feat(chromium): add IgnoreResourceHttpStatusDomains option to filter out resources based on their hostnames (#1434)

* Introduced `IgnoreResourceHttpStatusDomains` option to filter out resources based on their hostnames

- Introduced `IgnoreResourceHttpStatusDomains` option to filter out resources based on their hostnames during HTTP status code checks.
- Updated relevant functions to handle domain normalization and matching.
- Enhanced the form data handling to include the new option.
- Added integration test scenario to verify the functionality of ignoring specified domains.

* Updated the `normalizeDomains` function to initialize the `normalized` slice with a predefined capacity based on the input `domains` slice length, improving memory allocation efficiency.

pkg/modules/chromium/browser.go

@@ -376,6 +376,7 @@ func (b *chromiumBrowser) do(ctx context.Context, logger *zap.Logger, url string
 			invalidHttpStatusCode:           &invalidHttpStatusCode,
 			invalidHttpStatusCodeMu:         &invalidHttpStatusCodeMu,
 			failOnResourceOnHttpStatusCode:  options.FailOnResourceHttpStatusCodes,
+			ignoreResourceHttpStatusDomains: options.IgnoreResourceHttpStatusDomains,
 			invalidResourceHttpStatusCode:   &invalidResourceHttpStatusCode,
 			invalidResourceHttpStatusCodeMu: &invalidResourceHttpStatusCodeMu,
 		})

pkg/modules/chromium/chromium.go

@@ -108,6 +108,20 @@ type Options struct {
 	// status code from at least one resource matches with one if its entries.
 	FailOnResourceHttpStatusCodes []int64
 
+	// IgnoreResourceHttpStatusDomains excludes resources whose hostname matches
+	// one of these domains from the application of
+	// [Options.FailOnResourceHttpStatusCodes].
+	//
+	// A match happens if the hostname equals the domain or is a subdomain of it
+	// (e.g., "browser.sentry-cdn.com" matches "sentry-cdn.com").
+	//
+	// Values are normalized (trimmed, lowercased) and may be provided as:
+	// - "example.com"
+	// - "*.example.com" or ".example.com"
+	// - "example.com:443" (port is ignored)
+	// - "https://example.com/path" (scheme/path are ignored)
+	IgnoreResourceHttpStatusDomains []string
+
 	// FailOnResourceLoadingFailed sets if the conversion should fail like the
 	// main page if Chromium fails to load at least one resource.
 	FailOnResourceLoadingFailed bool
@@ -150,19 +164,20 @@ type Options struct {
 // DefaultOptions returns the default values for Options.
 func DefaultOptions() Options {
 	return Options{
-		SkipNetworkIdleEvent:          true,
-		FailOnHttpStatusCodes:         []int64{499, 599},
-		FailOnResourceHttpStatusCodes: nil,
-		FailOnResourceLoadingFailed:   false,
-		FailOnConsoleExceptions:       false,
-		WaitDelay:                     0,
-		WaitWindowStatus:              "",
-		WaitForExpression:             "",
-		Cookies:                       nil,
-		UserAgent:                     "",
-		ExtraHttpHeaders:              nil,
-		EmulatedMediaType:             "",
-		OmitBackground:                false,
+		SkipNetworkIdleEvent:            true,
+		FailOnHttpStatusCodes:           []int64{499, 599},
+		FailOnResourceHttpStatusCodes:   nil,
+		IgnoreResourceHttpStatusDomains: nil,
+		FailOnResourceLoadingFailed:     false,
+		FailOnConsoleExceptions:         false,
+		WaitDelay:                       0,
+		WaitWindowStatus:                "",
+		WaitForExpression:               "",
+		Cookies:                         nil,
+		UserAgent:                       "",
+		ExtraHttpHeaders:                nil,
+		EmulatedMediaType:               "",
+		OmitBackground:                  false,
 	}
 }
 

pkg/modules/chromium/events.go

@@ -4,7 +4,9 @@ import (
 	"context"
 	"fmt"
 	"net/http"
+	"net/url"
 	"slices"
+	"strings"
 	"sync"
 
 	"github.com/chromedp/cdproto/cdp"
@@ -143,6 +145,7 @@ type eventResponseReceivedOptions struct {
 	invalidHttpStatusCode           *error
 	invalidHttpStatusCodeMu         *sync.RWMutex
 	failOnResourceOnHttpStatusCode  []int64
+	ignoreResourceHttpStatusDomains []string
 	invalidResourceHttpStatusCode   *error
 	invalidResourceHttpStatusCodeMu *sync.RWMutex
 }
@@ -157,6 +160,8 @@ func listenForEventResponseReceived(
 	logger *zap.Logger,
 	options eventResponseReceivedOptions,
 ) {
+	normalizedIgnoreDomains := normalizeDomains(options.ignoreResourceHttpStatusDomains)
+
 	for _, code := range []int64{199, 299, 399, 499, 599} {
 		if slices.Contains(options.failOnHttpStatusCodes, code) {
 			for i := code - 99; i <= code; i++ {
@@ -190,6 +195,11 @@ func listenForEventResponseReceived(
 			logger.Debug(fmt.Sprintf("event EventResponseReceived fired for a resource: %+v", ev.Response))
 
 			if slices.Contains(options.failOnResourceOnHttpStatusCode, ev.Response.Status) {
+				if !shouldCheckResourceHttpStatusCode(ev.Response.URL, normalizedIgnoreDomains) {
+					logger.Debug(fmt.Sprintf("skip resource HTTP status code check for '%s' due to domain filtering", ev.Response.URL))
+					return
+				}
+
 				options.invalidResourceHttpStatusCodeMu.Lock()
 				defer options.invalidResourceHttpStatusCodeMu.Unlock()
 
@@ -202,6 +212,79 @@ func listenForEventResponseReceived(
 	})
 }
 
+func shouldCheckResourceHttpStatusCode(rawURL string, ignoreDomains []string) bool {
+	host := hostnameFromURL(rawURL)
+
+	if len(ignoreDomains) > 0 && matchesAnyDomain(host, ignoreDomains) {
+		return false
+	}
+
+	return true
+}
+
+func hostnameFromURL(rawURL string) string {
+	u, err := url.Parse(rawURL)
+	if err != nil {
+		return ""
+	}
+	return strings.ToLower(u.Hostname())
+}
+
+func normalizeDomains(domains []string) []string {
+	normalized := make([]string, 0, len(domains))
+
+	for _, domain := range domains {
+		d := normalizeDomain(domain)
+		if d == "" {
+			continue
+		}
+		normalized = append(normalized, d)
+	}
+
+	return normalized
+}
+
+func normalizeDomain(domain string) string {
+	d := strings.ToLower(strings.TrimSpace(domain))
+	if d == "" {
+		return ""
+	}
+
+	// Accept "example.com", "*.example.com", ".example.com", "https://example.com/path",
+	// or "example.com:443".
+	if strings.Contains(d, "://") || strings.HasPrefix(d, "//") {
+		u, err := url.Parse(d)
+		if err == nil && u.Hostname() != "" {
+			d = strings.ToLower(u.Hostname())
+		}
+	} else {
+		// Make it parseable as a URL to extract the hostname and drop any port/path.
+		u, err := url.Parse("https://" + d)
+		if err == nil && u.Hostname() != "" {
+			d = strings.ToLower(u.Hostname())
+		}
+	}
+
+	d = strings.TrimPrefix(d, "*.")
+	d = strings.TrimPrefix(d, ".")
+
+	return d
+}
+
+func matchesAnyDomain(host string, domains []string) bool {
+	if host == "" || len(domains) == 0 {
+		return false
+	}
+
+	for _, domain := range domains {
+		if host == domain || strings.HasSuffix(host, "."+domain) {
+			return true
+		}
+	}
+
+	return false
+}
+
 type eventLoadingFailedOptions struct {
 	loadingFailed           *error
 	loadingFailedMu         *sync.RWMutex

pkg/modules/chromium/events_test.go

@@ -0,0 +1,63 @@
+package chromium
+
+import "testing"
+
+func TestNormalizeDomain(t *testing.T) {
+	tests := []struct {
+		name string
+		in   string
+		want string
+	}{
+		{name: "empty", in: "", want: ""},
+		{name: "spaces", in: "   ", want: ""},
+		{name: "simple", in: "example.com", want: "example.com"},
+		{name: "mixed case", in: "ExAmPlE.Com", want: "example.com"},
+		{name: "leading wildcard", in: "*.example.com", want: "example.com"},
+		{name: "leading dot", in: ".example.com", want: "example.com"},
+		{name: "with scheme and path", in: "https://example.com/foo/bar", want: "example.com"},
+		{name: "with port", in: "example.com:443", want: "example.com"},
+		{name: "with scheme and port", in: "https://example.com:443/foo", want: "example.com"},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := normalizeDomain(tt.in); got != tt.want {
+				t.Fatalf("normalizeDomain(%q) = %q, want %q", tt.in, got, tt.want)
+			}
+		})
+	}
+}
+
+func TestMatchesAnyDomain(t *testing.T) {
+	host := "browser.sentry-cdn.com"
+
+	if !matchesAnyDomain(host, []string{"sentry-cdn.com"}) {
+		t.Fatalf("expected %q to match %q", host, "sentry-cdn.com")
+	}
+
+	if matchesAnyDomain("not-sentry-cdn.com", []string{"sentry-cdn.com"}) {
+		t.Fatalf("expected %q to not match %q", "not-sentry-cdn.com", "sentry-cdn.com")
+	}
+}
+
+func TestShouldCheckResourceHttpStatusCode_IgnoreDomains(t *testing.T) {
+	ignore := normalizeDomains([]string{"sentry.io"})
+
+	if shouldCheckResourceHttpStatusCode("https://sentry.io/api/123", ignore) {
+		t.Fatalf("expected ignored domain to be skipped")
+	}
+
+	if shouldCheckResourceHttpStatusCode("https://sub.sentry.io/api/123", ignore) {
+		t.Fatalf("expected ignored subdomain to be skipped")
+	}
+
+	if !shouldCheckResourceHttpStatusCode("https://other.com/api/123", ignore) {
+		t.Fatalf("expected non-ignored domain to be checked")
+	}
+}
+
+func TestShouldCheckResourceHttpStatusCode_NonHTTPURL(t *testing.T) {
+	if !shouldCheckResourceHttpStatusCode("data:text/plain,hello", nil) {
+		t.Fatalf("expected data: URL to be checked (no host filtering possible)")
+	}
+}

pkg/modules/chromium/routes.go

@@ -29,25 +29,37 @@ var sameSiteRegexp = regexp2.MustCompile(
 	regexp2.None,
 )
 
-// FormDataChromiumOptions creates [Options] from the form data. Fallback to
-// the default value if the considered key is not present.
+// FormDataChromiumOptions creates [Options] from the form data.
+//
+// It falls back to the default value if the considered key is not present.
+//
+// JSON-encoded fields:
+//   - failOnHttpStatusCodes: []int
+//   - failOnResourceHttpStatusCodes: []int
+//   - ignoreResourceHttpStatusDomains: []string
+//   - cookies: []Cookie
+//   - extraHttpHeaders: map[string]string
+//
+// Domain filtering only applies to resource checks triggered by
+// "failOnResourceHttpStatusCodes".
 func FormDataChromiumOptions(ctx *api.Context) (*api.FormData, Options) {
 	defaultOptions := DefaultOptions()
 
 	var (
-		skipNetworkIdleEvent          bool
-		failOnHttpStatusCodes         []int64
-		failOnResourceHttpStatusCodes []int64
-		failOnResourceLoadingFailed   bool
-		failOnConsoleExceptions       bool
-		waitDelay                     time.Duration
-		waitWindowStatus              string
-		waitForExpression             string
-		cookies                       []Cookie
-		userAgent                     string
-		extraHttpHeaders              []ExtraHttpHeader
-		emulatedMediaType             string
-		omitBackground                bool
+		skipNetworkIdleEvent            bool
+		failOnHttpStatusCodes           []int64
+		failOnResourceHttpStatusCodes   []int64
+		ignoreResourceHttpStatusDomains []string
+		failOnResourceLoadingFailed     bool
+		failOnConsoleExceptions         bool
+		waitDelay                       time.Duration
+		waitWindowStatus                string
+		waitForExpression               string
+		cookies                         []Cookie
+		userAgent                       string
+		extraHttpHeaders                []ExtraHttpHeader
+		emulatedMediaType               string
+		omitBackground                  bool
 	)
 
 	form := ctx.FormData().
@@ -78,6 +90,19 @@ func FormDataChromiumOptions(ctx *api.Context) (*api.FormData, Options) {
 
 			return nil
 		}).
+		Custom("ignoreResourceHttpStatusDomains", func(value string) error {
+			if value == "" {
+				ignoreResourceHttpStatusDomains = defaultOptions.IgnoreResourceHttpStatusDomains
+				return nil
+			}
+
+			err := json.Unmarshal([]byte(value), &ignoreResourceHttpStatusDomains)
+			if err != nil {
+				return fmt.Errorf("unmarshal ignoreResourceHttpStatusDomains: %w", err)
+			}
+
+			return nil
+		}).
 		Bool("failOnResourceLoadingFailed", &failOnResourceLoadingFailed, defaultOptions.FailOnResourceLoadingFailed).
 		Bool("failOnConsoleExceptions", &failOnConsoleExceptions, defaultOptions.FailOnConsoleExceptions).
 		Duration("waitDelay", &waitDelay, defaultOptions.WaitDelay).
@@ -203,19 +228,20 @@ func FormDataChromiumOptions(ctx *api.Context) (*api.FormData, Options) {
 		Bool("omitBackground", &omitBackground, defaultOptions.OmitBackground)
 
 	options := Options{
-		SkipNetworkIdleEvent:          skipNetworkIdleEvent,
-		FailOnHttpStatusCodes:         failOnHttpStatusCodes,
-		FailOnResourceHttpStatusCodes: failOnResourceHttpStatusCodes,
-		FailOnResourceLoadingFailed:   failOnResourceLoadingFailed,
-		FailOnConsoleExceptions:       failOnConsoleExceptions,
-		WaitDelay:                     waitDelay,
-		WaitWindowStatus:              waitWindowStatus,
-		WaitForExpression:             waitForExpression,
-		Cookies:                       cookies,
-		UserAgent:                     userAgent,
-		ExtraHttpHeaders:              extraHttpHeaders,
-		EmulatedMediaType:             emulatedMediaType,
-		OmitBackground:                omitBackground,
+		SkipNetworkIdleEvent:            skipNetworkIdleEvent,
+		FailOnHttpStatusCodes:           failOnHttpStatusCodes,
+		FailOnResourceHttpStatusCodes:   failOnResourceHttpStatusCodes,
+		IgnoreResourceHttpStatusDomains: ignoreResourceHttpStatusDomains,
+		FailOnResourceLoadingFailed:     failOnResourceLoadingFailed,
+		FailOnConsoleExceptions:         failOnConsoleExceptions,
+		WaitDelay:                       waitDelay,
+		WaitWindowStatus:                waitWindowStatus,
+		WaitForExpression:               waitForExpression,
+		Cookies:                         cookies,
+		UserAgent:                       userAgent,
+		ExtraHttpHeaders:                extraHttpHeaders,
+		EmulatedMediaType:               emulatedMediaType,
+		OmitBackground:                  omitBackground,
 	}
 
 	return form, options

test/integration/features/chromium_convert_html.feature

@@ -343,6 +343,16 @@ Feature: /forms/chromium/convert/html
       https://gethttpstatus.com/400 - 400: Bad Request
       """
 
+  Scenario: POST /forms/chromium/convert/html (Fail On Resource HTTP Status Codes - Ignore Domains)
+    Given I have a default Gotenberg container
+    When I make a "POST" request to Gotenberg at the "/forms/chromium/convert/html" endpoint with the following form data and header(s):
+      | files                           | testdata/feature-rich-html/index.html | file  |
+      | failOnResourceHttpStatusCodes   | [499,599]                             | field |
+      | ignoreResourceHttpStatusDomains | ["gethttpstatus.com"]                 | field |
+    Then the response status code should be 200
+    Then the response header "Content-Type" should be "application/pdf"
+    Then there should be 1 PDF(s) in the response
+
   Scenario: POST /forms/chromium/convert/html (Fail On Resource Loading Failed)
     Given I have a default Gotenberg container
     When I make a "POST" request to Gotenberg at the "/forms/chromium/convert/html" endpoint with the following form data and header(s):