Addressing inconsistencies in the dynamic rule matrix #544
Comments
There is an ambiguity, however I have many reservations about how to solve this, and even whether this needs "solving" at all. I will elaborate when I get time. |
|
No worries, I realise this isn't a high priority "bug" and could be considered a non-issue/not worth the effort if you assume the user reads/understands the wiki on dynamic filtering. Nonetheless, I think the yellow mixed indicator would clear things up a bit, but maybe I'm missing some corner cases. |
This is a quick demonstration of how inconsistencies caused by conflicting dynamic block/allow rules in the dynamic blocking matrix could be handled. A yellow-ish indicator is added to cells that are affected by more than one conflicting dynamic filtering rule. For example if 3p deny and 3p-scripts or 3p-frames allow/noop are selected, the current way of shading 3p cells red/gray does not accurately reflect the mixed state for those cells. For first-party cells, inline-script and 1p-script are affected. The display of image and all rules hasn't been changed. This is based on the idea outlined in my comment gorhill#527 (comment) refer to that for screenshots and further examples. gorhill#544
|
I often come back to this one to come up with a visual, eventually I will make a decision. I didn't want yellowish because it implies a mix of block and allow rules, which is not the case, it's a block rule partially affecting the cell. |
Following up on #527 (comment), I threw together a quick/very dirty demonstration for gorhill to play with. For now I ignored the
imagetype rule as I don't think it's as important in the context of the matrix.Adding a yellow indicator for cells that have conflicting dynamic rules (e.g. 3p block & 3p-scripts allow) applied to them in the filtering matrix would resolve most inconsistencies and the confusion they cause.
The yellow indicator is already used for hosts that have some requests allowed and some denied in the first column of the matrix, so it would be intuitive to apply that to (at least parts of) the third column.
The commit is here: seanrand@8eefa9f (only tested in Chromium 48.0.2539.0, by the way)
Thoughts?
P.S., sorry for not following contributing.md but I hope the comment and commit is enough to get my point across as to why I think this would be an improvement.
The text was updated successfully, but these errors were encountered: