TODO list for 2.0. These are considered release blockers, unlike the items in the Version_2.0_Wishlist.
- Ensure that 1.4 blatantly and quickly fails to load 2.0 data files
- Add a metadata version that we check for so we can cleanly fail when loading future versions
- Note: we are explicitly ignoring camera-provided metadata for the moment (will require modifications to the Core to tag these metadatas)
- Having to register again is wrong, probably. In general all settings
that were in Preferences are “lost”.
- This depends on if the first official 2.0 release is from UCSF vs. Open Imaging.
- More general problem with importing Preferences: how do we access the old Preferences node? org.micromanager.MMStudio no longer exists. See also this StackOverflow thread
- Auto-select previous profile at startup prompt? Entire profile system is maybe unclear.
- Add inheritance for user profiles from a central authority (c.f. remote login systems – talk with Mark)
- Image dimensions etc. move from main window to image window?
- In fullscreen mode would be nice if we could shove the image over to make room for the inspector.
- Multicam config file, 10ms exposure, start program, click “Live” in
main frame (with no pre-existing image window) -> tons of error
- Error message: “Error: this application, or a library it uses, has passed an invalid numeric value (NaN, or not-a-number) to CoreGraphics API. This is a serious error and contributes to an overall degradation of system stability and reliability. This notice is a courtesy: please fix this problem. It will become a fatal error in an upcoming update.”
- Scan through entire codebase for TODO tags and fix them if appropriate
- Autofocus API needs some attention (remove use of PropertyItem and MMException?)
- Merge against latest SVN
- NB that due to line ending changes, conflicts in multipage TIFF reader/writer couldn’t be detected. Other files (MMStudio, ScriptInterface, MMAcquisition, ChannelControlPanel, DisplayWindow, MetadataPanel, FileMenu,) got “deleted” (improperly moved, I guess) and thus similarly couldn’t detect changes.
- ASIdiSPIM and Magellan plugins are of course hopelessly nonfunctional.
- Script panel is hard to read in night mode.
- Wiki page for Netbeans setup needs to take into account Maven directory structure (“resources” and “scripts” directories need to be added to the “libraries” in project properties)
- Be sure to test non-square images from time to time
- Leave a Live display open for a few hours and see if the display “breaks”.
- Check for memory leaks
- Try running acquisitions on 32-bit systems (in particular be aware of memory constraints e.g. with direct byte buffers)
- Test on Windows
- Test on older Macs and on displays with varying pixel densities
- Test high-speed acquisitions
- Still have to select overlay plugins (e.g. Acquisition Tools -> Pattern Overlay) from the Plugins menu before they become available in the overlays tab; this ought to happen automatically on program start.
- Ensure loading of 1.4 files works properly, with fallbacks for properties that aren’t available but should be in 2.0 (e.g. display settings)
- Test importing using BioFormats importer in FIJI.
- Autostretching and linking interact poorly – autostretch should
apply to the currently-displayed image.
- So make autostretch be a per-channel property, and modify ChannelControlPanel to only use the linked min/max when autostretch is not set. Requires adding an autostretch toggle button to each channel.
- Option for centered text for scalebar (or different alignments in general)
- Option to change size of timestamp (incl. font size)
- Inspector window moves when size changes (want to keep top-left corner consistent)