feat: preparation of cloakbrowser and undetected-chromedriver for use with Selenium-wrapper from other languages like VBA

routines to download cloakbrowser and download+patch undetected-chroemdriver, backpassing location values via ini-files

Author: dornech

src/utils_msoffice/__init__.py

@@ -10,6 +10,7 @@
 - submodule with utilities for Microsoft Office in general
 - submodule with utilities for Microsoft Excel including "pythonic" access to ExcelAPI
 - submodule for interfacing calling COM scripting with Microsoft Office applications
+- submodule for making cloakbrowser available
 """
 
 
@@ -30,3 +31,5 @@
 import utils_msoffice.utils_office as UtilsOffice
 import utils_msoffice.utils_excel as UtilsExcel
 import utils_msoffice.runner_VBA as RunnerVBA
+from utils_msoffice.prepare_cloakbrowser import executeVBAcallee as prepare_cloakbrowser
+from utils_msoffice.prepare_undetectedchrome  import executeVBAcallee as prepare_undetectedchrome

src/utils_msoffice/prepare_cloakbrowser.py

@@ -0,0 +1,97 @@
+# prepare cloakbrowser for calling via VBA
+
+"""
+Download cloakbrowser to be called from other languages like VBA
+return location and call parameters via INI-file
+"""
+
+
+# ruff and mypy per file settings
+#
+# empty lines
+# ruff: noqa: E302, E303
+# naming conventions
+# ruff: noqa: N802, N812
+# others
+# ruff: noqa: PLW1514
+
+# fmt: off
+
+
+
+import os
+import tempfile
+import configparser
+
+from tap import Tap as TypedArgParse
+
+from cloakbrowser.config import get_default_stealth_args
+from cloakbrowser.download import ensure_binary
+
+import utils_mystuff as Utils
+
+
+
+# TAP for parameters
+class ParamsClass(TypedArgParse):
+    cloakbrowser_cache_dir: str | None = None
+    params_inifile: str | None = None
+
+
+
+# caller routines
+
+# test routine for PythonIDE debugging - parameters are normally set via CLI
+# hint: quotation marks surrounding parameter values are swallowed by CLI param transfer mechanism
+def executeStandaloneTest() -> None:
+
+    paramsparser = ParamsClass()
+    params = paramsparser.parse_args()
+    executeMain(params)
+
+
+# VBA caller with direct CLI parameters (basically an CLI caller)
+def executeVBAcallee() -> None:
+
+    Utils.log_cli_args()
+    paramsparser = ParamsClass(explicit_bool=True)
+    params = paramsparser.parse_args()
+    Utils.log_cli_params(params)
+
+    executeMain(params)
+
+
+
+#  main program logic
+
+# main caller
+def executeMain(params: ParamsClass) -> None:
+
+    # determine cache path for cloakbrowser
+    if params.cloakbrowser_cache_dir is None:
+        params.cloakbrowser_cache_dir = os.path.join(tempfile.gettempdir(), "cloakbrowser")
+    os.environ["CLOAKBROWSER_CACHE_DIR"] = params.cloakbrowser_cache_dir
+    # determine ini file for parameters calling cloakbrowser externally
+    if params.params_inifile is None:
+        params.params_inifile = os.path.join(tempfile.gettempdir(), "cloakbrowser.ini")
+
+    # determine content of ini file
+    config = configparser.ConfigParser()
+    config.add_section("cloakbrowser")
+    config["cloakbrowser"]["binary_path"] = ensure_binary()
+    config["cloakbrowser"]["stealth_args"] = ",".join(get_default_stealth_args())
+
+    # write back ini file
+    with open(params.params_inifile, 'w') as configfile:
+        config.write(configfile)
+
+
+
+# standalone call / test
+
+if __name__ == '__main__':
+
+    executeStandaloneTest()
+
+    # completion message
+    Utils.exitFinished("prepare_cloakbrowser")

src/utils_msoffice/prepare_undetectedchrome.py

@@ -0,0 +1,94 @@
+# prepare cloakbrowser for calling via VBA
+
+
+"""
+Download chromedriver and apply undetected-chromedriver patch to be called from other languages like VBA
+return location via INI-file
+"""
+
+
+# ruff and mypy per file settings
+#
+# empty lines
+# ruff: noqa: E302, E303
+# naming conventions
+# ruff: noqa: N802, N812
+# others
+# ruff: noqa: PLW1514
+
+# fmt: off
+
+
+
+import os
+import tempfile
+import configparser
+
+from tap import Tap as TypedArgParse
+
+from undetected_chromedriver.patcher import Patcher
+
+import utils_mystuff as Utils
+
+
+
+# TAP for parameters
+class ParamsClass(TypedArgParse):
+    params_inifile: str | None = None
+
+
+
+# caller routines
+
+# test routine for PythonIDE debugging - parameters are normally set via CLI
+# hint: quotation marks surrounding parameter values are swallowed by CLI param transfer mechanism
+def executeStandaloneTest() -> None:
+
+    paramsparser = ParamsClass()
+    params = paramsparser.parse_args()
+    executeMain(params)
+
+
+# VBA caller with direct CLI parameters (basically an CLI caller)
+def executeVBAcallee() -> None:
+
+    Utils.log_cli_args()
+    paramsparser = ParamsClass(explicit_bool=True)
+    params = paramsparser.parse_args()
+    Utils.log_cli_params(params)
+
+    executeMain(params)
+
+
+
+#  main program logic
+
+# main caller
+def executeMain(params: ParamsClass) -> None:
+
+    # determine ini file for parameters calling cloakbrowser externally
+    if params.params_inifile is None:
+        params.params_inifile = os.path.join(tempfile.gettempdir(), "undetectedchrome.ini")
+
+    patcher = Patcher()
+    patcher.auto()
+
+    # determine content of ini file
+    config = configparser.ConfigParser()
+    config.add_section("undetectedchrome")
+    config["undetectedchrome"]["binary_path"] = patcher.executable_path
+
+    # write back ini file
+    with open(params.params_inifile, 'w') as configfile:
+        config.write(configfile)
+
+
+
+# standalone call / test
+
+if __name__ == '__main__':
+
+    executeStandaloneTest()
+
+    # completion message
+    Utils.exitFinished("prepare_cloakbrowser")