on SDL, bugless code and testing

I would argue that 100% bugless code would be possible after several iterations of processors and run-times designed by software written by 99,999 bugless code, which could start an endless fight about future ... visions , AI et. , however , let get back to hands-on level :
- if the software producer defines a feature of his product as a set of functionalities, each of which could be mapped 1 on 1 with an automated test , should once all the tests be passed than the feature could be considered as bugless.

The bigger the software project, the greater the integration between the persons working on the artifacts for software requirements, documentation , development and testing.

I mean really persons should pay attention on the wording, software buyers should careful on all the wording ... by simply asking proof it !:

cd /tmp wget --no-check-certificate https://bash-stub.googlecode.com/files/bash-stub.zip

unzip -o bash-stub.zip -d .

sh /tmp/bash-stub/test-bash-stub.sh

I began to realize that each issue any of the tools, each defined functionality should be presented in a self-sustained manner, so that it should be mapped 1 on 1 on a test proving its validity. Being in the role of software developer it means in the beginning more work, but later on greater velocity and overall productivity improvement.