refactor(scrape): detect prose-style "must be X" discriminator values on variants

Sealed-interface union variants whose Type/Source field is declared as
bare prose (e.g. "Type of the result, must be article" or "Scope type,
must be all_private_chats") were skipped by extractEnumValues because
the existing patterns require curly-quoted values. The genapi emitter
already extracted these values via discBareRE for marshal-side
discriminator injection; lifting the same detection into the scraper
populates Field.EnumValues so planUnifiedUnionEnums folds them into
shared union-level enums automatically.

Unions newly unified (10): BotCommandScope, MenuButton, InputMedia,
InputPaidMedia, InputPollMedia, InputPollOptionMedia, InputProfilePhoto,
InputStoryContent, InlineQueryResult, PassportElementError.

InputMessageContent stays excluded — its variants dispatch
structurally on field presence and have no Type/Source field, so
planUnifiedUnionEnums correctly skips it.

Constants added: 60 typed enum constants across the 10 unions; the
corresponding variant struct fields are retyped from string to the
shared enum.

Internal call-site cleanups: 0 — no internal package referenced these
discriminator values via magic strings.

False positives the prose detector explicitly rejects: terminal
prose-word continuations like "must be sent", "must be shown above",
"must be specified", "must be paid", "must be active", "must be one
of 3, 6, or 12", "must be between 5 and 100000", "must be a Pay
button", "must be repainted". Guarded via terminal-position regex
anchor + closed-list isProseWord filter.

Determinism verified across two consecutive make regen-from-fixture
runs. go test -race ./..., go vet ./..., staticcheck ./... all clean.

cmd/scrape/enums.go

@@ -27,6 +27,33 @@ import (
 //     emits the canonical Markdown / MarkdownV2 / HTML triple.
 //
 // Returns nil when the description does not look like an enum.
+// extractEnumValues inspects a field-description string and returns the
+// list of wire-level string values when the description matches one of
+// the enum-like patterns Telegram uses in its docs. Order follows doc
+// order; duplicates are removed but order of first occurrence is kept.
+//
+// Handled patterns (curly quotes “…” are required to avoid false
+// positives on free-text quoting):
+//
+//   - "Type of the chat, can be either “private”, “group”, … or “channel”"
+//   - "Currently, can be “mention”, “hashtag”, …"
+//   - "Currently, one of “XTR” … or “TON” …"
+//   - "Currently, must be one of “XTR” …"
+//   - "Currently, it can be one of “pending”, “approved”, “declined”."
+//   - "Must be one of “danger” …, “success” …"
+//   - "Must be one of “image/jpeg”, “image/gif”, or “video/mp4”"
+//   - "Format … must be one of “static” …, “animated” …, “video” …"
+//   - "Currently, either “upgrade” …, “transfer” …, “resale” …"
+//   - "..., always “creator”"
+//   - parse_mode parameter special case ("Mode for parsing entities …")
+//     emits the canonical Markdown / MarkdownV2 / HTML triple.
+//   - bare prose discriminator at end of description, e.g.
+//     "Type of the result, must be article" or
+//     "Scope type, must be all_private_chats". Used by sealed-interface
+//     union variants whose Type/Source field carries a single literal
+//     value declared without curly quotes.
+//
+// Returns nil when the description does not look like an enum.
 func extractEnumValues(jsonName, desc string) []string {
 	if values := parseModeEnumValues(jsonName, desc); values != nil {
 		return values
@@ -34,12 +61,15 @@ func extractEnumValues(jsonName, desc string) []string {
 
 	trigger, triggerEnd, isAlways := findEnumTrigger(desc)
 	if trigger < 0 {
-		return nil
+		return extractProseDiscriminator(desc)
 	}
 	tail := desc[trigger:]
 
 	values := collectQuotedValues(tail)
 	if len(values) == 0 {
+		if v := extractProseDiscriminator(desc); v != nil {
+			return v
+		}
 		return nil
 	}
 	// First quoted value must sit close to the trigger phrase (e.g.
@@ -203,3 +233,59 @@ func dedupeStrings(in []string) []string {
 	}
 	return out
 }
+
+// proseDiscRE matches a terminal "must be <ident>" clause: the
+// discriminator value sits at the END of the description (optionally
+// followed by trailing punctuation/whitespace) so multi-clause prose
+// like "must be shown above the message" is not picked up.
+//
+// The identifier is a snake_case wire literal: lowercase letters, digits,
+// and underscores, starting with a letter. Numeric-only and prose words
+// are filtered separately by isProseWord.
+var proseDiscRE = regexp.MustCompile(`(?i)\bmust be\s+([a-z][a-z0-9_]*)\s*[.,]?\s*$`)
+
+// extractProseDiscriminator detects unambiguous single-value
+// discriminator declarations of the form "..., must be <ident>" used by
+// sealed-interface union variants (e.g. "Type of the result, must be
+// article" or "Scope type, must be all_private_chats"). Returns the
+// extracted value as a one-element slice or nil when no match is found.
+//
+// The terminal-position anchor is what protects against prose like
+// "must be shown above" or "must be one of 3, 6, or 12" — the candidate
+// must close the description.
+func extractProseDiscriminator(desc string) []string {
+	desc = strings.TrimSpace(desc)
+	if desc == "" {
+		return nil
+	}
+	m := proseDiscRE.FindStringSubmatch(desc)
+	if m == nil {
+		return nil
+	}
+	v := m[1]
+	if isProseWord(v) {
+		return nil
+	}
+	return []string{v}
+}
+
+// isProseWord rejects bare-prose continuations that pass the regex but
+// are clearly English filler ("must be sent", "must be available"). The
+// list is the closed set of words that empirically appear in the IR's
+// "must be …" tails outside the variant-discriminator pattern. Wire
+// identifiers are always single tokens with no English meaning, so any
+// match here is a free-text false positive.
+func isProseWord(s string) bool {
+	switch s {
+	case "a", "an", "the",
+		"sent", "shown", "set", "used", "passed", "specified", "available",
+		"applied", "supported", "assumed", "active", "paid", "between",
+		"of", "on", "in", "at", "by", "to", "from", "for", "with",
+		"and", "or", "no", "non",
+		"positive", "negative",
+		"administrator", "repainted",
+		"one", "exactly":
+		return true
+	}
+	return false
+}

cmd/scrape/enums_test.go

@@ -83,3 +83,56 @@ func TestExtractEnumValues_DedupeRepeatedValues(t *testing.T) {
 	got := extractEnumValues("currency", desc)
 	require.Equal(t, []string{"XTR"}, got)
 }
+
+func TestExtractEnumValues_ProseDiscriminator(t *testing.T) {
+	cases := []struct {
+		name string
+		desc string
+		want []string
+	}{
+		{"InlineQueryResultArticle", "Type of the result, must be article", []string{"article"}},
+		{"InlineQueryResultPhoto", "Type of the result, must be photo", []string{"photo"}},
+		{"InlineQueryResultMpeg4Gif", "Type of the result, must be mpeg4_gif", []string{"mpeg4_gif"}},
+		{"BotCommandScopeAllPrivateChats", "Scope type, must be all_private_chats", []string{"all_private_chats"}},
+		{"BotCommandScopeChat", "Scope type, must be chat", []string{"chat"}},
+		{"PassportElementErrorData", "Error source, must be data", []string{"data"}},
+		{"MenuButtonWebApp", "Type of the button, must be web_app", []string{"web_app"}},
+		{"InputProfilePhotoAnimated", "Type of the profile photo, must be animated", []string{"animated"}},
+		{"InputStoryContentVideo", "Type of the content, must be video", []string{"video"}},
+		{"InputPaidMediaPhoto", "Type of the media, must be photo", []string{"photo"}},
+	}
+	for _, tc := range cases {
+		t.Run(tc.name, func(t *testing.T) {
+			require.Equal(t, tc.want, extractEnumValues("type", tc.desc))
+		})
+	}
+}
+
+func TestExtractEnumValues_ProseFalsePositives(t *testing.T) {
+	cases := []struct {
+		name string
+		desc string
+	}{
+		{"available_only_for", "Optional. Bot-specified invoice payload. Can be available only for “invoice_payment” transactions."},
+		{"must_be_sent", "If True, the message must be sent immediately."},
+		{"must_be_shown_above", "Optional. True, if the link preview must be shown above the message text"},
+		{"must_be_specified", "The identifiers must be specified in a strictly increasing order."},
+		{"must_be_paid", "The number of Telegram Stars that must be paid to send the sticker"},
+		{"must_be_one_of_numbers", "Number of months the Telegram Premium subscription will be active for the user; must be one of 3, 6, or 12"},
+		{"must_be_between", "Currently, price in Telegram Stars must be between 5 and 100000"},
+		{"must_be_a_pay_button", "If not empty, the first button must be a Pay button."},
+		{"must_be_repainted", "True, if the sticker must be repainted to a text color in messages"},
+		{"must_be_active", "the subscription must be active up to the end of the current subscription period"},
+	}
+	for _, tc := range cases {
+		t.Run(tc.name, func(t *testing.T) {
+			require.Nil(t, extractEnumValues("type", tc.desc))
+		})
+	}
+}
+
+func TestExtractEnumValues_CanonicalMustBeOneOfStillWorks(t *testing.T) {
+	desc := "Currently, must be one of “Markdown”, “MarkdownV2”, “HTML”"
+	got := extractEnumValues("parse_mode_kind", desc)
+	require.Equal(t, []string{"Markdown", "MarkdownV2", "HTML"}, got)
+}