sheaf

IMO it would also be nice to adjust the `gi-pangocairo` package to have a nice monadic API like `gi-cairo-render`. I haven't looked closely, but it mostly seems like a case...

In that case perhaps it would be nice to have a `gi-pangocairo-render` package? I still haven't taken a close enough look at `gi-pangocairo` to have an opinion though...

> @sheaf, would you be able to help review #9578? I'm using `canonicalizePath` there for normalizing project imports. I haven't looked in depth, but `normalize` is sufficient for the purposes...

I have compiled [`clc-stackage`](https://github.com/Bodigrim/clc-stackage) using this branch. It showed up a single failure which is unrelated to this patch, which I filed separately as #9777 . I now consider this...

> It might make sense to split it into multiple reviewers by chunks, since it's an awfully big gulp for a single reviewer. I understand why you say that, but...

> This would be similar to the design in `hackage-security`. @andreabedini Thanks for the pointer, I wasn't aware that `hackage-security` [vendored its own `Path` abstraction](https://github.com/haskell/hackage-security/blob/f18ad437bcdf16e7b40e0e1785948cfcda42629c/hackage-security/src/Hackage/Security/TUF/Paths.hs). As for your comment about...

To reiterate, the concept that seems most useful, at least for the purposes of this working directory change, is the notion of a filepath which, if it is relative, is...

I have updated the SymbolicPath API thusly: - we no longer distinguish different kinds of files (in practice it didn't seem to catch any bugs, and made some things a...

I have rebased this MR, updated the API to make use of recent refactors (such as the refactor to file globbing in e2019f5a90213b8e8805f0e06fbe06227ebc5614), updated the documentation, and added a changelog...

Note that this includes PR #9518, which should be merged before this PR. [Edit by Mikolaj: that PR has already been merged.]