fix(parsers): stop coercing free-text fields (single-letter cue names); rc11

[ibiltari]

Closes ClickUp 869cqbpxa — "Error al querer poner el nombre de una cue de una sola letra".

The bug

There was never a length rule. NameStringType has always been minLength=1, and the frontend cue-name input has no validator at all.

CuemsParser.str_to_value() ran every scalar through int → float → strtobool → Uuid regardless of its key, so free text was silently rewritten on save. Because strtobool accepts the truth abbreviations, the failing characters were exactly its vocabulary:

Typed name	Persisted as
n, N, f, F	False
y, Y, t, T	True
0–9	the integer
a, b, x, Q, …	(correct)

18 of 62 alphanumeric single characters were corrupted, plus yes/no/true/false/on/off. Reproduced on test2 through the editor's real save path: a cue named n was written to XML as <name>False</name>.

Second failure class, same root cause

A cue named lowercase none/null hit the ['none','null',''] → None branch, serialised to <name/>, and failed minLength=1 — a hard XMLSchemaValidationError at save time, not silent corruption. The new short-circuit precedes that branch, so both are fixed together.

The fix

STRING_TYPED_KEYS + an optional key argument on str_to_value, threaded through all four call sites. Keeping key optional leaves existing single-argument callers unaffected.

id is deliberately excluded: the Uuid() branch inside str_to_value is the only thing producing Uuid objects on parse (parsers assign via raw dict.__setitem__ and never reach the property setters), so adding it would downgrade every cue/script/media id to a plain str. Consequence: DmxSceneType.id (script.xsd:403, declared xs:string) can't be covered by this mechanism — accepted, since DMX scene ids are system-assigned.

Scope — narrower than first assumed

Only name, description and file_name are reachable today. An earlier analysis claimed output_name was also affected; that is empirically false. output_name, parameter_name, icon and color never reach str_to_value, for three distinct reasons:

• outputsParser.parse() constructs output objects directly via self._class(item)
• fade_profileParser diverts parameters to _normalize_fade_parameters() before the scalar branch
• get_class('ui_properties') misses (the class is UI_properties) and falls back to GenericDict raw passthrough

They're kept in the allowlist as defensive entries so that fixing any of those bypasses can't silently reintroduce this bug. The comment in the source says so explicitly.

Testing

• New tests/test_name_coercion.py: exhaustive sweep over all 62 alphanumeric single characters per reachable key, the boolean/nullish word set, full XML round-trip, and negative tests pinning that enabled/autoload/timecode/loop still coerce and id still yields a Uuid.
• Full suite: 551 passed. One pre-existing unrelated failure (test_realtime_25fps_jitter, a ±1ms wall-clock tolerance test) — confirmed it fails identically on a clean tree without this change.
• Verified on test2 against the live stack: all 62 single characters clean through a full XML round-trip, none/null no longer error, and enabled 'true'→True / loop '1'→1 still correct.

Follow-ups (not in this PR)

• Projects already saved carry the corrupted name in their cue_script.xml; the original text is unrecoverable (n, no, N, off all collapse to False) and needs manual renaming.
• cuems-nodeconf has the identical bug at NodeXmlBuilders.py:80 (calls the inherited str_to_value without a key) — tracked as ClickUp 869eag7hv.
• Deriving string-typed keys from the XSD instead of a hand-maintained allowlist, and the get_parser_class/get_class case-sensitivity defects — each needs its own plan.

[cursor]

Bugbot is not enabled for your account, so this pull request was not reviewed.

Enable Bugbot in the Cursor dashboard to get automatic reviews on future PRs.

[codecov]

Welcome to Codecov 🎉

Once you merge this PR into your default branch, you're all set! Codecov will compare coverage reports and display results in all future pull requests.

ℹ️ You can also turn on project coverage checks and project coverage reporting on Pull Request comment

Thanks for integrating Codecov - We've got you covered ☂️