diff options
author | Dan Davison <dandavison7@gmail.com> | 2020-06-11 17:26:30 -0400 |
---|---|---|
committer | Dan Davison <dandavison7@gmail.com> | 2020-06-11 18:01:59 -0400 |
commit | 30c5fc9d361287765163c283520b152c26acbaaa (patch) | |
tree | 09ef441eaa223cdaab4a99f677e293b0e97cfaf5 | |
parent | 307e2f8e483e139769f75b1b1b60372a31e5ec3c (diff) |
Fix test: force tokenization algorithm to match previous behavior
-rw-r--r-- | src/edits.rs | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/src/edits.rs b/src/edits.rs index 35615dfb..474071ef 100644 --- a/src/edits.rs +++ b/src/edits.rs @@ -86,6 +86,9 @@ fn tokenize(line: &str) -> Vec<&str> { let mut tokens = Vec::new(); let mut offset = 0; for m in TOKENIZATION_REGEXP.find_iter(line) { + if offset == 0 && m.start() > 0 { + tokens.push(""); + } // Align separating text as multiple single-character tokens. for i in offset..m.start() { tokens.push(&line[i..i + 1]); @@ -94,6 +97,9 @@ fn tokenize(line: &str) -> Vec<&str> { offset = m.end(); } if offset < line.len() { + if offset == 0 { + tokens.push(""); + } for i in offset..line.len() { tokens.push(&line[i..i + 1]); } @@ -336,6 +342,7 @@ mod tests { assert_tokenize( " let col = Color::from_str(s).unwrap_or_else(|_| die());", &[ + "", " ", " ", " ", |