[Javstash Autofill] Scraped name never replaces the typed name; stash-box id is not stored; source needs a URL Stash already has

[dude-stash]

Plugin: Javstash Autofill v1.3 (plugins/javstashAutofill/javstash_autofill.py)

Three separate problems. I have a fix for all three and will open a PR.

1. The scraper's name never replaces the created name

With use scraper name as primary enabled, a performer whose name I typed in lowercase keeps the lowercase name. The rename is gated on:

prefer_scraper = bool(use_scraper_name) and bool(cand_name) and norm(cand_name) != norm(name)

norm() lowercases and strips spaces, so yua mikami and Yua Mikami compare equal and the rename is skipped. Only a genuinely different name triggers it — a casing or spacing difference, which is the common case when typing a name by hand, never does.

Expected: the primary name is corrected to the scraper's spelling. A casing-only correction is arguably safe even with the toggle off, since it is the same name.

2. The stash-box id is not stored

When the source is a stash-box, the matched performer's remote id is never written to stash_ids. The scrape query does not ask for it:

JAV_FIELDS = ("name gender birthdate death_date ethnicity country hair_color eye_color "
              "height weight measurements fake_tits career_length tattoos piercings "
              "details twitter instagram aliases images")

remote_site_id is missing, so the id is not available to the update at all. After autofill the performer still has no stash id, and Identify / the Tagger do not recognise it afterwards.

Related, same query: build_update() reads cand.get("url"), but neither url nor the modern urls list is requested, so scraped URLs other than twitter/instagram are always dropped.

3. Configuring the source is harder than it needs to be

Scraper (Identify) and Scraper (manual) want a full stash-box endpoint URL, which Stash already knows from Settings → Metadata Providers:

def build_source(source_str):
    # A URL is treated as a stash-box endpoint; anything else as a scraper_id.
    s = (source_str or JAV).strip()
    return {"stash_box_endpoint": s} if s.startswith("http") else {"scraper_id": s}

Retyping the URL is also a silent failure mode: resolveStashBox in stash only accepts an endpoint matching a configured one exactly (case-insensitively), so a trailing slash or a slightly wrong path fails with "stash box not found" and the hook only logs a scrape error.

Proposal: resolve the setting against configuration.general.stashBoxes and listScrapers(types: [PERFORMER]), so the value may be a stash-box name (StashDB), part of a name or endpoint (stashdb), a performer scraper name or id, or still a full URL. Empty falls back to javstash when configured, otherwise the first configured stash-box.

Steps to reproduce (1 and 2)

1. Configure a stash-box (e.g. StashDB) under Settings → Metadata Providers and set it as the plugin's manual source.
2. Turn on Manual - use scraper name as primary.
3. Create a performer manually, typing the name in lowercase (yua mikami).
4. The fields are filled from the scraper, but the name stays yua mikami and no stash id is added.