Skip to content
A Java parser to convert xlsx sheets to JSON
Java
Branch: master
Clone or download

Latest commit

Latest commit f2d66f7 Jan 2, 2020

Files

Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
example Fix a bug and update the monsters sheet example Jun 4, 2016
gradle/wrapper upgrade project files Apr 22, 2018
src
testcases 1. Support 'Date' type Sep 29, 2019
.gitattributes :neckbeard: Added .gitattributes & .gitignore files Jun 2, 2016
.gitignore update .gitignore Sep 29, 2019
.gitlab-ci.yml Add gitlab-ci Jun 2, 2016
README.md 1. Support 'Date' type Sep 29, 2019
build.gradle 1. Support 'Date' type Sep 29, 2019
gradlew Initial import Jun 2, 2016
gradlew.bat Initial import Jun 2, 2016
settings.gradle Initial import Jun 2, 2016

README.md

Xlsx2Json

Build Status

A Java parser to convert xlsx sheets to JSON

Supported platforms: Anywhere you can run a Java program

Quick start

  1. Download release or use Gradle to make a build
  2. Run the command line (See Usage section)
  3. The json file will be generated with the same filename

Usage

java -jar xlsx2json-x.x.jar target_name "sheet_name_1 sheet_name_2 ..." [true|false]

Example

java -jar xlsx2json-1.2.jar test.xlsx "monsters maps weapons" true

Arguments

  • The first argument is the Excel filename
  • The second argument is the sheet you want to export to json
  • The third argument indicates whether show the sheet names in generated json or not, will be set to false if omitted

e.g.

ture = {"sheet1":{...},"sheet2":{...} false = [{...},{...}]

Gradle build command

$ gradle clean

$ gradle fatJar

The Jar is created under the $project/build/libs/ folder.

Input file format

The first row should be column type definition (See Supported types section)

The second row should be column name definition

The following rows should be data

Especially, the first column should be Basic type so the parser could index it as the primary key

Example (Excel .xlsx file)

Monsters sheet

Integer String Basic Array<Double> Array<String> Reference Object
id weapon $flag nums words shiled@shield#$id objects
123 shield TRUE 1,2 hello,world 123 a:123,b:"45",c:false
sword FALSE null oh god a:11;b:"22",c:true
monsters sheet

Shields sheet

The type definition line is omitted because all columns are basic types

$id name forSale price
123 COPPER SHIELD TRUE 3600

shields sheet

Result:

{
   "monsters":[
      {
         "weapon":"shiled",
         "objects":{
            "a":123,
            "b":"45",
            "c":false
         },
         "words":[
            "hello",
            "world"
         ],
         "id":123,
         "nums":[
            1,
            2
         ],
         "shiled":{
            "forSale":true,
            "price":3600,
            "name":"COPPER SHIELD",
         }
      },
      {
         "weapon":"sword",
         "objects":{
            "a":11,
            "b":"22",
            "c":true
         },
         "words":[
            "oh god"
         ],
         "id":null,
         "nums":[

         ],
         "shiled":null
      }
   ]
}

Supported types

Basic Types

  • String
  • Integer
  • Float
  • Double
  • Boolean

You can use "Basic" to let the parser automatically detect types

Especially if all columns are Basic types, you can omit the type definition row

Time Type

Support all strings with format HH:mm:ss (directly out) or cell format Time (converted to a calander object and format with simple date format)

Date Type

Support string or numeric with format yyyyMMdd and it will format as yyyy-MM-dd in the json.

DateTIme support would be added in future versions, add an issue if you need it.

Array Types

  • Array<String>
  • Array<Boolean>
  • Array<Double>

The values should be divided using commas ","

You can use the Array<Double> to represent all numeric types like Integer/Float and so on

Object type

  • Object

Use this one to directly construct a JSON object using basic types, child should be divided using commas ","

For more complicated objects, see Reference type

Reference type

  • Reference

Use this type to insert a JSON object from another sheet, the format should be

name_of_this_column@sheet_name#column_name

and the value should be the column value of target.

Use @ to split column name and sheet name, use # to split target sheet name and target column name

Null values

  • Null

If a column is blank, will automatically generate a null value in the JSON file.

Hidden columns

  • $column_name

If a column's name starts with $ sign, then it won't appear in the result json

Especially, if you want to reference to a hidden column, you should also include the $ sign in reference column name

You can’t perform that action at this time.