Fix patterns of provide_context gets panicked

jmuk · jmuk · commit 41a02531e110 · 2018-03-12T17:29:57.000-07:00
This PR fixes two cases where provide_context panics even though
the previous call of `is_boundary` returns a PreContext error.

The primary cause is `provide_context` assumes its state as
Regional or Emoji; otherwise it gets panicked with 'invalid state!'
However:
- if the chunk starts with the cursor position, it requires
  pre-context for `cat_before` (and that's necessary for some
  cases indeed). so provide_context should fill `cat_before`
  in such case.
- if both `cat_before` and `cat_after` are RIS and the boundary
  is undecided by the chunk, it requires pre-context but it does
  not set its state to Regional. This is done by setting state
  within `handle_regional`.
diff --git a/src/grapheme.rs b/src/grapheme.rs
@@ -360,7 +360,10 @@ impl GraphemeCursor {
         match self.state {
             GraphemeState::Regional => self.handle_regional(chunk, chunk_start),
             GraphemeState::Emoji => self.handle_emoji(chunk, chunk_start),
-            _ => panic!("invalid state")
+            _ => if self.cat_before.is_none() && self.offset == chunk.len() + chunk_start {
+                let ch = chunk.chars().rev().next().unwrap();
+                self.cat_before = Some(gr::grapheme_category(ch));
+            },
         }
     }
 
@@ -406,6 +409,7 @@ impl GraphemeCursor {
             return;
         }
         self.pre_context_offset = Some(chunk_start);
+        self.state = GraphemeState::Regional;
     }
 
     fn handle_emoji(&mut self, chunk: &str, chunk_start: usize) {
@@ -428,6 +432,7 @@ impl GraphemeCursor {
             return;
         }
         self.pre_context_offset = Some(chunk_start);
+        self.state = GraphemeState::Emoji;
     }
 
     /// Determine whether the current cursor location is a grapheme cluster boundary.
@@ -659,3 +664,21 @@ impl GraphemeCursor {
         }
     }
 }
+
+#[test]
+fn test_grapheme_cursor_ris_precontext() {
+    let s = "\u{1f1fa}\u{1f1f8}\u{1f1fa}\u{1f1f8}\u{1f1fa}\u{1f1f8}";
+    let mut c = GraphemeCursor::new(8, s.len(), true);
+    assert_eq!(c.is_boundary(&s[4..], 4), Err(GraphemeIncomplete::PreContext(4)));
+    c.provide_context(&s[..4], 0);
+    assert_eq!(c.is_boundary(&s[4..], 4), Ok(true));
+}
+
+#[test]
+fn test_grapheme_cursor_chunk_start_require_precontext() {
+    let s = "\r\n";
+    let mut c = GraphemeCursor::new(1, s.len(), true);
+    assert_eq!(c.is_boundary(&s[1..], 1), Err(GraphemeIncomplete::PreContext(1)));
+    c.provide_context(&s[..1], 0);
+    assert_eq!(c.is_boundary(&s[1..], 1), Ok(false));
+}

Original file line number	Diff line number	Diff line change
`@@ -360,7 +360,10 @@ impl GraphemeCursor {`
`360`	`360`	`match self.state {`
`361`	`361`	`GraphemeState::Regional => self.handle_regional(chunk, chunk_start),`
`362`	`362`	`GraphemeState::Emoji => self.handle_emoji(chunk, chunk_start),`
`363`		`- _ => panic!("invalid state")`
	`363`	`+ _ => if self.cat_before.is_none() && self.offset == chunk.len() + chunk_start {`
	`364`	`+ let ch = chunk.chars().rev().next().unwrap();`
	`365`	`+ self.cat_before = Some(gr::grapheme_category(ch));`
	`366`	`+ },`
`364`	`367`	`}`
`365`	`368`	`}`
`366`	`369`
`@@ -406,6 +409,7 @@ impl GraphemeCursor {`
`406`	`409`	`return;`
`407`	`410`	`}`
`408`	`411`	`self.pre_context_offset = Some(chunk_start);`
	`412`	`+ self.state = GraphemeState::Regional;`
`409`	`413`	`}`
`410`	`414`
`411`	`415`	`fn handle_emoji(&mut self, chunk: &str, chunk_start: usize) {`
`@@ -428,6 +432,7 @@ impl GraphemeCursor {`
`428`	`432`	`return;`
`429`	`433`	`}`
`430`	`434`	`self.pre_context_offset = Some(chunk_start);`
	`435`	`+ self.state = GraphemeState::Emoji;`
`431`	`436`	`}`
`432`	`437`
`433`	`438`	`/// Determine whether the current cursor location is a grapheme cluster boundary.`
`@@ -659,3 +664,21 @@ impl GraphemeCursor {`
`659`	`664`	`}`
`660`	`665`	`}`
`661`	`666`	`}`
	`667`	`+`
	`668`	`+#[test]`
	`669`	`+fn test_grapheme_cursor_ris_precontext() {`
	`670`	`+ let s = "\u{1f1fa}\u{1f1f8}\u{1f1fa}\u{1f1f8}\u{1f1fa}\u{1f1f8}";`
	`671`	`+ let mut c = GraphemeCursor::new(8, s.len(), true);`
	`672`	`+ assert_eq!(c.is_boundary(&s[4..], 4), Err(GraphemeIncomplete::PreContext(4)));`
	`673`	`+ c.provide_context(&s[..4], 0);`
	`674`	`+ assert_eq!(c.is_boundary(&s[4..], 4), Ok(true));`
	`675`	`+}`
	`676`	`+`
	`677`	`+#[test]`
	`678`	`+fn test_grapheme_cursor_chunk_start_require_precontext() {`
	`679`	`+ let s = "\r\n";`
	`680`	`+ let mut c = GraphemeCursor::new(1, s.len(), true);`
	`681`	`+ assert_eq!(c.is_boundary(&s[1..], 1), Err(GraphemeIncomplete::PreContext(1)));`
	`682`	`+ c.provide_context(&s[..1], 0);`
	`683`	`+ assert_eq!(c.is_boundary(&s[1..], 1), Ok(false));`
	`684`	`+}`