Skip to content

Latest commit

 

History

History
155 lines (129 loc) · 6.54 KB

Q&A.md

File metadata and controls

155 lines (129 loc) · 6.54 KB
Raw

Q&A

Table of Contents

  1. Why use Translation Text as key?
  2. What is the difference between Variable Interpolation and Interpolation Variable ?
  3. Why doesn't Variable Interpolation support object property parsing?
  4. Is it necessary to separate Date and Time for Interpolation Variable type?
  5. Will rich text be supported?

1. Why use Translation Text as key?

To achieve the following goals

  • Automatically extract texts
  • automatic-translation
  • Automatically generate language pack

With Translation Text as key can we recognize all copywriting that requires translation through the script, so as to achieve the goal of Automatically extract texts . Of course, Automatically extract texts also laid a good foundation for the realization of subsequent goals

Usually, internationalization libraries recommend the following writing style

// Define a simple key
const text1 = t('hello')
// Define a key with module
const text2 = t('module.hello')

Format of corresponding language pack

// en.json
{
  "hello": "hello world",
  "module": {
    "hello": "hello xxx",
  }
}

// zh-CN.json
{
  "hello": "你好世界",
  "module": {
    "hello": "你好xxx",
  }
}

Current writing style of the library

const text1 = t('hello world')
const text2 = t('hello xxx')

Format of corresponding language pack

// zh-CN.json
{
  "hello world": "你好世界",
  "hello xxx": "你好xxx",
}

Compared to traditional writing, Translation Text as key, there are as follows

  • Not friendly to multiple translations of a single word
  • Generates larger language packages

However, it also has the following advantages:

  • Strong readability of source code
  • Translation Text language itself does not need to generate language packs

If you cannot accept the above shortcomings, then other internationalization solutions are more suitable for you; if you can accept the above shortcomings, then I believe i18n-pro will bring you a very good development experience

2. What is the difference between Variable Interpolation and Interpolation Variable ?

// Number Type
t('The number of users has reached {n0}', 100000000)

// Currency Type
t('The selling price is {c0}', 14999)

// Date Type
t(`Today's date is {d0}`, new Date())

// Time Type
t('Current time: {t0}', new Date())

// Plural Type
t('I have {p0 apple}, {p1 banana} and {p2 pear}', 5, 4, 3)

Variable Interpolation：Refers to the realization of t('hello {0}', 'world') → 'hello world' this feature
Interpolation Variable：Refers to the variable inserted into the text
For example, 100000000 , 14999 , and newDate() in the sample code above

  • Insertion position of Interpolation Variable
    • Similar to {0}、{1}、{2}, etc
  • Interpolation Variable type tag
    • Similar to {n0}、{c1}、{t2}、{d3}、{p4{0} apples}, etc

3. Why doesn't Variable Interpolation support object property parsing?

Sample code

// Object attribute resolution
i18n('我叫{name}，今年{age}岁，来自{base}，是一名{job}', {
  name: '王尼玛',
  age: 22,
  base: '火星',
  job: '码农',
})

// Resolution of subscripts in the current library
i18n('我叫{0}，今年{1}岁，来自{2}，是一名{3}',
  '王尼玛',
  '22',
  '火星',
  '码农',
)

The main reason is that the copy contains attribute names, which is not conducive to translation through third-party platforms. The example above is still okay when translating from Chinese to English. However, if translating from English to Chinese or other languages, the attribute names in Interpolation Variable will also be translated, which is the problem

Example of object attribute resolution

// Translation Text as Chinese
const zh = '我叫{name}，今年{age}岁，来自{base}，是一名{job}'

// Translated into English through Baidu-Translation, it seems OK
const zhToEn = `My name is {name}. I'm {age} years old. I'm from {base}. I'm a {job} `

// Then translate the above English into Chinese through Baidu-Translation, we can find that the translation of {job} has problems, and different translation platforms may have different problems
const enToZh = '我的名字是｛name｝。我{age}岁。我来自{base}。我是{工作}'

Let's take a look at the example of subscript parsing

// Translation Text as Chinese
const zh = '我叫{0}，今年{1}岁，来自{2}，是一名{3}'

// Translated into English through Baidu-Translation
const zhToEn = `My name is {0}. I'm {1} years old. I'm from {2}. I'm a {3}`

// Translate the above English into Chinese through Baidu-Translation, and the above parameters will not be mismatched
const enToZh = '我的名字是｛0｝。我是｛1｝岁。我来自｛2｝。我是｛3｝'

Although machine translation cannot achieve 100% accuracy, this method can avoid unnecessary errors as much as possible

4. Is it necessary to separate Date and Time for Interpolation Variable type?

Personally, I don't think it is necessary, but it has been implemented in the design. You can choose to use it flexibly at your discretion. Of course, it is not ruled out that some business scenarios will be more convenient to deal with separately

5. Will rich text be supported?

It will not be supported, because automatic translation is the core function of the library. The basic principle of achieving this function is Translation Text requires ordinary pure texts. Supporting rich texts and existing realizations will be logically conflict

In some scenarios, Variable Interpolation can be used to achieve rich text effects
For example, the text here is hello world , and world needs to be displayed as red and bold on the page
Option 1

t('hello {0}world{1}', '<b style="color:red;">', '</b>')

// The result of executing the t  function is: hello <b style="color:red;">world</b>

Option 2

t('Hello {0}', `<b style="color:red;">${t('world')}</b>`)

// The result of executing the t  function is: hello <b style="color:red;">world</b>

You can choose to use either of the above options based on the actual scenario