parsagon 0.16.0__tar.gz → 0.17.0__tar.gz

{parsagon-0.16.0 → parsagon-0.17.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: parsagon
-Version: 0.16.0
+Version: 0.17.0
 Summary: Allows you to create browser automations with natural language
 Author-email: Sandy Suh <sandy@parsagon.io>
 Project-URL: Homepage, https://parsagon.io
@@ -8,15 +8,14 @@ Classifier: Programming Language :: Python :: 3
 Classifier: Operating System :: OS Independent
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
-Requires-Dist: seleniumbase==4.38.0
-Requires-Dist: selenium>=4.9.1
+Requires-Dist: selenium==4.9.1
 Requires-Dist: lxml[html_clean]==5.3.2
 Requires-Dist: httpx>=0.27.0
 Requires-Dist: psutil==7.0.0
-Requires-Dist: rich>=13.6.0
+Requires-Dist: rich==13.6.0
 Requires-Dist: PyVirtualDisplay==3.0
 Requires-Dist: selenium-wire==5.1.0
-Requires-Dist: cssselect>=1.2.0
+Requires-Dist: cssselect==1.2.0
 Requires-Dist: undetected-chromedriver==3.5.5
 Requires-Dist: webdriver-manager==4.0.2
 Requires-Dist: jsonpath-ng==1.5.3

{parsagon-0.16.0 → parsagon-0.17.0}/pyproject.toml

@@ -16,7 +16,7 @@ line-length = 120
 
 [project]
 name = "parsagon"
-version = "0.16.0"
+version = "0.17.0"
 description = "Allows you to create browser automations with natural language"
 readme = "README.md"
 requires-python = ">=3.8"
@@ -29,17 +29,16 @@ classifiers = [
 ]
 
 dependencies = [
-    'seleniumbase == 4.38.0',
-    'selenium >= 4.9.1',
+    'selenium == 4.9.1',
     'lxml[html_clean] == 5.3.2',
     'httpx >= 0.27.0',
     'psutil == 7.0.0',
-    'rich >= 13.6.0',
+    'rich == 13.6.0',
 
     # Used only in pipeline code execution
     'PyVirtualDisplay==3.0',
     'selenium-wire==5.1.0',
-    'cssselect>=1.2.0',
+    'cssselect==1.2.0',
     'undetected-chromedriver==3.5.5',
     'webdriver-manager==4.0.2',
     'jsonpath-ng==1.5.3',

{parsagon-0.16.0 → parsagon-0.17.0}/src/parsagon/api.py

@@ -247,7 +247,7 @@ def get_pipelines():
     return _api_call(httpx.get, f"/pipelines/")
 
 
-def get_pipeline_code(pipeline_name, variables, headless, use_uc, optimize, use_proxy, pipeline_id=None):
+def get_pipeline_code(pipeline_name, variables, headless, use_cdp, use_uc, optimize, use_proxy, pipeline_id=None):
     if pipeline_id:
         return _api_call(
             httpx.post,
@@ -255,6 +255,7 @@ def get_pipeline_code(pipeline_name, variables, headless, use_uc, optimize, use_
             json={
                 "variables": variables,
                 "headless": headless,
+                "use_cdp": use_cdp,
                 "use_uc": use_uc,
                 "optimize": optimize,
                 "use_proxy": use_proxy,
@@ -268,6 +269,7 @@ def get_pipeline_code(pipeline_name, variables, headless, use_uc, optimize, use_
             json={
                 "variables": variables,
                 "headless": headless,
+                "use_cdp": use_cdp,
                 "use_uc": use_uc,
                 "optimize": optimize,
                 "use_proxy": use_proxy,

{parsagon-0.16.0 → parsagon-0.17.0}/src/parsagon/executor.py

@@ -17,11 +17,17 @@ from lxml import etree
 import lxml.html
 from lxml.html.clean import Cleaner
 from pypdf import PdfReader
-from seleniumbase import Driver
+from pyvirtualdisplay import Display
+import undetected_chromedriver as uc
+from selenium import webdriver
+from selenium.webdriver.chrome.service import Service as ChromeService
+from selenium.webdriver.chrome.options import Options
 from selenium.webdriver.common.action_chains import ActionChains
 from selenium.webdriver.common.by import By
 from selenium.webdriver.common.keys import Keys
 from selenium.webdriver.support.select import Select
+from webdriver_manager.chrome import ChromeDriverManager
+from webdriver_manager.core.driver_cache import DriverCacheManager
 
 from parsagon import settings
 from parsagon.api import (
@@ -76,6 +82,7 @@ class Executor:
         infer=False,
         use_uc=False,
         driver_path=None,
+        options=[],
         page_load_timeout=None,
         script_timeout=None,
         function_bank={},
@@ -83,8 +90,43 @@ class Executor:
         self.task = task
         self.headless = headless
         self.function_bank = function_bank
-        self.driver = Driver(external_pdf=True, undetectable=use_uc)
-        self.driver.maximize_window()
+        if self.headless:
+            self.display = Display(visible=False, size=(1280, 1050)).start()
+        if driver_path:
+            cache_manager = DriverCacheManager(root_dir=driver_path)
+            driver_executable_path = ChromeDriverManager(cache_manager=cache_manager).install()
+        else:
+            driver_executable_path = ChromeDriverManager().install()
+        if use_uc:
+            chrome_options = uc.ChromeOptions()
+            chrome_options.add_argument("--start-maximized")
+            for option in options:
+                chrome_options.add_argument(option)
+            chrome_options.add_experimental_option(
+                "prefs",
+                {
+                    "download.default_directory": os.getcwd(),
+                    "download.prompt_for_download": False,
+                    "download.directory_upgrade": True,
+                    "plugins.always_open_pdf_externally": True,
+                },
+            )
+            self.driver = uc.Chrome(driver_executable_path=driver_executable_path, options=chrome_options)
+        else:
+            chrome_options = webdriver.ChromeOptions()
+            chrome_options.add_argument("--start-maximized")
+            for option in options:
+                chrome_options.add_argument(option)
+            chrome_options.add_experimental_option(
+                "prefs",
+                {
+                    "download.default_directory": os.getcwd(),
+                    "download.prompt_for_download": False,
+                    "download.directory_upgrade": True,
+                    "plugins.always_open_pdf_externally": True,
+                },
+            )
+            self.driver = webdriver.Chrome(service=ChromeService(driver_executable_path), options=chrome_options)
         if page_load_timeout:
             self.driver.set_page_load_timeout(page_load_timeout)
         if script_timeout:
@@ -115,6 +157,7 @@ class Executor:
             "get_serp_data": get_serp_data,
             "get_pdf_text": self.get_pdf_text,
             "str_to_iso8601": self.str_to_iso8601,
+            "get_network_requests": self.get_network_requests,
         }
         self.custom_functions = {}
         self.infer = infer
@@ -671,6 +714,17 @@ class Executor:
                 s = " ".join(s.split()[:-1])
         return ""
 
+    def get_network_requests(window_id, delay=30):
+        requests = []
+
+        def handle_request(event):
+            requests.append(event)
+
+        driver.switch_to.window(window_id)
+        driver.add_cdp_listener("Network.requestWillBeSent", handle_request)
+        wait(delay)
+        return requests
+
     def execute(self, code):
         loc = {}
         try:

{parsagon-0.16.0 → parsagon-0.17.0}/src/parsagon/main.py

@@ -144,6 +144,11 @@ def get_args(argv):
         type=str,
         help="write the data to the given file path",
     )
+    parser_run.add_argument(
+        "--use_cdp",
+        action="store_true",
+        help="enable CDP (only in undetected mode)",
+    )
     parser_run.add_argument(
         "--undetected",
         action="store_true",

{parsagon-0.16.0 → parsagon-0.17.0}/src/parsagon/runs.py

@@ -28,6 +28,7 @@ def run(
     remote=False,
     output_log=False,
     output_file=None,
+    use_cdp=False,
     undetected=False,
     optimize=False,
     use_proxy=False,
@@ -83,7 +84,7 @@ def run(
 
     run = create_pipeline_run(pipeline_id, variables, True)
     code = get_pipeline_code(
-        program_name, variables, headless, undetected, optimize, use_proxy, pipeline_id=program_id
+        program_name, variables, headless, use_cdp, undetected, optimize, use_proxy, pipeline_id=program_id
     )["code"]
     start_time = datetime.datetime.now(datetime.timezone.utc).isoformat()
     run_data = {"start_time": start_time}

{parsagon-0.16.0 → parsagon-0.17.0}/src/parsagon.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: parsagon
-Version: 0.16.0
+Version: 0.17.0
 Summary: Allows you to create browser automations with natural language
 Author-email: Sandy Suh <sandy@parsagon.io>
 Project-URL: Homepage, https://parsagon.io
@@ -8,15 +8,14 @@ Classifier: Programming Language :: Python :: 3
 Classifier: Operating System :: OS Independent
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
-Requires-Dist: seleniumbase==4.38.0
-Requires-Dist: selenium>=4.9.1
+Requires-Dist: selenium==4.9.1
 Requires-Dist: lxml[html_clean]==5.3.2
 Requires-Dist: httpx>=0.27.0
 Requires-Dist: psutil==7.0.0
-Requires-Dist: rich>=13.6.0
+Requires-Dist: rich==13.6.0
 Requires-Dist: PyVirtualDisplay==3.0
 Requires-Dist: selenium-wire==5.1.0
-Requires-Dist: cssselect>=1.2.0
+Requires-Dist: cssselect==1.2.0
 Requires-Dist: undetected-chromedriver==3.5.5
 Requires-Dist: webdriver-manager==4.0.2
 Requires-Dist: jsonpath-ng==1.5.3

{parsagon-0.16.0 → parsagon-0.17.0}/src/parsagon.egg-info/requires.txt

@@ -1,12 +1,11 @@
-seleniumbase==4.38.0
-selenium>=4.9.1
+selenium==4.9.1
 lxml[html_clean]==5.3.2
 httpx>=0.27.0
 psutil==7.0.0
-rich>=13.6.0
+rich==13.6.0
 PyVirtualDisplay==3.0
 selenium-wire==5.1.0
-cssselect>=1.2.0
+cssselect==1.2.0
 undetected-chromedriver==3.5.5
 webdriver-manager==4.0.2
 jsonpath-ng==1.5.3