William Rusnack

Shrinking is Insanely Slow for `write_read`

7

Just one test took ~25 minutes due to shrinking. https://github.com/haskell/text/actions/runs/9201285542/job/25309228763?pr=590#step:6:1224 Should `noShrinking` be added?

shrinking takes forever on type [Text] closes #596

Fixed off by one for writeBlocksRaw

8

closes #587

`Data.Text.IO.Utf8.hPutStr` was implemented in #503 and has much better performance than `Data.Text.IO.hPutStr` when the encoding is `"UTF-8"` and the neline is `LF`, so I added it. Question: Is there a...

closes #446 Questions 1. If there is no encoding specified in haCodec, should should hPutBuilderUtf8 be used? 2. Does the write byte buffer actually need to be flushed? 1. Is...

So that hPutBuilder can be used until #586 or similar is pulled.

The test function [`write_read`](https://github.com/haskell/text/blob/f4c2b7ebf3443c0e5ee23a15c6409ac8c83d8b49/tests/Tests/QuickCheckUtils.hs#L243) does not very the encoding. This seems like an important parameter to very to sure that other encodings work properly.

[`writeBlocksRaw`](https://github.com/haskell/text/blob/3f2a71bb1dad5fd8a64f3743b58d765d985723e2/src/Data/Text/IO.hs#L255) uses the same bounds check as [`writeBlocksCRLF`](https://github.com/haskell/text/blob/3f2a71bb1dad5fd8a64f3743b58d765d985723e2/src/Data/Text/IO.hs#L239). `writeBlocksRaw` will always only write one character while `writeBlocksCRLF` can write two. Seems like it should actually be ```haskell | n >=...

Hover Info Cannot be Seen at Page Bottom

2

If you go to the bottom of a source page [example](https://hackage.haskell.org/package/ghc-internal-9.1001.0/docs/src/GHC.Internal.Control.Arrow.html) and hover over some of the annotated text then you will notice that the yellow extra information box is...