It's obvious the functionality list of Libreoffice is becomming overwhelming.
I wouldn't be surprised if maintaining the functionality list ended before the days of OOo 1.1.3.
There was functionality broken in 1.1.4 that neither QA nor the developers knew existed in 1.1.2.
(I'm not talking only about things like the functionality in OOo 1.1.3-ZA that was not present in any other version of OOo 1.1.x. )
Now wondering if that game has been removed from LibO.
I've forgotten how it was accessed. 
Regardless of the outcome of that discussion, *there is a potential HUGE conflict of interest looming on the horizon in the form of creating bugs so that outside companies can charge $$$ to fix them. *
As a potential issue, it exists only as long as organizations and individuals are mislead about FLOSS.
The biggest misconception being that FLOSS is gratis.
QUESTION:
Does there exist a set of LibreOffice TEST documents that are designed to test ALL OF THE FUNCTIONALITY of a given version of LibreOffice?
No.
Specifically, one that can automate test, perhaps through macros or self contained programs, all the functionality of Libreoffice?
Theoretically that can be automated, but due to a11y issues with LibO, it is not currently possible to do so.
SUGGESTION #3: If such a set of documents or test harness does NOT EXIST.
Perhaps the QA team could itemize all the functionality of a version of Libreoffice and structure a project for contribution by end users to contribute the tests.
i.e: A test suite that each user could download and then run and/or contribute more.
In theory, the documents that accompany bug reports are:
* Used to verify that bug fixes do fix the reported issue;
* Construct test documents to be incorporated into QA;
SUGGESTION #4: If a test harness is too complicated,
Assuming the a11y issues are solved (which is required, if LibO is to be deployed in corporate environment in a legal jurisdiction that takes a11y legislation as something more significant than a bone for voters), the stumbling block is the number of platforms each item of functionality can be tested on.
Even basic testing requires 91 platforms. (12 for Windows, 5 for Mac OS X, 10 for BSD distros, 2 for OpenSolaris, 10 each for Ubuntu, Debian, Red Hat, Mint, Arch, Puppy, and one each for Kali,TAILS & DoD.)
If "exotic" platforms are included, then the number of testing platforms increases by at least two orders of magnitude, and maybe even three.
>could the QA team make an itemized list of all the functionality of LibreOffice, inherited and otherwise
Methinks that LibO is like WangWriter during its prime.
C/Support and Sales: "No, you can not construct spreadsheets using this product".
Experienced users: "Yes you can. Here is how to do it."
Wang developers: "That won't work."
Wang tech support: "It does work, regardless of what your specifications claim".
*and create some sort of method to crowd source the testing in a manner designed to capture any lost functionality*?
That is what QA _currently_ does, albeit without the list of all of LibO's current functionality.
It might be a bit too late to compile a complete list of functionality.
What is possible, is to construct a list, based upon:
* Reported regression bugs;
* Deliberately added new functionality;
* Known, existing functionality;
* New functionality added by accident, serendipity, or unintentionally, and discovered by a user;
i.e.: a project website for each version LibreOffice with the 100,000 things that LibreOffice can do where the users can test each one and report the test?
Depending upon how "things" is defined, that figure could be closer to 10,000,000,000 than 100,000.
jonathon
![[TEST] Document Foundation Community](https://communitytest.documentfoundation.org/images/discourse-logo-sketch.png){kind=logo}
