parsagon 0.14.24__tar.gz → 0.14.26__tar.gz

{parsagon-0.14.24 → parsagon-0.14.26}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: parsagon
-Version: 0.14.24
+Version: 0.14.26
 Summary: Allows you to create browser automations with natural language
 Author-email: Sandy Suh <sandy@parsagon.io>
 Project-URL: Homepage, https://parsagon.io
@@ -16,10 +16,11 @@ Requires-Dist: rich==13.6.0
 Requires-Dist: PyVirtualDisplay==3.0
 Requires-Dist: selenium-wire==5.1.0
 Requires-Dist: cssselect==1.1.0
-Requires-Dist: undetected-chromedriver==3.5.3
+Requires-Dist: undetected-chromedriver==3.5.5
 Requires-Dist: webdriver-manager==4.0.1
 Requires-Dist: jsonpath-ng==1.5.3
 Requires-Dist: simplejson==3.19.1
+Requires-Dist: pypdf==4.2.0
 Provides-Extra: dev
 Requires-Dist: pytest==7.3.2; extra == "dev"
 Requires-Dist: pytest-mock==3.11.1; extra == "dev"

{parsagon-0.14.24 → parsagon-0.14.26}/pyproject.toml

@@ -16,7 +16,7 @@ line-length = 120
 
 [project]
 name = "parsagon"
-version = "0.14.24"
+version = "0.14.26"
 description = "Allows you to create browser automations with natural language"
 readme = "README.md"
 requires-python = ">=3.8"
@@ -39,10 +39,11 @@ dependencies = [
     'PyVirtualDisplay==3.0',
     'selenium-wire==5.1.0',
     'cssselect==1.1.0',
-    'undetected-chromedriver==3.5.3',
+    'undetected-chromedriver==3.5.5',
     'webdriver-manager==4.0.1',
     'jsonpath-ng==1.5.3',
     'simplejson==3.19.1',
+    'pypdf==4.2.0',
 ]
 
 [project.optional-dependencies]

{parsagon-0.14.24 → parsagon-0.14.26}/src/parsagon/api.py

@@ -235,7 +235,7 @@ def get_pipelines():
     return _api_call(httpx.get, f"/pipelines/")
 
 
-def get_pipeline_code(pipeline_name, variables, headless, use_uc, optimize):
+def get_pipeline_code(pipeline_name, variables, headless, use_uc, optimize, use_proxy):
     escaped_pipeline_name = double_quote(pipeline_name)
     with RaiseProgramNotFound(pipeline_name):
         return _api_call(
@@ -246,6 +246,7 @@ def get_pipeline_code(pipeline_name, variables, headless, use_uc, optimize):
                 "headless": headless,
                 "use_uc": use_uc,
                 "optimize": optimize,
+                "use_proxy": use_proxy,
             },
         )
 

{parsagon-0.14.24 → parsagon-0.14.26}/src/parsagon/executor.py

@@ -1,13 +1,16 @@
 from collections import defaultdict
 import copy
+import glob
 import json
 import logging
+import os
 from pathlib import Path
 import psutil
 import time
 from urllib.parse import urljoin
 
 import lxml.html
+from pypdf import PdfReader
 from pyvirtualdisplay import Display
 import undetected_chromedriver as uc
 from selenium import webdriver
@@ -92,12 +95,30 @@ class Executor:
             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)
@@ -124,6 +145,7 @@ class Executor:
             "get_str_about_data": get_str_about_data,
             "get_bool_about_data": get_bool_about_data,
             "get_json_about_data": get_json_about_data,
+            "get_pdf_text": self.get_pdf_text,
         }
         self.custom_functions = {}
         self.infer = infer
@@ -617,6 +639,19 @@ class Executor:
         self.add_custom_function(call_id, custom_function)
         return scraped_data
 
+    def get_pdf_text(self, url):
+        window_id = self.goto(url)
+        self.close_window(window_id)
+        files = glob.glob("*")
+        most_recent_file = max(files, key=os.path.getmtime)
+        reader = PdfReader(most_recent_file)
+        text = ""
+        for page in reader.pages:
+            text += page.extract_text(extraction_mode="layout", layout_mode_space_vertically=False)
+            text += "\n"
+        os.remove(most_recent_file)
+        return text
+
     def execute(self, code):
         loc = {}
         try:

{parsagon-0.14.24 → parsagon-0.14.26}/src/parsagon/main.py

@@ -142,6 +142,11 @@ def get_args(argv):
         action="store_true",
         help="run in optimized mode",
     )
+    parser_run.add_argument(
+        "--use_proxy",
+        action="store_true",
+        help="run with proxy",
+    )
     parser_run.set_defaults(func=run)
 
     # Delete

{parsagon-0.14.24 → parsagon-0.14.26}/src/parsagon/runs.py

@@ -29,6 +29,7 @@ def run(
     output_file=None,
     undetected=False,
     optimize=False,
+    use_proxy=False,
     verbose=False,
 ):
     """
@@ -80,7 +81,7 @@ def run(
                 time.sleep(5)
 
     run = create_pipeline_run(pipeline_id, variables, True)
-    code = get_pipeline_code(program_name, variables, headless, undetected, optimize)["code"]
+    code = get_pipeline_code(program_name, variables, headless, undetected, optimize, use_proxy)["code"]
     start_time = datetime.datetime.now(datetime.timezone.utc).isoformat()
     run_data = {"start_time": start_time}
 

{parsagon-0.14.24 → parsagon-0.14.26}/src/parsagon.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: parsagon
-Version: 0.14.24
+Version: 0.14.26
 Summary: Allows you to create browser automations with natural language
 Author-email: Sandy Suh <sandy@parsagon.io>
 Project-URL: Homepage, https://parsagon.io
@@ -16,10 +16,11 @@ Requires-Dist: rich==13.6.0
 Requires-Dist: PyVirtualDisplay==3.0
 Requires-Dist: selenium-wire==5.1.0
 Requires-Dist: cssselect==1.1.0
-Requires-Dist: undetected-chromedriver==3.5.3
+Requires-Dist: undetected-chromedriver==3.5.5
 Requires-Dist: webdriver-manager==4.0.1
 Requires-Dist: jsonpath-ng==1.5.3
 Requires-Dist: simplejson==3.19.1
+Requires-Dist: pypdf==4.2.0
 Provides-Extra: dev
 Requires-Dist: pytest==7.3.2; extra == "dev"
 Requires-Dist: pytest-mock==3.11.1; extra == "dev"

{parsagon-0.14.24 → parsagon-0.14.26}/src/parsagon.egg-info/requires.txt

@@ -6,10 +6,11 @@ rich==13.6.0
 PyVirtualDisplay==3.0
 selenium-wire==5.1.0
 cssselect==1.1.0
-undetected-chromedriver==3.5.3
+undetected-chromedriver==3.5.5
 webdriver-manager==4.0.1
 jsonpath-ng==1.5.3
 simplejson==3.19.1
+pypdf==4.2.0
 
 [dev]
 pytest==7.3.2