From 5774091741c39a4a965960d37e613fd77a661985 Mon Sep 17 00:00:00 2001
From: riddim-developer-bot <developer-bot@riddimsoftware.com>
Date: Sun, 14 Jun 2026 14:43:39 -0400
Subject: [PATCH] [EPAC-2298]: Persist clause-level bill diff data for backend
 route

---
 .../internal/adapter/legisinfo/fetcher.go     | 36 ++++++++++++++-
 .../adapter/legisinfo/fetcher_test.go         | 44 +++++++++++++++++++
 2 files changed, 79 insertions(+), 1 deletion(-)

diff --git a/backend/bills-indexer/internal/adapter/legisinfo/fetcher.go b/backend/bills-indexer/internal/adapter/legisinfo/fetcher.go
index 23e996fa..e712806e 100644
--- a/backend/bills-indexer/internal/adapter/legisinfo/fetcher.go
+++ b/backend/bills-indexer/internal/adapter/legisinfo/fetcher.go
@@ -158,6 +158,7 @@ func (f *Fetcher) fetchDetail(ctx context.Context, session domain.Session, numbe
 
 func (f *Fetcher) enrichVersions(ctx context.Context, session domain.Session, number string, pubs []publicationJSON) []domain.BillVersion {
 	versions := make([]domain.BillVersion, 0, len(pubs))
+	var firstXMLURL, firstPDFURL string
 	for i, pub := range pubs {
 		stage := firstNonEmpty(pub.PublicationTypeNameEn, pub.PublicationTypeName)
 		slug := publicationSlug(stage)
@@ -171,7 +172,21 @@ func (f *Fetcher) enrichVersions(ctx context.Context, session domain.Session, nu
 			Source:        "LEGISinfo publication",
 			SortOrder:     i + 1,
 		}
-		xmlURL, pdfURL := f.fetchDocumentLinks(ctx, htmlURL)
+
+		var xmlURL, pdfURL string
+		if firstXMLURL == "" {
+			xmlURL, pdfURL = f.fetchDocumentLinks(ctx, htmlURL)
+			if xmlURL != "" {
+				firstXMLURL = xmlURL
+			}
+			if pdfURL != "" {
+				firstPDFURL = pdfURL
+			}
+		} else {
+			xmlURL = constructXMLURL(firstXMLURL, i+1)
+			pdfURL = constructPDFURL(firstPDFURL, i+1)
+		}
+
 		version.XMLURL = xmlURL
 		version.PDFURL = pdfURL
 
@@ -194,6 +209,25 @@ func (f *Fetcher) enrichVersions(ctx context.Context, session domain.Session, nu
 	return versions
 }
 
+func constructXMLURL(firstURL string, sortOrder int) string {
+	if firstURL == "" {
+		return ""
+	}
+	return strings.Replace(firstURL, "_1/", fmt.Sprintf("_%d/", sortOrder), 1)
+}
+
+func constructPDFURL(firstURL string, sortOrder int) string {
+	if firstURL == "" {
+		return ""
+	}
+	res := strings.Replace(firstURL, "_1/", fmt.Sprintf("_%d/", sortOrder), 1)
+	res = strings.Replace(res, "_1.PDF", fmt.Sprintf("_%d.PDF", sortOrder), 1)
+	res = strings.Replace(res, "_1.pdf", fmt.Sprintf("_%d.pdf", sortOrder), 1)
+	res = strings.Replace(res, "_1.Pdf", fmt.Sprintf("_%d.Pdf", sortOrder), 1)
+	return res
+}
+
+
 func (f *Fetcher) fetchDocumentLinks(ctx context.Context, pageURL string) (string, string) {
 	body, err := f.getBytes(ctx, pageURL, "text/html")
 	if err != nil {
diff --git a/backend/bills-indexer/internal/adapter/legisinfo/fetcher_test.go b/backend/bills-indexer/internal/adapter/legisinfo/fetcher_test.go
index 40fbc6de..e77e3bfe 100644
--- a/backend/bills-indexer/internal/adapter/legisinfo/fetcher_test.go
+++ b/backend/bills-indexer/internal/adapter/legisinfo/fetcher_test.go
@@ -160,3 +160,47 @@ func TestFetcherBuildsRelationalBillRecordsFromLegisInfoExports(t *testing.T) {
 		t.Fatalf("committee meetings = %#v", stage.Meetings)
 	}
 }
+
+func TestConstructURL(t *testing.T) {
+	t.Run("XML construction", func(t *testing.T) {
+		first := "https://www.parl.ca/Content/Bills/451/Government/C-11/C-11_1/C-11_E.xml"
+		want := "https://www.parl.ca/Content/Bills/451/Government/C-11/C-11_2/C-11_E.xml"
+		got := constructXMLURL(first, 2)
+		if got != want {
+			t.Errorf("constructXMLURL = %q, want %q", got, want)
+		}
+
+		// Empty URL returns empty
+		if constructXMLURL("", 2) != "" {
+			t.Error("constructXMLURL with empty string should return empty string")
+		}
+
+		// If no _1/ exists, returns first URL unchanged
+		noMatch := "https://www.parl.ca/other/url.xml"
+		if constructXMLURL(noMatch, 2) != noMatch {
+			t.Errorf("constructXMLURL without matching prefix should return input unchanged")
+		}
+	})
+
+	t.Run("PDF construction", func(t *testing.T) {
+		first := "https://www.parl.ca/Content/Bills/451/Government/C-11/C-11_1/C-11_1.PDF"
+		want := "https://www.parl.ca/Content/Bills/451/Government/C-11/C-11_2/C-11_2.PDF"
+		got := constructPDFURL(first, 2)
+		if got != want {
+			t.Errorf("constructPDFURL = %q, want %q", got, want)
+		}
+
+		firstLower := "https://www.parl.ca/Content/Bills/451/Government/C-11/C-11_1/C-11_1.pdf"
+		wantLower := "https://www.parl.ca/Content/Bills/451/Government/C-11/C-11_2/C-11_2.pdf"
+		gotLower := constructPDFURL(firstLower, 2)
+		if gotLower != wantLower {
+			t.Errorf("constructPDFURL lower = %q, want %q", gotLower, wantLower)
+		}
+
+		// Empty URL returns empty
+		if constructPDFURL("", 2) != "" {
+			t.Error("constructPDFURL with empty string should return empty string")
+		}
+	})
+}
+