lyuanqi / record_merger Goto Github PK

Overview
--------
Given the 2 sample files: "first.html" and "second.csv" under the "data" directory,
write a Java program that performs the following:

1. In first.html it contains a table called "directory". Read and extract the table.

2. Read and extract the table from second.csv.

3. Combine and merge these 2 tables into 1 by consolidating duplicated columns and write the results in a file named "combined.csv".
   - For the merge operation, assume "ID" is unique.
   - The "ID" column in the output file "combined.csv" should be sorted in ascending order.

When in doubt, state your assumptions and proceed.

Your design must take into consideration of future support of new input file formats (for example xml files).


Note 1: DO NOT MODIFY THE GIVEN DATA FILES!
Note 2: you can include any open-source libraries you need in addition to what's provided in "lib" directory.
Note 3: Use the given RecordMerger class as a starting point.
Note 4: A header row is required in the final CSV file.


Expected Outcome
----------------
- Your program must generate "combined.csv"
- "combined.csv" must contain no duplicate IDs.
- the IDs in "combined.csv" must be sorted in ascending order.


Important!!!
------------
- Your program should be extensible and easy to maintain.
- Additional file types will need to be supported in the future (not yet). Your code must make it easy to
  add new file types.
- The sample data files are just samples. Your program must allow:
    a. any number of input files.
    b. any number of columns in the csv or html files.
    c. the data can be in any language (Chinese, French, English, etc).