Only show "This data was chached at..." when there's a problem #13
Comments
|
I'd actually just put a warning sign after the title of the back-end and put the description with the cache time in a tooltip. |
Like @m-mohr suggested: "Put a warning sign after the title of the back-end and put the description [...] in a tooltip". The text is a bit more generic though: It doesn't show the crawl time, but just the statement that the data is older than the configured threshold. Likewise, it just states that the backend was recently unavailable for crawling, not how often this was the case or when the last successful crawl was. This commit introduces a new config option `flagWhenOlderThanXHours`.
DataRetrievedNotice's output is now smaller and gray (instead of black) so that it is less prominent. UnsuccessfulCrawlNotice now has a clearer text and doesn't nudge the previous element anymore.
I like the idea to have a warning sign after the title. I implemented that:
It shows when
and/or
But I didn't delete the DataRetrievedNotice in general -- I think it's fair to tell users that the data they see is not live but cached. Since the backends are now collapsed by default, the text isn't that prominent anymore anyway. And on top of that I made it smaller and gray so that it is even less prominent. @m-mohr What do you think? |
|
This is much better than before. The only thing that bugs me is that we now have three places with very similar information. I'd merge the red and grey texts and only show it in case there is a problem. I'd expect from portals/search engines that data is not live but cached... (it's the same in Google and most other portals, too). |
The DataRetrievedNotice component was re-dedicated: So far it was *always* shown as an informative text, now it is displayed as a warning and only shown when the data is old. Thus the restyling from "small, italic and grey" to "normal font size and style but red color". In order to group similar information together, I moved all occurrences of `DataRetrievedNotice` next to the related `UnsuccessfulCrawlNotice`s. And then I moved the two notices "beyond the break", i.e. they are only visible once the backend/collection/process has been expanded. This is still eye-catching enough because their presence is indicated by a warning sign next to the backend name, which is always visible.
Oookay 😉 Done:
The warning texts are always "behind the fold" i.e. only shown when the backend/collection/process is expanded (which is still obvious enough as their presence is indicated by the warning sign next to the backend name). Whether they are shown is determined by the |
|
Better, but why don't we put it into a single paragraph? |
|
Because they are two different things that can happen independent of each other. |
|
The first issue implies the other, so a single message should be enough. It could be expanded though, for example:
|
|
I opened a new issue because "as openEO Hub could not [rough reason]" requires that the reason is known to the Hub itself. |
Currently, the "This data was cached at..." information is output quite prominently:
And that although it's usually not that interesting.
We could remove it from the output for the standard case (the user will assume the data is up-to-date anyway) and only show it when e.g. the data is older than 12 hours or the last crawl failed. And then we should display it with a warning sign like⚠️
The text was updated successfully, but these errors were encountered: