You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: ChangeLog.md
+4-2Lines changed: 4 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -6,6 +6,8 @@ Changes
6
6
- Default `string_view_type``operator std::basic_string<CharT,Traits,Allocator>() const` made explicit
7
7
to be consistent with `std::string_view`
8
8
9
+
- The virtual method `do_double_value` of `json_input_handler` and `json_output_handler` takes a `number_format` parameter
10
+
9
11
Performance improvements
10
12
11
13
- Faster json dump to string (avoids streams)
@@ -600,7 +602,7 @@ v0.97 Release
600
602
601
603
- Reversion of v0.96 change:
602
604
603
-
The virtual methods `do_float_value`, `do_integer_value`, and `do_unsigned_value` of `json_input_handler` and `json_outputhandler` have been restored to `do_double_value`, `do_longlong_value` and `do_ulonglong_value`, and their typedefed parameter types `float_type`, `integer_type`, and `unsigned_type` have been restored to `double`, `long long`, and `unsigned long long`.
605
+
The virtual methods `do_float_value`, `do_integer_value`, and `do_unsigned_value` of `json_input_handler` and `json_output_handler` have been restored to `do_double_value`, `do_longlong_value` and `do_ulonglong_value`, and their typedefed parameter types `float_type`, `integer_type`, and `unsigned_type` have been restored to `double`, `long long`, and `unsigned long long`.
604
606
605
607
The rationale for this reversion is that the change doesn't really help to make the software more flexible, and that it's better to leave out the typedefs. There will be future enhancements to support greater numeric precision, but these will not affect the current method signatures.
606
608
@@ -622,7 +624,7 @@ Breaking changes:
622
624
623
625
- Only json arrays now support `operator[](size_t)` to loop over values, this is no longer supported for `json` objects. Use a json object iterator instead.
624
626
625
-
- The virtual methods `do_double_value`, `do_integer_value` and `do_uinteger_value` of `json_input_handler` and `json_outputhandler` have been renamed to `do_float_value`, `do_integer_value`, and `do_unsigned_value`,
627
+
- The virtual methods `do_double_value`, `do_integer_value` and `do_uinteger_value` of `json_input_handler` and `json_output_handler` have been renamed to `do_float_value`, `do_integer_value`, and `do_unsigned_value`,
626
628
and their parameters have been changed from `double`, `long long`, and `unsigned long long` to typedefs `float_type`, `integer_type`, and `unsigned_type`.
627
629
The rationale for this change is to allow different configurations for internal number types (reversed in v0.97.)
0 commit comments