diff options
| author | Julian Kulesh <julian.kulesh@kupibilet.ru> | 2017-12-01 00:39:47 +0300 |
|---|---|---|
| committer | Julian Kulesh <julian.kulesh@kupibilet.ru> | 2017-12-01 00:39:47 +0300 |
| commit | 537f2a6e1e9823e22ead30a85fdd3a8438648bc4 (patch) | |
| tree | 8d2c4e071e567835f9e272b42b42b2d764a2fcfc /src/libsyntax/util | |
| parent | 6506f9c2532d05606d97629c20b08e856e880177 (diff) | |
| download | rust-537f2a6e1e9823e22ead30a85fdd3a8438648bc4.tar.gz rust-537f2a6e1e9823e22ead30a85fdd3a8438648bc4.zip | |
move comparator into +find_best_match_name+ function
Diffstat (limited to 'src/libsyntax/util')
| -rw-r--r-- | src/libsyntax/util/lev_distance.rs | 32 |
1 files changed, 27 insertions, 5 deletions
diff --git a/src/libsyntax/util/lev_distance.rs b/src/libsyntax/util/lev_distance.rs index 9307f3c58d4..e429791f2bd 100644 --- a/src/libsyntax/util/lev_distance.rs +++ b/src/libsyntax/util/lev_distance.rs @@ -44,23 +44,45 @@ pub fn lev_distance(a: &str, b: &str) -> usize { /// To find the best match for a given string from an iterator of names /// As a loose rule to avoid the obviously incorrect suggestions, it takes /// an optional limit for the maximum allowable edit distance, which defaults -/// to one-third of the given word +/// to one-third of the given word. +/// Besides Levenshtein, we use case insensitive comparison to improve accuracy on an edge case with +/// a lower(upper)case letters mismatch. pub fn find_best_match_for_name<'a, T>(iter_names: T, lookup: &str, dist: Option<usize>) -> Option<Symbol> where T: Iterator<Item = &'a Symbol> { let max_dist = dist.map_or_else(|| cmp::max(lookup.len(), 3) / 3, |d| d); - iter_names + + let (case_insensitive_match, levenstein_match) = iter_names .filter_map(|&name| { let dist = lev_distance(lookup, &name.as_str()); - if dist <= max_dist { // filter the unwanted cases + if dist <= max_dist { Some((name, dist)) } else { None } }) - .min_by_key(|&(_, val)| val) // extract the tuple containing the minimum edit distance - .map(|(s, _)| s) // and return only the string + // Here we are collecting the next structure: + // (case_insensitive_match, (levenstein_match, levenstein_distance)) + .fold((None, None), |result, (candidate, dist)| { + ( + if candidate.as_str().to_uppercase() == lookup.to_uppercase() { + Some(candidate) + } else { + result.0 + }, + match result.1 { + None => Some((candidate, dist)), + Some((c, d)) => Some(if dist < d { (candidate, dist) } else { (c, d) }) + } + ) + }); + + if let Some(candidate) = case_insensitive_match { + Some(candidate) // exact case insensitive match has a higher priority + } else { + if let Some((candidate, _)) = levenstein_match { Some(candidate) } else { None } + } } #[test] |
