Skip to content

lintingbin/csv2record

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

24 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

csv2record

A conversion tool for csv to erlang record.

Introduction

Usually a large number of configurations in the game server are stored through the csv file. In order to read these configurations, we will convert csv format into record format, and then write to erl file. The above is the main purpose of the tool.

Csv file format

In order for the tool to work properly, we need to define the correct csv file format. We need to use the first line of the csv file to declare the type of data, and then use the second line to declare the name of the record field. For example:

KN S NI B AS
MapId Name Type IsBool Mons
1 text1 1 True 1a;2;3
2 text2 1 False 4df;53f;6
3 text3 1 True 8;2;3asdf
4 text4 2 False 3;2;3fd;1

In the above csv table, the first line is the type(Can be combined), the second line is the name of the record field. As shown below, all types are single uppercase letters:

K I A N B S O
Key Index Array Number Boolean String Omit

Key, Index and Array can be used with Number, Boolean and String. A csv table has at least one Key and zero or more Index. The above csv table will eventually be converted to the following .hrl file and .erl file:

-ifndef(CSV_MAP_HRL_FILE).
-define(CSV_MAP_HRL_FILE, csv_map_hrl_file).

-record(csv_map, {mapid = 0, name = "", type = 0, isbool = false, mons = ""}).

-endif.
-module(csv_map).

-include("csv_map.hrl").

-export([get/1, get_all_keys/0, get_type_keys/1]).

get(1) -> 
  #csv_map{mapid = 1, name = "text1", type = 1, isbool = true, mons = ["1a", "2", "3"]};
get(2) -> 
  #csv_map{mapid = 2, name = "text2", type = 1, isbool = false, mons = ["4df", "53f", "6"]};
get(3) -> 
  #csv_map{mapid = 3, name = "text3", type = 1, isbool = true, mons = ["8", "2", "3asdf"]};
get(4) -> 
  #csv_map{mapid = 4, name = "text4", type = 2, isbool = false, mons = ["3", "2", "3fd", "1"]};
get(_) -> not_found.

get_all_keys() -> 
  [1, 2, 3, 4].

get_type_keys(1) -> 
  [1, 2, 3];
get_type_keys(2) -> 
  [4];
get_type_keys(_) -> 
 not_found.

Usage

Compile the project, and then execute the following command:

csv2record:generate(File).  
csv2record:generate(File, Option).  
% Types 
%   File = CsvFilePath | CsvDir
%   Option = List = [{Atom, String}| List]
%   Atom = 
%     hrl_dir(.hrl file output directory. Default value: "csv_hrl_dir") |
%     src_dir(.erl file output directory. Default value: "csv_src_dir") |
%     ebin_dir(.beam file output directory. Default value: "ebin") |
%     record_prefix(The prefix of record name. Default: "csv")

Or create a escript executable file:

rebar3 escriptize
_build/default/bin/csv2record $csv_dir $csv_opt  
An example of $csv_opt: "ebin_dir=ebin;hrl_dir=include;src_dir=src;"

Discussing