Update the info about macOS interactive mode

To reflect the fact that it now works in IPython/Jupyter notebooks.

Author: ctrueden

doc/01-Starting-PyImageJ.ipynb

@@ -59,8 +59,6 @@
     "| With Fiji plugins (specific version)                   | `ij = imagej.init('sc.fiji:fiji:2.14.0')`                          | YES                       |\n",
     "| From a local installation                              | `ij = imagej.init('/Applications/Fiji.app')`                      | DEPENDS                   |\n",
     "\n",
-    "*: _`mode='interactive'` is unavailalbe on macOS. Instead use `mode='gui'`. When set to `gui` mode the Python interpeter is blocked and no longer interactive. Check out the [initialization](Initialization.md) documentation for more information._\n",
-    "\n",
     "<sup>1</sup> pyimagej uses [`jgo`](https://github.com/scijava/jgo) internally to call up ImageJ, so all of these initializations are tied to the usage of `jgo`. You can read up on the [usage of `jgo`](https://github.com/scijava/jgo#usage) to find out more about this initialization.\n",
     "\n",
     "<sup>2</sup> ___Reproducible___ means code is stable, executing the same today, tomorrow, and in years to come. While it is convenient and elegant to depend on the newest version of a program, behavior may change when new versions are released—for the better if bugs are fixed; for the worse if bugs are introduced—and people executing your notebook at a later time may encounter broken cells, unexpected results, or other more subtle behavioral differences. You can help avoid this pitfall by pinning to a specific version of the software. The British Ecological Society published [Guide to Better Science: Reproducible Code](https://www.britishecologicalsociety.org/wp-content/uploads/2018/12/BES-Reproducible-Code.pdf) diving into the relevant challenges in more detail, including an [R](https://www.r-project.org/)-centric illustration of best practices. A web search for `reproducible python` also yields several detailed articles."

doc/06-Working-with-Images.ipynb

@@ -414,7 +414,7 @@
     "\n",
     "<table><tr><td>\n",
     "\n",
-    "This section does not currently work on macOS. See [imagej/pyimagej#197](https://github.com/imagej/pyimagej/issues/197) for details.\n",
+    "This section may not currently work on macOS. See [imagej/pyimagej#197](https://github.com/imagej/pyimagej/issues/197) for details.\n",
     "\n",
     "</td></tr></table>\n",
     "\n",

doc/Cellpose-StarDist-Segmentation.ipynb

@@ -393,13 +393,6 @@
     "    print(nuc_roi_tree.children().get(i).data())"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "⚠️**Warning**: The following section requires an [interactive](Initialization.md#ways-to-initialize) PyImageJ session. Unfortnately macOS does not support PyImageJ's interactive mode, thus this notebook can only be run in `headless` mode on macOS. Visit the [interactive mode on macOS](Troubleshooting.md#non-blocking-interactive-mode-on-macos) section in the PyImageJ troubleshooting documentation for more information."
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": 11,

doc/Initialization.md

@@ -118,8 +118,9 @@ This mode returns immediately after initializing ImageJ2, allowing you to
 "mix and match" operations performed via Python code with operations
 performed via GUI interactions.
 
-_**Note:** This mode does not work on macOS,
-due to a limitation in the macOS threading model._
+_**Note:** This mode does not work on macOS from plain Python, due to a
+limitation in the macOS threading model. It does work in IPython/Jupyter, as
+well as via the [Jaunch](https://github.com/apposed/jaunch#readme) launcher._
 
 ### Support for the original ImageJ
 

doc/Troubleshooting.md

@@ -65,12 +65,13 @@ you want to register or update.
 
 ### Non-blocking INTERACTIVE mode on macOS
 
-On macOS, the Cocoa event loop needs to be started from the main thread before
-any Java-AWT-specific functions can work. And doing so blocks the main thread.
-For this reason, PyImageJ includes two graphical modes, `GUI` and
-`INTERACTIVE`, with `GUI` blocking the `imagej.init` invocation, and
-`INTERACTIVE` returning immediately... but `INTERACTIVE` cannot work on macOS
-and is therefore not available, due to this OS-specific limitation.
+On macOS, the CoreFoundation/AppKit event loop needs to be started from the
+main thread before any Java-AWT-specific functions can work. And doing so
+blocks the main thread. For this reason, PyImageJ includes two graphical modes,
+`GUI` and `INTERACTIVE`, with `GUI` blocking the `imagej.init` invocation, and
+`INTERACTIVE` returning immediately... but `INTERACTIVE` can only work on macOS
+in certain scenarios. PyImageJ makes a best effort to check your environment
+and report when interactive cannot work, and how you might go about fixing it.
 
 ### Old versions of ImageJ2
 

src/imagej/__init__.py

@@ -1179,8 +1179,9 @@ def init(
             Start ImageJ2 with GUI support, but *not* displaying the GUI
             automatically, Does not block. To display the GUI in this mode,
             call ij.ui().showUI().
-            NB: This mode is not available on macOS, due to its application
-            threading model.
+            NB: On macOS, this mode only works in some scenarios. It will
+            not work from plain Python, but it works in IPython/Jupyter,
+            as well as when launching Python via the Jaunch launcher.
             NB: In this mode with add_legacy=True, the JVM and Python will
             both terminate when ImageJ closes!