fix(pdfengines): read metadata better webhook error - fixes #1148

pkg/modules/pdfengines/routes.go

@@ -500,6 +500,11 @@ func readMetadataRoute(engine gotenberg.PdfEngine) api.Route {
 
 			err = c.JSON(http.StatusOK, res)
 			if err != nil {
+				if strings.Contains(err.Error(), "request method or response status code does not allow body") {
+					// High probability that the user is using the webhook
+					// feature. It does not make sense for this route.
+					return api.ErrNoOutputFile
+				}
 				return fmt.Errorf("return JSON response: %w", err)
 			}
 

pkg/modules/webhook/middleware.go

@@ -184,11 +184,21 @@ func webhookMiddleware(w *Webhook) api.Middleware {
 						// Call the next middleware in the chain.
 						err := next(c)
 						if err != nil {
+							if errors.Is(err, api.ErrNoOutputFile) {
+								errNoOutputFile := fmt.Errorf("%w - the webhook middleware cannot handle the result of this route", err)
+								handleAsyncError(api.WrapError(
+									errNoOutputFile,
+									api.NewSentinelHttpError(
+										http.StatusBadRequest,
+										"The webhook middleware can only work with multipart/form-data routes that results in output files",
+									),
+								))
+								return
+							}
 							// The process failed for whatever reason. Let's send the
 							// details to the webhook.
 							ctx.Log().Error(err.Error())
 							handleAsyncError(err)
-
 							return
 						}
 
@@ -197,7 +207,6 @@ func webhookMiddleware(w *Webhook) api.Middleware {
 						if err != nil {
 							ctx.Log().Error(fmt.Sprintf("build output file: %s", err))
 							handleAsyncError(err)
-
 							return
 						}
 
@@ -205,7 +214,6 @@ func webhookMiddleware(w *Webhook) api.Middleware {
 						if err != nil {
 							ctx.Log().Error(fmt.Sprintf("open output file: %s", err))
 							handleAsyncError(err)
-
 							return
 						}
 
@@ -221,7 +229,6 @@ func webhookMiddleware(w *Webhook) api.Middleware {
 						if err != nil {
 							ctx.Log().Error(fmt.Sprintf("read header of output file: %s", err))
 							handleAsyncError(err)
-
 							return
 						}
 
@@ -229,7 +236,6 @@ func webhookMiddleware(w *Webhook) api.Middleware {
 						if err != nil {
 							ctx.Log().Error(fmt.Sprintf("get stat from output file: %s", err))
 							handleAsyncError(err)
-
 							return
 						}
 
@@ -237,7 +243,6 @@ func webhookMiddleware(w *Webhook) api.Middleware {
 						if err != nil {
 							ctx.Log().Error(fmt.Sprintf("reset output file reader: %s", err))
 							handleAsyncError(err)
-
 							return
 						}
 

test/integration/features/pdfengines_metadata.feature

@@ -233,6 +233,24 @@ Feature: /forms/pdfengines/{write|read}
       }
       """
 
+  Scenario: POST /forms/pdfengines/metadata/read (Webhook)
+    Given I have a default Gotenberg container
+    Given I have a webhook server
+    When I make a "POST" request to Gotenberg at the "/forms/pdfengines/metadata/read" endpoint with the following form data and header(s):
+      | files                       | testdata/page_1.pdf                          | file   |
+      | Gotenberg-Webhook-Url       | http://host.docker.internal:%d/webhook       | header |
+      | Gotenberg-Webhook-Error-Url | http://host.docker.internal:%d/webhook/error | header |
+    Then the response status code should be 204
+    When I wait for the asynchronous request to the webhook
+    Then the webhook request header "Content-Type" should be "application/json"
+    Then the webhook request body should match JSON:
+      """
+      {
+        "status": 400,
+        "message": "the webhook middleware can only work with multipart/form-data routes that results in output files"
+      }
+      """
+
   Scenario: POST /forms/pdfengines/metadata/write (Basic Auth)
     Given I have a Gotenberg container with the following environment variable(s):
       | API_ENABLE_BASIC_AUTH             | true |

test/integration/scenario/scenario.go

@@ -407,11 +407,7 @@ func (s *scenario) theBodyShouldMatchString(kind string, expectedDoc *godog.DocS
 	} else if s.server.req == nil {
 		return errors.New("no webhook request found")
 	} else {
-		body, err := io.ReadAll(s.server.req.Body)
+		actual = string(s.server.bodyCopy)
-		if err != nil {
-			return fmt.Errorf("read request body: %w", err)
-		}
-		actual = string(body)
 	}
 
 	expected := strings.ReplaceAll(expectedDoc.Content, "{version}", GotenbergVersion)
@@ -431,11 +427,7 @@ func (s *scenario) theBodyShouldContainString(kind string, expectedDoc *godog.Do
 	} else if s.server.req == nil {
 		return errors.New("no webhook request found")
 	} else {
-		body, err := io.ReadAll(s.server.req.Body)
+		actual = string(s.server.bodyCopy)
-		if err != nil {
-			return fmt.Errorf("read request body: %w", err)
-		}
-		actual = string(body)
 	}
 
 	expected := strings.ReplaceAll(expectedDoc.Content, "{version}", GotenbergVersion)
@@ -455,11 +447,7 @@ func (s *scenario) theBodyShouldMatchJSON(kind string, expectedDoc *godog.DocStr
 	} else if s.server.req == nil {
 		return errors.New("no webhook request found")
 	} else {
-		b, err := io.ReadAll(s.server.req.Body)
+		body = s.server.bodyCopy
-		if err != nil {
-			return fmt.Errorf("read request body: %w", err)
-		}
-		body = b
 	}
 
 	var expected, actual interface{}

test/integration/scenario/server.go

@@ -19,9 +19,10 @@ import (
 )
 
 type server struct {
-	srv     *echo.Echo
+	srv      *echo.Echo
-	req     *http.Request
+	req      *http.Request
-	errChan chan error
+	bodyCopy []byte
+	errChan  chan error
 }
 
 func newServer(ctx context.Context, workdir string) (*server, error) {
@@ -51,6 +52,8 @@ func newServer(ctx context.Context, workdir string) (*server, error) {
 			return webhookErr(fmt.Errorf("read request body: %w", err))
 		}
 
+		s.bodyCopy = body
+
 		cd := s.req.Header.Get("Content-Disposition")
 		if cd == "" {
 			return webhookErr(fmt.Errorf("no Content-Disposition header"))
@@ -125,6 +128,11 @@ func newServer(ctx context.Context, workdir string) (*server, error) {
 	}
 	webhookErrorHandler := func(c echo.Context) error {
 		s.req = c.Request()
+		body, err := io.ReadAll(s.req.Body)
+		if err != nil {
+			return webhookErr(fmt.Errorf("read request body: %w", err))
+		}
+		s.bodyCopy = body
 		return webhookErr(c.String(http.StatusOK, http.StatusText(http.StatusOK)))
 	}