Align the NUM2_ bitmask defines to make them easier to read.

This commit is contained in:
Reece H. Dunn
2020-04-12 10:53:04 +01:00
parent 62e8efad66
commit e8f774326d
+16 -17
View File
@@ -475,23 +475,22 @@ typedef struct {
int numbers;
#define NUM2_THOUSANDS_VAR1 0x40
#define NUM2_THOUSANDS_VAR2 0x80
#define NUM2_THOUSANDS_VAR3 0xc0
#define NUM2_THOUSANDS_VAR4 0x100
#define NUM2_THOUSANDS_VAR5 0x140
#define NUM2_SWAP_THOUSANDS 0x200
#define NUM2_ORDINAL_NO_AND 0x800
#define NUM2_MULTIPLE_ORDINAL 0x1000
#define NUM2_NO_TEEN_ORDINALS 0x2000
#define NUM2_MYRIADS 0x4000
#define NUM2_ENGLISH_NUMERALS 0x8000
#define NUM2_PERCENT_BEFORE 0x10000
#define NUM2_OMIT_1_HUNDRED_ONLY 0x20000
#define NUM2_ORDINAL_AND_THOUSANDS 0x40000
#define NUM2_ORDINAL_DROP_VOWEL 0x80000 // currently only for tens and units
#define NUM2_ZERO_TENS 0x100000
#define NUM2_THOUSANDS_VAR1 0x00000040
#define NUM2_THOUSANDS_VAR2 0x00000080
#define NUM2_THOUSANDS_VAR3 0x000000c0
#define NUM2_THOUSANDS_VAR4 0x00000100
#define NUM2_THOUSANDS_VAR5 0x00000140
#define NUM2_SWAP_THOUSANDS 0x00000200
#define NUM2_ORDINAL_NO_AND 0x00000800
#define NUM2_MULTIPLE_ORDINAL 0x00001000
#define NUM2_NO_TEEN_ORDINALS 0x00002000
#define NUM2_MYRIADS 0x00004000
#define NUM2_ENGLISH_NUMERALS 0x00008000
#define NUM2_PERCENT_BEFORE 0x00010000
#define NUM2_OMIT_1_HUNDRED_ONLY 0x00020000
#define NUM2_ORDINAL_AND_THOUSANDS 0x00040000
#define NUM2_ORDINAL_DROP_VOWEL 0x00080000 // currently only for tens and units
#define NUM2_ZERO_TENS 0x00100000
// bits 1-4 use variant form of numbers before thousands,millions,etc.
// bits 6-8 use different forms of thousand, million, etc (M MA MB)
// bit9=(LANG=rw) say "thousand" and "million" before its number, not after