but not at compile time

esphome/cpp_helpers.py

@@ -1,3 +1,4 @@
+from dataclasses import dataclass, field
 import logging
 
 from esphome.const import (
@@ -7,15 +8,130 @@ from esphome.const import (
     CONF_UPDATE_INTERVAL,
     KEY_PAST_SAFE_MODE,
 )
-from esphome.core import CORE, ID, coroutine
+from esphome.core import CORE, ID, CoroPriority, coroutine, coroutine_with_priority
 from esphome.coroutine import FakeAwaitable
-from esphome.cpp_generator import LogStringLiteral, add, add_define, get_variable
+from esphome.cpp_generator import (
+    RawStatement,
+    add,
+    add_define,
+    add_global,
+    get_variable,
+)
 from esphome.cpp_types import App
+from esphome.helpers import cpp_string_escape
 from esphome.types import ConfigFragmentType, ConfigType
 from esphome.util import Registry, RegistryEntry
 
 _LOGGER = logging.getLogger(__name__)
 
+_COMPONENT_SOURCE_DOMAIN = "component_source_pool"
+
+# Maximum unique component source names (8-bit index, 0 = not set)
+_MAX_COMPONENT_SOURCES = 0xFF  # 255
+
+
+@dataclass
+class ComponentSourcePool:
+    """Pool of component source names for PROGMEM lookup table.
+
+    Source names are registered during to_code() and assigned 1-based indices.
+    Index 0 means "not set" (returns LOG_STR("<unknown>")). At render time,
+    the pool generates a C++ PROGMEM table + lookup function.
+    """
+
+    sources: dict[str, int] = field(default_factory=dict)
+    table_registered: bool = False
+
+
+def _get_source_pool() -> ComponentSourcePool:
+    """Get or create the component source pool from CORE.data."""
+    if _COMPONENT_SOURCE_DOMAIN not in CORE.data:
+        CORE.data[_COMPONENT_SOURCE_DOMAIN] = ComponentSourcePool()
+    return CORE.data[_COMPONENT_SOURCE_DOMAIN]
+
+
+def _ensure_source_table_registered() -> None:
+    """Schedule the table generation job (once)."""
+    pool = _get_source_pool()
+    if pool.table_registered:
+        return
+    pool.table_registered = True
+    CORE.add_job(_generate_component_source_table)
+
+
+def register_component_source(name: str) -> int:
+    """Register a component source name and return its 1-based index.
+
+    Deduplicates: multiple components from the same source share one index.
+    """
+    if not name:
+        return 0
+    pool = _get_source_pool()
+    if name in pool.sources:
+        return pool.sources[name]
+    idx = len(pool.sources) + 1
+    if idx > _MAX_COMPONENT_SOURCES:
+        _LOGGER.warning(
+            "Too many unique component source names (max %d), '%s' will show as '<unknown>'",
+            _MAX_COMPONENT_SOURCES,
+            name,
+        )
+        return 0
+    pool.sources[name] = idx
+    _ensure_source_table_registered()
+    return idx
+
+
+def _generate_source_table_code(
+    table_var: str,
+    lookup_fn: str,
+    strings: dict[str, int],
+) -> str:
+    """Generate C++ PROGMEM table + LogString* lookup for component sources.
+
+    Same pattern as entity_helpers._generate_category_code but returns
+    const LogString* instead of const char* (needed for LOG_STR_ARG).
+    """
+    if not strings:
+        return ""
+
+    sorted_strings = sorted(strings.items(), key=lambda x: x[1])
+    count = len(sorted_strings)
+
+    # Emit individual PROGMEM char arrays so string data lives in flash on ESP8266
+    lines: list[str] = []
+    var_names: list[str] = []
+    for i, (s, _) in enumerate(sorted_strings):
+        var_name = f"{table_var}_STR_{i}"
+        var_names.append(var_name)
+        lines.append(
+            f"static const char {var_name}[] PROGMEM = {cpp_string_escape(s)};"
+        )
+
+    entries = ", ".join(var_names)
+    lines.append(f"static const char *const {table_var}[] PROGMEM = {{{entries}}};")
+    lines.append(f"const LogString *{lookup_fn}(uint8_t index) {{")
+    lines.append(f'  if (index == 0 || index > {count}) return LOG_STR("<unknown>");')
+    lines.append("  return reinterpret_cast<const LogString *>(")
+    lines.append(f"    progmem_read_ptr(&{table_var}[index - 1]));")
+    lines.append("}")
+    return "\n".join(lines) + "\n"
+
+
+@coroutine_with_priority(CoroPriority.FINAL)
+async def _generate_component_source_table() -> None:
+    """Generate the component source lookup table as a FINAL-priority job.
+
+    Runs after all component to_code() calls have registered their sources.
+    """
+    pool = _get_source_pool()
+    if code := _generate_source_table_code(
+        "COMP_SRC_TABLE", "component_source_lookup", pool.sources
+    ):
+        add_global(
+            RawStatement(f"namespace esphome {{\n{code}}}  // namespace esphome")
+        )
+
 
 async def gpio_pin_expression(conf):
     """Generate an expression for the given pin option.
@@ -77,7 +193,8 @@ async def register_component(var, config):
             "Error while finding name of component, please report this", exc_info=e
         )
     if name is not None:
-        add(var.set_component_source(LogStringLiteral(name)))
+        idx = register_component_source(name)
+        add(var.set_component_source_(idx))
 
     add(App.register_component_(var))
 

tests/components/esp8266/test.esp8266-ard.yaml

@@ -14,3 +14,6 @@ esphome:
         assert(x == 95);
         x = clamp_at_most(x, 40);
         assert(x == 40);
+    - lambda: |-
+        float value = 0.0f;
+        sscanf("3.14", "%f", &value);

tests/integration/test_uart_mock_ld2410.py

@@ -73,9 +73,16 @@ async def test_uart_mock_ld2410(
         ],
     )
 
-    # Signal when we see recovery frame values
+    # Signal when we see ALL recovery frame values to avoid race where some
+    # arrive after the waiter fires but before we index into the lists
     recovery_received = collector.add_waiter(
-        lambda: pytest.approx(50.0) in collector.sensor_states["moving_distance"]
+        lambda: (
+            pytest.approx(50.0) in collector.sensor_states["moving_distance"]
+            and pytest.approx(75.0) in collector.sensor_states["still_distance"]
+            and pytest.approx(100.0) in collector.sensor_states["moving_energy"]
+            and pytest.approx(80.0) in collector.sensor_states["still_energy"]
+            and pytest.approx(127.0) in collector.sensor_states["detection_distance"]
+        )
     )
 
     async with (

tests/unit_tests/components/test_esp8266.py

@@ -0,0 +1,62 @@
+"""Tests for ESP8266 component."""
+
+import pytest
+
+from esphome.components.esp8266 import lambdas_use_scanf_float
+from esphome.core import Lambda
+from esphome.types import ConfigType
+
+
+@pytest.mark.parametrize(
+    ("src", "expected"),
+    [
+        # Basic float formats
+        ('sscanf(buf, "%f", &v)', True),
+        ('sscanf(buf, "%F", &v)', True),
+        ('sscanf(buf, "%e", &v)', True),
+        ('sscanf(buf, "%E", &v)', True),
+        ('sscanf(buf, "%g", &v)', True),
+        ('sscanf(buf, "%G", &v)', True),
+        ('sscanf(buf, "%a", &v)', True),
+        ('sscanf(buf, "%A", &v)', True),
+        # With modifiers
+        ('sscanf(buf, "%lf", &v)', True),
+        ('sscanf(buf, "%Lf", &v)', True),
+        ('sscanf(buf, "%8lf", &v)', True),
+        ('sscanf(buf, "%*f")', True),
+        ('sscanf(buf, "%.2f", &v)', True),
+        # Mixed formats
+        ('sscanf(buf, "%d,%f", &a, &b)', True),
+        # fscanf and std::sscanf
+        ('fscanf(fp, "%f", &v)', True),
+        ('std::sscanf(buf, "%f", &v)', True),
+        # Multi-line
+        ('sscanf(buf,\n"%f", &v)', True),
+        # No float format
+        ('sscanf(buf, "%d", &v)', False),
+        ('sscanf(buf, "%s", s)', False),
+        # printf not scanf
+        ('printf("%f", val)', False),
+        # %f in a different statement after scanf
+        ('sscanf(buf, "%d", &x); printf("%f", val);', False),
+        # scanf %f in comment only
+        ('// sscanf(buf, "%f", &v)\nsscanf(buf, "%d", &x)', False),
+        ('/* sscanf(buf, "%f") */\nsscanf(buf, "%d", &x)', False),
+    ],
+)
+def test_lambdas_use_scanf_float(src: str, expected: bool) -> None:
+    """Test scanf float detection in lambda source."""
+    config: ConfigType = {"test": [Lambda(src)]}
+    assert lambdas_use_scanf_float(config) is expected
+
+
+def test_lambdas_use_scanf_float_no_lambdas() -> None:
+    """Test with config containing no lambdas."""
+    config: ConfigType = {"key": "value", "list": [1, 2]}
+    assert lambdas_use_scanf_float(config) is False
+
+
+def test_lambdas_use_scanf_float_nested() -> None:
+    """Test detection in deeply nested config."""
+    config: ConfigType = {"a": {"b": {"c": [Lambda('sscanf(buf, "%f", &v)')]}}}
+    assert lambdas_use_scanf_float(config) is True

tests/unit_tests/test_cpp_helpers.py

@@ -1,8 +1,10 @@
+import logging
 from unittest.mock import Mock
 
 import pytest
 
 from esphome import const, cpp_helpers as ch
+from esphome.cpp_helpers import ComponentSourcePool, register_component_source
 
 
 @pytest.mark.asyncio
@@ -23,7 +25,7 @@ async def test_register_component(monkeypatch):
     app_mock = Mock(register_component_=Mock(return_value=var))
     monkeypatch.setattr(ch, "App", app_mock)
 
-    core_mock = Mock(component_ids=["foo.bar"])
+    core_mock = Mock(component_ids=["foo.bar"], data={})
     monkeypatch.setattr(ch, "CORE", core_mock)
 
     add_mock = Mock()
@@ -59,7 +61,7 @@ async def test_register_component__with_setup_priority(monkeypatch):
     app_mock = Mock(register_component_=Mock(return_value=var))
     monkeypatch.setattr(ch, "App", app_mock)
 
-    core_mock = Mock(component_ids=["foo.bar"])
+    core_mock = Mock(component_ids=["foo.bar"], data={})
     monkeypatch.setattr(ch, "CORE", core_mock)
 
     add_mock = Mock()
@@ -78,3 +80,69 @@ async def test_register_component__with_setup_priority(monkeypatch):
     assert add_mock.call_count == 4
     app_mock.register_component_.assert_called_with(var)
     assert core_mock.component_ids == []
+
+
+def test_register_component_source_empty_name(monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.setattr(ch, "CORE", Mock(data={}))
+    assert register_component_source("") == 0
+
+
+def test_register_component_source_deduplicates(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    monkeypatch.setattr(ch, "CORE", Mock(data={}))
+    idx1 = register_component_source("wifi")
+    idx2 = register_component_source("api")
+    idx3 = register_component_source("wifi")
+    assert idx1 == 1
+    assert idx2 == 2
+    assert idx3 == 1  # deduplicated
+
+
+def test_generate_source_table_code_empty() -> None:
+    from esphome.cpp_helpers import _generate_source_table_code
+
+    assert _generate_source_table_code("TBL", "lookup", {}) == ""
+
+
+def test_generate_source_table_code_non_empty() -> None:
+    from esphome.cpp_helpers import _generate_source_table_code
+
+    code = _generate_source_table_code("TBL", "lookup", {"wifi": 1, "api": 2})
+    assert "PROGMEM" in code
+    assert "wifi" in code
+    assert "api" in code
+    assert "lookup" in code
+    assert "index == 0" in code
+    assert "progmem_read_ptr" in code
+    assert "index > 2" in code
+
+
+@pytest.mark.asyncio
+async def test_generate_component_source_table_empty_pool(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    """Test that _generate_component_source_table does nothing with an empty pool."""
+    from esphome.cpp_helpers import _generate_component_source_table
+
+    monkeypatch.setattr(ch, "CORE", Mock(data={}))
+    add_global_mock = Mock()
+    monkeypatch.setattr(ch, "add_global", add_global_mock)
+    await _generate_component_source_table()
+    add_global_mock.assert_not_called()
+
+
+def test_register_component_source_overflow_warns(
+    monkeypatch: pytest.MonkeyPatch, caplog: pytest.LogCaptureFixture
+) -> None:
+    # Pre-fill pool to max
+    pool = ComponentSourcePool(
+        sources={f"comp_{i}": i + 1 for i in range(0xFF)},
+        table_registered=True,
+    )
+    monkeypatch.setattr(ch, "CORE", Mock(data={ch._COMPONENT_SOURCE_DOMAIN: pool}))
+    with caplog.at_level(logging.WARNING):
+        idx = register_component_source("overflow_component")
+    assert idx == 0
+    assert "Too many unique component source names" in caplog.text
+    assert "overflow_component" in caplog.text