summaryrefslogtreecommitdiff
path: root/test/src/regex-emacs-tests.el
diff options
context:
space:
mode:
Diffstat (limited to 'test/src/regex-emacs-tests.el')
-rw-r--r--test/src/regex-emacs-tests.el6
1 files changed, 3 insertions, 3 deletions
diff --git a/test/src/regex-emacs-tests.el b/test/src/regex-emacs-tests.el
index 6a661afeff9..f9372e37b11 100644
--- a/test/src/regex-emacs-tests.el
+++ b/test/src/regex-emacs-tests.el
@@ -161,7 +161,7 @@ what failed, if anything; valid values are 'search-failed,
'compilation-failed and nil. I compare the beginning/end of each
group with their expected values. This is done with either
BOUNDS-REF or SUBSTRING-REF; one of those should be non-nil.
-BOUNDS-REF is a sequence \[start-ref0 end-ref0 start-ref1
+BOUNDS-REF is a sequence [start-ref0 end-ref0 start-ref1
end-ref1 ....] while SUBSTRING-REF is the expected substring
obtained by indexing the input string by start/end-ref.
@@ -327,7 +327,7 @@ emacs requires an extra symbol character"
(defun regex-tests-BOOST-frob-escapes (s ispattern)
"Mangle \\ the way it is done in frob_escapes() in
regex-tests-BOOST.c in glibc: \\t, \\n, \\r are interpreted;
-\\\\, \\^, \{, \\|, \} are unescaped for the string (not
+\\\\, \\^, \\{, \\|, \\} are unescaped for the string (not
pattern)"
;; this is all similar to (regex-tests-unextend)
@@ -505,7 +505,7 @@ differences in behavior.")
(cond
;; pattern
- ((save-excursion (re-search-forward "^/\\(.*\\)/\\(.*i?\\)$" nil t))
+ ((save-excursion (re-search-forward "^/\\(.*\\)/\\(.*\\)$" nil t))
(setq icase (string= "i" (match-string 2))
pattern (regex-tests-unextend (match-string 1))))