Full Description of the compilation of Scilab

Scilab 5 的编译过程全面转向 Autotools,这是一组简化跨平台项目概念、维护和移植性的工具。文章详细介绍了 autoconf、automake 和 libtool 的作用,以及如何使用它们来添加新的源文件或模块。

摘要生成于 C知道 ,由 DeepSeek-R1 满血版支持, 前往体验 >


  • Immutable Page
  • Comments
  • Info
  • Attachments

Full Description of the compilation of Scilab

Description

Scilab 4 and previous versions compilation process were using an outdated compilation method based on the include of series of Makefiles, I was hard and painful to maintain but also to port the Scilab build process on new platform. Scilab 5 is fully based on the autotools. There are programs which greatly simplify the conception, maintenance and portability of a multi-platform project.

Autotools are composed by a few programs :

  • autoconf (at least version 2.61)

  • automake (at least version 1.10)

  • libtool

Each one is designed for a specific task.

autoconf

From configure.ac (formely configure.in), autoconf creates the configure script which will check if libraries are available on the computer (or not), set a few flags for the compiler... In theory, if the configure is successful, that ensure that all the mandatory dependencies of Scilab 5 are available on the operating system.

automake

From Makefile.am and indirectly from configure.ac, for each Makefile.am declared in the configure.ac (see the macro AC_CONFIG_FILES) , automake will create the file Makefile.in. It specifies which file must be compiled, with which libs... The autogenerated makefiles contain the directive for the compilation for many compilers, operating systems & archs.

libtool

libtool, as its name says, is the tool used to create libraries (static and dynamic) managing the inherent constraints.

Advantages

There are many advantages to this migration:

  • LSB & FHS compatibly works are greatly simplified

  • Simplify the portage to new architectures and compilers
  • Facilitate the work of distribution packagers

Technical description

Important files

File

Description

configure.ac

Transformed by autoconf to create the famous ./configure, this program handles:
* all the --with-xxx & --without-xxx options
* check if headers are available and usable
* check the versions
* Manages compilation flags
* ...

m4/*

Contains all the various macros used to perform dedicated checks

Makefile.am

Defines a few important operations:
* Create the actual Scilab binary
* Build all Scilab macros
* Build the Scilab documentation
* Build the localization
* Launch all the tests

Makefile.incl.am

This file is included in each Makefile.am of a Scilab module.
It contains target for:
* Scilab macros build
* java build
* swig/giws generation
* installation of data files
* Launch tests of a specific module

modules/Makefile.am

This file contains the list of the module to build, the creation of the libscilab library

How to regenerate all the files

The best way is to call autoreconf, this command will call all the other commands (aclocal, autoconf, automake...). Otherwise, when working on the source tree, adding the option --enable-maintainer-mode to the ./configure will automatically rebuild Makefile & configure when any changes occurs

Howto

Add a new source file for compilation

  • Go into the module you want to edit
  • Edit Makefile.am and add the file where it belongs (*_C_SOURCES, *_FORTRAN_SOURCES, GATEWAY_C_SOURCES, GATEWAY_FORTRAN_SOURCES...)

  • Go into the root of the scilab source tree (cd \$SCI)

  • Type automake

This command should rebuild Makefile.in from Makefile.am

Add a new module to Scilab

  • Create your tree in \$SCI/modules/

  • Copy an already existing Makefile.am from an other module

  • Modify it in order to match your changes (sources, libraries dependencies...)
  • Edit \$SCI/modules/Makefile.am

  • Add in the variable SUBDIRS the name of your directory you just created & and add the library to the variable ALL_LIBS

  • In configure.ac, look for AC_CONFIG_FILES and add your future Makefile

  • Go into the root of the Scilab source tree (cd \$SCI)

  • Type autoreconf (make sure that you have the right versions of the autotools).

  • Relaunch the ./configure file in order to create your Makefile from your brand new Makefile.in

public: Full Description of the compilation of Scilab (last edited 2011-03-30 16:18:07 by localhost)

资源下载链接为: https://pan.quark.cn/s/3d8e22c21839 随着 Web UI 框架(如 EasyUI、JqueryUI、Ext、DWZ 等)的不断发展与成熟,系统界面的统一化设计逐渐成为可能,同时代码生成器也能够生成符合统一规范的界面。在这种背景下,“代码生成 + 手工合并”的半智能开发模式正逐渐成为新的开发趋势。通过代码生成器,单表数据模型以及一对多数据模型的增删改查功能可以被直接生成并投入使用,这能够有效节省大约 80% 的开发工作量,从而显著提升开发效率。 JEECG(J2EE Code Generation)是一款基于代码生成器的智能开发平台。它引领了一种全新的开发模式,即从在线编码(Online Coding)到代码生成器生成代码,再到手工合并(Merge)的智能开发流程。该平台能够帮助开发者解决 Java 项目中大约 90% 的重复性工作,让开发者可以将更多的精力集中在业务逻辑的实现上。它不仅能够快速提高开发效率,帮助公司节省大量的人力成本,同时也保持了开发的灵活性。 JEECG 的核心宗旨是:对于简单的功能,可以通过在线编码配置来实现;对于复杂的功能,则利用代码生成器生成代码后,再进行手工合并;对于复杂的流程业务,采用表单自定义的方式进行处理,而业务流程则通过工作流来实现,并且可以扩展出任务接口,供开发者编写具体的业务逻辑。通过这种方式,JEECG 实现了流程任务节点和任务接口的灵活配置,既保证了开发的高效性,又兼顾了项目的灵活性和可扩展性。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值