Improvement: change of data fetching method

[baron-chiu]

Currently the data is fetched by static URLs, and the results are limited by the api.

taiwan-holiday.js/lib/opendata-tw.js

Line 26 in 5c7801d

OpendataTW.url = 'https://data.ntpc.gov.tw/api/datasets/308DCD75-6434-45BC-A95F-584DA4FED251/json?size=2000';

PR #6 doesn't solve it either. It just increases the limit to 2000.

Proposal:

1. Remove the url in opendata-tw.js. Instead, provide a Get method of the url with offset and limit.
2. Fetch the data multiple times and asynchronously.

The fetch batch maybe determined by count in the response of the first 1000 results.
If count is more then 1000, then creates a batch of requests to fetch the remaining data and merge them with the first 1000 result.

API reference: 臺北市政府行政機關辦公日曆表

[baron-chiu]

It seems my javascript is very limited and I have problem implementing the chaining of all the Promises.

Does it have a better approach?

[baron-chiu]

#10 is open to add the feature. However the date range provided by OpenData is still limited to the latest year, and the cache will expire eventually.

Should the library provide a cache update method for users to trigger manually?