luwrap: replace Nim unicode maps with libunicode

Instead of using the built-in (and outdated, and buggy) tables, we now
use libunicode from QJS.  This shaves some bytes off the executable,
though far less than I had imagined it would.

Also, a surprising effect of this change: because libunicode's tables
aren't glitched out, kanji properly gets classified as alpha.  I found
this greatly annoying because `w' in Japanese text would now jump
through whole sentences.  As a band-aid solution I added an extra
Han category, but I wish we had a more robust solution that could
differentiate between *all* scripts.

TODO: I suspect that separately loading the tables for every rune in
breaksViWordCat is rather inefficient.  Using some context object (at
least per operation) would probably be beneficial.

1 files changed, 1 insertions, 12 deletions

diff --git a/src/types/url.nim b/src/types/url.nim
index 54d6f8ed..abb76f56 100644
--- a/src/types/url.nim
+++ b/src/types/url.nim
@@ -245,17 +245,6 @@ func endsInNumber(input: string): bool =
       inc i
   return true
 
-type u32pair {.packed.} = object
-  a: uint32
-  b: uint32
-
-func cmpRange(x: u32pair; y: uint32): int =
-  if x.a > y:
-    return 1
-  elif x.b < y:
-    return -1
-  return 0
-
 type
   IDNATableStatus = enum
     itsValid, itsIgnored, itsMapped, itsDeviation, itsDisallowed
@@ -306,7 +295,7 @@ func processIdna(str: string; beStrict: bool): Option[string] =
     of itsDeviation: mapped &= r
     of itsValid: mapped &= r
   if mapped.len == 0: return
-  mapped.mnormalize()
+  mapped = mapped.normalize()
   var cr: CharRange
   {.cast(noSideEffect).}:
     cr_init(addr cr, nil, passRealloc)