mikk150
Repos
112
Followers
7

Events

branch namer sanitizes slashes in prefix

Is there an existing issue for this?

  • [X] I have searched the existing issues

Code improvement description

I would like to use git-flow style branch names so for this I have

version: 2
updates:
  - package-ecosystem: "npm"
    directory: /
    schedule:
      interval: "weekly"
      # Check for npm updates on Mondays
      day: "monday"
      time: "09:00"
    pull-request-branch-name:
      prefix: 'feature/dependabot'

but since branch namer sanitazes prefix part too, I cannot have slashes in my branch name

https://github.com/dependabot/dependabot-core/blob/6ba5296f550ade045187a1bd8425912faed8009b/common/lib/dependabot/pull_request_creator/branch_namer.rb#L43

Created at 3 days ago
create branch
mikk150 create branch make-ability-to-change-node-name
Created at 5 days ago
Created at 5 days ago
issue comment
[🐛 Bug]: Performance regression of Selenium 4.4

also happens to me on 4.3.0

Created at 1 week ago

add android to list as well

Created at 1 week ago
create branch
mikk150 create branch playwright-android
Created at 1 week ago
issue comment
check if browser is W3C instead of Android

Hi @mikk150 so this check would be sufficient for both android and iOS right? Just out of curiosity.

if webdriverio reports iOS as W3C browser(Which it should, because it is really Appium), then I do not see why not

Created at 1 week ago
pull request opened
check if browser is W3C instead of Android

Motivation/Description of the PR

My first implementation to this was stupid, this is way better one

  • Resolves #3317

Applicable helpers:

  • [x] WebDriver
  • [ ] Puppeteer
  • [ ] Nightmare
  • [ ] REST
  • [ ] FileHelper
  • [ ] Appium
  • [ ] Protractor
  • [ ] TestCafe
  • [ ] Playwright

Applicable plugins:

  • [ ] allure
  • [ ] autoDelay
  • [ ] autoLogin
  • [ ] customLocator
  • [ ] pauseOnFail
  • [ ] coverage
  • [ ] retryFailedStep
  • [ ] screenshotOnFail
  • [ ] selenoid
  • [ ] stepByStepReport
  • [ ] stepTimeout
  • [ ] wdio
  • [ ] subtitles

Type of change

  • [ ] :fire: Breaking changes
  • [ ] :rocket: New functionality
  • [ ] :bug: Bug fix
  • [ ] :clipboard: Documentation changes/updates
  • [ ] :hotsprings: Hot fix
  • [ ] :hammer: Markdown files fix - not related to source code
  • [x] :nail_care: Polish code

Checklist:

  • [ ] Tests have been added
  • [ ] Documentation has been added (Run npm run docs)
  • [ ] Lint checking (Run npm run lint)
  • [ ] Local tests are passed (Run npm test)
Created at 1 week ago
create branch
mikk150 create branch use-w3c-reporting-instead-of-comparison
Created at 1 week ago

Update codecept.Playwright.js (#3319)

Docs: Improved example for PW grabBrowserLogs() (#3315)

Playwright: Fix grabbing of browser console messages and window resize in new tab (#3325)

  • Docs: Fixed property name for plugin "retryFailedStep"

  • Playwright: Fixed grabbing of browser console messages in new tab and correct resize of window in new tab Also fixed an indentation issue I got by pre-commit hook (not related to my changes)

Co-authored-by: Jaromir Obr jobr@spork.tech

FIX-issue-3147 incorporates pollInterval into retryTo plugin as documented. (#3351)

Added capability to have customLocator attribute arrays; Updated chai-as-promised to v7; Fixed codecept.workers test; Fixed linting in codecept.playwright test; Fixed locator_test to.be.thrown call (#3348)

Playwright: connect over cdp option (#3309)

  • cdp connection option

  • update docs with cdp connection option

  • update docs in helper file

  • fix annotation

Co-authored-by: Hoffer Mihály mihaly.hoffer@shoprenter.hu

try to fix playwright tests (#3352)

  • try to fix playwright tests

  • fixed TestCafe tests

  • bump playwright

implemented secret for object (#3355)

use W3C standards when interfacing with appium. (#3317)

  • use W3C standards when interfacing with appium. It may still be broken when interfacing Ios, because I have no idea what IOS platform name is

  • platform name is in capabilities

improve: jsonresponse validation (#3316)

  • WIP

  • fix: handle the returned response is array

  • fix: code reviews

  • fix: code reviews

3.3.4 release (#3357)

  • updated changelog

  • fixed tests, added changelog and docs

  • version bump, docs updated

fix: TS complains secret args (#3360)

  • fix: TS complains secret args

  • fix: TS complains secret args

  • fix: TS complains secret args

fix: don't report files as existing when non-directory is in path (#3374)

fileExists would return true when testing "somedir/somefile" if somedir existed but was a file, not a directory. This was caused by checking for ENOENT only, when in that case the error code returned by "statSync" is ENOTDIR.

Replacing the use of "statSync" by a call to "fs.existsSync" avoids the problem while working correctly for all other cases.

Added codecept-tesults plugin documentation (#3365)

fix(3367): Android Click Broken When Platform Definition Capitalized (#3368)

  • fix(3367): Android Click Broken When Platform Definition Capitalized

  • fix(3367): Android Click Broken When Platform Definition Capitalized

  • fix: wrong method name is called

fix(3341): handleDownloads function is broken with Puppeteer 14.4 and onwards (#3372)

  • fix(3341): handleDownloads function is broken with Puppeteer 14.4 and onwards

  • fix: CR

fix(3370): els[0]._scrollIntoViewIfNeeded is broken with pupputeer > 14.4.0 (#3371)

  • fix: els[0]._scrollIntoViewIfNeeded while upgrading Puppeteer version from 10.2.0 to 15.4.0

  • change back the puppeteer version

fix(3384): scenario outline parameter substitution (#3385)

Fix Wrong Order with async MetaSteps (#3393)

  • Add async metastep handling

  • test(runner): Fix missing async/await

Feat/init codeceptjs with ts (#3364)

  • feat: option to init codeceptjs with typescript

  • feat: option to init codeceptjs with typescript

  • feat: option to init codeceptjs with typescript

  • fix: conflicts

  • fix: conflicts

  • fix: conflicts

  • addressed the CR

  • fix: CR

  • fix: CR

  • fix: CR

Created at 1 week ago

Update codecept.Playwright.js (#3319)

Docs: Improved example for PW grabBrowserLogs() (#3315)

Playwright: Fix grabbing of browser console messages and window resize in new tab (#3325)

  • Docs: Fixed property name for plugin "retryFailedStep"

  • Playwright: Fixed grabbing of browser console messages in new tab and correct resize of window in new tab Also fixed an indentation issue I got by pre-commit hook (not related to my changes)

Co-authored-by: Jaromir Obr jobr@spork.tech

FIX-issue-3147 incorporates pollInterval into retryTo plugin as documented. (#3351)

Added capability to have customLocator attribute arrays; Updated chai-as-promised to v7; Fixed codecept.workers test; Fixed linting in codecept.playwright test; Fixed locator_test to.be.thrown call (#3348)

Playwright: connect over cdp option (#3309)

  • cdp connection option

  • update docs with cdp connection option

  • update docs in helper file

  • fix annotation

Co-authored-by: Hoffer Mihály mihaly.hoffer@shoprenter.hu

try to fix playwright tests (#3352)

  • try to fix playwright tests

  • fixed TestCafe tests

  • bump playwright

implemented secret for object (#3355)

use W3C standards when interfacing with appium. (#3317)

  • use W3C standards when interfacing with appium. It may still be broken when interfacing Ios, because I have no idea what IOS platform name is

  • platform name is in capabilities

improve: jsonresponse validation (#3316)

  • WIP

  • fix: handle the returned response is array

  • fix: code reviews

  • fix: code reviews

3.3.4 release (#3357)

  • updated changelog

  • fixed tests, added changelog and docs

  • version bump, docs updated

fix: TS complains secret args (#3360)

  • fix: TS complains secret args

  • fix: TS complains secret args

  • fix: TS complains secret args

fix: don't report files as existing when non-directory is in path (#3374)

fileExists would return true when testing "somedir/somefile" if somedir existed but was a file, not a directory. This was caused by checking for ENOENT only, when in that case the error code returned by "statSync" is ENOTDIR.

Replacing the use of "statSync" by a call to "fs.existsSync" avoids the problem while working correctly for all other cases.

Added codecept-tesults plugin documentation (#3365)

fix(3367): Android Click Broken When Platform Definition Capitalized (#3368)

  • fix(3367): Android Click Broken When Platform Definition Capitalized

  • fix(3367): Android Click Broken When Platform Definition Capitalized

  • fix: wrong method name is called

fix(3341): handleDownloads function is broken with Puppeteer 14.4 and onwards (#3372)

  • fix(3341): handleDownloads function is broken with Puppeteer 14.4 and onwards

  • fix: CR

fix(3370): els[0]._scrollIntoViewIfNeeded is broken with pupputeer > 14.4.0 (#3371)

  • fix: els[0]._scrollIntoViewIfNeeded while upgrading Puppeteer version from 10.2.0 to 15.4.0

  • change back the puppeteer version

fix(3384): scenario outline parameter substitution (#3385)

Fix Wrong Order with async MetaSteps (#3393)

  • Add async metastep handling

  • test(runner): Fix missing async/await

Feat/init codeceptjs with ts (#3364)

  • feat: option to init codeceptjs with typescript

  • feat: option to init codeceptjs with typescript

  • feat: option to init codeceptjs with typescript

  • fix: conflicts

  • fix: conflicts

  • fix: conflicts

  • addressed the CR

  • fix: CR

  • fix: CR

  • fix: CR

Created at 1 week ago
composer.lock is now in .gitignore

What steps will reproduce the problem?

  1. Start a new project using this template composer create-project --prefer-dist yiisoft/yii2-app-advanced test-yii2

What's expected?

  • composer.lock should not be in .gitignore to keep track of my composer locs

What do you get instead?

  • composer.lock is in .gitignore

Additional info

| Q | A | ---------------- | --- | Yii version | | PHP version | | Operating system |

Created at 3 weeks ago

Implement InMemoryFilesystem

Created at 4 weeks ago

implement local filesystem adapter

Created at 1 month ago
issue comment
Copying frames from HEVC block to H264 block impossible due to incompatible pixel format?

Sorry to get you excited - the working transcode is just 8-bit unscaled. I've not built the v4l2 ISP rescale filter yet, but I'll put that on my todo.

Nah, I'm not exited, I know that there are hardware limitations in place (and why this is really hard, possibly impossible) to do. So I am not holding my breath.. Just a thing that would be really awesome to see(Get over hardware limitations in possibly hacky ways)

Created at 1 month ago
issue comment
Copying frames from HEVC block to H264 block impossible due to incompatible pixel format?

could only be started from outside X as I haven't worked out how to get a DRM fd that will let be get at writeback from within X

would be fine, because transcode servers do not even need X(Or any working output for that matter, could possibly even disable HDMI output, and people would still rejoice)

I suspect would use enough memory bandwidth to break real-time 1080p60 transcode.

Bummer, as UHD 10bit to FHD is really far out of reach then

BTW for what it is worth, h/w 8-bit H265 -> H264 transcode does now work.

does it do UHD H265 to FHD H264? if so, this already is really big stepping stone

Created at 1 month ago
issue comment
Copying frames from HEVC block to H264 block impossible due to incompatible pixel format?

Sadly the only h/w that can consume (and convert) SAND30 on a Pi4 is the HVS. Under KMS that is given to the exclusive use of the display. In theory you can get get DRM (owner of the HVS) to render to a special connector rather than the screen and that would do what you want, but I've yet to see that done and getting hold of a DRM lease/fd that will let you do even try requires (AFAIK) whoever is holding it at that time (the display) to give it up which probably isn't what you want as it will take the display away.

BTW, if it took blanking display to transcode HEVC to H264, it would actually work, as many people who want it I presume are trying to use RPI as transcode server(as it is cheap, power efficient and can decode HEVC 10bit and encode H264)

Created at 1 month ago

make them work

Created at 1 month ago
create branch
mikk150 create branch use-v2-flysystem
Created at 1 month ago

fix Dropbox adapter

this is never used anywhere

documnt new ways to authendicate

Merge pull request #50 from mikk150/fix-49-dropbox

fix Dropbox adapter

Created at 1 month ago
there is no way to upload files using other transport method than built-in method

@cubiclesoft reopen as yes, I can call startupload() and pass false, but now it does not show progress of uploading

Created at 1 month ago
issue comment
Add ability to send PSR-7 streams from response

So, now when 2.1 is pretty much deleted, maybe we should reopen it??

Created at 2 months ago
there is no way to upload files using other transport method than built-in method

I would like to do everything using Axios, unfortunately oninitupload does not have a way to tell this plugin that upload has finished

Created at 2 months ago
issue comment
class_exists() error in yii2-debug after upgrade php to 8.1

it is fixed in 634b9cc5e89bcc65091fe25ea2b5538b5bc5ce76, so when new release comes out it will be fixed, in the meantime you can use "yiisoft/yii2-debug": "dev-master#5d9ab363e96869e1ff55d85e4aaa942f63d2cc85",

Created at 2 months ago
Created at 2 months ago
pull request opened
make it compatible with ramsey uuid library version 4
Created at 2 months ago
mikk150 create branch update_ramsey_uuid
Created at 2 months ago

what was on my mind when I did that

Merge pull request #3 from mikk150/do-not-add-namespace-to-migration

what was on my mind when I did that

fixed saving json fields

Merge pull request #4 from rinu/bugfix/json

fixed saving json fields

Created at 2 months ago
issue comment
Squad (393380)

We can hope this ongoing work will solve the VRAM issue: doitsujin/dxvk#2663

all seems to be D3D9 related, squad is D3D11

Created at 2 months ago
issue comment
Squad (393380)

So, I tried it - performance was as good as DX11 mode with no jitter(Last time I tried it I remember I had bad jitter), I did not see more than 7GB of VRAM usage... I still doubt that it can help for horrible performance tho

okay, i ll try it again soon on new 3080/10G, maybe it will be any better. any VRAM leaks on maps change? what GPU do you have? for some reason i never saw "DX12" any good on Proton even when same GPU was okay on Win10.

So, I do see telltale VRAM leaks, and I see 11GB of VRAM usage on Fallujah, so I do not think you will have great experience on your 3080. I have RX6900XT with 16GB of VRAM and I have not experienced these performace issues ever since I upgraded from Vega64 with 8GB of VRAM. When I had Vega64 and I had these issues, I could see that VRAM usage was pretty much 100%, so I am really certain, it is VRAM usage issue

Created at 2 months ago