Merge branch 'main' of github.com:jina-ai/url2text

backend/firebase.json

@@ -13,7 +13,8 @@
         ".git",
         "*.log",
         "*.local",
-        ".secret.*"
+        ".secret.*",
+        ".firebase-emu"
       ],
       "predeploy": [
         "npm --prefix \"$RESOURCE_DIR\" run build:clean",

backend/functions/package-lock.json

@@ -29,7 +29,10 @@
         "puppeteer": "^22.6.3",
         "puppeteer-extra": "^3.3.6",
         "puppeteer-extra-plugin-block-resources": "^2.4.3",
+        "puppeteer-extra-plugin-page-proxy": "^2.0.0",
         "puppeteer-extra-plugin-stealth": "^2.11.2",
+        "puppeteer-page-proxy": "^1.3.0",
+        "set-cookie-parser": "^2.6.0",
         "stripe": "^11.11.0",
         "tiktoken": "^1.0.10",
         "turndown": "^7.1.3",
@@ -42,6 +45,7 @@
         "@types/cors": "^2.8.17",
         "@types/generic-pool": "^3.8.1",
         "@types/node": "^18",
+        "@types/set-cookie-parser": "^2.4.7",
         "@typescript-eslint/eslint-plugin": "^5.12.0",
         "@typescript-eslint/parser": "^5.12.0",
         "eslint": "^8.9.0",
@@ -1986,6 +1990,17 @@
       "dev": true,
       "peer": true
     },
+    "node_modules/@sindresorhus/is": {
+      "version": "4.6.0",
+      "resolved": "https://registry.npmjs.org/@sindresorhus/is/-/is-4.6.0.tgz",
+      "integrity": "sha512-t09vSN3MdfsyCHoFcTRCH/iUtG7OJ0CsjzB8cjAmKc/va/kIgeDI/TxsigdncE/4be734m0cvIYwNaV4i2XqAw==",
+      "engines": {
+        "node": ">=10"
+      },
+      "funding": {
+        "url": "https://github.com/sindresorhus/is?sponsor=1"
+      }
+    },
     "node_modules/@sinonjs/commons": {
       "version": "3.0.1",
       "resolved": "https://registry.npmjs.org/@sinonjs/commons/-/commons-3.0.1.tgz",
@@ -2006,6 +2021,17 @@
         "@sinonjs/commons": "^3.0.0"
       }
     },
+    "node_modules/@szmarczak/http-timer": {
+      "version": "4.0.6",
+      "resolved": "https://registry.npmjs.org/@szmarczak/http-timer/-/http-timer-4.0.6.tgz",
+      "integrity": "sha512-4BAffykYOgO+5nzBWYwE3W90sBgLJoUPRWWcL8wlyiM8IB8ipJz3UMJ9KXQd1RKQXpKp8Tutn80HZtWsu2u76w==",
+      "dependencies": {
+        "defer-to-connect": "^2.0.0"
+      },
+      "engines": {
+        "node": ">=10"
+      }
+    },
     "node_modules/@tootallnate/once": {
       "version": "2.0.0",
       "resolved": "https://registry.npmjs.org/@tootallnate/once/-/once-2.0.0.tgz",
@@ -2091,6 +2117,17 @@
         "@types/node": "*"
       }
     },
+    "node_modules/@types/cacheable-request": {
+      "version": "6.0.3",
+      "resolved": "https://registry.npmjs.org/@types/cacheable-request/-/cacheable-request-6.0.3.tgz",
+      "integrity": "sha512-IQ3EbTzGxIigb1I3qPZc1rWJnH0BmSKv5QYTalEwweFvyBDLSAe24zP0le/hyi7ecGfZVlIVAg4BZqb8WBwKqw==",
+      "dependencies": {
+        "@types/http-cache-semantics": "*",
+        "@types/keyv": "^3.1.4",
+        "@types/node": "*",
+        "@types/responselike": "^1.0.0"
+      }
+    },
     "node_modules/@types/caseless": {
       "version": "0.12.5",
       "resolved": "https://registry.npmjs.org/@types/caseless/-/caseless-0.12.5.tgz",
@@ -2161,6 +2198,11 @@
         "@types/node": "*"
       }
     },
+    "node_modules/@types/http-cache-semantics": {
+      "version": "4.0.4",
+      "resolved": "https://registry.npmjs.org/@types/http-cache-semantics/-/http-cache-semantics-4.0.4.tgz",
+      "integrity": "sha512-1m0bIFVc7eJWyve9S0RnuRgcQqF/Xd5QsUZAZeQFr1Q3/p9JWoQQEqmVy+DPTNpGXwhgIetAoYF8JSc33q29QA=="
+    },
     "node_modules/@types/http-errors": {
       "version": "2.0.4",
       "resolved": "https://registry.npmjs.org/@types/http-errors/-/http-errors-2.0.4.tgz",
@@ -2213,6 +2255,14 @@
         "@types/node": "*"
       }
     },
+    "node_modules/@types/keyv": {
+      "version": "3.1.4",
+      "resolved": "https://registry.npmjs.org/@types/keyv/-/keyv-3.1.4.tgz",
+      "integrity": "sha512-BQ5aZNSCpj7D6K2ksrRCTmKRLEpnPvWDiLPfoGyhZ++8YtiK9d/3DBKPJgry359X/P1PfruyYwvnvwFjuEiEIg==",
+      "dependencies": {
+        "@types/node": "*"
+      }
+    },
     "node_modules/@types/lodash": {
       "version": "4.17.0",
       "resolved": "https://registry.npmjs.org/@types/lodash/-/lodash-4.17.0.tgz",
@@ -2294,6 +2344,14 @@
         "node": ">= 0.12"
       }
     },
+    "node_modules/@types/responselike": {
+      "version": "1.0.3",
+      "resolved": "https://registry.npmjs.org/@types/responselike/-/responselike-1.0.3.tgz",
+      "integrity": "sha512-H/+L+UkTV33uf49PH5pCAUBVPNj2nDBXTN+qS1dOwyyg24l3CcicicCA7ca+HMvJBZcFgl5r8e+RR6elsb4Lyw==",
+      "dependencies": {
+        "@types/node": "*"
+      }
+    },
     "node_modules/@types/semver": {
       "version": "7.5.8",
       "resolved": "https://registry.npmjs.org/@types/semver/-/semver-7.5.8.tgz",
@@ -2319,6 +2377,15 @@
         "@types/send": "*"
       }
     },
+    "node_modules/@types/set-cookie-parser": {
+      "version": "2.4.7",
+      "resolved": "https://registry.npmjs.org/@types/set-cookie-parser/-/set-cookie-parser-2.4.7.tgz",
+      "integrity": "sha512-+ge/loa0oTozxip6zmhRIk8Z/boU51wl9Q6QdLZcokIGMzY5lFXYy/x7Htj2HTC6/KZP1hUbZ1ekx8DYXICvWg==",
+      "dev": true,
+      "dependencies": {
+        "@types/node": "*"
+      }
+    },
     "node_modules/@types/stack-utils": {
       "version": "2.0.3",
       "resolved": "https://registry.npmjs.org/@types/stack-utils/-/stack-utils-2.0.3.tgz",
@@ -3447,6 +3514,45 @@
         "node": ">= 6.0.0"
       }
     },
+    "node_modules/cacheable-lookup": {
+      "version": "5.0.4",
+      "resolved": "https://registry.npmjs.org/cacheable-lookup/-/cacheable-lookup-5.0.4.tgz",
+      "integrity": "sha512-2/kNscPhpcxrOigMZzbiWF7dz8ilhb/nIHU3EyZiXWXpeq/au8qJ8VhdftMkty3n7Gj6HIGalQG8oiBNB3AJgA==",
+      "engines": {
+        "node": ">=10.6.0"
+      }
+    },
+    "node_modules/cacheable-request": {
+      "version": "7.0.4",
+      "resolved": "https://registry.npmjs.org/cacheable-request/-/cacheable-request-7.0.4.tgz",
+      "integrity": "sha512-v+p6ongsrp0yTGbJXjgxPow2+DL93DASP4kXCDKb8/bwRtt9OEF3whggkkDkGNzgcWy2XaF4a8nZglC7uElscg==",
+      "dependencies": {
+        "clone-response": "^1.0.2",
+        "get-stream": "^5.1.0",
+        "http-cache-semantics": "^4.0.0",
+        "keyv": "^4.0.0",
+        "lowercase-keys": "^2.0.0",
+        "normalize-url": "^6.0.1",
+        "responselike": "^2.0.0"
+      },
+      "engines": {
+        "node": ">=8"
+      }
+    },
+    "node_modules/cacheable-request/node_modules/get-stream": {
+      "version": "5.2.0",
+      "resolved": "https://registry.npmjs.org/get-stream/-/get-stream-5.2.0.tgz",
+      "integrity": "sha512-nBF+F1rAZVCu/p7rjzgA+Yb4lfYXrpl7a6VmJrU8wF9I1CKvP/QwPNZHnOlwbTkY6dvtFIzFMSyQXbLoTQPRpA==",
+      "dependencies": {
+        "pump": "^3.0.0"
+      },
+      "engines": {
+        "node": ">=8"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/call-bind": {
       "version": "1.0.7",
       "resolved": "https://registry.npmjs.org/call-bind/-/call-bind-1.0.7.tgz",
@@ -3642,6 +3748,25 @@
         "node": ">=0.10.0"
       }
     },
+    "node_modules/clone-response": {
+      "version": "1.0.3",
+      "resolved": "https://registry.npmjs.org/clone-response/-/clone-response-1.0.3.tgz",
+      "integrity": "sha512-ROoL94jJH2dUVML2Y/5PEDNaSHgeOdSDicUyS7izcF63G6sTc/FTjLub4b8Il9S8S0beOfYt0TaA5qvFK+w0wA==",
+      "dependencies": {
+        "mimic-response": "^1.0.0"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
+    "node_modules/clone-response/node_modules/mimic-response": {
+      "version": "1.0.1",
+      "resolved": "https://registry.npmjs.org/mimic-response/-/mimic-response-1.0.1.tgz",
+      "integrity": "sha512-j5EctnkH7amfV/q5Hgmoal1g2QHFJRraOtmx0JpIqkxhBhI/lJSl1nMpQ45hVarwNETOoWEimndZ4QK0RHxuxQ==",
+      "engines": {
+        "node": ">=4"
+      }
+    },
     "node_modules/co": {
       "version": "4.6.0",
       "resolved": "https://registry.npmjs.org/co/-/co-4.6.0.tgz",
@@ -4003,6 +4128,17 @@
         "url": "https://github.com/sponsors/sindresorhus"
       }
     },
+    "node_modules/decompress-response/node_modules/mimic-response": {
+      "version": "3.1.0",
+      "resolved": "https://registry.npmjs.org/mimic-response/-/mimic-response-3.1.0.tgz",
+      "integrity": "sha512-z0yWI+4FDrrweS8Zmt4Ej5HdJmky15+L2e6Wgn3+iK5fWzb6T3fhNFq2+MeTRb064c6Wr4N/wv0DzQTjNzHNGQ==",
+      "engines": {
+        "node": ">=10"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/dedent": {
       "version": "1.5.3",
       "resolved": "https://registry.npmjs.org/dedent/-/dedent-1.5.3.tgz",
@@ -4046,6 +4182,14 @@
         "node": ">=0.10.0"
       }
     },
+    "node_modules/defer-to-connect": {
+      "version": "2.0.1",
+      "resolved": "https://registry.npmjs.org/defer-to-connect/-/defer-to-connect-2.0.1.tgz",
+      "integrity": "sha512-4tvttepXG1VaYGrRibk5EwJd1t4udunSOVMdLSAL6mId1ix438oPwPZMALY41FCijukO1L0twNcGsdzS7dHgDg==",
+      "engines": {
+        "node": ">=10"
+      }
+    },
     "node_modules/define-data-property": {
       "version": "1.1.4",
       "resolved": "https://registry.npmjs.org/define-data-property/-/define-data-property-1.1.4.tgz",
@@ -5837,6 +5981,30 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/got": {
+      "version": "11.8.6",
+      "resolved": "https://registry.npmjs.org/got/-/got-11.8.6.tgz",
+      "integrity": "sha512-6tfZ91bOr7bOXnK7PRDCGBLa1H4U080YHNaAQ2KsMGlLEzRbk44nsZF2E1IeRc3vtJHPVbKCYgdFbaGO2ljd8g==",
+      "dependencies": {
+        "@sindresorhus/is": "^4.0.0",
+        "@szmarczak/http-timer": "^4.0.5",
+        "@types/cacheable-request": "^6.0.1",
+        "@types/responselike": "^1.0.0",
+        "cacheable-lookup": "^5.0.3",
+        "cacheable-request": "^7.0.2",
+        "decompress-response": "^6.0.0",
+        "http2-wrapper": "^1.0.0-beta.5.2",
+        "lowercase-keys": "^2.0.0",
+        "p-cancelable": "^2.0.0",
+        "responselike": "^2.0.0"
+      },
+      "engines": {
+        "node": ">=10.19.0"
+      },
+      "funding": {
+        "url": "https://github.com/sindresorhus/got?sponsor=1"
+      }
+    },
     "node_modules/graceful-fs": {
       "version": "4.2.11",
       "resolved": "https://registry.npmjs.org/graceful-fs/-/graceful-fs-4.2.11.tgz",
@@ -6024,6 +6192,11 @@
         "node": ">= 0.6"
       }
     },
+    "node_modules/http-cache-semantics": {
+      "version": "4.1.1",
+      "resolved": "https://registry.npmjs.org/http-cache-semantics/-/http-cache-semantics-4.1.1.tgz",
+      "integrity": "sha512-er295DKPVsV82j5kw1Gjt+ADA/XYHsajl82cGNQG2eyoPkvgUhX+nDIyelzhIWbbsXP39EHcI6l5tYs2FYqYXQ=="
+    },
     "node_modules/http-errors": {
       "version": "2.0.0",
       "resolved": "https://registry.npmjs.org/http-errors/-/http-errors-2.0.0.tgz",
@@ -6067,6 +6240,18 @@
         "node": ">= 14"
       }
     },
+    "node_modules/http2-wrapper": {
+      "version": "1.0.3",
+      "resolved": "https://registry.npmjs.org/http2-wrapper/-/http2-wrapper-1.0.3.tgz",
+      "integrity": "sha512-V+23sDMr12Wnz7iTcDeJr3O6AIxlnvT/bmaAAAP/Xda35C90p9599p0F1eHR/N1KILWSoWVAiOMFjBBXaXSMxg==",
+      "dependencies": {
+        "quick-lru": "^5.1.1",
+        "resolve-alpn": "^1.0.0"
+      },
+      "engines": {
+        "node": ">=10.19.0"
+      }
+    },
     "node_modules/https-proxy-agent": {
       "version": "5.0.1",
       "resolved": "https://registry.npmjs.org/https-proxy-agent/-/https-proxy-agent-5.0.1.tgz",
@@ -7391,8 +7576,7 @@
     "node_modules/json-buffer": {
       "version": "3.0.1",
       "resolved": "https://registry.npmjs.org/json-buffer/-/json-buffer-3.0.1.tgz",
-      "integrity": "sha512-4bV5BfR2mqfQTJm+V5tPPdf+ZpuhiIvTuAB5g8kcrXOZpTT/QwwVRWBywX1ozr6lEuPdbHxwaJlm9G6mI2sfSQ==",
+      "integrity": "sha512-4bV5BfR2mqfQTJm+V5tPPdf+ZpuhiIvTuAB5g8kcrXOZpTT/QwwVRWBywX1ozr6lEuPdbHxwaJlm9G6mI2sfSQ=="
-      "dev": true
     },
     "node_modules/json-parse-even-better-errors": {
       "version": "2.3.1",
@@ -7550,7 +7734,6 @@
       "version": "4.5.4",
       "resolved": "https://registry.npmjs.org/keyv/-/keyv-4.5.4.tgz",
       "integrity": "sha512-oxVHkHR/EJf2CNXnWxRLW6mg7JyCCUcG0DtEGmL2ctUo1PNTin1PUil+r/+4r5MpVgC/fn1kjsx7mjSujKqIpw==",
-      "dev": true,
       "dependencies": {
         "json-buffer": "3.0.1"
       }
@@ -7869,6 +8052,14 @@
       "integrity": "sha512-BFRuQUqc7x2NWxfJBCyUrN8iYUYznzL9JROmRz1gZ6KlOIgmoD+njPVbb+VNn2nGMKggMsK79iUNErillsrx7w==",
       "optional": true
     },
+    "node_modules/lowercase-keys": {
+      "version": "2.0.0",
+      "resolved": "https://registry.npmjs.org/lowercase-keys/-/lowercase-keys-2.0.0.tgz",
+      "integrity": "sha512-tqNXrS78oMOE73NMxK4EMLQsQowWf8jKooH9g7xPavRT706R6bkQJ6DY2Te7QukaZsulxa30wQ7bk0pm4XiHmA==",
+      "engines": {
+        "node": ">=8"
+      }
+    },
     "node_modules/lru-cache": {
       "version": "5.1.1",
       "resolved": "https://registry.npmjs.org/lru-cache/-/lru-cache-5.1.1.tgz",
@@ -8054,17 +8245,6 @@
         "node": ">=6"
       }
     },
-    "node_modules/mimic-response": {
-      "version": "3.1.0",
-      "resolved": "https://registry.npmjs.org/mimic-response/-/mimic-response-3.1.0.tgz",
-      "integrity": "sha512-z0yWI+4FDrrweS8Zmt4Ej5HdJmky15+L2e6Wgn3+iK5fWzb6T3fhNFq2+MeTRb064c6Wr4N/wv0DzQTjNzHNGQ==",
-      "engines": {
-        "node": ">=10"
-      },
-      "funding": {
-        "url": "https://github.com/sponsors/sindresorhus"
-      }
-    },
     "node_modules/minimatch": {
       "version": "3.1.2",
       "resolved": "https://registry.npmjs.org/minimatch/-/minimatch-3.1.2.tgz",
@@ -8423,6 +8603,17 @@
         "node": ">=0.10.0"
       }
     },
+    "node_modules/normalize-url": {
+      "version": "6.1.0",
+      "resolved": "https://registry.npmjs.org/normalize-url/-/normalize-url-6.1.0.tgz",
+      "integrity": "sha512-DlL+XwOy3NxAQ8xuC0okPgK46iuVNAK01YN7RueYBqqFeGsBjV9XmCAzAdgt+667bCl5kPh9EqKKDwnaPG1I7A==",
+      "engines": {
+        "node": ">=10"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/npm-run-path": {
       "version": "4.0.1",
       "resolved": "https://registry.npmjs.org/npm-run-path/-/npm-run-path-4.0.1.tgz",
@@ -8632,6 +8823,14 @@
         "node": ">= 0.8.0"
       }
     },
+    "node_modules/p-cancelable": {
+      "version": "2.1.1",
+      "resolved": "https://registry.npmjs.org/p-cancelable/-/p-cancelable-2.1.1.tgz",
+      "integrity": "sha512-BZOr3nRQHOntUjTrH8+Lh54smKHoHyur8We1V8DSMVrl5A2malOOwuJRnKRDjSnkoeBh4at6BwEnb5I7Jl31wg==",
+      "engines": {
+        "node": ">=8"
+      }
+    },
     "node_modules/p-limit": {
       "version": "3.1.0",
       "resolved": "https://registry.npmjs.org/p-limit/-/p-limit-3.1.0.tgz",
@@ -9243,6 +9442,11 @@
       "resolved": "https://registry.npmjs.org/pseudomap/-/pseudomap-1.0.2.tgz",
       "integrity": "sha512-b/YwNhb8lk1Zz2+bXXpS/LK9OisiZZ1SNsSLxN1x2OXVEhW2Ckr/7mWE5vrC1ZTiJlD9g19jWszTmJsB+oEpFQ=="
     },
+    "node_modules/psl": {
+      "version": "1.9.0",
+      "resolved": "https://registry.npmjs.org/psl/-/psl-1.9.0.tgz",
+      "integrity": "sha512-E/ZsdU4HLs/68gYzgGTkMicWTLPdAftJLfJFlLUAAKZGkStNU72sZjT66SnMDVOfOWY/YAoiD7Jxa9iHvngcag=="
+    },
     "node_modules/pump": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/pump/-/pump-3.0.0.tgz",
@@ -9367,6 +9571,58 @@
         }
       }
     },
+    "node_modules/puppeteer-extra-plugin-page-proxy": {
+      "version": "2.0.0",
+      "resolved": "https://registry.npmjs.org/puppeteer-extra-plugin-page-proxy/-/puppeteer-extra-plugin-page-proxy-2.0.0.tgz",
+      "integrity": "sha512-G8pvHdHK1dO1dgFvYL+dJIlykUKjLkGUvPjzHE3R/eurqAkD4VZ9lWOU/CxYiKPhK2JxlG9QmjGjhxR6IOuP7w==",
+      "dependencies": {
+        "debug": "^4.1.1",
+        "got": "^11.8.5",
+        "http-proxy-agent": "^5.0.0",
+        "https-proxy-agent": "^5.0.1",
+        "puppeteer-extra-plugin": "^3.2.3",
+        "socks-proxy-agent": "^7.0.0",
+        "tough-cookie": "^4.1.2"
+      },
+      "peerDependencies": {
+        "playwright-extra": "*",
+        "puppeteer-extra": "*"
+      },
+      "peerDependenciesMeta": {
+        "playwright-extra": {
+          "optional": true
+        },
+        "puppeteer-extra": {
+          "optional": true
+        }
+      }
+    },
+    "node_modules/puppeteer-extra-plugin-page-proxy/node_modules/http-proxy-agent": {
+      "version": "5.0.0",
+      "resolved": "https://registry.npmjs.org/http-proxy-agent/-/http-proxy-agent-5.0.0.tgz",
+      "integrity": "sha512-n2hY8YdoRE1i7r6M0w9DIw5GgZN0G25P8zLCRQ8rjXtTU3vsNFBI/vWK/UIeE6g5MUUz6avwAPXmL6Fy9D/90w==",
+      "dependencies": {
+        "@tootallnate/once": "2",
+        "agent-base": "6",
+        "debug": "4"
+      },
+      "engines": {
+        "node": ">= 6"
+      }
+    },
+    "node_modules/puppeteer-extra-plugin-page-proxy/node_modules/socks-proxy-agent": {
+      "version": "7.0.0",
+      "resolved": "https://registry.npmjs.org/socks-proxy-agent/-/socks-proxy-agent-7.0.0.tgz",
+      "integrity": "sha512-Fgl0YPZ902wEsAyiQ+idGd1A7rSFx/ayC1CQVMw5P+EQx2V0SgpGtf6OKFhVjPflPUl9YMmEOnmfjCdMUsygww==",
+      "dependencies": {
+        "agent-base": "^6.0.2",
+        "debug": "^4.3.3",
+        "socks": "^2.6.2"
+      },
+      "engines": {
+        "node": ">= 10"
+      }
+    },
     "node_modules/puppeteer-extra-plugin-stealth": {
       "version": "2.11.2",
       "resolved": "https://registry.npmjs.org/puppeteer-extra-plugin-stealth/-/puppeteer-extra-plugin-stealth-2.11.2.tgz",
@@ -9457,6 +9713,44 @@
         }
       }
     },
+    "node_modules/puppeteer-page-proxy": {
+      "version": "1.3.0",
+      "resolved": "https://registry.npmjs.org/puppeteer-page-proxy/-/puppeteer-page-proxy-1.3.0.tgz",
+      "integrity": "sha512-PDpLjJfcUKiLvUZ3yQJeUcP1d+7nW17s2LZIrKH0gyxEN4zTGkCvB9/HwquPgYq5YcVi8QugsvBckP/K9Vn/iw==",
+      "dependencies": {
+        "got": "^11.8.5",
+        "http-proxy-agent": "^5.0.0",
+        "https-proxy-agent": "^5.0.1",
+        "socks-proxy-agent": "^7.0.0",
+        "tough-cookie": "^4.1.2"
+      }
+    },
+    "node_modules/puppeteer-page-proxy/node_modules/http-proxy-agent": {
+      "version": "5.0.0",
+      "resolved": "https://registry.npmjs.org/http-proxy-agent/-/http-proxy-agent-5.0.0.tgz",
+      "integrity": "sha512-n2hY8YdoRE1i7r6M0w9DIw5GgZN0G25P8zLCRQ8rjXtTU3vsNFBI/vWK/UIeE6g5MUUz6avwAPXmL6Fy9D/90w==",
+      "dependencies": {
+        "@tootallnate/once": "2",
+        "agent-base": "6",
+        "debug": "4"
+      },
+      "engines": {
+        "node": ">= 6"
+      }
+    },
+    "node_modules/puppeteer-page-proxy/node_modules/socks-proxy-agent": {
+      "version": "7.0.0",
+      "resolved": "https://registry.npmjs.org/socks-proxy-agent/-/socks-proxy-agent-7.0.0.tgz",
+      "integrity": "sha512-Fgl0YPZ902wEsAyiQ+idGd1A7rSFx/ayC1CQVMw5P+EQx2V0SgpGtf6OKFhVjPflPUl9YMmEOnmfjCdMUsygww==",
+      "dependencies": {
+        "agent-base": "^6.0.2",
+        "debug": "^4.3.3",
+        "socks": "^2.6.2"
+      },
+      "engines": {
+        "node": ">= 10"
+      }
+    },
     "node_modules/pure-rand": {
       "version": "6.1.0",
       "resolved": "https://registry.npmjs.org/pure-rand/-/pure-rand-6.1.0.tgz",
@@ -9523,6 +9817,11 @@
         "url": "https://github.com/sponsors/sindresorhus"
       }
     },
+    "node_modules/querystringify": {
+      "version": "2.2.0",
+      "resolved": "https://registry.npmjs.org/querystringify/-/querystringify-2.2.0.tgz",
+      "integrity": "sha512-FIqgj2EUvTa7R50u0rGsyTftzjYmv/a3hO345bZNrqabNqjtgiDMgmo4mkUjd+nzU5oF3dClKqFIPUKybUyqoQ=="
+    },
     "node_modules/queue-microtask": {
       "version": "1.2.3",
       "resolved": "https://registry.npmjs.org/queue-microtask/-/queue-microtask-1.2.3.tgz",
@@ -9554,6 +9853,17 @@
       "integrity": "sha512-tYC1Q1hgyRuHgloV/YXs2w15unPVh8qfu/qCTfhTYamaw7fyhumKa2yGpdSo87vY32rIclj+4fWYQXUMs9EHvg==",
       "optional": true
     },
+    "node_modules/quick-lru": {
+      "version": "5.1.1",
+      "resolved": "https://registry.npmjs.org/quick-lru/-/quick-lru-5.1.1.tgz",
+      "integrity": "sha512-WuyALRjWPDGtt/wzJiadO5AXY+8hZ80hVpe6MyivgraREW751X3SbhRvG3eLKOYN+8VEvqLcf3wdnt44Z4S4SA==",
+      "engines": {
+        "node": ">=10"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/range-parser": {
       "version": "1.2.1",
       "resolved": "https://registry.npmjs.org/range-parser/-/range-parser-1.2.1.tgz",
@@ -9709,6 +10019,11 @@
         "node": ">=0.10.0"
       }
     },
+    "node_modules/requires-port": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/requires-port/-/requires-port-1.0.0.tgz",
+      "integrity": "sha512-KigOCHcocU3XODJxsu8i/j8T9tzT4adHiecwORRQ0ZZFcp7ahwXuRU1m+yuO90C5ZUyGeGfocHDI14M3L3yDAQ=="
+    },
     "node_modules/resolve": {
       "version": "1.22.8",
       "resolved": "https://registry.npmjs.org/resolve/-/resolve-1.22.8.tgz",
@@ -9726,6 +10041,11 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/resolve-alpn": {
+      "version": "1.2.1",
+      "resolved": "https://registry.npmjs.org/resolve-alpn/-/resolve-alpn-1.2.1.tgz",
+      "integrity": "sha512-0a1F4l73/ZFZOakJnQ3FvkJ2+gSTQWz/r2KE5OdDY0TxPm5h4GkqkWWfM47T7HsbnOtcJVEF4epCVy6u7Q3K+g=="
+    },
     "node_modules/resolve-cwd": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/resolve-cwd/-/resolve-cwd-3.0.0.tgz",
@@ -9767,6 +10087,17 @@
         "node": ">=10"
       }
     },
+    "node_modules/responselike": {
+      "version": "2.0.1",
+      "resolved": "https://registry.npmjs.org/responselike/-/responselike-2.0.1.tgz",
+      "integrity": "sha512-4gl03wn3hj1HP3yzgdI7d3lCkF95F21Pz4BPGvKHinyQzALR5CapwC8yIi0Rh58DEMQ/SguC03wFj2k0M/mHhw==",
+      "dependencies": {
+        "lowercase-keys": "^2.0.0"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/retry": {
       "version": "0.13.1",
       "resolved": "https://registry.npmjs.org/retry/-/retry-0.13.1.tgz",
@@ -10029,6 +10360,11 @@
       "resolved": "https://registry.npmjs.org/set-blocking/-/set-blocking-2.0.0.tgz",
       "integrity": "sha512-KiKBS8AnWGEyLzofFfmvKwpdPzqiy16LvQfK3yv/fVH7Bj13/wl3JSR1J+rfgRE9q7xUJK4qvgS8raSOeLUehw=="
     },
+    "node_modules/set-cookie-parser": {
+      "version": "2.6.0",
+      "resolved": "https://registry.npmjs.org/set-cookie-parser/-/set-cookie-parser-2.6.0.tgz",
+      "integrity": "sha512-RVnVQxTXuerk653XfuliOxBP81Sf0+qfQE73LIYKcyMYHG94AuH0kgrQpRDuTZnSmjpysHmzxJXKNfa6PjFhyQ=="
+    },
     "node_modules/set-function-length": {
       "version": "1.2.2",
       "resolved": "https://registry.npmjs.org/set-function-length/-/set-function-length-1.2.2.tgz",
@@ -10767,6 +11103,28 @@
         "node": ">=0.6"
       }
     },
+    "node_modules/tough-cookie": {
+      "version": "4.1.3",
+      "resolved": "https://registry.npmjs.org/tough-cookie/-/tough-cookie-4.1.3.tgz",
+      "integrity": "sha512-aX/y5pVRkfRnfmuX+OdbSdXvPe6ieKX/G2s7e98f4poJHnqH3281gDPm/metm6E/WRamfx7WC4HUqkWHfQHprw==",
+      "dependencies": {
+        "psl": "^1.1.33",
+        "punycode": "^2.1.1",
+        "universalify": "^0.2.0",
+        "url-parse": "^1.5.3"
+      },
+      "engines": {
+        "node": ">=6"
+      }
+    },
+    "node_modules/tough-cookie/node_modules/universalify": {
+      "version": "0.2.0",
+      "resolved": "https://registry.npmjs.org/universalify/-/universalify-0.2.0.tgz",
+      "integrity": "sha512-CJ1QgKmNg3CwvAv/kOFmtnEN05f0D/cn9QntgNOQlQF9dgvVTHj3t+8JPdjqawCHk7V/KA+fbUqzZ9XWhcqPUg==",
+      "engines": {
+        "node": ">= 4.0.0"
+      }
+    },
     "node_modules/tr46": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/tr46/-/tr46-3.0.0.tgz",
@@ -11127,6 +11485,15 @@
         "punycode": "^2.1.0"
       }
     },
+    "node_modules/url-parse": {
+      "version": "1.5.10",
+      "resolved": "https://registry.npmjs.org/url-parse/-/url-parse-1.5.10.tgz",
+      "integrity": "sha512-WypcfiRhfeUP9vvF0j6rw0J3hrWrw6iZv3+22h6iRMJ/8z1Tj6XfLP4DsUix5MhMPnXpiHDoKyoZ/bdCkwBCiQ==",
+      "dependencies": {
+        "querystringify": "^2.1.1",
+        "requires-port": "^1.0.0"
+      }
+    },
     "node_modules/urlpattern-polyfill": {
       "version": "10.0.0",
       "resolved": "https://registry.npmjs.org/urlpattern-polyfill/-/urlpattern-polyfill-10.0.0.tgz",

backend/functions/package.json

@@ -49,7 +49,10 @@
     "puppeteer": "^22.6.3",
     "puppeteer-extra": "^3.3.6",
     "puppeteer-extra-plugin-block-resources": "^2.4.3",
+    "puppeteer-extra-plugin-page-proxy": "^2.0.0",
     "puppeteer-extra-plugin-stealth": "^2.11.2",
+    "puppeteer-page-proxy": "^1.3.0",
+    "set-cookie-parser": "^2.6.0",
     "stripe": "^11.11.0",
     "tiktoken": "^1.0.10",
     "turndown": "^7.1.3",
@@ -62,6 +65,7 @@
     "@types/cors": "^2.8.17",
     "@types/generic-pool": "^3.8.1",
     "@types/node": "^18",
+    "@types/set-cookie-parser": "^2.4.7",
     "@typescript-eslint/eslint-plugin": "^5.12.0",
     "@typescript-eslint/parser": "^5.12.0",
     "eslint": "^8.9.0",

backend/functions/src/cloud-functions/crawler.ts

@@ -1,51 +1,25 @@
-import { assignTransferProtocolMeta, marshalErrorLike, RPCHost, RPCReflection, AssertionFailureError, ParamValidationError } from 'civkit';
+import {
+    assignTransferProtocolMeta, marshalErrorLike,
+    RPCHost, RPCReflection,
+    HashManager,
+    AssertionFailureError, ParamValidationError,
+} from 'civkit';
 import { singleton } from 'tsyringe';
-import { CloudHTTPv2, Ctx, Logger, OutputServerEventStream, RPCReflect } from '../shared';
+import { CloudHTTPv2, Ctx, FirebaseStorageBucketControl, Logger, OutputServerEventStream, RPCReflect } from '../shared';
 import _ from 'lodash';
-import { PageSnapshot, PuppeteerControl } from '../services/puppeteer';
+import { PageSnapshot, PuppeteerControl, ScrappingOptions } from '../services/puppeteer';
 import { Request, Response } from 'express';
 import normalizeUrl from "@esm2cjs/normalize-url";
 import { AltTextService } from '../services/alt-text';
 import TurndownService from 'turndown';
+import { parseString as parseSetCookieString } from 'set-cookie-parser';
+import { CookieParam } from 'puppeteer';
+import { Crawled } from '../db/crawled';
+import { tidyMarkdown } from '../utils/markdown';
+import { cleanAttribute } from '../utils/misc';
+import { randomUUID } from 'crypto';
 
-function tidyMarkdown(markdown: string): string {
+const md5Hasher = new HashManager('md5', 'hex');
-
-    // Step 1: Handle complex broken links with text and optional images spread across multiple lines
-    let normalizedMarkdown = markdown.replace(/\[\s*([^]+?)\s*\]\s*\(\s*([^)]+)\s*\)/g, (match, text, url) => {
-        // Remove internal new lines and excessive spaces within the text
-        text = text.replace(/\s+/g, ' ').trim();
-        url = url.replace(/\s+/g, '').trim();
-        return `[${text}](${url})`;
-    });
-
-    normalizedMarkdown = normalizedMarkdown.replace(/\[\s*([^!]*?)\s*\n*(?:!\[([^\]]*)\]\((.*?)\))?\s*\n*\]\s*\(\s*([^)]+)\s*\)/g, (match, text, alt, imgUrl, linkUrl) => {
-        // Normalize by removing excessive spaces and new lines
-        text = text.replace(/\s+/g, ' ').trim();
-        alt = alt ? alt.replace(/\s+/g, ' ').trim() : '';
-        imgUrl = imgUrl ? imgUrl.replace(/\s+/g, '').trim() : '';
-        linkUrl = linkUrl.replace(/\s+/g, '').trim();
-        if (imgUrl) {
-            return `[${text} ![${alt}](${imgUrl})](${linkUrl})`;
-        } else {
-            return `[${text}](${linkUrl})`;
-        }
-    });
-
-    // Step 2: Normalize regular links that may be broken across lines
-    normalizedMarkdown = normalizedMarkdown.replace(/\[\s*([^\]]+)\]\s*\(\s*([^)]+)\)/g, (match, text, url) => {
-        text = text.replace(/\s+/g, ' ').trim();
-        url = url.replace(/\s+/g, '').trim();
-        return `[${text}](${url})`;
-    });
-
-    // Step 3: Replace more than two consecutive empty lines with exactly two empty lines
-    normalizedMarkdown = normalizedMarkdown.replace(/\n{3,}/g, '\n\n');
-
-    // Step 4: Remove leading spaces from each line
-    normalizedMarkdown = normalizedMarkdown.replace(/^[ \t]+/gm, '');
-
-    return normalizedMarkdown.trim();
-}
 
 @singleton()
 export class CrawlerHost extends RPCHost {
@@ -53,12 +27,29 @@ export class CrawlerHost extends RPCHost {
 
     turnDownPlugins = [require('turndown-plugin-gfm').tables];
 
+    cacheRetentionMs = 1000 * 3600 * 24 * 7;
+    cacheValidMs = 1000 * 300;
+    urlValidMs = 1000 * 3600 * 4;
+
     constructor(
         protected globalLogger: Logger,
         protected puppeteerControl: PuppeteerControl,
         protected altTextService: AltTextService,
+        protected firebaseObjectStorage: FirebaseStorageBucketControl,
     ) {
         super(...arguments);
+
+        puppeteerControl.on('crawled', async (snapshot: PageSnapshot, options: ScrappingOptions & { url: URL; }) => {
+            if (!snapshot.title?.trim()) {
+                return;
+            }
+            if (options.cookies?.length) {
+                // Potential privacy issue, dont cache if cookies are used
+                return;
+            }
+
+            await this.setToCache(options.url, snapshot);
+        });
     }
 
     override async init() {
@@ -67,16 +58,51 @@ export class CrawlerHost extends RPCHost {
         this.emit('ready');
     }
 
-    async formatSnapshot(snapshot: PageSnapshot, nominalUrl?: string) {
+    async formatSnapshot(mode: string | 'markdown' | 'html' | 'text' | 'screenshot', snapshot: PageSnapshot & {
+        screenshotUrl?: string;
+    }, nominalUrl?: URL) {
+        if (mode === 'screenshot') {
+            if (snapshot.screenshot && !snapshot.screenshotUrl) {
+                const fid = `instant-screenshots/${randomUUID()}`;
+                await this.firebaseObjectStorage.saveFile(fid, snapshot.screenshot, {
+                    metadata: {
+                        contentType: 'image/png',
+                    }
+                });
+                snapshot.screenshotUrl = await this.firebaseObjectStorage.signDownloadUrl(fid, Date.now() + this.urlValidMs);
+            }
+
+            return {
+                screenshotUrl: snapshot.screenshotUrl,
+                toString() {
+                    return this.screenshotUrl;
+                }
+            };
+        }
+        if (mode === 'html') {
+            return {
+                html: snapshot.html,
+                toString() {
+                    return this.html;
+                }
+            };
+        }
+        if (mode === 'text') {
+            return {
+                text: snapshot.text,
+                toString() {
+                    return this.text;
+                }
+            };
+        }
+
         const toBeTurnedToMd = snapshot.parsed?.content;
         let turnDownService = new TurndownService();
         for (const plugin of this.turnDownPlugins) {
             turnDownService = turnDownService.use(plugin);
         }
-
+        const urlToAltMap: { [k: string]: string | undefined; } = {};
-        let contentText = '';
+        if (snapshot.imgs?.length) {
-        if (toBeTurnedToMd) {
-            const urlToAltMap: { [k: string]: string | undefined; } = {};
             const tasks = (snapshot.imgs || []).map(async (x) => {
                 const r = await this.altTextService.getAltText(x).catch((err: any) => {
                     this.logger.warn(`Failed to get alt text for ${x.src}`, { err: marshalErrorLike(err) });
@@ -88,25 +114,27 @@ export class CrawlerHost extends RPCHost {
             });
 
             await Promise.all(tasks);
-            let imgIdx = 0;
+        }
-
+        let imgIdx = 0;
-            turnDownService.addRule('img-generated-alt', {
+        turnDownService.addRule('img-generated-alt', {
-                filter: 'img',
+            filter: 'img',
-                replacement: (_content, node) => {
+            replacement: (_content, node) => {
-                    const src = (node.getAttribute('src') || '').trim();
+                const src = (node.getAttribute('src') || '').trim();
-                    const alt = cleanAttribute(node.getAttribute('alt'));
+                const alt = cleanAttribute(node.getAttribute('alt'));
-                    if (!src) {
+                if (!src) {
-                        return '';
+                    return '';
-                    }
-                    const mapped = urlToAltMap[src];
-                    imgIdx++;
-                    if (mapped) {
-                        return `![Image ${imgIdx}: ${mapped || alt}](${src})`;
-                    }
-                    return `![Image ${imgIdx}: ${alt}](${src})`;
                 }
-            });
+                const mapped = urlToAltMap[src];
+                imgIdx++;
+                if (mapped) {
+                    return `![Image ${imgIdx}: ${mapped || alt}](${src})`;
+                }
+                return `![Image ${imgIdx}: ${alt}](${src})`;
+            }
+        });
 
+        let contentText = '';
+        if (toBeTurnedToMd) {
             try {
                 contentText = turnDownService.turndown(toBeTurnedToMd).trim();
             } catch (err) {
@@ -141,7 +169,7 @@ export class CrawlerHost extends RPCHost {
 
         const formatted = {
             title: (snapshot.parsed?.title || snapshot.title || '').trim(),
-            url: nominalUrl || snapshot.href?.trim(),
+            url: nominalUrl?.toString() || snapshot.href?.trim(),
             content: cleanText,
             publishedTime: snapshot.parsed?.publishedTime || undefined,
 
@@ -171,6 +199,7 @@ ${this.content}
             timeoutSeconds: 540,
             concurrency: 4,
         },
+        tags: ['Crawler'],
         httpMethod: ['get', 'post'],
         returnType: [String, OutputServerEventStream],
     })
@@ -181,6 +210,57 @@ ${this.content}
             concurrency: 21,
             maxInstances: 476,
         },
+        openapi: {
+            operation: {
+                parameters: {
+                    'Accept': {
+                        description: `Specifies your preference for the response format. \n\n` +
+                            `Supported formats:\n` +
+                            `- text/event-stream\n` +
+                            `- application/json  or  text/json\n` +
+                            `- text/plain`
+                        ,
+                        in: 'header',
+                        schema: { type: 'string' }
+                    },
+                    'X-No-Cache': {
+                        description: `Ignores internal cache if this header is specified with a value.`,
+                        in: 'header',
+                        schema: { type: 'string' }
+                    },
+                    'X-Respond-With': {
+                        description: `Specifies the form factor of the crawled data you prefer. \n\n` +
+                            `Supported formats:\n` +
+                            `- markdown\n` +
+                            `- html\n` +
+                            `- text\n` +
+                            `- screenshot\n\n` +
+                            `Defaults to: markdown`
+                        ,
+                        in: 'header',
+                        schema: { type: 'string' }
+                    },
+                    'X-Proxy-Url': {
+                        description: `Specifies your custom proxy if you prefer to use one. \n\n` +
+                            `Supported protocols:\n` +
+                            `- http\n` +
+                            `- https\n` +
+                            `- socks4\n` +
+                            `- socks5\n\n` +
+                            `For authentication, https://user:pass@host:port`,
+                        in: 'header',
+                        schema: { type: 'string' }
+                    },
+                    'X-Set-Cookie': {
+                        description: `Sets cookie(s) to the headless browser for your request. \n\n` +
+                            `Syntax is the same with standard Set-Cookie`,
+                        in: 'header',
+                        schema: { type: 'string' }
+                    },
+                }
+            }
+        },
+        tags: ['Crawler'],
         httpMethod: ['get', 'post'],
         returnType: [String, OutputServerEventStream],
     })
@@ -207,27 +287,41 @@ ${this.content}
                 path: 'url'
             });
         }
-        const screenshotEnabled = Boolean(ctx.req.headers['x-screenshot']);
+
-        const noCache = Boolean(ctx.req.headers['x-no-cache']);
+        const customMode = ctx.req.get('x-respond-with') || 'markdown';
+        const noCache = Boolean(ctx.req.get('x-no-cache'));
+        const cookies: CookieParam[] = [];
+        const setCookieHeaders = ctx.req.headers['x-set-cookie'];
+        if (Array.isArray(setCookieHeaders)) {
+            for (const setCookie of setCookieHeaders) {
+                cookies.push({
+                    ...parseSetCookieString(setCookie, { decodeValues: false }) as CookieParam,
+                    domain: urlToCrawl.hostname,
+                });
+            }
+        } else if (setCookieHeaders) {
+            cookies.push({
+                ...parseSetCookieString(setCookieHeaders, { decodeValues: false }) as CookieParam,
+                domain: urlToCrawl.hostname,
+            });
+        }
+
+        const crawlOpts: ScrappingOptions = {
+            proxyUrl: ctx.req.get('x-proxy-url'),
+            cookies,
+        };
 
         if (!ctx.req.accepts('text/plain') && ctx.req.accepts('text/event-stream')) {
             const sseStream = new OutputServerEventStream();
             rpcReflect.return(sseStream);
 
             try {
-                for await (const scrapped of this.puppeteerControl.scrap(urlToCrawl.toString(), noCache)) {
+                for await (const scrapped of this.cachedScrap(urlToCrawl, crawlOpts, noCache)) {
                     if (!scrapped) {
                         continue;
                     }
 
-                    const formatted = await this.formatSnapshot(scrapped, urlToCrawl?.toString());
+                    const formatted = await this.formatSnapshot(customMode, scrapped, urlToCrawl);
-
-                    if (scrapped.screenshot && screenshotEnabled) {
-                        sseStream.write({
-                            event: 'screenshot',
-                            data: scrapped.screenshot.toString('base64'),
-                        });
-                    }
 
                     sseStream.write({
                         event: 'data',
@@ -235,7 +329,7 @@ ${this.content}
                     });
                 }
             } catch (err: any) {
-                this.logger.error(`Failed to crawl ${urlToCrawl.toString()}`, { err: marshalErrorLike(err) });
+                this.logger.error(`Failed to crawl ${urlToCrawl}`, { err: marshalErrorLike(err) });
                 sseStream.write({
                     event: 'error',
                     data: marshalErrorLike(err),
@@ -249,13 +343,13 @@ ${this.content}
 
         let lastScrapped;
         if (!ctx.req.accepts('text/plain') && (ctx.req.accepts('text/json') || ctx.req.accepts('application/json'))) {
-            for await (const scrapped of this.puppeteerControl.scrap(urlToCrawl.toString(), noCache)) {
+            for await (const scrapped of this.cachedScrap(urlToCrawl, crawlOpts, noCache)) {
                 lastScrapped = scrapped;
                 if (!scrapped?.parsed?.content || !(scrapped.title?.trim())) {
                     continue;
                 }
 
-                const formatted = await this.formatSnapshot(scrapped, urlToCrawl?.toString());
+                const formatted = await this.formatSnapshot(customMode, scrapped, urlToCrawl);
 
                 return formatted;
             }
@@ -264,16 +358,22 @@ ${this.content}
                 throw new AssertionFailureError(`No content available for URL ${urlToCrawl}`);
             }
 
-            return await this.formatSnapshot(lastScrapped, urlToCrawl?.toString());
+            return await this.formatSnapshot(customMode, lastScrapped, urlToCrawl);
         }
 
-        for await (const scrapped of this.puppeteerControl.scrap(urlToCrawl.toString(), noCache)) {
+        for await (const scrapped of this.cachedScrap(urlToCrawl, crawlOpts, noCache)) {
             lastScrapped = scrapped;
             if (!scrapped?.parsed?.content || !(scrapped.title?.trim())) {
                 continue;
             }
 
-            const formatted = await this.formatSnapshot(scrapped, urlToCrawl?.toString());
+            const formatted = await this.formatSnapshot(customMode, scrapped, urlToCrawl);
+            if (customMode === 'screenshot' && Reflect.get(formatted, 'screenshotUrl')) {
+
+                return assignTransferProtocolMeta(`${formatted}`,
+                    { code: 302, envelope: null, headers: { Location: Reflect.get(formatted, 'screenshotUrl') } }
+                );
+            }
 
             return assignTransferProtocolMeta(`${formatted}`, { contentType: 'text/plain', envelope: null });
         }
@@ -282,12 +382,111 @@ ${this.content}
             throw new AssertionFailureError(`No content available for URL ${urlToCrawl}`);
         }
 
-        return `${await this.formatSnapshot(lastScrapped, urlToCrawl?.toString())}`;
+        const formatted = await this.formatSnapshot(customMode, lastScrapped, urlToCrawl);
+        if (customMode === 'screenshot' && Reflect.get(formatted, 'screenshotUrl')) {
+
+            return assignTransferProtocolMeta(`${formatted}`,
+                { code: 302, envelope: null, headers: { Location: Reflect.get(formatted, 'screenshotUrl') } }
+            );
+        }
+
+        return assignTransferProtocolMeta(`${formatted}`, { contentType: 'text/plain', envelope: null });
     }
 
+    getUrlDigest(urlToCrawl: URL) {
+        const normalizedURL = new URL(urlToCrawl);
+        normalizedURL.hash = '';
+        const normalizedUrl = normalizedURL.toString().toLowerCase();
+        const digest = md5Hasher.hash(normalizedUrl.toString());
+
+        return digest;
+    }
+
+    async queryCache(urlToCrawl: URL) {
+        const digest = this.getUrlDigest(urlToCrawl);
+
+        const cache = (await Crawled.fromFirestoreQuery(Crawled.COLLECTION.where('urlPathDigest', '==', digest).orderBy('createdAt', 'desc').limit(1)))?.[0];
+
+        if (cache) {
+            const age = Date.now() - cache.createdAt.valueOf();
+            const stale = cache.createdAt.valueOf() > (Date.now() - this.cacheValidMs);
+            this.logger.info(`${stale ? 'Only stale ' : ''}Cache exists for ${urlToCrawl}, normalized digest: ${digest}, ${age}ms old`, {
+                url: urlToCrawl, digest, age, stale
+            });
+
+            const r = cache.snapshot;
+
+            return {
+                isFresh: !stale,
+                snapshot: {
+                    ...r,
+                    screenshot: undefined,
+                    screenshotUrl: cache.screenshotAvailable ?
+                        await this.firebaseObjectStorage.signDownloadUrl(`screenshots/${cache._id}`, Date.now() + this.urlValidMs) : undefined,
+                } as PageSnapshot & { screenshotUrl?: string; }
+            };
+        }
+
+        return undefined;
+    }
+
+    async setToCache(urlToCrawl: URL, snapshot: PageSnapshot) {
+        const digest = this.getUrlDigest(urlToCrawl);
+
+        this.logger.info(`Caching snapshot of ${urlToCrawl}...`, { url: urlToCrawl, digest, title: snapshot?.title, href: snapshot?.href });
+        const nowDate = new Date();
+
+        const cache = Crawled.from({
+            _id: randomUUID(),
+            url: urlToCrawl.toString(),
+            createdAt: nowDate,
+            expireAt: new Date(nowDate.valueOf() + this.cacheRetentionMs),
+            urlPathDigest: digest,
+            snapshot: {
+                ...snapshot,
+                screenshot: null
+            },
+        });
+
+        if (snapshot.screenshot) {
+            await this.firebaseObjectStorage.saveFile(`screenshots/${cache._id}`, snapshot.screenshot, {
+                metadata: {
+                    contentType: 'image/png',
+                }
+            });
+            cache.screenshotAvailable = true;
+        }
+        const r = await Crawled.save(cache.degradeForFireStore()).catch((err) => {
+            this.logger.error(`Failed to save cache for ${urlToCrawl}`, { err: marshalErrorLike(err) });
+
+            return undefined;
+        });
+
+        return r;
+    }
+
+    async *cachedScrap(urlToCrawl: URL, crawlOpts: ScrappingOptions, noCache: boolean = false) {
+        let cache;
+        if (!noCache && !crawlOpts.cookies?.length) {
+            cache = await this.queryCache(urlToCrawl);
+        }
+
+        if (cache?.isFresh) {
+            yield cache.snapshot;
+
+            return;
+        }
+
+        try {
+            yield* this.puppeteerControl.scrap(urlToCrawl, crawlOpts);
+        } catch (err: any) {
+            if (cache) {
+                this.logger.warn(`Failed to scrap ${urlToCrawl}, but a stale cache is available. Falling back to cache`, { err: marshalErrorLike(err) });
+                yield cache.snapshot;
+                return;
+            }
+            throw err;
+        }
+    }
 
 }
-
-function cleanAttribute(attribute: string) {
-    return attribute ? attribute.replace(/(\n+\s*)+/g, '\n') : '';
-}

backend/functions/src/db/crawled.ts

@@ -1,6 +1,7 @@
 import { Also, parseJSONText, Prop } from 'civkit';
 import { FirestoreRecord } from '../shared/lib/firestore';
 import _ from 'lodash';
+import type { PageSnapshot } from '../services/puppeteer';
 
 @Also({
     dictOf: Object
@@ -21,7 +22,10 @@ export class Crawled extends FirestoreRecord {
     urlPathDigest!: string;
 
     @Prop()
-    snapshot!: any;
+    snapshot!: PageSnapshot & { screenshot: never; };
+
+    @Prop()
+    screenshotAvailable?: boolean;
 
     @Prop()
     createdAt!: Date;

backend/functions/src/services/puppeteer.ts

@@ -1,13 +1,17 @@
-import { AssertionFailureError, AsyncService, Defer, HashManager, marshalErrorLike } from 'civkit';
-import { container, singleton } from 'tsyringe';
-import type { Browser, Page } from 'puppeteer';
-import { Logger } from '../shared/services/logger';
-import genericPool from 'generic-pool';
 import os from 'os';
 import fs from 'fs';
-import { Crawled } from '../db/crawled';
+import { container, singleton } from 'tsyringe';
+import genericPool from 'generic-pool';
+import { AsyncService, Defer, marshalErrorLike, AssertionFailureError } from 'civkit';
+import { Logger } from '../shared/services/logger';
+
+import type { Browser, CookieParam, Page } from 'puppeteer';
 import puppeteer from 'puppeteer-extra';
 
+import puppeteerBlockResources from 'puppeteer-extra-plugin-block-resources';
+import puppeteerPageProxy from 'puppeteer-extra-plugin-page-proxy';
+
+
 const READABILITY_JS = fs.readFileSync(require.resolve('@mozilla/readability/Readability.js'), 'utf-8');
 
 export interface ImgBrief {
@@ -42,7 +46,12 @@ export interface PageSnapshot {
     screenshot?: Buffer;
     imgs?: ImgBrief[];
 }
-const md5Hasher = new HashManager('md5', 'hex');
+
+export interface ScrappingOptions {
+    proxyUrl?: string;
+    cookies?: CookieParam[];
+}
+
 
 const puppeteerStealth = require('puppeteer-extra-plugin-stealth');
 puppeteer.use(puppeteerStealth());
@@ -51,9 +60,13 @@ puppeteer.use(puppeteerStealth());
 //     userAgent: `Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; GPTBot/1.0; +https://openai.com/gptbot)`,
 //     platform: `Linux`,
 // }))
-const puppeteerBlockResources = require('puppeteer-extra-plugin-block-resources');
+
 puppeteer.use(puppeteerBlockResources({
     blockedTypes: new Set(['media']),
+    interceptResolutionPriority: 1,
+}));
+puppeteer.use(puppeteerPageProxy({
+    interceptResolutionPriority: 1,
 }));
 
 @singleton()
@@ -74,7 +87,7 @@ export class PuppeteerControl extends AsyncService {
             return page.browser().connected && !page.isClosed();
         }
     }, {
-        max: Math.max(1 + Math.floor(os.freemem() / (1024 * 1024 * 1024)), 16),
+        max: Math.max(1 + Math.floor(os.totalmem() / (384 * 1024 * 1024)), 16),
         min: 1,
         acquireTimeoutMillis: 60_000,
         testOnBorrow: true,
@@ -88,7 +101,7 @@ export class PuppeteerControl extends AsyncService {
 
     override async init() {
         await this.dependencyReady();
-
+        this.logger.info(`PuppeteerControl initializing with pool size ${this.pagePool.max}`, { poolSize: this.pagePool.max });
         this.pagePool.start();
 
         if (this.browser) {
@@ -128,7 +141,10 @@ export class PuppeteerControl extends AsyncService {
         // preparations.push(page.setUserAgent(`Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; GPTBot/1.0; +https://openai.com/gptbot)`));
         preparations.push(page.setBypassCSP(true));
         preparations.push(page.setViewport({ width: 1024, height: 1024 }));
-        preparations.push(page.exposeFunction('reportSnapshot', (snapshot: any) => {
+        preparations.push(page.exposeFunction('reportSnapshot', (snapshot: PageSnapshot) => {
+            if (snapshot.href === 'about:blank') {
+                return;
+            }
             page.emit('snapshot', snapshot);
         }));
         preparations.push(page.evaluateOnNewDocument(READABILITY_JS));
@@ -166,40 +182,39 @@ function giveSnapshot() {
         const elem = document.createElement('div');
         elem.innerHTML = parsed.content;
         r.imgs = briefImgs(elem);
+    } else {
+        const allImgs = briefImgs();
+        if (allImgs.length === 1) {
+            r.imgs = allImgs;
+        }
     }
 
     return r;
 }
 `));
-        preparations.push(page.evaluateOnNewDocument(() => {
+        preparations.push(page.evaluateOnNewDocument(`
-            let aftershot: any;
+let aftershot = undefined;
-            const handlePageLoad = () => {
+const handlePageLoad = () => {
-                // @ts-expect-error
+    if (document.readyState !== 'complete') {
-                if (document.readyState !== 'complete' && document.readyState !== 'interactive') {
+        return;
-                    return;
+    }
-                }
+    const parsed = giveSnapshot();
-                // @ts-expect-error
+    window.reportSnapshot(parsed);
-                const parsed = giveSnapshot();
+    if (!parsed.text) {
-                if (parsed) {
+        if (aftershot) {
-                    // @ts-expect-error
+            clearTimeout(aftershot);
-                    window.reportSnapshot(parsed);
+        }
-                } else {
+        aftershot = setTimeout(() => {
-                    if (aftershot) {
+            const r = giveSnapshot();
-                        clearTimeout(aftershot);
+            if (r && r.text) {
-                    }
+                window.reportSnapshot(r);
-                    aftershot = setTimeout(() => {
+            }
-                        // @ts-expect-error
+        }, 500);
-                        window.reportSnapshot(giveSnapshot());
+    }
-                    }, 500);
+};
-                }
+document.addEventListener('readystatechange', handlePageLoad);
-            };
+document.addEventListener('load', handlePageLoad);
-            // setInterval(handlePageLoad, 1000);
+`));
-            // @ts-expect-error
-            document.addEventListener('readystatechange', handlePageLoad);
-            // @ts-expect-error
-            document.addEventListener('load', handlePageLoad);
-        }));
-
         await Promise.all(preparations);
 
         // TODO: further setup the page;
@@ -207,41 +222,23 @@ function giveSnapshot() {
         return page;
     }
 
-    async *scrap(url: string, noCache: string | boolean = false): AsyncGenerator<PageSnapshot | undefined> {
+    async *scrap(parsedUrl: URL, options: ScrappingOptions): AsyncGenerator<PageSnapshot | undefined> {
-        const parsedUrl = new URL(url);
         // parsedUrl.search = '';
-        parsedUrl.hash = '';
+        const url = parsedUrl.toString();
-        const normalizedUrl = parsedUrl.toString().toLowerCase();
+
-        const digest = md5Hasher.hash(normalizedUrl);
+        this.logger.info(`Scraping ${url}`, { url });
-        this.logger.info(`Scraping ${url}, normalized digest: ${digest}`, { url, digest });
 
         let snapshot: PageSnapshot | undefined;
         let screenshot: Buffer | undefined;
 
-        if (!noCache) {
+        const page = await this.pagePool.acquire();
-            const cached = (await Crawled.fromFirestoreQuery(Crawled.COLLECTION.where('urlPathDigest', '==', digest).orderBy('createdAt', 'desc').limit(1)))?.[0];
+        if (options.proxyUrl) {
-
+            await page.useProxy(options.proxyUrl);
-            if (cached && cached.createdAt.valueOf() > (Date.now() - 1000 * 300)) {
+        }
-                const age = Date.now() - cached.createdAt.valueOf();
+        if (options.cookies) {
-                this.logger.info(`Cache hit for ${url}, normalized digest: ${digest}, ${age}ms old`, { url, digest, age });
+            await page.setCookie(...options.cookies);
-                snapshot = {
-                    ...cached.snapshot
-                };
-                if (snapshot) {
-                    delete snapshot.screenshot;
-                }
-
-                screenshot = cached.snapshot?.screenshot ? Buffer.from(cached.snapshot.screenshot, 'base64') : undefined;
-                yield {
-                    ...cached.snapshot,
-                    screenshot: cached.snapshot?.screenshot ? Buffer.from(cached.snapshot.screenshot, 'base64') : undefined
-                };
-
-                return;
-            }
         }
 
-        const page = await this.pagePool.acquire();
         let nextSnapshotDeferred = Defer();
         let finalized = false;
         const hdl = (s: any) => {
@@ -262,48 +259,43 @@ function giveSnapshot() {
                     cause: err,
                 }));
             }).finally(async () => {
-                finalized = true;
                 if (!snapshot?.html) {
+                    finalized = true;
                     return;
                 }
-                screenshot = await page.screenshot({
-                    type: 'jpeg',
-                    quality: 75,
-                });
                 snapshot = await page.evaluate('giveSnapshot()') as PageSnapshot;
+                screenshot = await page.screenshot();
                 if (!snapshot.title || !snapshot.parsed?.content) {
                     const salvaged = await this.salvage(url, page);
                     if (salvaged) {
-                        screenshot = await page.screenshot({
-                            type: 'jpeg',
-                            quality: 75,
-                        });
                         snapshot = await page.evaluate('giveSnapshot()') as PageSnapshot;
+                        screenshot = await page.screenshot();
                     }
                 }
-                this.logger.info(`Snapshot of ${url} done`, { url, digest, title: snapshot?.title, href: snapshot?.href });
+                finalized = true;
-                const nowDate = new Date();
+                this.logger.info(`Snapshot of ${url} done`, { url, title: snapshot?.title, href: snapshot?.href });
-                Crawled.save(
+                this.emit(
-                    Crawled.from({
+                    'crawled',
-                        url,
+                    { ...snapshot, screenshot },
-                        createdAt: nowDate,
+                    { ...options, url: parsedUrl }
-                        expireAt: new Date(nowDate.valueOf() + 1000 * 3600 * 24 * 7),
+                );
-                        urlPathDigest: digest,
-                        snapshot: { ...snapshot, screenshot: screenshot?.toString('base64') || '' },
-                    }).degradeForFireStore()
-                ).catch((err) => {
-                    this.logger.warn(`Failed to save snapshot`, { err: marshalErrorLike(err) });
-                });
             });
 
         try {
+            let lastHTML = snapshot?.html;
             while (true) {
                 await Promise.race([nextSnapshotDeferred.promise, gotoPromise]);
                 if (finalized) {
                     yield { ...snapshot, screenshot } as PageSnapshot;
                     break;
                 }
-                yield snapshot;
+                if (snapshot?.title && snapshot?.html !== lastHTML) {
+                    screenshot = await page.screenshot();
+                    lastHTML = snapshot.html;
+                }
+                if (snapshot || screenshot) {
+                    yield { ...snapshot, screenshot } as PageSnapshot;
+                }
             }
         } finally {
             gotoPromise.finally(() => {
@@ -333,6 +325,8 @@ function giveSnapshot() {
             this.logger.warn(`Page salvation did not fully succeed.`, { err: marshalErrorLike(err) });
         });
 
+        this.logger.info(`Salvation completed.`);
+
         return true;
     }
 }

backend/functions/src/utils/markdown.ts

@@ -0,0 +1,39 @@
+
+export function tidyMarkdown(markdown: string): string {
+
+    // Step 1: Handle complex broken links with text and optional images spread across multiple lines
+    let normalizedMarkdown = markdown.replace(/\[\s*([^]+?)\s*\]\s*\(\s*([^)]+)\s*\)/g, (match, text, url) => {
+        // Remove internal new lines and excessive spaces within the text
+        text = text.replace(/\s+/g, ' ').trim();
+        url = url.replace(/\s+/g, '').trim();
+        return `[${text}](${url})`;
+    });
+
+    normalizedMarkdown = normalizedMarkdown.replace(/\[\s*([^!]*?)\s*\n*(?:!\[([^\]]*)\]\((.*?)\))?\s*\n*\]\s*\(\s*([^)]+)\s*\)/g, (match, text, alt, imgUrl, linkUrl) => {
+        // Normalize by removing excessive spaces and new lines
+        text = text.replace(/\s+/g, ' ').trim();
+        alt = alt ? alt.replace(/\s+/g, ' ').trim() : '';
+        imgUrl = imgUrl ? imgUrl.replace(/\s+/g, '').trim() : '';
+        linkUrl = linkUrl.replace(/\s+/g, '').trim();
+        if (imgUrl) {
+            return `[${text} ![${alt}](${imgUrl})](${linkUrl})`;
+        } else {
+            return `[${text}](${linkUrl})`;
+        }
+    });
+
+    // Step 2: Normalize regular links that may be broken across lines
+    normalizedMarkdown = normalizedMarkdown.replace(/\[\s*([^\]]+)\]\s*\(\s*([^)]+)\)/g, (match, text, url) => {
+        text = text.replace(/\s+/g, ' ').trim();
+        url = url.replace(/\s+/g, '').trim();
+        return `[${text}](${url})`;
+    });
+
+    // Step 3: Replace more than two consecutive empty lines with exactly two empty lines
+    normalizedMarkdown = normalizedMarkdown.replace(/\n{3,}/g, '\n\n');
+
+    // Step 4: Remove leading spaces from each line
+    normalizedMarkdown = normalizedMarkdown.replace(/^[ \t]+/gm, '');
+
+    return normalizedMarkdown.trim();
+}

backend/functions/src/utils/misc.ts

@@ -0,0 +1,3 @@
+export function cleanAttribute(attribute: string) {
+    return attribute ? attribute.replace(/(\n+\s*)+/g, '\n') : '';
+}