From 247b87bbfea1f2a5e8200f76eeeb32abd1f4d7f6 Mon Sep 17 00:00:00 2001
From: gongzhenxing <gongzhenxing@costrip.cn>
Date: Wed, 20 Dec 2023 18:51:54 +0800
Subject: [PATCH 1/2] Support for multiple cookies

---
 src/config.ts | 14 ++++++++++----
 src/core.ts   | 16 +++++++++-------
 2 files changed, 19 insertions(+), 11 deletions(-)

diff --git a/src/config.ts b/src/config.ts
index 7e5f5fbf..e62340aa 100644
--- a/src/config.ts
+++ b/src/config.ts
@@ -37,10 +37,16 @@ export const configSchema = z.object({
   outputFileName: z.string(),
   /** Optional cookie to be set. E.g. for Cookie Consent */
   cookie: z
-    .object({
-      name: z.string(),
-      value: z.string(),
-    })
+    .union([
+      z.object({
+       name: z.string(),
+       value: z.string(),
+      }),
+      z.array(z.object({
+       name: z.string(),
+       value: z.string(),
+      })),
+     ])
     .optional(),
   /** Optional function to run for each page found */
   onVisitPage: z
diff --git a/src/core.ts b/src/core.ts
index 8e03bbe5..1f2e8f90 100644
--- a/src/core.ts
+++ b/src/core.ts
@@ -56,13 +56,15 @@ export async function crawl(config: Config) {
       // Use the requestHandler to process each of the crawled pages.
       async requestHandler({ request, page, enqueueLinks, log, pushData }) {
         if (config.cookie) {
-          // Set the cookie for the specific URL
-          const cookie = {
-            name: config.cookie.name,
-            value: config.cookie.value,
-            url: request.loadedUrl,
-          };
-          await page.context().addCookies([cookie]);
+          const cookies = (Array.isArray(config.cookie) ? config.cookie : [config.cookie])
+            .map((cookie)=>{
+              return {
+                name:cookie.name,
+                value:cookie.value,
+                url:request.loadedUrl
+              }
+          });
+          await page.context().addCookies(cookies);
         }
 
         const title = await page.title();

From 581cc7706c2c65485d0aa36853d9c0b97e4d48c0 Mon Sep 17 00:00:00 2001
From: gongzhenxing <gongzhenxing@costrip.cn>
Date: Mon, 25 Dec 2023 09:46:43 +0800
Subject: [PATCH 2/2] Set cookies in advance and modify code style

---
 src/config.ts | 16 +++++++++-------
 src/core.ts   | 26 +++++++++++++-------------
 2 files changed, 22 insertions(+), 20 deletions(-)

diff --git a/src/config.ts b/src/config.ts
index e62340aa..e5230b29 100644
--- a/src/config.ts
+++ b/src/config.ts
@@ -39,14 +39,16 @@ export const configSchema = z.object({
   cookie: z
     .union([
       z.object({
-       name: z.string(),
-       value: z.string(),
+        name: z.string(),
+        value: z.string(),
       }),
-      z.array(z.object({
-       name: z.string(),
-       value: z.string(),
-      })),
-     ])
+      z.array(
+        z.object({
+          name: z.string(),
+          value: z.string(),
+        }),
+      ),
+    ])
     .optional(),
   /** Optional function to run for each page found */
   onVisitPage: z
diff --git a/src/core.ts b/src/core.ts
index 1f2e8f90..93179b0d 100644
--- a/src/core.ts
+++ b/src/core.ts
@@ -55,18 +55,6 @@ export async function crawl(config: Config) {
     const crawler = new PlaywrightCrawler({
       // Use the requestHandler to process each of the crawled pages.
       async requestHandler({ request, page, enqueueLinks, log, pushData }) {
-        if (config.cookie) {
-          const cookies = (Array.isArray(config.cookie) ? config.cookie : [config.cookie])
-            .map((cookie)=>{
-              return {
-                name:cookie.name,
-                value:cookie.value,
-                url:request.loadedUrl
-              }
-          });
-          await page.context().addCookies(cookies);
-        }
-
         const title = await page.title();
         pageCounter++;
         log.info(
@@ -110,12 +98,24 @@ export async function crawl(config: Config) {
       // headless: false,
       preNavigationHooks: [
         // Abort requests for certain resource types
-        async ({ page, log }) => {
+        async ({ request, page, log }) => {
           // If there are no resource exclusions, return
           const RESOURCE_EXCLUSTIONS = config.resourceExclusions ?? [];
           if (RESOURCE_EXCLUSTIONS.length === 0) {
             return;
           }
+          if (config.cookie) {
+            const cookies = (
+              Array.isArray(config.cookie) ? config.cookie : [config.cookie]
+            ).map((cookie) => {
+              return {
+                name: cookie.name,
+                value: cookie.value,
+                url: request.loadedUrl,
+              };
+            });
+            await page.context().addCookies(cookies);
+          }
           await page.route(`**\/*.{${RESOURCE_EXCLUSTIONS.join()}}`, (route) =>
             route.abort("aborted"),
           );