feat(outbound): support authenticated proxy from environment variables

pkg/modules/libreoffice/api/api.go

@@ -336,6 +336,7 @@ func (a *Api) Descriptor() gotenberg.ModuleDescriptor {
 			fs.StringSlice("libreoffice-deny-list", []string{}, "Set the denied URLs for LibreOffice outbound fetches using regular expressions - supports multiple values")
 			fs.Bool("libreoffice-deny-private-ips", false, "Reject LibreOffice outbound URLs whose host resolves to a non-public IP address (loopback, RFC1918, link-local, unique-local). Enable on deployments that accept untrusted documents to mitigate SSRF against internal services")
 			fs.Bool("libreoffice-deny-public-ips", false, "Reject LibreOffice outbound URLs whose host resolves to a public IP address. Enable on air-gapped or data-governed deployments to prevent outbound traffic from leaving a private network")
+			fs.Bool("libreoffice-enable-environment-proxy", false, "Route LibreOffice outbound fetches through the proxy defined by the standard HTTP_PROXY, HTTPS_PROXY, and NO_PROXY variables, including credentials")
 
 			return fs
 		}(),
@@ -363,10 +364,11 @@ func (a *Api) Provision(ctx *gotenberg.Context) error {
 		unoBinPath:   unoBinPath,
 		startTimeout: flags.MustDuration("libreoffice-start-timeout"),
 		proxyOptions: outboundProxyOptions{
-			allowList:      flags.MustRegexpSlice("libreoffice-allow-list"),
-			denyList:       flags.MustRegexpSlice("libreoffice-deny-list"),
-			denyPrivateIPs: flags.MustBool("libreoffice-deny-private-ips"),
-			denyPublicIPs:  flags.MustBool("libreoffice-deny-public-ips"),
+			allowList:              flags.MustRegexpSlice("libreoffice-allow-list"),
+			denyList:               flags.MustRegexpSlice("libreoffice-deny-list"),
+			denyPrivateIPs:         flags.MustBool("libreoffice-deny-private-ips"),
+			denyPublicIPs:          flags.MustBool("libreoffice-deny-public-ips"),
+			enableEnvironmentProxy: flags.MustBool("libreoffice-enable-environment-proxy"),
 		},
 	}
 

pkg/modules/libreoffice/api/proxy.go

@@ -15,16 +15,18 @@ import (
 	"time"
 
 	"github.com/dlclark/regexp2"
+	"golang.org/x/net/http/httpproxy"
 
 	"github.com/gotenberg/gotenberg/v8/pkg/gotenberg"
 )
 
 // outboundProxyOptions configures a [libreOfficeProxy].
 type outboundProxyOptions struct {
-	allowList      []*regexp2.Regexp
-	denyList       []*regexp2.Regexp
-	denyPrivateIPs bool
-	denyPublicIPs  bool
+	allowList              []*regexp2.Regexp
+	denyList               []*regexp2.Regexp
+	denyPrivateIPs         bool
+	denyPublicIPs          bool
+	enableEnvironmentProxy bool
 }
 
 // libreOfficeProxy is an HTTP/HTTPS forward proxy that LibreOffice routes
@@ -45,6 +47,12 @@ type libreOfficeProxy struct {
 	opts     outboundProxyOptions
 	logger   *slog.Logger
 
+	// upstreamProxy resolves the upstream (corporate) proxy for a destination
+	// URL from the standard proxy environment variables, or returns a nil URL
+	// to connect directly. Nil unless the operator opted into proxy-
+	// environment honoring. See https://github.com/gotenberg/gotenberg/issues/1592.
+	upstreamProxy func(*url.URL) (*url.URL, error)
+
 	stopOnce sync.Once
 }
 
@@ -64,10 +72,15 @@ func newLibreOfficeProxy(logger *slog.Logger, opts outboundProxyOptions) (*libre
 
 	p := &libreOfficeProxy{
 		listener: listener,
-		client:   gotenberg.NewOutboundHttpClient(0, opts.allowList, opts.denyList, decideOpts...),
+		client:   gotenberg.NewOutboundHttpClient(0, opts.allowList, opts.denyList, opts.enableEnvironmentProxy, decideOpts...),
 		opts:     opts,
 		logger:   logger.With(slog.String("logger", "libreoffice-proxy")),
 	}
+	if opts.enableEnvironmentProxy {
+		// Honor the standard proxy environment variables, credentials
+		// included. httpproxy reads the environment now and applies NO_PROXY.
+		p.upstreamProxy = httpproxy.FromEnvironment().ProxyFunc()
+	}
 	p.server = &http.Server{
 		Handler:           p,
 		ReadHeaderTimeout: 10 * time.Second,
@@ -182,8 +195,25 @@ func (p *libreOfficeProxy) handleConnect(w http.ResponseWriter, r *http.Request)
 		return
 	}
 
+	// When the operator routes egress through an authenticated proxy, soffice
+	// cannot supply the credentials, so the proxy performs the CONNECT (and
+	// authentication) upstream. The decision above still gated the destination.
+	var proxyURL *url.URL
+	if p.upstreamProxy != nil {
+		proxyURL, err = p.upstreamProxy(&url.URL{Scheme: "https", Host: net.JoinHostPort(host, port)})
+		if err != nil {
+			p.logger.WarnContext(r.Context(), fmt.Sprintf("LibreOffice proxy resolve upstream proxy for '%s': %s", rawURL, err))
+			http.Error(w, "proxy: upstream proxy error", http.StatusBadGateway)
+			return
+		}
+	}
+
 	var dest net.Conn
 	switch {
+	case proxyURL != nil:
+		dest, err = gotenberg.DialThroughProxy(r.Context(), proxyURL, r.Host, func(ctx context.Context, network, addr string) (net.Conn, error) {
+			return net.DialTimeout(network, addr, 10*time.Second)
+		})
 	case len(decision.Pinned) > 0:
 		dest, err = gotenberg.DialPinned(r.Context(), "tcp", decision.Pinned, port)
 	default: