Permalink
Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
431 lines (249 sloc) 11.5 KB

biicode.conf: configure your biicode projects

biicode.conf is a configuration file to --wait for it-- configure your biicode projects.

Place it into your block, next to your source code:

|-- my_project/
|    +-- bii/
|    +-- bin/
|    +-- blocks
|    |    +-- myuser
|    |    |     +-- my_block
|    |    |     |     |-- main.cpp
|    |    |     |     |-- biicode.conf

biicode.conf has 9 different sections to configure your project.

Here's a biicode.conf example:

# Biicode configuration file
[requirements]
        # Blocks and versions this block depends on e.g.
        # user/depblock1: 3
        # user2/depblock2(track) @tag

[parent]
    # The parent version of this block. Must match folder name. E.g.
    # user/block  # No version number means not published yet
    # You can change it to publish to a different track, and change version, e.g.
    # user/block(track): 7

[paths]
    # Local directories to look for headers (within block)
    # /
    # include

[dependencies]
    # Manual adjust file implicit dependencies, add (+), remove (-), or overwrite (=)
    # hello.h + hello_imp.cpp hello_imp2.cpp
    # *.h + *.cpp

[mains]
    # Manual adjust of files that define an executable
    # !main.cpp  # Do not build executable from this file
    # main2.cpp # Build it (it doesnt have a main() function, but maybe it includes it)

[tests]
    # Manual adjust of files that define a CTest test
    # test/* pattern to evaluate this test/ folder sources like tests

[hooks]
    # These are defined equal to [dependencies],files names matching bii*stage*hook.py
    # will be launched as python scripts at stage = {post_process, clean}
    # CMakeLists.txt + bii/my_post_process1_hook.py bii_clean_hook.py

[includes]
    # Mapping of include patterns to external blocks
    # hello*.h: user3/depblock  # includes will be processed as user3/depblock/hello*.h

[data]
    # Manually define data files dependencies, that will be copied to bin for execution
    # By default they are copied to bin/user/block/... which should be taken into account
    # when loading from disk such data
    # image.cpp + image.jpg  # code should write open("user/block/image.jpg")

[requirements]

[requirements] section is fullfiled after executing bii find with the blocks and versions your block depends on.

You can manually specify the block to depend on with its corresponding version or override a dependency just writing the version you want and executing bii build after that.

[requirements]
    # required blocks (with version)
    erincatto/box2d: 10

Take a look at the :ref:`docs about dependencies <cpp_dependencies>` to know more.

[parent]

[parent] section tells you "who is your parent version". Indicates the version of the remote block being edited and looks like this:

[parent]
     myuser/myblock: 0

This section is fullfilled automatically when publishing or opening a block and comes in handy while :ref:`publishing <cpp_publishing>` take a look at it.

When publishing a new block this section should be blank or referenced as -1 version:

[parent]
    # Comments like this are ignored

or

[parent]
    myuser/my_new_block: -1

[paths]

Use [paths] section to tell biicode in which folders it has to look for the local files specified in your #includes. You only need to specify this when your project has non-file-relative #include (s).

Common use case example

Libraries usually have a folder structure like this one

|-- library
|    +-- include
|    |    |-- tool.h
|    +-- test
|    |    |-- main1.cpp (#include "tool.h")

In which main1.cpp includes: #include "tool.h" that it is truly located into /include folder. The proper #include would be #include "../include/tool.h"

If we execute bii deps on this example, we'll see #include "tool.h" as unresolved. Why is this happening? Biicode can't find the tool.h file unless we specify where they can find it.

Let's fix this write into the [paths] section:

[paths]
        # Local directories to look for headers (within block)
        /include

Root directory example

Let's imagine now that we have a folder with the following structure into it

|-- mylib.h
|-- mylib.cpp
|    +-- examples
|    |    |-- main.cpp (#include "mylib.h")

If we execute bii deps on this example, we'll see mylib.h as unresolved. Why is this happening? Biicode, considers the #include(s) relative to their location. So if there isn't a root folder they can refer to, when looking for mylib.h they will search it in the examples folder and they won't be able to find it.

What should we write on the [paths] section?

[paths]
        # Local directories to look for headers (within block)
        /

Write / in [paths] section and biicode will know that it has to include the root directory on its search.

[dependencies]

Biicode knows how the source code files connect to each other. It parses the source code files and deduces some things. But sometimes, this mechanism can detect non existent dependencies or can fail detecting existent dependencies.

Use [dependencies] section to manually define rules to adjust file implicit dependencies.

[dependencies] rules match the following pattern:

#dependent_file_name [operator] NULL|[[!]dependency_file ]

The Operator establishes the meaning of each rule:

  • - operator to delete all specified dependencies from their dependent file.
  • + operator to add all specified dependencies to their dependent file.
  • = operator to overwrite all specified dependencies with existing dependencies.

You can declare that a file has no dependencies using the NULL keyword.

Mark a dependency with a ! symbol to declare a dependency, but excude it from the building process. This is sometimes used to define license files that must be downloaded along with your code, but shouldn't be included in the compilation process.

The dependent_file_name may be defined using Unix filename pattern matching.

Pattern Meaning
* Matches everything
? Matches a single character
[seq] Matches any character in seq
[!seq] Matches any character not in seq

Examples

Let's see a few examples:

  • matrix32.h is dependency of the main.cpp file.
[dependencies]
    main.cpp + matrix32.h
  • Delete matrix16.h dependency to main.cpp.
[dependencies]
    main.cpp - matrix16.h
  • test.cpp depends on both example.h and LICENSE. And LICENSE will be excluded from the compilation process.
[dependencies]
    test.cpp + example.h !LICENSE
  • All files with .cpp extension depend on the README file, but this dependency won't be compiled.
[dependencies]
     *.cpp + !README
  • example.h = NULL tells biicode that example.h has no dependencies (even if it truly has).
[dependencies]
 example.h = NULL
  • Both solver.h and type.h are the only dependencies of calculator.cpp, overwriting any existing implicit dependencies.
[dependencies]
    calculator.cpp = solver.h type.h

[mains]

Use [mains] section to define entry points in your code.

Biicode automatically detects entry points to your programs by examining which files contain a main function definition. But when that's not enough you can explicitly tell biicode where are your entry points.

[mains] has the following structure:

[[!]file ]

An example:

  • Write the name of the file you want to be the entry point.
  • Exclude an entry point writing an exclamation mark, ! before the name of the file.
[mains]
        funct.cpp
        !no_main.cpp

[tests]

Tests section is useful to define specific tests for your code. Adjust files manually that define a CTest test.

Indicate the patter to your test files:

[tests]
    test/*
    test/simple_test.cpp

Those test are excluded from the normal building and are built and executed only when doing :ref:`bii test<bii_test_command>`.

[hooks]

Use [hooks] section to link to certain python scripts that will be executed, for example, before building your project. They can be used to download and install a package needed.

This scripts have ".py" extension and name matches:

  • bii*post_process*hook.py: For scripts that will be launched before project building (bii build or bii configure)
  • bii*clean*hook.py: For scripts that will be launched before a bii clean command.

These are defined like :ref:`[dependencies] <dependencies_conf>`.

In the following example we define that CMakeLists.txt depends on two hooks:

[hooks]
    CMakeLists.txt + bii/my_post_process1_hook.py bii_clean_hook.pyw

Use bii variable inside hook scripts to:

  • Print text:
bii.out.debug("error_msg")
bii.out.info("error_msg")
bii.out.warn("error_msg")
bii.out.error("error_msg")
  • Download files:
bii.download(url, tmp_path)
  • Read your project settings:
bii.settings.cpp.cross_build

Check an example in this block: Maria/bitscope.

[includes]

Enables mapping include patterns to external blocks.

  • For example you can tell biicode: Whenever you read uv.h in my code, it means lasote/libuv/include/uv.h:
[requirements]
    lasote/libuv(v1.0): 0

[includes]
    uv.h: lasote/libuv/include

In the previous example, the [requirements] section has a line specifying a dependency to lasote/libuv(v1.0): 0 version, so, lasote/libuv #includes will be matched against these block.

  • You can also specify complex patterns. To process hello*.h #includes as user3/depblock/hello*.h
[includes]
    hello*.h: user3/depblock

This is pretty useful when using already existing libraries and you don't want to change all the includes.

[data]

Use [data] to specify a link with any file (.h, .cpp, ...) with any data (.txt, .jpg, ...) in your block.

Once [data] section is specified and the code is built (bii build), the data files will be saved, by default, in your project/bin/user/block folder.

Example:

You have in your main code this line:

main.cpp

CImg<unsigned char> image("phil/cimg_example/lena.jpg")

Then, add to your configuration file:

[data]
    main.cpp + lena.jpg

This will copy lena.jpg to project/bin/user/block/ when main.cpp is built.

Any doubts? Do not hesitate to contact us visit our forum and feel free to ask any questions.