From ce0e788ce28c675808a6bc9a34f8db1199cd7cef Mon Sep 17 00:00:00 2001
From: itchyny <itchyny@cybozu.co.jp>
Date: Sat, 2 Mar 2024 00:33:17 +0900
Subject: [PATCH] improve tonumber/0 performance by parsing input as number
 literal

Previously, the tonumber/0 filter parses the input as JSON values, but
this is less-performant on large non-number strings. Parsing the input
string as number literal fixes the performance issue. Also, this fix
changes the filter to reject numbers with white spaces.
---
 src/builtin.c | 21 +++++++++++++++++----
 tests/jq.test |  4 ++--
 2 files changed, 19 insertions(+), 6 deletions(-)

diff --git a/src/builtin.c b/src/builtin.c
index 9aebd1f2d2..393fac0ded 100644
--- a/src/builtin.c
+++ b/src/builtin.c
@@ -43,6 +43,8 @@ void *alloca (size_t);
 #include "locfile.h"
 #include "jv_unicode.h"
 #include "jv_alloc.h"
+#include "jv_dtoa.h"
+#include "jv_dtoa_tsd.h"
 #include "jv_private.h"
 #include "util.h"
 
@@ -464,11 +466,22 @@ static jv f_tonumber(jq_state *jq, jv input) {
     return input;
   }
   if (jv_get_kind(input) == JV_KIND_STRING) {
-    jv parsed = jv_parse(jv_string_value(input));
-    if (!jv_is_valid(parsed) || jv_get_kind(parsed) == JV_KIND_NUMBER) {
-      jv_free(input);
-      return parsed;
+    const char* s = jv_string_value(input);
+#ifdef USE_DECNUM
+    jv number = jv_number_with_literal(s);
+    if (jv_get_kind(number) == JV_KIND_INVALID) {
+      return type_error(input, "cannot be parsed as a number");
+    }
+#else
+    char *end = 0;
+    double d = jvp_strtod(tsd_dtoa_context_get(), s, &end);
+    if (end == 0 || *end != 0) {
+      return type_error(input, "cannot be parsed as a number");
     }
+    jv number = jv_number(d);
+#endif
+    jv_free(input);
+    return number;
   }
   return type_error(input, "cannot be parsed as a number");
 }
diff --git a/tests/jq.test b/tests/jq.test
index b94f29d245..60715f693a 100644
--- a/tests/jq.test
+++ b/tests/jq.test
@@ -2000,8 +2000,8 @@ null
 2
 
 .[] |= try tonumber
-["1", "2a", "3", 4]
-[1, 3, 4]
+["1", "2a", "3", " 4 ", "5.67", ".89", "-876", "+5.43", 21]
+[1, 3, 5.67, 0.89, -876, 5.43, 21]
 
 # Also 1859, but from 2073
 any(keys[]|tostring?;true)