shithub: mc

Info  •  Files  •  Log  •  Branches

Add more syntaxes for uncode regexes.

--- a/interp.myr

+++ b/interp.myr

@@ -207,9 +207,18 @@

 }

 const die = {re, thr, msg

+        /*

+ 	  we can have die called on a thread

+	  multiple times, eg, if it has a bad

+	  range *and* end in a state that another

+	  thread is in. We should only decrement

+	  the number of threads for that once.

+	 */

 	trace(re, thr, "\t\tdie %z: %s\n", thr.tid, msg)

+        if !thr.dead

+		re.nthr--

+        ;;

 	thr.dead = true

-	re.nthr--

 }

 const finish = {re, thr

--- a/test/data/regex-unicode-expected

+++ b/test/data/regex-unicode-expected

@@ -3,3 +3,9 @@

 Matched Aabæc%!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! via (\p{L}*)bæc\P{L}* : 2

 	match 0: Aabæc%!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

 	match 1: Aa

+Matched Aabæc%!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! via (\pL*)bæc\PL* : 2

+	match 0: Aabæc%!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

+	match 1: Aa

+Matched Aabæc%!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! via (\p{Letter}*)bæc\P{Uppercase_Letter}* : 2

+	match 0: Aabæc%!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

+	match 1: Aa

--- a/test/regex-unicode.myr

+++ b/test/regex-unicode.myr

@@ -6,4 +6,7 @@

 const main = {

 	testmatch(".*bæc", "Abæc")

 	testmatch("(\\p{L}*)bæc\\P{L}*", "Aabæc%!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!")

+        /* test various syntaxen */

+	testmatch("(\\pL*)bæc\\PL*", "Aabæc%!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!")

+	testmatch("(\\p{Letter}*)bæc\\P{Uppercase_Letter}*", "Aabæc%!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!")

 }