parsagon 0.14.33__tar.gz → 0.14.35__tar.gz

{parsagon-0.14.33 → parsagon-0.14.35}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: parsagon
-Version: 0.14.33
+Version: 0.14.35
 Summary: Allows you to create browser automations with natural language
 Author-email: Sandy Suh <sandy@parsagon.io>
 Project-URL: Homepage, https://parsagon.io
@@ -21,6 +21,8 @@ Requires-Dist: webdriver-manager==4.0.2
 Requires-Dist: jsonpath-ng==1.5.3
 Requires-Dist: simplejson==3.19.1
 Requires-Dist: pypdf==4.2.0
+Requires-Dist: python-dateutil==2.9.0.post0
+Requires-Dist: html2text==2024.2.26
 Provides-Extra: dev
 Requires-Dist: pytest==7.3.2; extra == "dev"
 Requires-Dist: pytest-mock==3.11.1; extra == "dev"

{parsagon-0.14.33 → parsagon-0.14.35}/pyproject.toml

@@ -16,7 +16,7 @@ line-length = 120
 
 [project]
 name = "parsagon"
-version = "0.14.33"
+version = "0.14.35"
 description = "Allows you to create browser automations with natural language"
 readme = "README.md"
 requires-python = ">=3.8"
@@ -44,6 +44,8 @@ dependencies = [
     'jsonpath-ng==1.5.3',
     'simplejson==3.19.1',
     'pypdf==4.2.0',
+    'python-dateutil==2.9.0.post0',
+    'html2text==2024.2.26',
 ]
 
 [project.optional-dependencies]

{parsagon-0.14.33 → parsagon-0.14.35}/src/parsagon/executor.py

@@ -1,5 +1,6 @@
 from collections import defaultdict
 import copy
+import dateutil.parser
 import glob
 import json
 import logging
@@ -9,6 +10,7 @@ import psutil
 import time
 from urllib.parse import urljoin
 
+import html2text
 from lxml import etree
 import lxml.html
 from pypdf import PdfReader
@@ -150,6 +152,7 @@ class Executor:
             "get_json_about_data": get_json_about_data,
             "get_serp_data": get_serp_data,
             "get_pdf_text": self.get_pdf_text,
+            "str_to_iso8601": self.str_to_iso8601,
         }
         self.custom_functions = {}
         self.infer = infer
@@ -587,6 +590,12 @@ class Executor:
         elem = self._id_to_elem(elem_id)
         return elem.get_attribute("innerText")
 
+    def get_inner_markdown(self, window_id):
+        if self.driver.current_window_handle != window_id:
+            self.driver.switch_to.window(window_id)
+        html = self.driver.execute_script("return document.body.innerHTML;")
+        return html2text.html2text(html)
+
     def scrape_data(self, schema, window_id, call_id):
         """
         Scrapes data from the current page.
@@ -671,6 +680,14 @@ class Executor:
         os.remove(most_recent_file)
         return text
 
+    def str_to_iso8601(self, s):
+        while s:
+            try:
+                return dateutil.parser.parse(s).isoformat()
+            except dateutil.parser._parser.ParserError:
+                s = " ".join(s.split()[:-1])
+        return ""
+
     def execute(self, code):
         loc = {}
         try:

{parsagon-0.14.33 → parsagon-0.14.35}/src/parsagon.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: parsagon
-Version: 0.14.33
+Version: 0.14.35
 Summary: Allows you to create browser automations with natural language
 Author-email: Sandy Suh <sandy@parsagon.io>
 Project-URL: Homepage, https://parsagon.io
@@ -21,6 +21,8 @@ Requires-Dist: webdriver-manager==4.0.2
 Requires-Dist: jsonpath-ng==1.5.3
 Requires-Dist: simplejson==3.19.1
 Requires-Dist: pypdf==4.2.0
+Requires-Dist: python-dateutil==2.9.0.post0
+Requires-Dist: html2text==2024.2.26
 Provides-Extra: dev
 Requires-Dist: pytest==7.3.2; extra == "dev"
 Requires-Dist: pytest-mock==3.11.1; extra == "dev"

{parsagon-0.14.33 → parsagon-0.14.35}/src/parsagon.egg-info/requires.txt

@@ -11,6 +11,8 @@ webdriver-manager==4.0.2
 jsonpath-ng==1.5.3
 simplejson==3.19.1
 pypdf==4.2.0
+python-dateutil==2.9.0.post0
+html2text==2024.2.26
 
 [dev]
 pytest==7.3.2