When using save_interview_answers() it should be possible to set the answers to encrypted/unencrypted
#625
Labels
bug
Something isn't working
Comments
|
Also note we can add a |
|
What are some of the benefits of this change? I know currently existing sessions get shared to legal aid helpers, multi-signature stuff, etc., but my impression was that answer sets were primarily supposed to be for the one person filling it out; encrypting answer sets seems like a good choice there. Is this about making sure |
|
If a legal aid or law firm uses an intake interview, answer sets are an easy way to let people fill-in the intake with anonymous answers and still let the attorney re-use the answers on various pleadings or letters the attorney then wants to file. Edit: it makes sense to allow the interview author configure this, therefore. But given the full answers associated with the user's interview are generally not encrypted--encryption is turned off for all AssemblyLine interviews--I'm not sure encrypting just the answer sets is a good default? |
nonprofittechy
added a commit
that referenced
this issue
Nov 29, 2022
plocket
added a commit
that referenced
this issue
Jan 18, 2023
* Fix #296 * Provide some placeholders to allow dynamically updating state list - #281 * Format phone numbers * 🎨 Format Python code with psf/black * Check if the `suffix_to_append` attribute exists On older objects, the attribute might not exist, so make sure it's updated to the standard default. * 🎨 Format Python code with psf/black * Bump version: 2.15.0 → 2.16.0 * Bump version: 2.16.0 → 2.17.0 * Update CHANGELOG.md Remove accidental 2.16 * Fix #585 - improve usability of signature question * Fix #578 fix #571 - let author filter irrelevant sessions and remove sessions at step 1 by default * Incorporate feedback from PR * Fix #573 Fix #572 * reformat with black * Improve #571 - add to global config * Fix #592 - redirect if user logged out * language changes from @CaroRob * User user_role, not users * more using user_role * Use `user_started_case`, not `user_role` The latter isn't actually used in AL core much, so even though it is the better variable name, changing the logic of other variables is more likely to break others interviews. * Fix #598 * 🎨 Format Python code with psf/black * Added black hook + black opts to pyproject.toml Also moved mypy configs to the same pyproject.toml instead of a separate mypy.ini * Add note about `exclude_newly_started_sessions` * Decrease the size of images in tables Commonly in review tables. * Added a max size feature to the ExhibitDocument (#604) * Added a max size feature to the ExhibitDocument Also give suggested remaining upload size in human readable file sizes in errors and in between upload screens. * Teams bump hook, in a github action (#602) * Teams bump hook, in a github action Fixes #547 * Don't check if the ref equals the version on branches, only tags * Use `|` instead of `/`, otherwise sed can't use repo names and URLs. * If address is empty, `on_one_line()` will be empty (#605) * If address is empty, `on_one_line()` will be empty Originally, was ", ", which is def a bug * Mypy changed defaults about implicit optionals * No more implicit optionals in function signatures Apparently not allowed by PEP-484. Personally, I've gotten used to it, but I remember being confused by it, and the change isn't too bad. * Fix #243, don't cut off words for addendum in mid-sentence * 🎨 Format Python code with psf/black * Multi-line return value * Bump version to v2.18.0 * failure was in a name, not an if * Tag URL had extra https:// * Improve calculation of max lines, better test cases, consistent overflow and safe_value handling of whitespace * 🎨 Format Python code with psf/black * Test case * Some correctness fixes for whitespace, working tests without 'define()' * Update docassemble/AssemblyLine/al_document.py * Add preserve_words option to alternate API * Remove words 'the' and 'project' from opening screen - Fix #611 * Bump version * Update CHANGELOG.md * Change OCR to use ocrmypdf in the background Triggers an event in `al_document.yml`, which calls back into a module. * Add config option for OCRMyPDF * Added OCRmyPDF to changelog * Add visible trigger variable to the developer-only heads up display (#617) * Fix #616 * Update docassemble/AssemblyLine/data/questions/al_visual.yml Co-authored-by: Bryce Willey <Bryce.Steven.Willey@gmail.com> Co-authored-by: Bryce Willey <Bryce.Steven.Willey@gmail.com> * First version of DACatchAll, #618 * Add css to all catchall qs * Progress on #620 * WIP * Fix #615, additional refinement of #620 * Cleanup error page a bit * Add focus shadow for back button Before this change, you aren't able to see that you are keyboard tabbing over the back button, as the value bootstrap uses to color the shadow doesn't have a value for the `daquestionbackbutton` class, only the `btn-primary` class. TBH, I'm not a huge fan of the default focus indicators in bootstrap, they don't have enough contrast. But that will likely have to wait for a different PR, as it's a much larger change. * Put 'download work in progress' behind a config option- anyone w/ a paywall won't want this turned on * `will_send_to_real_court` false if debug == True Necessary to safely test on any new server being spun up. Debug: true shouldn't be on in production servers. * Fix #625 - make answer sets unencrypted to match interview state * Put actions behind config + individual interview option * Try to cache python deps when running unittests Also bump to python 3.9: closer to still working with python 3.8 for older docassemble versions, but still on by default in ubuntu 22.04, the new test runners on github. * Quick remove old blank step * Trying for easier unittest workflow * Add wheel package * Start using --use-pep517? * Can't actually use `--use-pep517` yet One of the packages fails with an obscure error message, so I'll just have to turn it off now. ``` Collecting guess-language-spirit==0.5.3 Using cached guess_language-spirit-0.5.3.tar.bz2 (81 kB) Installing build dependencies: started Installing build dependencies: finished with status 'done' Getting requirements to build wheel: started Getting requirements to build wheel: finished with status 'done' ERROR: Could not install packages due to an OSError: [Errno 2] No such file or directory: '/tmp/tmpdpiv1dse/output.json' ``` * Move everything to a different action * Address SuffolkLITLab/docassemble-MassAccess#8 * Remove unnecessary 'logo' from alt text * Fix #636 * Add simple service of process questions, Fix #597, fix #597 * Add marriage questions, fix #630 * Update docassemble/AssemblyLine/data/questions/ql_baseline.yml Co-authored-by: Bryce Willey <Bryce.Steven.Willey@gmail.com> * Update docassemble/AssemblyLine/data/questions/ql_baseline.yml Co-authored-by: Bryce Willey <Bryce.Steven.Willey@gmail.com> * PR feedback * Fix #637 - better API for how a lot of states organize courts, by county. Added some useful features like searching by keyword too. * Reformat with black * Fix #296 * A little extra safety for existing sessions that might be in the wild * Fix #260 * set country at a better point in interview; remove commented out section for now * Readd newlines between each step on intro screen 2 or more trailing whitespaces in markdown will add a line break (<br>) to the line. The trailing whitespaces that were separating out the 3 introduction steps for each AL interview were mistakenly removed in 57df4ab#diff-55dff7108dded7bbc3ce8459dd669a79baec6674c1558af86f66dbc878268288L44. This PR replaces those trailing whitespaces with a more explicit `<br>`. * Address some linting feedback * Bump version * Update CHANGELOG.md * Fix issue where al_sessions_to_exclude kept growing * Update docassemble/AssemblyLine/al_general.py Co-authored-by: Bryce Willey <Bryce.Steven.Willey@gmail.com> * Adjust comments per review + add docstring * 🎨 Format Python code with psf/black * Cleanup appearance of error action page, make technical information optional * Split if statement for trace and history Co-authored-by: Quinten Steenhuis <qsteenhuis@gmail.com> Co-authored-by: nonprofittechy <nonprofittechy@users.noreply.github.com> Co-authored-by: Bryce Willey <bryce.steven.willey@gmail.com> Co-authored-by: BryceStevenWilley <BryceStevenWilley@users.noreply.github.com> Co-authored-by: Quinten Steenhuis <qsteenhuis@suffolk.edu>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Currently, we remove
multi_userfrom the answers. Probably, it should be an explicit interview author choice whether or not to encrypt the answer set that doesn't depend on the encryption settings of the source interview.The text was updated successfully, but these errors were encountered: