Nidium Tools

Several tools that were specially crafted for the nidium project

Contents:

• Konstructor: tool to manage dependencies and build configurations
• Styleguidor: tool to rate the c/c++ code
• Dokumentor: tool to keep the documentation, examples and code in sync.

Konstructor

Very, and effective way for the developer to keep the dependencies and the build configuration in sync with the gyp settings of the nidium project. This uses the :

• deps.py : module internally to dependenies
• release.py to create a release
• installer/* to create installation package

Show me the code

Every project has a configure python script, that uses this.

PYTHONPATH=~/NativeTools/src/ ./configure --debug --third-party=../../third-party/

Basically you pass --third-party the thirdparty path and PYTHONPATH to the path where konstruktor.py is.

Styleguidor

Simple, but effective way for the developer to check if the code is more or less compliant with the project's styleguide. This is simply a set of regexes, that rate and print each line. Tool is by no way ideal, but better then nothing.

Show me the code

You could write a styleguide rule like this:

    'if-(':         {'pattern': '\bif\(', 
                     #Essential keyword
                     'text': 'Whitespace needed between if and parenthesis', 
                     'weight': 0.25},

And by pointing src/styleguidor.py to the file's directory it could create:

...
src/NativeASCII.h:6 Whitespace must follow a comma
...
The code got a score of 4.5.

Dokumentor

Simple, but effective way to for the developer to keep the documentation and the code in sync without to much hassle.

While programming C/C++ to expose Javascript, you write currently python code inside '/$ comments $/'. These comments will be extracted and transformed with dokumentor.py into markdown, json or just the examples as test cases.

Show me the code

You basically write python code like this:

from dokumentor import *
FunctionDoc( "HTTPRequest.write", "Respond to a client that made a request to the webserver",
    [ SeeDoc( "HTTPListener" ), SeeDoc( "HTTPRequest.write"), SeeDoc( "HTTPRequest.end" ), SeeDoc( "HTTPRequest.writeHead" ) ],
    NO_Examples,
    IS_Dynamic, IS_Public, IS_Fast,
    [ ParamDoc( "data", "The data to send out", "string|ArrayBuffer", NO_Default, IS_Obligated ) ],
    NO_Returns
);

And by running ./src/dokumentor.py markdown the-amazing-raw-docs-are-in-this-directory-or-files it could output:

_HTTPRequest.write__

Respond to a client that made a request to the webserver.

__Public__: true

__Static__: false

__Sees__:

__HTTPListener__
__HTTPRequest.end__
__HTTPRequest.writeHead__

__Parameters__:

data    'string', 'ArrayBuffer'    false    The data to send out.

createModule.py

Tool to assist with boilerplate creation for modules.

Show me the code

./createModule.py --help
usage: createModule.py [-h] [--configuration --CONFIGURATION] [--verbose]
                       [--ignore-build --IGNORE-BUILD] [--force --FORCE]
                       [--force-build --FORCE-BUILD]
                       [--force-download --FORCE-DOWNLOAD] [--path --PATH]
                       [--classname --CLASSNAME] [--name --NAME]

optional arguments:
  -h, --help            show this help message and exit
  --configuration --CONFIGURATION
  --verbose
  --ignore-build --IGNORE-BUILD
  --force --FORCE
  --force-build --FORCE-BUILD
  --force-download --FORCE-DOWNLOAD
  --path --PATH
  --classname --CLASSNAME
  --name --NAME         What's the module name