in dashboard you can monitor usage when using the workers binding API right now. we are in process f
in dashboard you can monitor usage when using the workers binding API right now. we are in process for adding for the REST API. so please stay tuned
Mgot it, thank you!
DI’m really happy to see Cloudflare add support for Playwright with Browser Rendering.
I’m attempting to use it on a project, but I’m running in to errors when attempting to deploy.
As soon as I include the “@Cloudflare/playwright” package in my build, deploys fail.
My guess is that there could be something in that package that isn’t supported in the Workers run time?
I’m attempting to use it on a project, but I’m running in to errors when attempting to deploy.
As soon as I include the “@Cloudflare/playwright” package in my build, deploys fail.
My guess is that there could be something in that package that isn’t supported in the Workers run time?
RDoes browser rendering support proxies if I want to use? If I need a specific location
JIt looks like browser content requests using the REST api append
cf-biso-request-id and cf-biso-devtools request headers when fetching the url. Is there any way to override/remove them? (setExtraHTTPHeaders did not seem to work for these)
JAny examples how can i integrate browser rendering api in my hono app
KGitHub
Cloudflare Worker to analyze SEO of input website using Cloudflare browser rendering and Llama-3.2 hosted on Workers AI - elizabethsiegle/seo-analysis-browser-rendering-worker
CHi lovely folks - I am trying to use with the sdk:
https://developers.cloudflare.com/api/node/resources/browser_rendering/subresources/screenshot/
What is the response attribute which gives me back an image buffer (or base64) when I call
The response type def seems to contain { status, errors? } but no option for the actual image.
https://developers.cloudflare.com/api/node/resources/browser_rendering/subresources/screenshot/
What is the response attribute which gives me back an image buffer (or base64) when I call
await client.browserRendering.screenshot.create()The response type def seems to contain { status, errors? } but no option for the actual image.
Interact with Cloudflare's products and services via the Cloudflare API
CI do seem to get back some data as screenshot response but it isnt base64 or a buffer that is valid image. What am I doing wrong?
SThe API docs are wrong, they say the binding type is
browser_rendering but that results in a 400. The wrangler source code uses "browser": https://github.com/cloudflare/workers-sdk/blob/main/packages/wrangler/src/deployment-bundle/create-worker-upload-form.ts#L470-L471
TAm I correct to assume that the Rest API for creating PDFs does not allow specifying format (e.g. A4 instead of US Letter)?
Tgod damn
T
Jyea you would think that endpoint would take a
it's not in the public docs at least
PDFOptions https://pptr.dev/api/puppeteer.pdfoptionsit's not in the public docs at least
Valid options to configure PDF generation via Page.pdf().
Tfor sure it works when accessing puppeteer in workers directly, but I wanted to avoid that for simple pdf rendering because of all the local/remote dev shenigans
Jagreed, these endpoints are useful, as long as enough of the params are supported. Maybe someone from the cf team can chime in on the idea of supporting
PDFOptions for that pdf endpoint at some point
Fhey, team, there is a bug on /links, it will respose the wrong links now, I have used this feature for several weeks,you must have changed something wrong
Fthe /docs are doubled in the links
Tare there known issues with font rendering using browser rendering in a worker? this works locally but here you can see the kerning is completely off - not sure how to debug
TThis is using am embedded woff2 variable font
Fwhere should we report this issue
Fit can be easily reproduced
Ki'm able to reproduce. have reported bug to team
Kwe've got something in the works that should fix this month. please hold tight
Kthanks for the feedback @John Spurlock and @Tobias . will discuss with the team. aret here any other params you would expect to be included in the simple endpoints?
Tthis is the only one i noticed so far, not used anything else
Klooking into this
Fthanks!
MHello Everyone, I am getting error Unable to regenerate API key. Invalid password
For Workers & Pages, what is the name of the domain?
NA
What is the error number?
1071
What is the error message?
Unable to regenerate API key. Invalid password
What is the issue or error you’re encountering
Trying to get the global API to work with the Cloudflare API, but while clicking on view, it asks for the password, which I don’t have, as I have logged in using Google auth
What steps have you taken to resolve the issue?
I tried adding the custom API as a password at that time, but it started giving this error.
For Workers & Pages, what is the name of the domain?
NA
What is the error number?
1071
What is the error message?
Unable to regenerate API key. Invalid password
What is the issue or error you’re encountering
Trying to get the global API to work with the Cloudflare API, but while clicking on view, it asks for the password, which I don’t have, as I have logged in using Google auth
What steps have you taken to resolve the issue?
I tried adding the custom API as a password at that time, but it started giving this error.
CYou can logout and use the "Forgot your password?" option on the login screen, to get it set and then generate it
Using Global API Key isn't recommended though, scoped api tokens are more secure
Using Global API Key isn't recommended though, scoped api tokens are more secure
Chey, any idea how the caching works for the REST API? IE: If I do a call to
/markdown, then immediately after do a /scrape call to get the title element, will that generally have to re-scrape or can it pull from the cache?
Ajust checking in on this after a while... did the weird as heck billing ever get sorted out with this?
Fhey, Kathy, this issue has not been fixed yet. I think it is a bug obviously, when will your team fix this.
Chey guys, your docs are messed up for the scrape endpoint. https://developers.cloudflare.com/api/resources/browser_rendering/subresources/scrape/
The result shape is
The result shape is
{result: [{ results: [{ ... }] }]}{result: {result: {}}}Interact with Cloudflare's products and services via the Cloudflare API
CAnother serious issue. Links endpoint works with 
urlhtml/contentAnother thing... I would really like to be able to get the title of the page, I'm not sure if scraping for the title element will be good enough for SPAs... I tried to do this by injecting a script that would create a new dom element with document.title content in it but that didn't seem to work.
CRunning into this as well
Kit's in progress. hopefully fixed tmrw
Kwhat do you mean? we have not yet started charging, however we have raised limits in the meantime and added browser rendering to workers free
Kthanks for raising these bugs + feature request.
MHi, I am evaluating browser rendering API for an AI agent with browsing capability and persistent session (logged in user). Is this a good solution? (I also looking at Browserbase)
CI'm receiving a ton of 429s, any chance I can get some help raising the rate limits? I'm happy to do a user interview call too with our use cases or if you're thinking through pricing
Rhello
We have a issue with puppeteer setUserAgent
we do a classical await page.setUserAgent(userAgent);
but the navigator.userAgent always returns Cloudflare-Workers, ignoring what we used. Is that expected, a known issue or something?
This is blocking us for leveraging the browser api properly since sites we analyze recognize this fact and respond differently.
We have a issue with puppeteer setUserAgent
we do a classical await page.setUserAgent(userAgent);
but the navigator.userAgent always returns Cloudflare-Workers, ignoring what we used. Is that expected, a known issue or something?
This is blocking us for leveraging the browser api properly since sites we analyze recognize this fact and respond differently.
AHello. Why i'm getting: {"error":"Unable to create new browser: code: 429: message: Browser time limit exceeded for today"}, i bought worker paid plan to avoid this...
ANo responses? 
MHello , I have issue when I using Browser rendering restfull api , I want to render url to pdf , first requests it works fine , then all request return this error "{"success":false,"errors":[{"message":"Authentication failed","code":10000}]}" the api token is valid and has permission on browser rendering
Sexactly the same issue
not sure why
not sure why
Si tried account and user both the tokens as well
Stried with "Bearer Token" as well
am i doing anything wrong here?
am i doing anything wrong here?
MI think there an issue with cloudflare
