1
0
mirror of https://github.com/whippyshou/mastodon synced 2024-12-15 23:28:22 +09:00
whippy-edition/app/javascript/mastodon/locales/locale-data/README.md
Koala Yeung 7ac092513c Fix Occitan's locale data plural function (#2957)
* Fix the pluralRuleFunction of the locale data.
* Added more descriptions about how to write it.
2017-05-10 14:44:48 +02:00

222 lines
6.0 KiB
Markdown

# Custom Locale Data
This folder is used to store custom locale data. These custom locale data are
not yet provided by [Unicode Common Locale Data Repository](http://cldr.unicode.org/development/new-cldr-developers)
and hence not provided in [react-intl/locale-data/*](https://github.com/yahoo/react-intl).
The locale data should support [Locale Data APIs](https://github.com/yahoo/react-intl/wiki/API#locale-data-apis)
of the react-intl library.
It is recommended to start your custom locale data from this sample English
locale data ([*](#plural-rules)):
```javascript
/*eslint eqeqeq: "off"*/
/*eslint no-nested-ternary: "off"*/
export default [
{
locale: "en",
pluralRuleFunction: function(e, a) {
var n = String(e).split("."),
l = !n[1],
o = Number(n[0]) == e,
t = o && n[0].slice(-1),
r = o && n[0].slice(-2);
return a ? 1 == t && 11 != r ? "one" : 2 == t && 12 != r ? "two" : 3 == t && 13 != r ? "few" : "other" : 1 == e && l ? "one" : "other"
},
fields: {
year: {
displayName: "year",
relative: {
0: "this year",
1: "next year",
"-1": "last year"
},
relativeTime: {
future: {
one: "in {0} year",
other: "in {0} years"
},
past: {
one: "{0} year ago",
other: "{0} years ago"
}
}
},
month: {
displayName: "month",
relative: {
0: "this month",
1: "next month",
"-1": "last month"
},
relativeTime: {
future: {
one: "in {0} month",
other: "in {0} months"
},
past: {
one: "{0} month ago",
other: "{0} months ago"
}
}
},
day: {
displayName: "day",
relative: {
0: "today",
1: "tomorrow",
"-1": "yesterday"
},
relativeTime: {
future: {
one: "in {0} day",
other: "in {0} days"
},
past: {
one: "{0} day ago",
other: "{0} days ago"
}
}
},
hour: {
displayName: "hour",
relativeTime: {
future: {
one: "in {0} hour",
other: "in {0} hours"
},
past: {
one: "{0} hour ago",
other: "{0} hours ago"
}
}
},
minute: {
displayName: "minute",
relativeTime: {
future: {
one: "in {0} minute",
other: "in {0} minutes"
},
past: {
one: "{0} minute ago",
other: "{0} minutes ago"
}
}
},
second: {
displayName: "second",
relative: {
0: "now"
},
relativeTime: {
future: {
one: "in {0} second",
other: "in {0} seconds"
},
past: {
one: "{0} second ago",
other: "{0} seconds ago"
}
}
}
}
}
]
```
## Notes
### Plural Rules
The function `pluralRuleFunction()` should return the key to proper string of
a plural form(s). The purpose of the function is to provide key of translate
strings of correct plural form according. The different forms are described in
[CLDR's Plural Rules][cldr-plural-rules],
[cldr-plural-rules]: http://cldr.unicode.org/index/cldr-spec/plural-rules
#### Quick Overview on CLDR Rules
Let's take English as an example.
When you describe a number, you can be either describe it as:
* Cardinals: 1st, 2nd, 3rd ... 11th, 12th ... 21st, 22nd, 23nd ....
* Ordinals: 1, 2, 3 ...
In any of these cases, the nouns will reflect the number with singular or plural
form. For example:
* in 0 days
* in 1 day
* in 2 days
The `pluralRuleFunction` receives 2 parameters:
* `e`: a string representation of the number. Such as, "`1`", "`2`", "`2.1`".
* `a`: `true` if this is "cardinal" type of description. `false` for ordinal and other case.
#### How you should write `pluralRuleFunction`
The first rule to write pluralRuleFunction is never translate the output string
into your language. [Plural Rules][cldr-plural-rules] specified you should use
these as the return values:
* "`zero`"
* "`one`" (singular)
* "`two`" (dual)
* "`few`" (paucal)
* "`many`" (also used for fractions if they have a separate class)
* "`other`" (required—general plural form—also used if the language only has a single form)
Again, we'll use English as the example here.
Let's read the `return` statement in the pluralRuleFunction above:
```javascript
return a ? 1 == t && 11 != r ? "one" : 2 == t && 12 != r ? "two" : 3 == t && 13 != r ? "few" : "other" : 1 == e && l ? "one" : "other"
```
This nested ternary is hard to read. It basically means:
```javascript
// e: the number variable to examine
// a: "true" if cardinals
// l: "true" if the variable e has nothin after decimal mark (e.g. "1.0" would be false)
// o: "true" if the variable e is an integer
// t: the "ones" of the number. e.g. "3" for number "9123"
// r: the "ones" and "tens" of the number. e.g. "23" for number "9123"
if (a == true) {
if (t == 1 && r != 11) {
return "one"; // i.e. 1st, 21st, 101st, 121st ...
} else if (t == 2 && r != 12) {
return "two"; // i.e. 2nd, 22nd, 102nd, 122nd ...
} else if (t == 3 && r != 13) {
return "few"; // i.e. 3rd, 23rd, 103rd, 123rd ...
} else {
return "other"; // i.e. 4th, 11th, 12th, 24th ...
}
} else {
if (e == 1 && l) {
return "one"; // i.e. 1 day
} else {
return "other"; // i.e. 0 days, 2 days, 3 days
}
}
```
If your language, like French, do not have complicated cardinal rules, you may
use the French's version of it:
```javascript
function (e, a) {
return a ? 1 == e ? "one" : "other" : e >= 0 && e < 2 ? "one" : "other";
}
```
If your language, like Chinese, do not have any pluralization rule at all you
may use the Chinese's version of it:
```javascript
function (e, a) {
return "other";
}
```