| Age | Commit message (Collapse) | Author |
|---|
|
Pos objects are somewhat wasteful as they duplicate the origin file name and
input type for each object. on files that produce more than one Pos when parsed
this a sizeable waste of memory (one pointer per Pos). the same goes for
ptr<Pos> on 64 bit machines: parsing enough source to require 8 bytes to locate
a position would need at least 8GB of input and 64GB of expression memory. it's
not likely that we'll hit that any time soon, so we can use a uint32_t index to
locate positions instead.
|
|
Given flake:
```nix
{ description = "nix json error provenance";
inputs = {};
outputs = { self }: {
jsonFunction = _: "function";
json = builtins.toJSON (_: "function");
};
}
```
- Before:
```console
❯ nix eval --json .#jsonFunction
error: cannot convert a function to JSON
```
- After:
```console
❯ nix eval --json .#jsonFunction
error: cannot convert a function to JSON
at /nix/store/b7imf1c2j4jnkg3ys7fsfbj02s5j0i4f-source/testflake/flake.nix:4:5:
3| outputs = { self }: {
4| jsonFunction = _: "function";
| ^
5| json = builtins.toJSON (_: "function");
```
|
|
Also, factor out JSON generation from value-to-json.{cc,hh}, and
support producing indented JSON.
|
|
|
|
|
|
|
|
|
