Edit inline automations on flat singleton components (sun:, mqtt:)

[bdraco]

What does this implement/fix?

The automation editor's parse / upsert / delete only handled list-domain
component instances (switch:, time: — YAML lists). Flat singleton
components that host inline on_*: triggers — sun: (on_sunrise/on_sunset),
mqtt: (on_message/on_connect/…) — are single mappings, and were silently
skipped: their inline automations couldn't be shown, edited, or deleted in the
Visual Editor, even though the wizard already offered sun.on_sunrise as
addable (the available surface and the parse/write surface disagreed).

Changes:

• Parser walks both list instances and flat singleton mappings; a singleton
  keys on its declared id: or the domain name when id-less.
• Locator routes a flat-mapping body to a singleton span (the whole block)
  instead of hunting for - list items, so upsert and delete both work.
• Domain recovery resolves an id-less singleton's domain before the ambiguous
  catalog guess, so a shared trigger key (mqtt.on_connect vs wifi.on_connect)
  attaches under the configured domain.
• get_available surfaces singletons as targetable AvailableComponentInstance
  entries so the available surface agrees with parse/write.

No new location kind is needed — ComponentOnLocation(..., index=None) already
models the single-handler form. writing_lists.py is untouched: flat singletons
are always single-handler (index=None) and route through the inline path, never
the list-index path.

DRY: the shared "is this block body a list?" check is hoisted into
block_body_is_list, replacing a duplicated inline scan in the multi-conf
normaliser.

Related issue or feature (if applicable):

• fixes Visual Editor: inline automations on flat singleton components (sun:, mqtt:) aren't parsed or editable #1138

Types of changes

• Bugfix (non-breaking change which fixes an issue) — bugfix
• New feature (non-breaking change which adds functionality) — new-feature
• Enhancement to an existing feature — enhancement
• Breaking change (fix or feature that would cause existing functionality to not work as expected) — breaking-change
• Refactor (no behaviour change) — refactor
• Documentation only — docs
• Maintenance / chore — maintenance
• CI / workflow change — ci
• Dependencies bump — dependencies

Frontend coordination

• No frontend change needed
• Companion frontend PR: esphome/device-builder-frontend#

This backend PR makes flat singletons parse/edit/delete and surfaces them in
get_available. A coordinated frontend PR will add the affordance for adding
a new automation to a flat singleton (frontend #552 deliberately left flat
singletons out of scope pending this backend support).

Checklist

• The code change is tested and works locally.
• Pre-commit hooks pass (ruff, codespell, yaml/json/python checks).
• Tests have been added or updated under tests/ where applicable.
• components.index.json / definitions/components/*.json have not been hand-edited (regenerate via script/sync_components.py if a sync is needed).
• Architecture-level changes are reflected in docs/ARCHITECTURE.md and/or docs/API.md.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.38%. Comparing base (6f0805c) to head (cdd8725).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1139   +/-   ##
=======================================
  Coverage   99.38%   99.38%           
=======================================
  Files         209      209           
  Lines       15376    15399   +23     
=======================================
+ Hits        15282    15305   +23     
  Misses         94       94           

Flag	Coverage Δ
py3.12	99.37% <100.00%> (+<0.01%)	⬆️
py3.14	99.38% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...vice_builder/controllers/automations/controller.py	100.00% <100.00%> (ø)
..._device_builder/controllers/automations/parsing.py	100.00% <100.00%> (ø)
..._device_builder/controllers/automations/writing.py	98.92% <100.00%> (+<0.01%)	⬆️
esphome_device_builder/helpers/yaml/component.py	100.00% <100.00%> (ø)
esphome_device_builder/helpers/yaml/inline.py	100.00% <100.00%> (ø)
esphome_device_builder/helpers/yaml/scalar.py	100.00% <100.00%> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[codspeed-hq]

Merging this PR will not alter performance

✅ 27 untouched benchmarks
⏩ 1 skipped benchmark¹

---

_{Comparing fix-flat-singleton-automations (cdd8725) with main (6f0805c)}

Footnotes

1. 1 benchmark was skipped, so the baseline result was used instead. If it was deleted from the codebase, click here and archive it to remove it from the performance reports. ↩

[Copilot]

Pull request overview

This PR fixes the automations editor’s inability to parse/edit/delete inline on_* handlers declared under flat singleton component blocks (e.g. sun:, mqtt:), bringing parse/write behavior in line with what get_available and the trigger catalog already expose.

Changes:

• Extend automation parsing to walk both list-based component instances and singleton mapping blocks, keying singleton instances by id: (or by domain when id-less).
• Update inline YAML instance location to treat singleton blocks as a single editable span (so upsert/delete work reliably).
• Update get_available scoping to surface singleton component blocks as targetable AvailableComponentInstances, and add coverage for parse/write/available behavior.

Reviewed changes

Copilot reviewed 13 out of 13 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
esphome_device_builder/controllers/automations/parsing.py	Parse inline triggers from both list instances and singleton mapping blocks; add shared instance-trigger parser helper.
esphome_device_builder/controllers/automations/writing.py	Improve domain recovery for id-less singleton instances (domain-name keyed) before falling back to catalog guessing.
esphome_device_builder/controllers/automations/controller.py	Include singleton component instances in get_available and refactor instance construction to a helper.
esphome_device_builder/helpers/yaml/inline.py	Detect non-list domain bodies and locate singleton blocks as a single instance span for inline handler edits.
esphome_device_builder/helpers/yaml/scalar.py	Add block_body_is_list() helper to centralize list-vs-mapping detection for YAML blocks.
esphome_device_builder/helpers/yaml/component.py	Reuse block_body_is_list() to simplify multi-conf normalization’s list-body detection.
tests/test_automations_parse.py	Add parsing tests for singleton inline triggers (sun, mqtt) including trigger-params handling.
tests/test_automations_writer.py	Add upsert/delete/round-trip tests for singleton inline handlers, including sibling key preservation.
tests/test_automations_controller.py	Add get_available test coverage ensuring singleton instances are surfaced as targets.
tests/fixtures/automation_yamls/inline_sun_on_sunrise.yaml	New fixture: id-less sun: with on_sunrise.
tests/fixtures/automation_yamls/inline_sun_idd.yaml	New fixture: id’d sun: with on_sunrise/on_sunset.
tests/fixtures/automation_yamls/inline_sun_empty.yaml	New fixture: sun: block with config keys only (no handlers).
tests/fixtures/automation_yamls/inline_mqtt_singleton.yaml	New fixture: singleton mqtt: with on_message trigger params and on_connect.

[esphbot]

PR Review — Edit inline automations on flat singleton components (sun:, mqtt:)

Clean, well-scoped bugfix — flat singletons now parse/edit/delete consistently with get_available. Merge-ready.

• block_body_is_list extraction is behavior-preserving vs the old inline scan in _normalize_multi_conf_block (first non-blank/non-comment line decides; empty body → not-a-list → same fall-through).
• Domain recovery for id-less singletons is correctly placed after the literal id: scan and before the ambiguous catalog guess; id'd singletons still resolve via the id-scan, so mqtt.on_connect vs wifi.on_connect attaches under the configured domain.
• No double-parsing of esphome.on_*: those are device-level and excluded from _component_trigger_domains, handled by the separate _parse_device_level path.
• _locate_singleton_instance mirrors the existing stale-id refusal pattern (returns None → clean not-found), and the unknown-id path is tested.
• Two minor, non-blocking suggestions: a possible non-display name: for singletons, and an optional multi-handler delete test.

---

🟢 Suggestions

1. Consider a delete test on a multi-handler singleton (`tests/test_automations_writer.py`, L264-272)

test_delete_inline_handler_on_sun_block deletes from the single-handler inline_sun_on_sunrise.yaml. The inline_sun_idd.yaml fixture already has both on_sunrise and on_sunset, so a delete test against it would pin the more interesting edge: that removing one handler leaves the sibling handler and the block's config keys intact (the handler_end scan in remove_inline_handler should stop at on_sunset's column).

Line coverage is already 100%, so this is purely a behavioral-edge nicety, not a gap that blocks merge.

---

Checklist

• List-vs-mapping detection refactor preserves behavior
• No double-parsing of device-level esphome triggers
• Domain recovery ordering correct (id-scan → synthetic → domain-name → catalog)
• Stale/unknown id refused cleanly
• Tests verify observable behavior, not source text
• Edge case coverage (empty body, bodyless singleton, ambiguous keys)
• No injection / unsafe deserialization (SafeLoader via make_yaml)

---

Automated review by Kōan6b05109