paliad/internal/services/submission_render_test.go

package services

import (
	"archive/zip"
	"bytes"
	"io"
	"strings"
	"testing"
)

// minimalDOCX builds a tiny .docx zip with one document.xml that
// contains the given body. Just enough to exercise the renderer
// without depending on Word's full OOXML scaffolding.
func minimalDOCX(t *testing.T, documentBody string) []byte {
	t.Helper()
	var buf bytes.Buffer
	zw := zip.NewWriter(&buf)
	w, err := zw.Create("word/document.xml")
	if err != nil {
		t.Fatalf("create document.xml: %v", err)
	}
	if _, err := io.WriteString(w, documentBody); err != nil {
		t.Fatalf("write document.xml: %v", err)
	}
	// Drop in a stub Content-Types so the bytes look more like a real
	// .docx for any downstream sanity checks; Word doesn't care about
	// the content during our unit tests but the shape stays honest.
	w2, err := zw.Create("[Content_Types].xml")
	if err != nil {
		t.Fatalf("create content types: %v", err)
	}
	if _, err := io.WriteString(w2, `<?xml version="1.0"?><Types/>`); err != nil {
		t.Fatalf("write content types: %v", err)
	}
	if err := zw.Close(); err != nil {
		t.Fatalf("close zip: %v", err)
	}
	return buf.Bytes()
}

// readDocumentXML pulls word/document.xml out of a rendered .docx.
func readDocumentXML(t *testing.T, b []byte) string {
	t.Helper()
	zr, err := zip.NewReader(bytes.NewReader(b), int64(len(b)))
	if err != nil {
		t.Fatalf("open rendered zip: %v", err)
	}
	for _, f := range zr.File {
		if f.Name != "word/document.xml" {
			continue
		}
		rc, err := f.Open()
		if err != nil {
			t.Fatalf("open document.xml: %v", err)
		}
		defer rc.Close()
		body, err := io.ReadAll(rc)
		if err != nil {
			t.Fatalf("read document.xml: %v", err)
		}
		return string(body)
	}
	t.Fatal("rendered .docx had no word/document.xml")
	return ""
}

// TestRender_SingleRunPlaceholder covers the 99% case: a placeholder
// that sits inside a single <w:t> text node.
func TestRender_SingleRunPlaceholder(t *testing.T) {
	doc := `<w:document><w:body><w:p><w:r><w:t>{{firm.name}}</w:t></w:r></w:p></w:body></w:document>`
	tmpl := minimalDOCX(t, doc)
	r := NewSubmissionRenderer()
	out, err := r.Render(tmpl, PlaceholderMap{"firm.name": "HLC"}, nil)
	if err != nil {
		t.Fatalf("render: %v", err)
	}
	body := readDocumentXML(t, out)
	if !strings.Contains(body, ">HLC<") {
		t.Errorf("expected HLC in body, got %q", body)
	}
	if strings.Contains(body, "{{") {
		t.Errorf("unreplaced placeholder marker in body: %q", body)
	}
}

// TestRender_MultiplePlaceholdersPerRun is the case go-docx fails on
// — sibling placeholders inside the same <w:t> run. The in-house
// renderer must handle them.
func TestRender_MultiplePlaceholdersPerRun(t *testing.T) {
	doc := `<w:document><w:body><w:p><w:r><w:t>{{parties.claimant.name}}, vertreten durch {{parties.claimant.representative}}</w:t></w:r></w:p></w:body></w:document>`
	tmpl := minimalDOCX(t, doc)
	r := NewSubmissionRenderer()
	out, err := r.Render(tmpl, PlaceholderMap{
		"parties.claimant.name":           "Acme Inc.",
		"parties.claimant.representative": "Kanzlei Müller",
	}, nil)
	if err != nil {
		t.Fatalf("render: %v", err)
	}
	body := readDocumentXML(t, out)
	if !strings.Contains(body, "Acme Inc.") || !strings.Contains(body, "Kanzlei Müller") {
		t.Errorf("expected both party values, got %q", body)
	}
	if strings.Contains(body, "{{") {
		t.Errorf("unreplaced placeholder marker in body: %q", body)
	}
}

// TestRender_MissingMarker confirms unbound placeholders render the
// missing-value marker instead of failing the request.
func TestRender_MissingMarker(t *testing.T) {
	doc := `<w:document><w:body><w:p><w:r><w:t>{{project.case_number}}</w:t></w:r></w:p></w:body></w:document>`
	tmpl := minimalDOCX(t, doc)
	r := NewSubmissionRenderer()
	out, err := r.Render(tmpl, PlaceholderMap{}, DefaultMissingMarker("de"))
	if err != nil {
		t.Fatalf("render: %v", err)
	}
	body := readDocumentXML(t, out)
	if !strings.Contains(body, "[KEIN WERT: project.case_number]") {
		t.Errorf("expected KEIN WERT marker, got %q", body)
	}
	outEN, err := r.Render(tmpl, PlaceholderMap{}, DefaultMissingMarker("en"))
	if err != nil {
		t.Fatalf("render en: %v", err)
	}
	bodyEN := readDocumentXML(t, outEN)
	if !strings.Contains(bodyEN, "[NO VALUE: project.case_number]") {
		t.Errorf("expected NO VALUE marker, got %q", bodyEN)
	}
}

// TestRender_CrossRunPlaceholder simulates Word fragmenting a
// placeholder across runs (autocorrect or post-edit run-split).
// Pass 2 must catch it.
func TestRender_CrossRunPlaceholder(t *testing.T) {
	doc := `<w:document><w:body><w:p><w:r><w:t>Hello {{</w:t></w:r><w:r><w:t>project</w:t></w:r><w:r><w:t>.case_number}}!</w:t></w:r></w:p></w:body></w:document>`
	tmpl := minimalDOCX(t, doc)
	r := NewSubmissionRenderer()
	out, err := r.Render(tmpl, PlaceholderMap{"project.case_number": "7 O 1234/26"}, nil)
	if err != nil {
		t.Fatalf("render: %v", err)
	}
	body := readDocumentXML(t, out)
	if !strings.Contains(body, "7 O 1234/26") {
		t.Errorf("expected case number after cross-run merge, got %q", body)
	}
	if strings.Contains(body, "{{") {
		t.Errorf("orphan placeholder marker remained: %q", body)
	}
}

// TestRender_XMLEscaping verifies special characters in placeholder
// values are escaped so they don't corrupt the document XML.
func TestRender_XMLEscaping(t *testing.T) {
	doc := `<w:document><w:body><w:p><w:r><w:t>{{user.display_name}}</w:t></w:r></w:p></w:body></w:document>`
	tmpl := minimalDOCX(t, doc)
	r := NewSubmissionRenderer()
	out, err := r.Render(tmpl, PlaceholderMap{
		"user.display_name": `Müller & Söhne <GmbH> "Special"`,
	}, nil)
	if err != nil {
		t.Fatalf("render: %v", err)
	}
	body := readDocumentXML(t, out)
	if !strings.Contains(body, "Müller &amp; Söhne &lt;GmbH&gt; &quot;Special&quot;") {
		t.Errorf("expected escaped value, got %q", body)
	}
}

// TestRender_PreservesNonWordEntries leaves the rest of the .docx
// untouched so any styles / theme / settings parts come through bit-
// for-bit.
func TestRender_PreservesNonWordEntries(t *testing.T) {
	doc := `<w:document><w:body><w:p><w:r><w:t>{{firm.name}}</w:t></w:r></w:p></w:body></w:document>`
	tmpl := minimalDOCX(t, doc)
	r := NewSubmissionRenderer()
	out, err := r.Render(tmpl, PlaceholderMap{"firm.name": "HLC"}, nil)
	if err != nil {
		t.Fatalf("render: %v", err)
	}
	zr, err := zip.NewReader(bytes.NewReader(out), int64(len(out)))
	if err != nil {
		t.Fatalf("open rendered: %v", err)
	}
	var sawTypes bool
	for _, f := range zr.File {
		if f.Name == "[Content_Types].xml" {
			sawTypes = true
		}
	}
	if !sawTypes {
		t.Error("rendered .docx lost [Content_Types].xml")
	}
}

// TestPlaceholderRegex_Boundaries pins the placeholder grammar.
func TestPlaceholderRegex_Boundaries(t *testing.T) {
	tests := []struct {
		in       string
		matches  []string
	}{
		{"plain text", nil},
		{"{{foo}}", []string{"{{foo}}"}},
		{"{{ foo }}", []string{"{{ foo }}"}},
		{"{{foo.bar}}", []string{"{{foo.bar}}"}},
		{"{{ foo.bar_baz }}", []string{"{{ foo.bar_baz }}"}},
		{"{{1bad}}", nil}, // must start with a letter
		{"{{ foo }} and {{ bar }}", []string{"{{ foo }}", "{{ bar }}"}},
	}
	for _, tc := range tests {
		t.Run(tc.in, func(t *testing.T) {
			got := placeholderRegex.FindAllString(tc.in, -1)
			if len(got) != len(tc.matches) {
				t.Fatalf("got %d matches, want %d (in=%q)", len(got), len(tc.matches), tc.in)
			}
			for i := range got {
				if got[i] != tc.matches[i] {
					t.Errorf("match %d: got %q, want %q", i, got[i], tc.matches[i])
				}
			}
		})
	}
}

// TestFamilyOf covers the proceeding-family extraction used by the
// template registry's fallback chain.
func TestFamilyOf(t *testing.T) {
	tests := map[string]string{
		"de.inf.lg.erwidg": "de.inf.lg",
		"upc.inf.cfi.soc":  "upc.inf.cfi",
		"dpma.opp.dpma":    "", // only three segments → no family
		"de.inf.lg":        "",
		"":                 "",
	}
	for in, want := range tests {
		t.Run(in, func(t *testing.T) {
			got := familyOf(in)
			if got != want {
				t.Errorf("familyOf(%q) = %q, want %q", in, got, want)
			}
		})
	}
}

// TestLegalSourcePretty covers the prefix table.
func TestLegalSourcePretty(t *testing.T) {
	tests := []struct {
		src, lang, want string
	}{
		{"DE.ZPO.276.1", "de", "§ 276 Abs. 1 ZPO"},
		{"DE.ZPO.276.1", "en", "Section 276(1) ZPO"},
		{"DE.ZPO.253", "de", "§ 253 ZPO"},
		{"DE.ZPO.253", "en", "Section 253 ZPO"},
		{"UPC.RoP.23.1", "de", "Regel 23.1 VerfO UPC"},
		{"UPC.RoP.23.1", "en", "Rule 23.1 RoP UPC"},
		{"UPC.RoP.198", "de", "Regel 198 VerfO UPC"},
		{"DE.PatG.83", "de", "§ 83 PatG"},
		{"EPC.123", "de", "Art. 123 EPÜ"},
		{"EPC.123", "en", "Art. 123 EPC"},
		// Unknown prefix → pass-through unchanged.
		{"FOO.BAR.123", "de", "FOO.BAR.123"},
		{"", "de", ""},
	}
	for _, tc := range tests {
		t.Run(tc.src+"/"+tc.lang, func(t *testing.T) {
			got := legalSourcePretty(tc.src, tc.lang)
			if got != tc.want {
				t.Errorf("legalSourcePretty(%q, %q) = %q, want %q", tc.src, tc.lang, got, tc.want)
			}
		})
	}
}

// TestOurSideTranslations pins the our_side enum → DE/EN prose
// mapping used by addProjectVars. Post t-paliad-222: seven sub-role
// values + the gender-neutral "-Seite" / "-Partei" suffix shape on
// DE. Legacy 'court' / 'both' yield "" (the column no longer accepts
// them after mig 110, but the function defensively handles stale
// in-memory values from older callers).
func TestOurSideTranslations(t *testing.T) {
	cases := []struct {
		in, wantDE, wantEN string
	}{
		{"claimant", "Klägerseite", "Claimant"},
		{"defendant", "Beklagtenseite", "Defendant"},
		{"applicant", "Antragstellerseite", "Applicant"},
		{"appellant", "Berufungsklägerseite", "Appellant"},
		{"respondent", "Antragsgegnerseite", "Respondent"},
		{"third_party", "Drittpartei", "Third Party"},
		{"other", "sonstige Verfahrensbeteiligte", "other party"},
		{"court", "", ""},
		{"both", "", ""},
		{"", "", ""},
		{"unknown", "", ""},
	}
	for _, tc := range cases {
		t.Run(tc.in, func(t *testing.T) {
			if got := ourSideDE(tc.in); got != tc.wantDE {
				t.Errorf("ourSideDE(%q) = %q, want %q", tc.in, got, tc.wantDE)
			}
			if got := ourSideEN(tc.in); got != tc.wantEN {
				t.Errorf("ourSideEN(%q) = %q, want %q", tc.in, got, tc.wantEN)
			}
		})
	}
}

// TestTemplateRegistry_Candidates verifies the fallback-chain order
// matches the m-locked Q4 decision (firm → base/code → base/family →
// skeleton).
func TestTemplateRegistry_Candidates(t *testing.T) {
	r := NewTemplateRegistry("", "HLC")
	got := r.candidates("de.inf.lg.erwidg")
	want := []string{
		"templates/HLC/de.inf.lg.erwidg.docx",
		"templates/_base/de.inf.lg.erwidg.docx",
		"templates/_base/de.inf.lg.docx",
		"templates/_base/_skeleton.docx",
	}
	if len(got) != len(want) {
		t.Fatalf("candidates = %v, want %v", got, want)
	}
	for i := range got {
		if got[i] != want[i] {
			t.Errorf("candidate[%d] = %q, want %q", i, got[i], want[i])
		}
	}
}

// TestTemplateRegistry_Candidates_NoFamily covers submission codes
// without a family suffix (only three dot-segments).
func TestTemplateRegistry_Candidates_NoFamily(t *testing.T) {
	r := NewTemplateRegistry("", "HLC")
	got := r.candidates("dpma.opp.dpma")
	want := []string{
		"templates/HLC/dpma.opp.dpma.docx",
		"templates/_base/dpma.opp.dpma.docx",
		"templates/_base/_skeleton.docx",
	}
	if len(got) != len(want) {
		t.Fatalf("candidates = %v, want %v", got, want)
	}
	for i := range got {
		if got[i] != want[i] {
			t.Errorf("candidate[%d] = %q, want %q", i, got[i], want[i])
		}
	}
}

// TestTemplateRegistry_Tiers labels each candidate slot. Must stay
// 1:1 with candidates().
func TestTemplateRegistry_Tiers(t *testing.T) {
	r := NewTemplateRegistry("", "HLC")
	codes := []string{"de.inf.lg.erwidg", "dpma.opp.dpma"}
	for _, code := range codes {
		c := r.candidates(code)
		ts := r.tiers(code)
		if len(c) != len(ts) {
			t.Fatalf("candidate/tier mismatch for %q: %d vs %d", code, len(c), len(ts))
		}
	}
}

// TestPatentNumberUPC covers the kind-code parenthesisation that UPC
// briefs use (t-paliad-215 Slice 2, design §22 Q-S2-4).
func TestPatentNumberUPC(t *testing.T) {
	tests := []struct {
		in, want string
	}{
		// EP variants — the common case.
		{"EP 1 234 567 B1", "EP 1 234 567 (B1)"},
		{"EP 4 056 049 A1", "EP 4 056 049 (A1)"},
		// DE national number with kind code.
		{"DE 10 2020 123 456 A1", "DE 10 2020 123 456 (A1)"},
		// No kind code → pass-through unchanged.
		{"EP 1 234 567", "EP 1 234 567"},
		// Leading + trailing whitespace trimmed.
		{"  EP 1 234 567 B1  ", "EP 1 234 567 (B1)"},
		// Empty input.
		{"", ""},
		// Slash-separated forms (WO publication numbers) don't match
		// the kind-code shape → pass through.
		{"WO/2023/123456", "WO/2023/123456"},
		// Two-digit kind code (e.g. B12) doesn't match the single-digit
		// pattern; pass through. This is intentional — real EP kind
		// codes are single-letter + single-digit.
		{"EP 1 234 567 B12", "EP 1 234 567 B12"},
	}
	for _, tc := range tests {
		t.Run(tc.in, func(t *testing.T) {
			got := patentNumberUPC(tc.in)
			if got != tc.want {
				t.Errorf("patentNumberUPC(%q) = %q, want %q", tc.in, got, tc.want)
			}
		})
	}
}