## Issue
4609: If claim search on latter page fails, don't clear out all previous data
## Note
This requires an accompanying change in `lbry-redux` to mark `claimSearchResultLastPageReached` during timeout and to not purge past results.
## Change
The previous flow was: increment the page, and expect the next render to either be 0 (timeout + purge) or fill in at least 1 new item.
Since it is now possible for no results to be returned for the new page (due to timeout + no purging), the code continues to make another query. This commit stops the perpetual querying.
## Test Case:
1. https://lbry.tv/$/discover?t=imherelbry&content=video&order=top&fresh=year
2. Scroll down a few pages until it timeout happens. Past results are gone.
## Issue
Fixes 2675 `Claim search with exactly 20 results keep fetching next page`
## Changes
Use `claimSearchResultLastPageReached` (I didn't know it existed back then).
The unnecessary "loading" refresh will still happen one time, but at least now it doesn't happen every time we scroll to the bottom (and incorrectly incrementing 'page').
This is a follow-up on PR-4392.
## Issue:
The previous PR covered "pop from within the same page but with different query/order", but missed "pop after entering a claim".
1. Click `Your Tags`.
2. Set order to `New`.
3. Restart app or reload (crucial step)
4. Click `Your Tags`.
5. Click any claim.
6. `Back`. The order incorrectly resets to `Trending` -- it should stay at `New`.
## Fixes:
3071: "Infinite scroll stops working when navigating to file page / back"
## The Issue:
In the POP operation, the `page` value is back to 1 due to the initializer `useState(1)`. If the results cache already contained more than 1 page's worth, then the rest of the logic thinks there's nothing to do.
## The Fix:
Previous fixes to Inf-Scroll added a "page correction" code to handle the mismatch. This fix simply adds this scenario to the list of scenarios to perform the correction.
--- Issues:
(1) #-2669: Page does not restore to top when navigating new tags.
(2) "Encountered children with the same key" error (duplicate entries shown).
https://github.com/lbryio/lbry-desktop/issues/4367#issuecomment-645449206
--- Changes:
(1) Ignore the history if it's a new query (i.e. explicitly clicked). The BACK history will still behave as normal (doesn't reset to top).
(2) Previously, the `page` variable will continue to increment as you scroll and stay within the page (e.g. Trending vs New, or clicked another Tag). As you move between queries, we hit a scenario where `page` is significantly under or over the latest retrieved `claimSearchResult.length`. This messes up the rest of the code.
Fix by correcting the value of `page` according to the current `claimSearchResult.length` when necessary.
--- The bad scenario:
A less popular tag like 'kanji' yields only 23 results today. The code continues to increase the page count. We'll either see some blank page glitches at the bottom, or repeated entries being shown.
--- The fix:
Assume that an unfilled page means "no more results" and stop incrementing the page. This seems true based on empirical evidence.
---The issue:
When switching between tags, the selector defaults back to Trending even though you had another option already selected.
---Changes:
- 'orderParamUser' will store the last user state persistently. The persistent state is also made unique for each page (i.e. Your Tags and All Content will be unique).
- If the parent component passes in a specific order, that will be respected and will also become the new persisted value. One example is "Your Following", where it always starts at 'New'.
- Handled navigation history correctly
The test case:
- Enter "Your Tags" (assume start at 'Trending')
- Click 'New'
- Click 'Top'
- Back
- Back (it should return to 'Trending')
As the top page history does not have any "?order=" value, we ended up with a no-op for the last Back. 'orderParamEntry' is added to handle this.
To improve search performance, assume we want content from the past year when viewing new and trending. Also pass rounded values to top so queries don't re-execute incorrectly.
