Skip to main content

Use Pandas to find and interact with handles, windows, and elements

Project description

Use Pandas to find and interact with handles, windows, and elements

This library doesn't need Pywin32, it uses ctypes!

Inspired by pywinauto, but Pandas is used to localize

elements and call functions. Sounds weird, but it is

extremely efficient and will save you a lot of time. Try it.

# How to get the DataFrames



from invo import (mkey, get_window_frame, get_automation_frame, get_automation_frame_from_pid,

                  get_automation_frame_from_filenames, get_automation_frame_from_all_pids, )

mkey.enable_failsafekill("ctrl+e") #Always use a failsafe 



# Here are some way you can use to get the DataFrame

df1 = get_automation_frame_from_pid(pids=[12720, 3684], uia=True, screenshot_folder='f:\\screenshotstestx1', timeout=30)

df2 = get_automation_frame_from_filenames(["chrome.exe", "mspaint.exe"], uia=True, screenshot_folder=None, timeout=30)

df3 = get_automation_frame_from_all_pids(uia=True, screenshot_folder=None)

df4 = get_window_frame(pid=0)

df5 = get_automation_frame(df4,uia=True, screenshot_folder=None)
# How to click on an element without moving the mouse:

# technically it is not a mouse click, but the results are the same 

# You might encounter elements that don't work with these methods:



# 1) Sometimes the results are better using:

df.iloc[1].ff_Invoke()

df1.dropna(subset='ff_Invoke').iloc[0].ff_Invoke() # another example



# 2) and other times, this function works better (Uses PostMessage)

df.d_loc_no_exception('str.contains','Cancel',na=False,flags=re.I).aa_direct_click.iloc[0]()
# How to click on an element with the mouse cursor:

# 1)

# The click does always work since it uses the coordinates instead of the handle.

# You don't have to pass the coordinates. The module takes care of it. 

# Be careful: The function is executed at maximum speed. 

df.mm_left_click_xy_direct.iloc[0]()

# If you want, you can pass an offset (from the center):

df.mm_left_click_xy_direct.iloc[0](offset_x=100,offset_y=100)

#-------------------------------------------------------------------------------------------#

# 2) 

# If you prefer a slower, natural mouse movement:

df.mm_left_click_xy_natural.iloc[0](    

    delay=0.1, # duration of the mouse click (down - up)

    min_variation=-3, # a random value will be added to each pixel  - define the minimum here

    max_variation=3, # a random value will be added to each pixel  - define the maximum here

    use_every=4, # use every nth pixel

    sleeptime=(0.005, 0.009), # delay between each coordinate

    print_coords=True, # console output

    percent=90, # if you want to speed up the execution, lower this number

    offset_x=50, # offset from the center 

    offset_y=100,) # offset from the center 



# Also available for middle/right mouse click:

# Some examples:

df.mm_middle_click_xy_natural.iloc[0](offset_x=50, offset_y=100)

df.mm_right_click_xy_natural.iloc[0](offset_x=50, offset_y=100)

df.mm_right_click_xy_direct.iloc[0](offset_x=100,offset_y=100)
# How to move the cursor to an element 

df.mm_move_to_natural.iloc[0](offset_x=50, offset_y=100) # takes longer 

df.mm_move_to_natural.iloc[0](use_every=10,offset_x=100, offset_y=100) # if you want to speed it up, lower the value of use_every

df.mm_moveto.iloc[0](offset_x=50, offset_y=100) # immediately
# How to save all screenshots to the HDD:

df1.screenshot_save.apply(lambda x:x())
# How to send keystrokes 

# 1) 

# The Window cannot be minimized if you use this function. 

# You can find all valid keys here: mkey.show_all_keys,

df.mm_press_key.iloc[0]('b',delay=1) 

#-------------------------------------------------------------------------------------------#

# 2)

# This function works when the window is minimized,

# it restores the window, and sets focus, but it also moves the mouse.

df.mm_press_key_force_activate.iloc[0]('a',1)

#-------------------------------------------------------------------------------------------#

# 3)

# Code based on:

# https://pywinauto.readthedocs.io

# The key codes are the same. 

df.mm_send_keystrokes.iloc[0]('%{F4}') # Closing the window: '%{F4}' (alt+f4)

#-------------------------------------------------------------------------------------------#

# 4) Send as many keystrokes as you want simultaneously

# Use force activate to make sure that the strokes go to the right window

df.mm_press_keys_simultaneously.iloc[0](["alt", "f4"], 1.1) 

df.mm_press_keys_simultaneously_force_activate.iloc[0](["alt", "f4"], 1.1) # better 

#-------------------------------------------------------------------------------------------#

# 5) Send Unicode

df.mm_send_unicode.iloc[0]('HGax&&öp')

# Use force activate to make sure that it goes to the right window

df.mm_send_unicode_force_activate.iloc[0]('HGax&&öp')
# How to activate (set focus to) a window 



# works most of the time and doesn't change mouse position

df.mm_activate_window.iloc[0]()

#-------------------------------------------------------------------------------------------#



# This function has been working for me without any problem. 

# It always activates the window and sets focus, 

# But it changes the mouse position

df.mm_force_activate_window.iloc[0]()



# You can also try this:

df.ff_Select.iloc[0](1)

More useful stuff:

# How to close the window/app

df.dropna(subset='ff_Close').iloc[0].ff_Close()

##############################################################################################

# move the window to (10,10) without resizing

df.dropna(subset='ff_Move').iloc[1].ff_Move(10,10)

##############################################################################################

# get a DataFrame with all children from an element:

df.dropna(subset='aa_all_children').iloc[1].aa_show_children()

##############################################################################################

# If you want to search in the whole DataFrame without getting any Exception ;-) :

df.d_loc_no_exception('str.contains','Cancel',na=False,flags=re.I).aa_direct_click.iloc[0]()

##############################################################################################

# maximize a window

df.window_maximize.iloc[0]()

##############################################################################################

# minimize a window

df.window_minimize.iloc[0]()

##############################################################################################

# restore windows

df.window_restore.iloc[0]()

##############################################################################################

# hide window

df.window_hide.iloc[0]()

##############################################################################################

# show window

df.window_show.iloc[0]()

##############################################################################################

# window to normal size

df.window_normal.iloc[0]()

##############################################################################################

# force minimize

df.window_forceminimize.iloc[0]()

##############################################################################################

# resizing a window:

df.iloc[0].resize_window( (0,0,100,100))

##############################################################################################

Here is an example with Notepad:

from pdwinauto import (

   mkey,

   click_on_main_menu,

   click_on_submenu,

   get_automation_frame_from_pid,

   sleep

)

mkey.enable_failsafekill("ctrl+e")

n1,n2 = "&File","&Save\tCtrl+S"

filepath = "F:\\tesa\\testtext1.txt"

text = "Hallo, mein Freund, wie geht es dir?"

pid = 11660

gdf = lambda: get_automation_frame_from_pid(pid)

df = gdf()

df.mm_send_unicode_force_activate.iloc[0](text)

sleep(1)

df4 = df.dropna(subset="aa_menu_items")

click_on_main_menu(df4[:1], menu1text=n1)

sleep(1)

df2 = gdf()

click_on_submenu(df2.dropna(subset="aa_menu_items"

                           ), menu1text=n1, menu2text=n2)

while True:

   sleep(2)

   try:

       df = gdf()

       f1 = df.d_loc_no_exception("str.contains",r"\.txt",

                                  regex=True)

       f2 = f1.d_loc_no_exception("str.contains",r"Edit")

       f2.iloc[0].mm_send_unicode_force_activate(filepath)

       sleep(2)

       df.loc[df.ff_CurrentDefaultAction ==

              "Save"].iloc[0].ff_DoDefaultAction()

       sleep(1)

       (df.loc[~df.ff_Close.isna() &

              (df.aa_title == 'Notepad')]

        .iloc[0].ff_Close())

       break

   except Exception as fe:

       continue

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pdwinauto-0.10.tar.gz (20.4 kB view details)

Uploaded Source

Built Distribution

pdwinauto-0.10-py3-none-any.whl (20.9 kB view details)

Uploaded Python 3

File details

Details for the file pdwinauto-0.10.tar.gz.

File metadata

  • Download URL: pdwinauto-0.10.tar.gz
  • Upload date:
  • Size: 20.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.9.13

File hashes

Hashes for pdwinauto-0.10.tar.gz
Algorithm Hash digest
SHA256 2a2a32bf9dd111c046df2863b927967adc276343e89df4717620af7e5b486940
MD5 97dd64a1283d424c60ab2037ef5b66e3
BLAKE2b-256 1667a27278610605b73c4cc81dd099c5b04b2e8774fc0fceccadbc81e2d8f5bf

See more details on using hashes here.

File details

Details for the file pdwinauto-0.10-py3-none-any.whl.

File metadata

  • Download URL: pdwinauto-0.10-py3-none-any.whl
  • Upload date:
  • Size: 20.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.9.13

File hashes

Hashes for pdwinauto-0.10-py3-none-any.whl
Algorithm Hash digest
SHA256 2786d4660f3e339c13ab1dff5ddf94dd816f180183cd0429b5aca42101a5f43a
MD5 3fce6932725fbdfa15371dd42dfff639
BLAKE2b-256 a36b68e05e50d6747e91c124e4e96c885daa5a17ca94e6d4fe7ccafb76d405e1

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page