NaN floatFormat with clang_cl (#12910)

* clang_cl nan floatFormat * format
nim-lang · Dec 17, 2019 · 7650617 · 7650617
1 parent e8f4586
commit 7650617
Showing 1 changed file with 2 additions and 2 deletions.
diff --git a/lib/system/formatfloat.nim b/lib/system/formatfloat.nim
@@ -44,10 +44,10 @@ proc writeFloatToBuffer*(buf: var array[65, char]; value: BiggestFloat): int =
     result = n + 2
   else:
     result = n
-  # On Windows nice numbers like '1.#INF', '-1.#INF' or '1.#NAN'
+  # On Windows nice numbers like '1.#INF', '-1.#INF' or '1.#NAN' or 'nan(ind)'
   # of '-1.#IND' are produced.
   # We want to get rid of these here:
-  if buf[n-1] in {'n', 'N', 'D', 'd'}:
+  if buf[n-1] in {'n', 'N', 'D', 'd', ')'}:
     writeToBuffer(buf, "nan")
     result = 3
   elif buf[n-1] == 'F':