Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
* fix bug where completing audio in one exp-lookit-dialogue-page frame was carrying through to other frames so users were not actually required to play audio; only require users to play audio rather than completing it for character audio * display images centered on a specified left value, rather than left-aligned at left value, to avoid overlapping images for polcon study * fix animation of images to work with new formatting * update troubleshooting info for exp-config frame * Add skeleton of observation frame. Displays webcam, allows user to show/hide webcam. * add dependencies * add placeholder survey frame and get dependencies working * fix green outline formatting for selected choice broken with previous formatting update * Wiggle the currently-speaking character so kids can tell who's saying what! * Get survey frame working. Allow specification of arbitrary fields, validation, repopulation when returning to frame. * Record transition to previous frame as event * Update exp-video-config formatting and troubleshooting text to be easier to follow. Include instructions to use Chrome or Firefox on a computer, not a mobile device, at the top. Break down instructions by browser. Don't send people to outside webpages unless necessary, just show screenshots for on-screen fixes. Edit wording on main instructions. * add time limit for recording; show/hide recorder smoothly; add video recording; add option to record starting immediately * Set up video recorder to work using Pipe instead of HDFVR. This is an initial proof of concept only. Current status: recorder shows up correctly formatted, w/o record button shown, in correct element; we can proceed from one frame to another without re-confirming settings; we can programmatically start and stop recording. (NOT YET: pause, wait for upload/allow overlapping recordings, warn if no audio, destroy recorder, display consent video to user, wait until recording is ready/started, edit troubleshooting directions, check video naming convention, security review.) Note that html5 is only used on secure origins, so for local development need to use https://localhost. * Remove requirement that user reloads webcam before finishing exp-video-config. * remove unnecessary _SWFId data * Make sure that new exp-lookit-dialogue-page styles don't affect other frames when in fullscreen! Fixes bug in fullscreen display for exp-lookit-story-page. * add instructions in exp-video-config-quality to adjust camera height based on child over shoulder * check that volume exceeds minimal level during exp-video-config frame * destroy recorder after frame ends * add keypress handler for enter -> next * revert attempt to add source selection to exp-video-config frame * only call finish() once in exp-video-consent * make pause/resume into placeholder dummies for backwards compatibility * only stop video recorder if it's currently recording * fix bug where exp-video-config called stop and, since recording hadn't ever been started, stop was deferred until 3s later by which point another frame might have a recorder active * have exp-lookit-observation use pause (placeholder) rather than stop until implementing multiple video clip solution for pausing * reversion of earlier changes to ensure proper playback - not sure whether this is necessary though * edit Next button on exp-lookit-dialogue-page to say "or press Enter" as well * store list of videos per frame; embed webrtc player in physics frame; store pipe account hash in env; dummy fns for pause/resume for now * add observation frame, using webrtc * make recorder elements & functions more consistent across frames in prep for generalizing * set up exp-lookit-story-page to work with new recorder * avoid error on leaving FS when frame has been destroyed * don't destroy recorder when upload completes; handled elsewhere (and leave flexibility for future multiple recordings) * adapt exp-lookit-dialogue-page to work with new recorder * remove unnecessary fields from exp-lookit-story-page * restore enter == next on exp-lookit-dialogue-page (not sure why pulling from develop is not reinstating this but other commits appear to be here) * Mark geometry frame as not adapted for webrtc. * remove unused frames for clarity when dealing with codebase * move stopping/destroying video recorder to mixin instead of individual frames * use frame name directly when recording event, rather than manually adding a boilerplate modification of makeTimeEvent to every frame * don't show timer on pipe recorder, to avoid some errors where it's not found * restore radio-group formatting, accidentally removed earlier * format next button so it's not flush at bottom of page * refactor to make recorder setup & triggering of actions once ready to record part of the VideoRecord mixin. Some cleanup of willDestroy refactor from earlier. * actually require at least 3 s recording on consent video * ignore compiled docs * have video-config check for both camera and mic, and allow reload * remove unused 'hidden' attribute from video-recorder service * stop tracking docs * move recorder hide/show functions to mixin and get working * set up webcam warnings for physics study - before starting test will pause and indicate if there's no webcam available. Doesn't currently require sound because that's not critical during study. * update the troubleshooting directions for the exp-video-config frame for webrtc player * load Pipe environment ID from env file * use loaded pipe environment ID * update docs for use of this mixin * keep original naming scheme since it's used by lookit-api * remove Flash player config * add directions about selecting mic/camera devices in Chrome to exp-video-config frame * remove unused video-pause mixin * Revert "ignore compiled docs" This reverts commit 2ac3366. * update yuidocs * fix linting errors
- Loading branch information