mdebray/tvl-depot
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

feat(tvix/eval): context-aware split

Browse source 
Nix does something like:

```cpp
        NixStringContext context;
        const auto str = state.forceString(*args[1], context, pos, "while evaluating the second argument passed to builtins.split");
```

And then do nothing with that context, therefore, we follow them and
make `split` aware of the context but still do nothing with it.

Change-Id: I4fee1936600ce86d99d00893ca3f64013213935b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10428
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: raitobezarius <tvl@lahfa.xyz>
This commit is contained in:
Ryan Lahfa 2023-12-26 00:40:55 +01:00 committed by clbot
parent 556e52c9cb
commit f44ac2a594
1 changed files with 14 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

19
tvix/eval/src/builtins/mod.rs
View file

@ -1060,7 +1060,7 @@ mod pure_builtins {
#[builtin("split")] #[builtin("split")]
async fn builtin_split(co: GenCo, regex: Value, str: Value) -> Result<Value, ErrorKind> { async fn builtin_split(co: GenCo, regex: Value, str: Value) -> Result<Value, ErrorKind> {
let s = str.to_str()?; let s = str.to_contextful_str()?;
let text = s.as_str(); let text = s.as_str();
let re = regex.to_str()?; let re = regex.to_str()?;
let re: Regex = Regex::new(re.as_str()).unwrap(); let re: Regex = Regex::new(re.as_str()).unwrap();
@ -1071,7 +1071,10 @@ mod pure_builtins {
while let Some(thematch) = re.captures_read_at(&mut capture_locations, text, pos) { while let Some(thematch) = re.captures_read_at(&mut capture_locations, text, pos) {
// push the unmatched characters preceding the match // push the unmatched characters preceding the match
ret.push_back(Value::from(&text[pos..thematch.start()])); ret.push_back(Value::from(NixString::new_inherit_context_from(
&s,
&text[pos..thematch.start()],
)));
// Push a list with one element for each capture // Push a list with one element for each capture
// group in the regex, containing the characters // group in the regex, containing the characters
@ -1080,8 +1083,12 @@ mod pure_builtins {
let v: imbl::Vector<Value> = (1..num_captures) let v: imbl::Vector<Value> = (1..num_captures)
.map(|i| capture_locations.get(i)) .map(|i| capture_locations.get(i))
.map(|o| { .map(|o| {
o.map(|(start, end)| Value::from(&text[start..end])) o.map(|(start, end)| {
.unwrap_or(Value::Null) // Here, a surprising thing happens: we silently discard the original
// context. This is as intended, Nix does the same.
Value::from(&text[start..end])
})
.unwrap_or(Value::Null)
}) })
.collect(); .collect();
ret.push_back(Value::List(NixList::from(v))); ret.push_back(Value::List(NixList::from(v)));
@ -1089,6 +1096,8 @@ mod pure_builtins {
} }
// push the unmatched characters following the last match // push the unmatched characters following the last match
// Here, a surprising thing happens: we silently discard the original
// context. This is as intended, Nix does the same.
ret.push_back(Value::from(&text[pos..])); ret.push_back(Value::from(&text[pos..]));
Ok(Value::List(NixList::from(ret))) Ok(Value::List(NixList::from(ret)))
@ -1134,7 +1143,7 @@ mod pure_builtins {
span, span,
) )
.await?; .await?;
Ok(Value::Integer(s.to_str()?.as_str().len() as i64)) Ok(Value::Integer(s.to_contextful_str()?.as_str().len() as i64))
} }
#[builtin("sub")] #[builtin("sub")]