compdb is a command line tool to manipulates compilation databases. A compilation database is a database for compile options, it has records of which compile options are used to build the files in a project. An example of compilation database is the JSON Compilation Database
compdb aims to make it easier for you to run tools on your codebase
by spoon-feeding you the right compile options.
compdb is not so much about generating the initial compilation database,
this, is left to other tools, such as cmake and ninja.
It is only a glue between the initial compilation database and your tool(s).
With the proliferation of Clang-based tools, it has become apparent that the compile options are no longer useful uniquely to the compiler.
Standalone tools such as clang-tidy or text editors with libclang integration have to deal with compile options.
Examples of such tools, dealing with compilation databases are: irony-mode, rtags and ycmd.
Based on this evidence, compdb came to life.
A tool that has knowledge of the compile options and can share it
both to inform the text editor and to run clang based tool from the shell.
Install with pip:
pip install compdb
From Github, as user:
pip install --user git+https://github.com/Sarcasm/compdb.git#egg=compdb
Assuming a build directory build/, containing a compile_commands.json,
a new compilation database, containing the header files,
can be generated with:
compdb -p build/ list > compile_commands.json
To run the tests, type:
python -m tests
Or:
tox --skip-missing-interpreters
For regression tests on a few open source projects:
cd tests/regression/headerdb make [all|help]
Contributions are always welcome!
Try to be consistent with the actual code, it will ease the review.
This project is licensed under the MIT License. See LICENSE.txt for details.
- repo: for its ubiquitous command line interface, which served as an inspiration
- scan-build: for the clear Python package design
- git: for the
git-configAPI - julio.meroh.net: for the interesting article serie on CLI design