Currently unmaintained - call for contributors and maintainers!

[Granitosaurus]

Edit from Mattwmaster58: This repo is unmaintained and has been outside of minor changes for a long time. Please consider playwright-python as an alternative.

The original repository has been unmaitained for a while and pyppeteer turned out to be quite a hit so I've established and organization and a fork - calling for contributors and additional maintainers!

Join us at matrix room: #pyppeteer@matrix.org or https://matrix.to/#/!ScehqfCSdMAUhZoeDC:matrix.org?via=matrix.org

[marksteward]

Count me in, pyppeteer's great and I'd be happy to help share the maintenance load.

[Granitosaurus]

Thanks @marksteward - I've set up a matrix channel to communicate this easier :)
(see original post edit)

[enquora]

A non-specific call for contributors/maintainers is likely to be less than successful, in my experience. The code base is complex and a non-specific approach risks never gaining traction.

A more effective approach might be the creation of a short list of triaged issues, then looking for existing solutions (in the original repository and on the web generally).
Perhaps the triaged issues list could be grouped into Admin (transferring issues from original repository), relatively easy technical and more difficult technical (addressing the inability to maintain connections and manage platform resources for long-running usage)

[Granitosaurus]

@enquora you raise a good point!

I did start few projects (projects tab of the repo) but the code base seems to be really bloated and really hard to onboard. I'm myself only partially familiar with it so the current goal is to fully onboard myself and improve onboarding experience for potential contributors.

You can see more over at https://github.com/pyppeteer/pyppeteer2/projects/1

[marksteward]

I assume you know it's basically a translation of puppeteer? What do you mean by bloated?

[Granitosaurus]

@marksteward I worded it a bit incorrectly - the package itself isn't bloated but everything that comes with it is. Makefiles, doit files, obsene setup.py instruction, tox with 100 lines that doesn't even build, py35 support etc. etc. Mostly things that make it difficult for maintainers not coders per se - onboarding as a maintainer or even a deeper contributor is a bit of a struggle with such setups.

[marksteward]

Hmm, I'm not interested in tearing stuff out immediately. There's a lot of code updates to make first.

[Granitosaurus]

@marksteward PRs are welcome! :)
None of the "tear out" work would affect PRs.

See https://github.com/pyppeteer/pyppeteer2/projects/2
There are few good PRs in other forks too that can be ported here.

[enquora]

Any work on removing code which doesn't directly and immediately impact performance or reliability of the codebase as currently used is likely to be interpreted negatively by seasoned potential contributors.

More than a few simple and immediate things need attention. The following come to mind:

• the PyPi package page doesn't have a useable link to this repository!
• the current included version of Chromium is woefully out of date and needs updating
• patches exist elsewhere which should be triaged, assessed, updated and merged where appropriate

The easiest thing in the world is to do what one wants and knows how to - rather than what needs doing. Work on other things may well prove useful in the long or even medium term but without attracting a viable support team in the near term, they won't matter, are a distraction for public discussion, and should be considered a private endeavour.

I'll reiterate - the current needs of the codebase are twofold:

• updating the included version of Chromium and keeping it current
• determining the cause of existing memory leaks and interprocess connection failures, and working steadily towards a fix.

Seasoned potential contributors are likely to use the visibility of immediate and practical steps forward as a litmus test for their own willingness to contribute. That is how projects gain traction.

[flyte]

I'd also be wary of deviating too far from Puppeteer that this is based on. I expect a lot of the issues this library has have already been solved by Puppeteer. I guess you need to make a decision to forge ahead with your own issues and fixes, or maintain parity with the "upstream" JavaScript project.

[DanEdens]

I agree with flyte, I would only be interested in helping to keep it caught up with Puppeteer. Google is already doing most of the work.
I'm in tho. This is a project worth maintaining. I'm worried about the visibility. Would be far better if we could take over the original.
I will go through the 300 odd forks and see what else has already been attempted, maybe I'll find some more people.

[steven-81]

I'm looking at moving from Selenium to Pyppeteer and once I become more familiar with the package would be interested in participating.

Agree with the comments above that the priority feels like it will be achieving parity with the latest versions of Puppeteer and Chromium.

[Granitosaurus]

A lot of good points were raised here however it seems that most of the existing patches and changes are moot without catching up with newest pupetteer version.

Right now this is the priority - catching up with puppeteer
https://github.com/pyppeteer/pyppeteer2/projects/3
#16

It's rather tedious task as javascript is really tough to read/follow and as the spirit of this project goes things are being tried to be kept as close to puppeteer as possible which results in some really ugly unpythonic code.

If anyone would like to contribute feel free to pickup any modules that are in the todo list and/or come join me on matrix for cooperation :)

[Mattwmaster58]

Right now the project mentioned by @Granitosaurus is being implemented in #16 (branch pup2.1.1). Ideally any contributor would work off of that branch for now to avoid duplicating our efforts.

[gyermolenko]

Because of pup2.1.1 branch, transition from forked repo (docs, notes, CI etc) - It is hard to see from "outside" where I can try to help.

[Granitosaurus]

You can come join us on matrix and we'll help you find something to do!

…

-------- Original Message --------

On Apr 23, 2020, 4:38 PM, Grygorii Iermolenko wrote: Because of pup2.1.1 branch, transition from forked repo (docs, notes, CI etc) - It is hard to see from "outside" where I can try to help. — You are receiving this because you were mentioned. Reply to this email directly, [view it on GitHub](#1 (comment)), or [unsubscribe](https://github.com/notifications/unsubscribe-auth/ABJY6RHPZ7Q3SA2NLE3E5PLROAEA3ANCNFSM4KSMUZLQ).

[FredHappyface]

Regarding the linked issue miyakogi/pyppeteer#295

I think you could go ahead trying to reclaim the name by opening an issue here pypi/support#145

There are a few prerequisites here https://www.python.org/dev/peps/pep-0541/#how-to-request-a-name-transfer

Though it has to be said that this is a pretty slow process (think you are looking at a minimum of 6 weeks)

Btw would be interested in contributing to this. Python knowledge is reasonable but puppeteer knowledge will most certainly need improving

Thank you for keeping this awesome project alive 👍

[Mattwmaster58]

@FredHappyface Thanks for the suggestion, I will definitely have to take a look into that.

WRT contributing, we're pretty active on the Matrix channel so I'd encourage you to join there, there are quite a few tasks that only require a intermediate level of understanding of puppeteer/JS.

[Mattwmaster58]

@FredHappyface The issue has been filed and of right now we are awaiting a response: pypi/support#344

[cgarciae]

Nice! Thanks for continuing the project :D
It might be a good idea to put an issue on the other project pointing here.

[asher-lab]

I also would like to volunteer. Count me in as well :)

[csreddy]

I would like to contribute as well. Not sure how to setup the project as I'm new to Python projects but comfortable with Python language. It would be great if there is a readme for instructions on how to run the project, unit tests, debugging etc.

[asoundarya96]

@Mattwmaster58 Is this Issue#123 still open?
When userDataDir is set, browser.close() tries to shut browser gracefully but chromeClosed property is not set to true. Hence it throws exception atexit. Can i raise PR on this?

[Mattwmaster58]

@Soundarya1996

Can i raise PR on this?

Nobody has time to review or make a release, sorry. Although a PR may help others apply their own fixes