Repository navigation
docs(authz-playground): show the conditions fix and the boundary-hour fix - #18
Conversation
… fix The playground now leads with a timing chart and has a fifth engine, the Zanzibar model with an OpenFGA condition. Swap the old map and projection screenshots for the chart and the two ReBAC graphs, describe the conditional model and what it costs, and say how the boundary-hour gap was closed.
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars.
 |
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
|
Findings on
|
The raw main-branch URLs 404 until merge, which failed the link check. Relative paths are what devto-cli rewrites on publish, and the ?v= hashes make the CDN fetch the new images.
|
On the two points outside the diff: "Eight of 72 requests split" is still what the page reports. The on-page map samples 08, 09, 17 and 18 (3 users x 3 actions x 2 documents x 4 hours = 72); the 8-hour enumeration is the test harness, not the map. "All fourteen tests passed" describes the suite at the time of the bug, which is the point of that paragraph. On four vs five: the fifth row is introduced as the same Zanzibar model with a condition, and "the other three" in that paragraph refers to Cedar, Rego and Casbin, which are the three it contrasts with. The title stays, since the article is about the four engines and the fifth is that fix. |
|
Relative image paths and the cache-bust query strings match the README asset rules. Two factual-consistency problems remain in
I did not check the new images' contents. There is no JA version of this article. |
|
|
||
|  | ||
|
|
||
| Add a document and Casbin needs its policy table regenerated, while ReBAC just gains a tuple and the model never moves. |
There was a problem hiding this comment.
This sentence compared the four engines' policy shapes and leaned on the proj.png side-by-side image ("The same rules side by side…"). Both the lead-in and the image are gone. It now sits under the graphs image, which shows only the classic and conditional ReBAC models at 08:00, so the Casbin remark has nothing to point at. Either restore a comparison or fold the sentence into the surrounding prose.
|
Content findings on articles/authz-playground.md, all from the four-to-five-engine change:
No EN/JA parity issue, since there is no JA file in the tree. The relative image paths with |
The playground was redesigned and gained a fifth engine, the Zanzibar model with an OpenFGA condition (0-draft/authz-playground#29, #30). This replaces the two screenshots of the old page with the new timing chart and the classic and conditional ReBAC graphs, adds a short passage on the conditional model and its cost, and says how the boundary-hour gap described under What broke was closed.