fix nonportable regular expression control character ranges

NEWS

@@ -28,4 +28,13 @@ opening fence.
 
 *** Link reference definitions
 
-Link destinations can be enclosed in an opening < and > characters.
+Link destinations can be enclosed in an opening < and > 
+
+** Bug fixes
+
+*** Fix nonportable range expression for control characters
+
+glibc 2.28 made some changes to the collation of Unicode characters
+which affect regular expression ranges. This caused the regular
+expression range for control characters to match charaacters it
+shouldn't.

commonmark/common.scm

@@ -20,6 +20,7 @@
   #:use-module (commonmark entities)
   #:export (ascii-punctuation-characters
             escaped-characters
+            control-characters
             regular-characters
             in-parens-no-space
             link-destination
@@ -34,7 +35,9 @@
 ;; to match ']'
 (define ascii-punctuation-characters "]!\"#$%&'()*+,-./:;<=>?@[\\^_`{|}~")
 (define escaped-characters (string-append "\\\\[" ascii-punctuation-characters "]"))
-(define regular-characters "[^\x01-\x19 ()\\\\]")
+;; can't use range as it is not portable
+(define control-characters "\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19")
+(define regular-characters (string-append "[^" control-characters " ()\\\\]"))
 (define in-parens-no-space (string-append "\\((" regular-characters "|" escaped-characters "|\\\\)*\\)"))
 (define link-destination (string-append "((" regular-characters "+|"
                                         escaped-characters "|"

commonmark/inlines.scm

@@ -36,7 +36,8 @@
 (define re-link-destination (make-regexp link-destination))
 (define re-link-title (make-regexp link-title))
 (define re-link-label (make-regexp link-label))
-(define re-autolink (make-regexp "^<([a-zA-Z][a-zA-Z0-9+.-]{1,31}:[^ \t\n<>\x01-\x19]*)>"))
+(define re-autolink (make-regexp (string-append "^<([a-zA-Z][a-zA-Z0-9+.-]{1,31}:[^ \t\n<>"
+                                                control-characters "]*)>")))
 (define re-email-autolink (make-regexp
                            (string-append "^<([a-zA-Z0-9.!#$%&'*+/=?^_`{|}~-]+@"
                                           "[a-zA-Z0-9]([a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?"