Skip to content

Latest commit

 

History

History
301 lines (218 loc) · 9.73 KB

CHANGELOG.md

File metadata and controls

301 lines (218 loc) · 9.73 KB
Raw

History

Version 2.0.0

  • Bump pyee version, which removes support for Python 3.7
  • Bumped included browser version to revision 1181205. It may not match the base puppeteer version, but at least it runs
  • Fix invalid escape sequence (#453)
  • Fix deprecated asyncio wait in page.py (#451)
  • Remove version_info. version is still present and can be parsed if necessary

Version 1.0.2

  • Fix circular import as result of 1.0.1 (#344)

Version 1.0.1

  • don't configure logging ourselves (#343)
  • make logging better for chromium downloader

Version 1.0.0

  • hotfix: websockets 10 for python 3.10 (#321, #327)
    • removes support for python 3.6./
  • remove Page.craete

Version 0.2.6

  • Change build backend to poetry-core (allows for faster PEP517 package building) #262 @fabaff
  • Chromium download fixes (file not found) #245 @mborsetti
  • Do not try to set an exception on finished futures #216 @polyfloyd
  • Add HTTPException to caught exceptions in launch #293 @raymondguo-db
  • Fix encoding error #226 @aleksei140888
  • support websockets 9.0 #252 @mborsetti
  • Fix tqdm exception when NO_PROGRESS_BAR is True #224 @mborsetti
  • fix(browser): Clean up coroutine Browser._targetCreated() #271 @H--o-I

Version 0.2.5

  • Match package version and __version__ (🤦‍♂️)
  • Use importlib_metadata so this isn't a problem in the future

Version 0.2.4

  • Update pyee dependency breaking build failures on NixOS + Fedora packaging systems (#207)

Version 0.2.3

  • Hotfix: random freezes from sending stdout to PIPE instead of DEVNULL
  • Fix tests package being installed for no reason

Version 0.0.26

  • Add $PYPPETEER_NO_PROGRESS_BAR environment variable
  • pyppeteer.defaultArgs now accepts that help infer chromium command-line flags.
  • pyppeteer.launch() argument ignoreDefaultArgs now accepts a list of flags to ignore.
  • Page.type() now supports typing emoji
  • Page.pdf() accepts a new argument preferCSSPageSize
  • Add new option defaultViewport to launch() and connect()
  • Add BrowserContext.pages() method

Version 0.0.25 (2018-09-27)

  • Fix miss-spelled methods and functions
    • Change Browser.isIncognite to Browser.isIncognito
    • Change Browser.createIncogniteBrowserContext to Browser.createIncognitoBrowserContext
    • Change chromium_excutable to chromium_executable
    • Remove craete function in page.py

Version 0.0.24 (2018-09-12)

Catch up puppeteer v1.6.0

  • Add ElementHandle.isIntersectingViewport()
  • Add reportAnonymousScript option to Coverage.startJSCoverage()
  • Add Page.waitForRequest and Page.waitForResponse methods
  • Now possible to attach to extension background pages with Target.page()
  • Improved reliability of clicking with Page.click() and ElementHandle.click()

Version 0.0.23 (2018-09-10)

Catch up puppeteer v1.5.0

  • Add BrowserContext class
  • Add Worker class
  • Change CDPSession.send to a normal function which returns awaitable value
  • Add Page.isClosed method
  • Add ElementHandle.querySelectorAllEval and ElementHandle.JJeval
  • Add Target.opener
  • Add Request.isNavigationRequest

Version 0.0.22 (2018-09-06)

Catch up puppeteer v1.4.0

  • Add pyppeteer.DEBUG variable
  • Add Page.browser
  • Add Target.browser
  • Add ElementHandle.querySelectorEval and ElementHandle.Jeval
  • Add runBeforeUnload option to Page.close method
  • Change Page.querySelectorEval to raise ElementHandleError when element which matches selector is not found
  • Report 'Log' domain entries as 'console' events
  • Fix Page.goto to return response when page pushes new state
  • (OS X) Suppress long log when extracting chromium

Version 0.0.21 (2018-08-21)

Catch up puppeteer v1.3.0

  • Add pyppeteer-install command
  • Add autoClose option to launch function
  • Add loop option to launch function (experimental)
  • Add Page.setBypassCSP method
  • Page.tracing.stop returns result data
  • Rename documentloaded to domcontentloaded on waitUntil option
  • Fix slowMo option
  • Fix anchor navigation
  • Fix to return response via redirects
  • Continue to find WS URL while process is alive

Version 0.0.20 (2018-08-11)

  • Run on msys/cygwin, anyway

  • Raise error correctly when connection failed (PR#91)

  • Change browser download location and temporary user data directory to:

    • If $PYPPETEER_HOME environment variable is defined, use this location
    • Otherwise, use platform dependent locations, based on appdirs:
      • 'C:\Users\<username>\AppData\Local\pyppeteer' (Windows)
      • '/Users/<username>/Library/Application Support/pyppeteer' (OS X)
      • '/home/<username>/.local/share/pyppeteer' (Linux)
        • or in '$XDG_DATA_HOME/pyppeteer' if $XDG_DATA_HOME is defined

  • Introduce $PYPPETEER_CHROMIUM_REVISION

  • Introduce $PYPPETEER_HOME

  • Add logLevel option to launch and connect functions

  • Add page close event

  • Add ElementHandle.boxModel method

  • Add an option to disable timeout for waitFor functions

Version 0.0.19 (2018-07-05)

Catch up puppeteer v1.2.0

  • Add ElementHandle.contentFrame method
  • Add Request.redirectChain method
  • Page.addScriptTag accepts a new option type

Version 0.0.18 (2018-07-04)

Catch up puppeteer v1.1.1

  • Add Page.waitForXPath and Frame.waitForXPath
  • Page.waitFor accepts xpath string which starts with //
  • Add Response.fromCache and Response.fromServiceWorker
  • Add SecurityDetails class and response.securityDetails
  • Add Page.setCacheEnabled method
  • Add ExecutionContext.frame
  • Add dumpio option to launch function
  • Add slowMo option to connect function
  • launcher.connect can be access from package top
    • from pyppeteer import connect is now valid
  • Add Frame.evaluateHandle
  • Add Page.Events.DOMContentLoaded

Version 0.0.17 (2018-04-02)

  • Mark as alpha

  • Gracefully terminate browser process

  • Request.method and Request.postData return None if no data

  • Change Target.url and Target.type to properties

  • Change Dialog.message and Dialog.defaultValue to properties

  • Fix: properly emit Browser.targetChanged events

  • Fix: properly emit Browser.targetDestroyed events

Version 0.0.16 (2018-03-23)

  • BugFix: Skip SIGHUP option on windows (windows does not support this signal)

Version 0.0.15 (2018-03-22)

Catch up puppeteer v1.0.0

  • Support raf and mutation polling for waitFor* methods
  • Add Page.coverage to support JS and CSS coverage
  • Add XPath support with Page.xpath, Frame.xpath, and ElementHandle.xpath
  • Add Target.createCDPSession to work with raw Devtools Protocol
  • Change Frame.executionContext from property to coroutine
  • Add ignoreDefaultArgs option to pyppeteer.launch
  • Add handleSIGINT/handleSIGTERM/handleSIGHUP options to pyppeteer.launch
  • Add Page.setDefaultNavigationTimeout method
  • Page.waitFor* methods accept JSHandle as argument
  • Implement Frame.content and Frame.setContent methods
  • page.tracing.start accepts custom tracing categories option
  • Add Browser.process property
  • Add Request.frame property

Version 0.0.14 (2018-03-14)

  • Read WS endpoint from web interface instead of stdout

  • Pass environment variables of python process to chrome by default

  • Do not limit size of websocket frames

  • BugFix:

    • Keyboard.type
    • Page.Events.Metrics

Version 0.0.13 (2018-03-10)

Catch up puppeteer v0.13.0

  • pyppeteer.launch() is now coroutine

  • Implement connect function

  • PYPPETEER_DOWNLOAD_HOST env variable specifies host part of URL to download chromium

  • Rename setRequestInterceptionEnable to setRequestInterception

  • Rename Page.getMetrics to Page.metrics

  • Implement Browser.pages to access all pages

    • Add Target class and some new method on Browser

  • Add ElementHandle.querySelector and ElementHandle.querySelectorAll

  • Refactor NavigatorWatcher

    • add documentloaded, networkidle0, and networkidle2 options

  • Request.abort accepts error code

  • addScriptTag and addStyleTag return ElementHandle

  • Add force_expr option to evaluate method

  • Page.select returns selected values

  • Add pyppeteer.version and pyppeteer.version_info

  • BugFix:

    • Do not change original options dictionary
    • Page.frames
    • Page.queryObjects
    • Page.exposeFunction
    • Request interception
    • Console API
    • websocket error on closing browser (#24)

Version 0.0.12 (2018-03-01)

  • BugFix (#33)

Version 0.0.11 (2018-03-01)

Catch up puppeteer v0.12.0

  • Remove ElementHandle.evaluate
  • Remove ElementHandle.attribute
  • Deprecate Page.plainText
  • Deprecate Page.injectFile
  • Add Page.querySelectorAllEval
  • Add Page.select and Page.type
  • Add ElementHandle.boundingBox and ElementHandle.screenshot
  • Add ElementHandle.focus, ElementHandle.type, and ElementHandle.press
  • Add getMetrics method
  • Add offlineMode

Version 0.0.10 (2018-02-27)

  • Enable to import launch from package root
  • Change browser.close to coroutine function
  • Catch up puppeteer v0.11.0

Version 0.0.9 (2017-09-09)

  • Delete temporary user data directory when browser closed
  • Fix bug to fail extracting zip on mac

Version 0.0.8 (2017-09-03)

  • Change chromium revision
  • Support steps option of Mouse.move()
  • Experimentally supports python 3.5 by py-backwards

Version 0.0.7 (2017-09-03)

  • Catch up puppeteer v0.10.2
    • Add Page.querySelectorEval (Page.$eval in puppeteer)
    • Deprecate ElementHandle.attribute
    • Add Touchscreen class and implement Page.tap and ElementHandle.tap

Version 0.0.6 (2017-09-02)

  • Accept keyword arguments for options
  • Faster polling on waitFor* functions
  • Fix bugs

Version 0.0.5 (2017-08-30)

  • Implement pdf printing
  • Implement waitFor* functions

Version 0.0.4 (2017-08-30)

  • Register PyPI