disable utf8 regex check for arbitrary bytes (#136)

Author: nitely

src/regex.nim

@@ -1466,6 +1466,7 @@ when isMainModule:
     block:
       let flags = {regexArbitraryBytes}
       doAssert match("\xff", re2(r"\xff", flags))
+      doAssert match("\xff", re2("\xff", flags))
       doAssert replace("\xff", re2(r"\xff", flags), "abc") == "abc"
       doAssert match("\xff\xff", re2(r"\xff\xff", flags))
       doAssert replace("\xff\xff", re2(r"\xff\xff", flags), "abc") == "abc"

src/regex/compiler.nim

@@ -9,7 +9,7 @@ when defined(regexDotDir):
   import ./dotgraph
 
 func reImpl*(s: string, flags: RegexFlags = {}): Regex {.inline.} =
-  if verifyUtf8(s) != -1:
+  if regexArbitraryBytes notin flags and verifyUtf8(s) != -1:
     raise newException(RegexError, "Invalid utf-8 regex")
   var groups: GroupsCapture
   let rpn = s

tests/tests2.nim

@@ -3073,11 +3073,19 @@ when not defined(js) or NimMajor >= 2:
     check match("a", re2(r"a", flags))
     check(not match("b", re2(r"a", flags)))
     check match("\xff", re2(r"\xff", flags))
+    check match("\xff", re2("\xff", flags))
+    check match("\xf8\xa1\xa1\xa1\xa1", re2(r"\xf8\xa1\xa1\xa1\xa1", flags))
+    check match("\xf8\xa1\xa1\xa1\xa1", re2("\xf8\xa1\xa1\xa1\xa1", flags))
     check replace("\xff", re2(r"\xff", flags), "abc") == "abc"
+    check replace("\xff", re2("\xff", flags), "abc") == "abc"
     check match("\xff\xff", re2(r"\xff\xff", flags))
+    check match("\xff\xff", re2("\xff\xff", flags))
     check replace("\xff\xff", re2(r"\xff\xff", flags), "abc") == "abc"
+    check replace("\xff\xff", re2("\xff\xff", flags), "abc") == "abc"
     check match("\xff\xff", re2(r"\xff+", flags))
+    check match("\xff\xff", re2("\xff+", flags))
     check replace("\xff\xff", re2(r"\xff", flags), "abc") == "abcabc"
+    check replace("\xff\xff", re2("\xff", flags), "abc") == "abcabc"
     check(not match("\xf0", re2(r"\xff", flags)))
     check replace("\xf0", re2(r"\xff", flags), "abc") == "\xf0"
     check match("弢", re2(r"弢", flags))