netIdle cleanup + better default for pages where networkIdle timesout (#916)

- set default networkIdle to 2
- add netIdleMaxRequests as an option, default to 1 (in case of long
running requests)
- further fix for #913 
- avoid accidental logging

---------

Co-authored-by: Tessa Walsh <tessa@bitarchivist.net>

docs/docs/user-guide/cli-options.md

@@ -221,11 +221,13 @@ Options:
                                             interrupted, don't run post-crawl pr
                                             ocesses on interrupt
                                                       [boolean] [default: false]
-      --netIdleWait                         if set, wait for network idle after
+      --netIdleWait                         number of seconds to wait for networ
-                                            page load and after behaviors are do
+                                            k idle after page load and after beh
-                                            ne (in seconds). if -1 (default), de
+                                            aviors are done (default: 2)
-                                            termine based on scope
+                                                           [number] [default: 2]
-                                                          [number] [default: -1]
+      --netIdleMaxRequests                  max active requests allowed for netw
+                                            ork to be considered idle
+                                                                    [default: 1]
       --lang                                if set, sets the language used by th
                                             e browser, should be ISO 639 languag
                                             e[-country] code            [string]
@@ -318,6 +320,10 @@ Options:
       --sshProxyKnownHostsFile              path to SSH known hosts file for SOC
                                             KS5 over SSH proxy connection
                                                                         [string]
+      --extraChromeArgs                     Extra arguments to pass directly to
+                                            the Chrome instance (space-separated
+                                             or multiple --extraChromeArgs)
+                                                           [array] [default: []]
       --config                              Path to YAML config file
 ```
 

src/crawler.ts

@@ -2329,6 +2329,7 @@ self.__bx_behaviors.selectMainBehavior();
     try {
       await this.browser.waitForNetworkIdle(page, {
         timeout: this.params.netIdleWait * 1000,
+        concurrency: this.params.netIdleMaxRequests,
       });
     } catch (e) {
       logger.debug("waitForNetworkIdle timed out, ignoring", details);

src/util/argParser.ts

@@ -516,9 +516,15 @@ class ArgParser {
 
         netIdleWait: {
           describe:
-            "if set, wait for network idle after page load and after behaviors are done (in seconds). if -1 (default), determine based on scope",
+            "number of seconds to wait for network idle after page load and after behaviors are done (default: 2)",
           type: "number",
-          default: -1,
+          default: 2,
+        },
+
+        netIdleMaxRequests: {
+          describe:
+            "max active requests allowed for network to be considered idle",
+          default: 1,
         },
 
         lang: {
@@ -837,15 +843,6 @@ class ArgParser {
 
     argv.selectLinks = selectLinks;
 
-    if (argv.netIdleWait === -1) {
-      if (argv.scopeType === "page" || argv.scopeType === "page-spa") {
-        argv.netIdleWait = 15;
-      } else {
-        argv.netIdleWait = 2;
-      }
-      //logger.debug(`Set netIdleWait to ${argv.netIdleWait} seconds`);
-    }
-
     if (isQA && !argv.qaSource) {
       logger.fatal("--qaSource required for QA mode");
     }

src/util/browser.ts

@@ -23,8 +23,11 @@ import puppeteer, {
   LaunchOptions,
   Viewport,
   CookieData,
+  WaitForNetworkIdleOptions,
+  CDPSession,
+  Target,
+  Browser as PptrBrowser,
 } from "puppeteer-core";
-import { CDPSession, Target, Browser as PptrBrowser } from "puppeteer-core";
 import { Recorder } from "./recorder.js";
 import { timedRun } from "./timing.js";
 import assert from "node:assert";
@@ -244,6 +247,7 @@ export class Browser {
     try {
       child_process.execSync("rm ./Singleton*", {
         cwd: this.profileDir,
+        stdio: "ignore",
       });
     } catch (e) {
       // ignore
@@ -700,7 +704,7 @@ export class Browser {
     page.on("request", callback);
   }
 
-  async waitForNetworkIdle(page: Page, params: { timeout?: number }) {
+  async waitForNetworkIdle(page: Page, params: WaitForNetworkIdleOptions) {
     return await page.waitForNetworkIdle(params);
   }