edgemesh/crux.run

The Chrome Real User Experience (CrUX) Database ... but faster

Edgemesh Client

The purpose of this repository is to serve as a knowledge base for Edgemesh Client development. Here you can request features, report bugs and find links to all the necessary resources to get started with Edgemesh Client.

Please check all the closed issues and wikis for solutions to your problems before opening new issues. You can also join our discord for community support, or email support directly.

To learn more about Edgemesh's product offerings go to edgemesh.com. For documentation please visit docs.edgemesh.com. For partnership inquiries send an email to [email protected].

crux.run's Issues

Offset parameter?

One more question if that’s okay. Is it possible to add something to offset what month of data you want to pull?

For example, you can use months=2 to pull the past two months. But if I want to compare to 6 months ago I have to pull 6 months of data rather than just being able to specify I want 1 months worth of data from 6 months ago.

Maybe something like an ?offset=6 parameter or something would be handy?

Then I could do something like:

https://api.crux.run/v1/summary/overview/?domain=www.google.com&months=1

For last month and then:

https://api.crux.run/v1/summary/overview/?domain=www.google.com&months=1&offset=6

For 6 months ago.

Combined SSL/Non-SSL details need more clarity

By default we combine metrics across both the http and https endpoints. This was done to ensure that customers could easily see the whole picture. We should add some descriptions on the results pages to explain this (and point folks to the SSL API)

Bar and density charts are not matching

At the moment it seems like the two charts for the measures are not matching. For example for Web.dev FCP, August bar chart shows 34% fast, the density chart shows roighly 80% fast:

Add ConnectionType Filter for Dashboard

It would be helpful to allow users to change the dashboard (filter) based on the ConnectionType. We already have this pivot available so it's just exposing that in the dashboard

New CruX metrics

Hey,

Are there plans to update this to support new metrics in CruX such as cumulative layout shift and largest contentful paint?

CrUX.run need Geo Breakdown

The CrUX dataset has an optional set of tables broken down by country.

CrUX.run could incorporate these into the model (an additional aggregate column).
This would allow us to display histograms and metrics by geo (similar to Edgemesh.
.

Ballpark cost estimates of ~another 300-400GB per month but we can likely reduce this down further (need to add 2 additional arrays into cruxRaw)

We could also then consider adding the Edgemesh style violin charts to show the histograms for the metrics

API: CSV Export for Methods

It would be great if we could expose CSV exports to the API.
This can all be done in the backend.

README Examples respond http404

You may need to update the README Examples and/or point to the documentation itselft and/or add a visible redirect from the versionless routes to the v1 ones.

White text on white background for "Sponsored by"

On every [data-page="api.md"] route, the EdgeMesh link is white on white.

Dotted lines that separate "fast-average-slow" areas are not in place.

I think that the lines that separate the three areas: fast, average and slow are not drawn at the correct points. Attached is an image of the result, and of what it should be, according to the data. As you can see, there are "fast loading" time points within the "average loading time" range; and "average" points in the "slow" range.

Current drawing:

Corrected drawing:

API endpoints don't work

Hello,

is this project still active? API doesn't work.
For example https://api.crux.run/v1/check/?domain=www.google.com I've got Error 1016

API : Ability to flip (transpose) API fields

Currently, the JSON responses are optimal for transmission but not for viewing.
It would be nice to have a flag to return the JSON in a flipped format for easy import into things like excel. Ideally we could abandon the JSON format here altogether and embrace the CSV :)