1.0.0

crates/odonata-base/src/infra/quote.rs

@@ -0,0 +1,120 @@
+/// based loosely on Splitty
+/// https://github.com/Canop/splitty (license MIT)
+
+pub struct QuoteParser<'t, F> {
+    text:         &'t str,
+    strip_quotes: bool,
+    pat:          F,
+}
+
+impl<'t, F> QuoteParser<'t, F>
+where
+    F: FnMut(char) -> bool,
+{
+    pub fn new(text: &'t str, pat: F) -> Self {
+        Self {
+            text,
+            strip_quotes: false,
+            pat,
+        }
+    }
+
+    pub fn without_quotes(mut self) -> Self {
+        self.strip_quotes = true;
+        self
+    }
+}
+
+#[inline(always)]
+fn is_quote(c: char) -> bool {
+    c == '"' || c == '\''
+}
+
+impl<'t, F> Iterator for QuoteParser<'t, F>
+where
+    F: FnMut(char) -> bool,
+{
+    type Item = &'t str;
+
+    fn next(&mut self) -> Option<&'t str> {
+        // we ignore delimitors at the start
+        self.text = self.text.trim_start_matches(&mut self.pat);
+
+        let mut char_indices = self.text.char_indices();
+        let (_, _c) = char_indices.next()?;
+
+        let mut quote_mode = false;
+        for (i, c) in self.text.char_indices() {
+            if is_quote(c) {
+                quote_mode = !quote_mode;
+            }
+            if quote_mode {
+                continue;
+            }
+            if (self.pat)(c) {
+                let token = &self.text[..i];
+                self.text = &self.text[i..];
+                match self.strip_quotes {
+                    true => return Some(token.trim_matches(is_quote)),
+                    false => return Some(token),
+                }
+            }
+        }
+        // we havent found an ending delimiter
+        #[allow(clippy::overly_complex_bool_expr)]
+        let token = match self.strip_quotes {
+            true => self.text.trim_matches(is_quote),
+            false => self.text,
+        };
+        self.text = &self.text[0..0]; // flag ends
+        Some(token)
+    }
+}
+
+fn split_into_tags(text: &str) -> impl Iterator<Item = &str> {
+    QuoteParser::new(text, |c| c == ';')
+}
+
+#[cfg(test)]
+mod tests {
+    use itertools::Itertools as _;
+    use test_log::test;
+
+    use super::*;
+
+    #[test]
+    fn test_quote_parser() {
+        let vec = QuoteParser::new(r#"bm e4"#, char::is_whitespace).collect_vec();
+        assert_eq!(vec, vec!["bm", "e4"]);
+
+        let vec = QuoteParser::new(r#"id "my name is bob""#, char::is_whitespace).collect_vec();
+        assert_eq!(vec, vec!["id", "\"my name is bob\""]);
+
+        let vec = QuoteParser::new(r#"id "my name is bob""#, char::is_whitespace)
+            .without_quotes()
+            .collect_vec();
+        assert_eq!(vec, vec!["id", "my name is bob"]);
+
+        let vec = QuoteParser::new(r#"id 'my name is bob'"#, char::is_whitespace).collect_vec();
+        assert_eq!(vec, vec!["id", "'my name is bob'"]);
+
+        let vec = split_into_tags(r#"cat"meo;w";"mouse";"toad;;;;;;" ;zebra;"#).collect_vec();
+        assert_eq!(vec, vec!["cat\"meo;w\"", "\"mouse\"", "\"toad;;;;;;\" ", "zebra"]);
+
+        let vec = split_into_tags(r#"cat'meo;w';'mouse';'toad;;;;;;' ;zebra;"#).collect_vec();
+        assert_eq!(vec, vec!["cat\'meo;w\'", "\'mouse\'", "\'toad;;;;;;\' ", "zebra"]);
+
+        let vec = split_into_tags(r#";cat;mouse;toad;;;;;;sheep;zebra"#).collect_vec();
+        assert_eq!(vec, vec!["cat", "mouse", "toad", "sheep", "zebra"]);
+
+        // OK, but not desirable (unmatched quote parsing)
+        let vec = split_into_tags(r#";ca't;mouse;"#).collect_vec();
+        assert_eq!(vec, vec!["ca't;mouse;"]);
+
+        let text = r#"id 'PIN.01'; c0 'Pins'; Sq c3 f6;"#;
+        let vec = split_into_tags(text).collect_vec();
+        assert_eq!(vec, vec!["id 'PIN.01'", " c0 'Pins'", " Sq c3 f6"]);
+
+
+    }
+}