I would like to test a patched implementation of juptyerlab. I was hoping to use selenium to execute "hello world"
in a code cell. So far I can log in and create a new notebook:
from selenium import webdriver
driver = webdriver.Firefox()
# assume jupyterlab is running and serving on localhost at port 8888
driver.get("http://localhost:8888")
elem = driver.find_element_by_id("password_input")
password = ""
elem.send_keys(password)
elem = driver.find_element_by_id("login_submit")
elem.click()
elem = driver.find_element_by_css_selector(".jp-Launcher-cwd+ .jp-Launcher-section .jp-LauncherCard")
elem.click()
This creates a new notebook, but now I'm stuck at the point of entering some code in a cell and running it. If I view the page source I don't see any html elements for the cells. But if I enter print("test")
in a cell, then driver.page_source
contains this (it's pretty nested in other stuff I've omitted too):
<div class="CodeMirror cm-s-jupyter CodeMirror-wrap jp-mod-readOnly">
<div style="overflow: hidden; position: relative; width: 3px; height: 0px; top: 0px; left: 0px;">
<textarea
style="position: absolute; bottom: -1em; padding: 0px; width: 1px; height: 1em; outline: currentcolor none medium;"
autocorrect="off" autocapitalize="off"
spellcheck="false" tabindex="0"
wrap="off"></textarea></div>
<div class="CodeMirror-vscrollbar" tabindex="-1"
cm-not-content="true"
style="display: block; bottom: 0px;">
<div style="min-width: 1px; height: 33px;"></div>
</div>
<div class="CodeMirror-hscrollbar" tabindex="-1"
cm-not-content="true">
<div style="height: 100%; min-height: 1px; width: 0px;"></div>
</div>
<div class="CodeMirror-scrollbar-filler"
cm-not-content="true"></div>
<div class="CodeMirror-gutter-filler"
cm-not-content="true"></div>
<div class="CodeMirror-scroll" tabindex="-1" draggable="true">
<div class="CodeMirror-sizer"
style="margin-left: 0px; padding-right: 0px; padding-bottom: 0px;">
<div style="position: relative;">
<div class="CodeMirror-lines" role="presentation">
<div style="position: relative; outline: currentcolor none medium;"
role="presentation">
<div class="CodeMirror-measure">
<pre><span>xxxxxxxxxx</span></pre>
</div>
<div class="CodeMirror-measure">
<pre class="CodeMirror-line"
role="presentation"><span
role="presentation"><span
class="cm-builtin">print</span>(<span
class="cm-string">"test"</span>)</span></pre>
</div>
<div style="position: relative; z-index: 1;"></div>
<div class="CodeMirror-cursors"></div>
<div class="CodeMirror-code"
role="presentation"></div>
</div>
</div>
</div>
</div>
<div style="position: absolute; height: 30px; width: 1px; border-bottom: 0px solid transparent;"></div>
<div class="CodeMirror-gutters"
style="display: none;"></div>
</div>
</div>
I can see where the text for print("text")
is (i.e. the deepest nested elements in the above html snippet), but I can't figure out which element here I would be able to send text to or send keys to.
I came across robotframework-jupyterlibrary and it has some clues such as this and this . From those links I see
${JLAB CSS ACTIVE INPUT} ${JLAB CSS ACTIVE CELL} .CodeMirror
and
Add and Run JupyterLab Code Cell
[Arguments] ${code}=print("hello world")
[Documentation] Add a ``code`` cell to the currently active notebook and run it.
Click Element css:${JLAB CSS NB TOOLBAR} ${JLAB CSS ICON ADD}
Sleep 0.1s
${cell} = Get WebElement css:${JLAB CSS ACTIVE INPUT}
Click Element ${cell}
Set CodeMirror Value ${JLAB CSS ACTIVE INPUT} ${code}
Run Current JupyterLab Code Cell
Click Element ${cell}
which makes me think if I select the .CodeMirror
element, then I just need to figure out what Get WebElement
does in that weird language and how to do it in selenium.
Any ideas?
I've also tried (based on https://stackoverflow.com/a/48723135/1011724 and https://stackoverflow.com/a/50279295/1011724):
from selenium.webdriver.common.action_chains import ActionChains
actions = action_chains.ActionChains(driver)
textarea = driver.find_elements_by_css_selector('.CodeMirror textarea')[0] # tried for [0], [1] ,[2] and [3] which is all of them.
actions.move_to_element(textarea).click().send_keys("testing...").perform()
but I keep getting the error
selenium.common.exceptions.WebDriverException: Message: TypeError: rect is undefined
Code below tested with Chrome, Firefox and jupyterlab latest versions:
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.support.wait import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.common.by import By
driver = webdriver.Chrome()
wait = WebDriverWait(driver, 10)
driver.get("http://localhost:8888")
token = "0107216930d05db8a7c36ad6a73573dd5349c3dd56fee852"
wait.until(EC.element_to_be_clickable((By.ID, "password_input"))).send_keys(token, Keys.ENTER)
# wait for "Python 3" Notebook menu or CodeMirror element if already launched.
wait.until(EC.presence_of_element_located(
(By.CSS_SELECTOR, "[title='Python 3'][data-category='Notebook'], .jp-Notebook .CodeMirror")))
# if "Python 3" Notebook menu found click to open new Notebook
if len(driver.find_elements_by_css_selector("[title='Python 3'][data-category='Notebook']")) > 0:
driver.find_element_by_css_selector("[title='Python 3'][data-category='Notebook']").click()
# wait for CodeMirror and click to focus
code_mirror = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, ".jp-Notebook .CodeMirror")))
code_mirror.click()
code_mirror.find_element_by_tag_name("textarea").send_keys("print('Hello World!')")
driver.find_element_by_css_selector("[data-icon='run']").click()
output = wait.until(EC.visibility_of_element_located((By.CSS_SELECTOR, ".jp-OutputArea-output")))
print(output.text)
assert output.text.strip() == "Hello World!"
driver.quit()
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With