aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorThomas Voss <mail@thomasvoss.com> 2023-10-20 21:40:42 +0200
committerThomas Voss <mail@thomasvoss.com> 2023-10-20 21:40:42 +0200
commit18576ed3375cc9c18e1e99f3b8762f590708dbca (patch)
tree0a74ab2fd48d99d34fcf7bbcc71f58b0a512f90d
parente367b214765232a1efea49ac8240dc8eecb87fee (diff)
Improve the implementation of peekRune()
-rw-r--r--parser/reader.go29
1 files changed, 12 insertions, 17 deletions
diff --git a/parser/reader.go b/parser/reader.go
index d77d1f5..6b077fc 100644
--- a/parser/reader.go
+++ b/parser/reader.go
@@ -2,6 +2,7 @@ package parser
import (
"bufio"
+ "errors"
"fmt"
"io"
"unicode"
@@ -23,25 +24,19 @@ type reader struct {
pos position
}
-func (reader *reader) peekRune() (r rune, err error) {
- bytes := make([]byte, 0, 4)
+// peakRune returns the next rune in the buffer without actually moving the
+// parser forwards.
+func (reader *reader) peekRune() (rune, error) {
+ bytes, _ := reader.r.Peek(4)
+ r, size := utf8.DecodeRune(bytes)
- // Peeking the next rune is annoying. We want to get the next rune
- // which could be the next 1–4 bytes. Normally we can just call
- // reader.r.Peek(4) but that doesn’t work here as the last rune in a
- // file could be a 1–3 byte rune, so we would fail with an EOF error.
- for i := 4; i > 0; i-- {
- if bytes, err = reader.r.Peek(i); err == io.EOF {
- continue
- } else if err != nil {
- return
- } else {
- r, _ = utf8.DecodeRune(bytes)
- return
- }
+ switch {
+ case r == utf8.RuneError && size == 0:
+ return 0, io.EOF
+ case r == utf8.RuneError && size == 1:
+ return 0, errors.New("Tried to decode malformed UTF-8")
}
-
- return 0, io.EOF
+ return r, nil
}
func (reader *reader) unreadRune() error {