Skip to main content
Version: Next

changelog

  • change log

    ##2022

    2022.7.26​

    • Added cpp's joint multi-primary key code generation support

    2022.07.19​

    • Added bidx, which is the index file output in binary format. Convenient and optimized on-demand loading

    2022.07.18​

    • Added bson export format
    • Added support for nested containers (currently only supports c# language)

    2022.06.09​

    • Added export support for gdscript language

    2022.03.12​

    • Added data_yaml yaml export format support

    2022.02.16​

    • Support externaltype mapping of vector2, 3, 4 and datetime types

    2022.02.12​

    • Excel, json, xml and other data formats use field names like $type to read polymorphic type names

    2022.02.11​

    • The ref of the container element type also generates the corresponding ref variable and resolve code, currently only supports c#

    2022.02.10​

    • Remove automatic sep for simple native data list types like "list,int"
    • Unify the usage of # and &, and no longer allow confusing usage due to compatibility
    • Adjust externaltype design

    2022.01.29​

    • Allow cross-module inheritance of beans! Bigger Changed Features!

    2022.01.28​

    • Support for defining polymorphic types in excel format

    2022.01.26​

    • root.xml new configuration item option
    • gen_types added code_cs_unity_editor_json
    • enum supports empty cells to take the default value, but it is required to contain items with a value of 0

    2022.01.06​

    • Path verification supports sub-resource file formats, such as Assets/Icons/GoodsIcon.spriteatlas[10000]

    2021​

    2021.12.22​

    • Added python27 support

    2021.12.18​

    • Added Unity asset format support, which can parse the target data from the asset
    • Allows specifying parameters for catalog data sources that will be applied to all child data files
    • Added --output:tables, --output:include_tables, --output:exclude_tables to specify to include or exclude certain tables
    • Added convert template support

    2021.12.09​

    • Add [xxx, xxx] this matching format to identify a multi-column field, which is convenient for csv, a format that does not support cell merging, to identify multi-column fields

    2021.12.08​

    • Added protobuf3 schema generation and corresponding json format export

    2021.12.06​

    • Added msgpack export format support
    • Added flatbuffers definition generation and corresponding json export format support (binary format is not supported temporarily, but a conversion command file is specially generated, you can use flat to convert json format to binary)

    2021.12.05​

    • Added refgroup support, which is convenient for the situation where many fields refer to the same group of tables
    • Adjust sep design, move from column name to type, more flexible and unified

    2021.12.04​

    • Add gen_types type code_template, you can add custom template types
    • Added parameters matching code_template --template:data:dir
    • no longer generate empty template generation result files

    2021.12.02​

    • Added external enum and bean support. Great Features!

    2021.12.01​

    • Add table.mode='list' type, this type of table supports no primary key, combined multiple primary keys, independent multiple primary keys three modes.
    • ref is adjusted accordingly, suitable for the newly added list type table mode

    2021.11.30​

    • Added protobuf definition and data format export. excellent! ! !
    • Added --disable_cache mode in Luban.Server, which disables the generated cache, which is convenient for debugging templates
    • The excel format supports that when the list is expanded horizontally, each element specifies a multi-level header separately, which is convenient for filling in

    2021.11.22​

    • Added validator: set . Checks if a value is in a set.

    2021.11.20​

    • Added "unchecked"Tag to instruct the validator not to check the record
    • remove "no"Tag

    2021.11.18​

    • Add size validator

    2021.11.16​

    • Considering that ##field is easy to write wrong, add ##var and ##var as labels for subfield lines
    • Both lua and yaml support reading data from subfields like json

    2021.11.15​

    • Added datetime alias time
    • Fix the bug that the datetime calculation is wrong due to the adjustment of TimeZone related to .net 6
    • Luban.Server and Luban.ClientServer add --timezone parameter to specify the default time zone
    • The json data source supports reading beans or lists and beans from json subfields (that is, the data of multiple tables can be allocated to a json file, although this is rarely done in practice), and supports using *@xxx.json to convert json as read as a list of records data in the form of list, bean
    • The type polymorphic id field name is renamed from ID to ID, to avoid compilation errors caused by conflicts with common field names

    2021.11.09​

    Framework upgrade to .net 6

    2021.11.05​

    • Use ClosedXml to replace the SpreedSheet library to solve the problem of convert_xlsx failure under mac
    • Added --input_convert_data_dir, use the conversion data convert_xxxx to generate the file instead of the file specified in table.input as the data source
    • In the file generated by convert_xlsx, records are sorted by key

    2021.11.2​

    • Added --generateonly option, generate but not download the result, for simple configuration validity check

    2021.1.29​

    • Relax the format requirements for enum, you can fill in the integer corresponding to the enumeration value

    2021.10.28​

    2021.10.27​

    • Refactor the excel data source format. More clear and elegant

    2021.10.25​

    • Added --naming_convention_xxxx related options, you can specify the naming style for the generated code (camelCase, PascalCase, under_scores, etc.)
    • Optimize the error message when parsing the excel format error, and give the exact error location

    2021.10.22​

    • Add --output_compact_json, generate compact json data
    • Greatly optimized the loading and saving speed of LubanAssistant. (Slow as an ox to fast as lightning)

    2021.10.20​

    • LubanAssistant supports multiline lists

    2021.10.18​

    • The output attribute is added to the table, and the generated file name can be specified

    2021.10.13​

    • Add LubanAssistant plugin

    2021.10.11​

    • Removed --export_test_data and added --export_exclude_tags option, allowing more flexible filtering of export data

    2021.9.1​

    • Added --data_file_extension option, allowing to specify the suffix of the generated data file

    2021.8.28​

    • Added erlang export data format

    2021.8.27​

    • Support data template generation, add gen_types type data_template and corresponding matching option --template_name
    • Add the default option of the excel field, allowing to specify the default address for the blank cell

    2021.8.26​

    • Add attrs attributes to most objects such as enum, bean, table, etc., to facilitate custom processing in code templates
    • Luban.Server's command line argument -t semantics adjusted for additional template search paths. If specified, this path will be searched first, and then the default Templates path will be searched.
    • Support dynamic localization

    2021.8.25​

    • Added the Luban.ClientServer integrated program, which can be generated with one program without deploying Luban.Server.

    2021.8.24​

    • Support code template

    2021.8.13​

    • Added data_json2 format. The difference from data_json is that the map of data_json is exported as [[k1,v1], ...] while the map of data_json2 is exported in the format of { k1:v1, ... }

    2021.8.12​

    • Added yaml data source support

    2021.8.9​

    • Automatically disable the quick edit mode of the console under win

    2021.8.4​

    • Support bean definition in excel
    • Support defining enum type in excel
    • Optimize cfg log, no longer print scary stack information

    2021.8.3​

    • Added --cs:use_unity_vector command line parameter to allow using UnityEngine.Vector{2,3,4} instead of System.Numerics.Vector{2,3,4}

    2021.8.2​

    • Solve the problem of generating different json and lua data files (line breaks) under different platforms, unified to \n
    • Support reading Table list definition from excel. In this way, in most cases, adding a new table can be done only by modifying the excel file.

    2021.7.30​

    • Support reading the definition directly from the header of the excel file, and the definition and data are integrated. Simplifies the hassle of writing record definitions

    2021.7.29​

    • Solve the problem that Luban.Server and Luban.Client run error logs garbled in non-mainland areas

    2021.7.8​

    • Added go language json data loading support (code_go_json)
    • Fix go's support for nullable variables

    2021.6.19​

    • Added the Excel2TextDiff tool item, which is used to convert excel to text or directly convert two excel files to text and then diff
    • excel, lua, xml, json support tag filtering

    2021.6.17​

    • Support multi-branch data merging of main version + branch version, which is extremely useful for releasing multi-regional versions
    • Support static localization

    2021.5.8​

    • Luban.Client supports the watch mechanism, which automatically regenerates after monitoring related definitions or data changes. Save iteration time.

    2021.4.12​

    • Optimize the exported json data file. Unicode characters are exported as raw characters instead of \uxxxx.

    2021.4.8​

    • Added gen_type type data_json_monolithic: generate all json files into one file