-
-
Notifications
You must be signed in to change notification settings - Fork 165
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Develop multi answer fix #767
Conversation
…plification and clean up.
and comparison between old and new ways of logging past answers
Counting the number of scores gives the number of answer groups (or answer evaluators) but not the number of answers.
Also changed the calling hierarchal structure of WeBWorK::ContentGenerator::Problem. The previous one seemed devious.
…eate_ans_str_from_responses in GatewayQuiz This is a first pass. The problems are gone through twice once to record sticky answers and a second time to record passed answers. These could probably be combined but at the moment the order of presentation is different.
This is companion pull request to pg PR #302. It addresses bugs that occur in multiAnswer and matrix answer evaluators. Many of the bugs arose because of the change to using the answer evaluator as the primary the primary object for a given answer —rather than the answer blank. (This helps regularize the increasing number of answer evaluators that use more than one answer blank — e.g. matrices and multiAnswer. At the moment I don’t know of any answer evaluators that share an answer blank — if that comes up there are mechanisms to deal with it on a case by case basis — but no organized protocol.) |
bug #3003 regarding preserving matrix answers beyond the first answer is fixed. You can test on bug #3812 is fixed by this PR and the companion #302 for PG. bug #3822 is fixed by this PR |
I believe this PR is ready for extensive checking. It should be checked on questions that call multiAnswer() and check matrices. I have attached the set definition file containing the questions I used for checking for both multiAnswer.def and multiAnsQuiz. You will need to change the extension from .txt to .def before uploading it to your webwork course. |
What is the time line for a next release? I could do extensive checking, but it might creep into July before I would be finished. |
This eliminates error messages at the end of quizzes. I'm not sure if this just hides other errors or not, however initializing $answerScore=0 and $studentAnswer='' seems safe.
I’m hoping to get a release ww2.13 out towards the end of June. I would leave out some of the current pull requests — in particular
the ones involving localization or making the ww3 interface ready for prime time. Those two will take a long time to check.
I would pull multiAnswer fix — it will also take a little time to check — but it fixes a number of bugs that have been outstanding
for two or three years. (If you have any time to do checking that is what I would recommend.)
I’m going to put together a formal plan for which PR to include in ww2.13 — hopefully it will be ready later today or tomorrow.
If you have orthogonal features that you would like included in ww2.13 that are relevant to PreTeX (MBX) now would be
a good time to submit them. If they are sufficiently orthogonal that the testing is not too extreme I will try to include them as well.
Once 2.13 is split off we can work hard on the localization and ww3 projects. I would like to at least make substantial
progress on those so that we can have a 2.14 release in late August.
Take care,
Mike
… On Jun 6, 2017, at 11:00 AM, Alex Jordan ***@***.***> wrote:
What is the time line for a next release? I could do extensive checking, but it might creep into July before I would be finished.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub <https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_openwebwork_webwork2_pull_767-23issuecomment-2D306513995&d=DwMCaQ&c=kbmfwr1Yojg42sGEpaQh5ofMHBeTl9EI2eaqQZhHbOU&r=C6Pt5AGtImanmAdcooarL-JZO8M5dSFPfs3VweYXYkE&m=kn01OqNQrZ1VXGAoiAf7LOOJ-p6lmkAqWxv06xex-pQ&s=SDz7ZXVLgYPqY4yYI8eqv-sLprUAiPqRIfc7B1wgIPU&e=>, or mute the thread <https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_ABJdfF9qgMU7YZL3klnSQInhUQVBYMxuks5sBWmdgaJpZM4NlHER&d=DwMCaQ&c=kbmfwr1Yojg42sGEpaQh5ofMHBeTl9EI2eaqQZhHbOU&r=C6Pt5AGtImanmAdcooarL-JZO8M5dSFPfs3VweYXYkE&m=kn01OqNQrZ1VXGAoiAf7LOOJ-p6lmkAqWxv06xex-pQ&s=xdMXnyvJ_HOV2wJ3yS0FMC3XgAcoycWjSKq41cfL9Os&e=>.
|
OK, PreTeXt related PR's should wait until the August merge then. They are a mess on my local branches of pg, webwork2, and mathbook. It will take me the summer to organize, commit, and test. I'll try to prioritize doing some testing here before July. |
Here is a better homework test suite for multianswer questions. This one includes the files for some of the questions which were modified and stored locally rather than being taken from the OPL.
|
Here's some questions/comments
All of the matrix questions have the problem similar to
|
I thought I was using the pg branch too. Anyway, no errors now. I've also tested about 10 other multiAnswer problems from the library and seems to be working. If you update the missing problems (13, 17, 18) then I'll check those, but this looks good. |
I've uploaded updated versions of the multiAnswer homework test suite. All of the question files are either standard entries in the OPL or have been placed in the templates/setmultiAnswer/ directory.
|
Not seeing any error with any of these problems and have done testing on others. I will pull this and the PG version in. |
This branch fixes reported bugs in which only the first answer of a multiAnswer question was recorded for "sticky answers" and into the pastAnswer log. It still needs extensive testing.