CC-CEDICT as a service

Demo: https://cedict.herokuapp.com/

CC-CEDICT: http://cc-cedict.org/wiki/start

Technologies used:

• Spring Boot
• Maven
• JPA
• H2
• Heroku

Info:

• This repository contains database file (> 60 MB) for H2 so the application is already populated with CC-CEDICT entries
• The entries were filtered for performance to contain no more than 255 characters; some lengthy English explanations from original CEDICT were skipped for this reason
• The application is integrated with Heroku PaaS - can be deployed with a single command (git push heroku master)
• Because of the embedded database configuration used and the fact that it is a part of repository, the data size is not restricted by platform database limits - can even run on Heroku free dynos despite total rows count over 430.000 (currently free plan allows up to 10.000 rows).
• The application code contains controller used for populating the database (LoaderController), but it is commented out (annotation) for safety
• data is exposed through REST API
• only simplified Chinese is supported
• numeric pinyin is replaced by diacritics (tone markers over the letters)

API:

1) Translate Chinese word specified as Chinese characters only

[base URL]/api/zh/[Chinese characters]

Returns list of all possible Chinese words (pairs of characters + pinyin) matching this characters.

Example

Request:

https://cedict.herokuapp.com/api/zh/好 GET

Response:

[
    {
        "pinyin": "hǎo",
        "text": "好",
        "translations": [
            "good",
            "well",
            "proper",
            "good to",
            "easy to",
            "very",
            "so",
            "(suffix indicating completion or readiness)",
            "(of an unmarried couple) to be close",
            "to be keen on each other"
        ]
    },
    {
        "pinyin": "hào",
        "text": "好",
        "translations": [
            "to be fond of",
            "to have a tendency to",
            "to be prone to"
        ]
    }
]

This method return list of pairs (characters + pinyin) with respective English translations, as in general single written form of Chinese word can be related to multiple pronouncations and each of them has different translations associated.

2) Translate Chinese word specified as Chinese characters and pinyin

[base URL]/api/zh/[Chinese characters]/[pinyin]

Returns Chinese word and associated translations (characters + pinyin) matching this characters and pinyin.

Example

Request:

https://cedict.herokuapp.com/api/zh/好/hào GET

Response:

{
    "pinyin": "hào",
    "text": "好",
    "translations": [
        "to be fond of",
        "to have a tendency to",
        "to be prone to"
    ]
}

This method matches at most one Chinese word with translations as a pair (characters + pinyin) uniquely identifies a word

3) Translate English word

[base URL]/api/en/[English word]

Return English word and Chinese translations associated with it.

Example

Request:

https://cedict.herokuapp.com/api/en/Europe GET

Response:

{
    "text": "Europe",
    "translations": [
        [
            "欧",
            "Ōu"
        ],
        [
            "欧洲",
            "Ōu zhōu"
        ],
        [
            "欧罗巴",
            "Ōu luó bā"
        ],
        [
            "欧罗巴洲",
            "Ōu luó bā Zhōu"
        ]
    ]
}