mirror of
https://github.com/fastfloat/fast_float.git
synced 2026-10-29 02:17:15 +08:00
Update performance link in README
Updated the result description with a new link for performance improvements.
This commit is contained in:
parent
edd182f509
commit
d727092daf
@ -83,7 +83,7 @@ Introduced new optional macros to minimize overhead when certain parsing feature
|
||||
---
|
||||
|
||||
✅ **Result:**
|
||||
Smaller, [faster by default](https://github.com/fastfloat/fast_float/pull/307#issuecomment-3775676197), and more configurable builds — now library can be used for **internal numeric parsers** or **embedded environments** — while maintaining full compatibility, functionality, and performance improvements in the default configuration.
|
||||
Smaller, [faster by default](https://github.com/fastfloat/fast_float/pull/307#issuecomment-4352796857), and more configurable builds — now library can be used for **internal numeric parsers** or **embedded environments** — while maintaining full compatibility, functionality, and performance improvements in the default configuration.
|
||||
|
||||
|
||||
🔥 **Motivation:**
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user