Fix propagation of fragments across multiple slides/subslides.

IPython/nbconvert/preprocessors/revealhelp.py

@@ -1,62 +1,68 b''
 """Module that pre-processes the notebook for export via Reveal."""
 
 # Copyright (c) IPython Development Team.
 # Distributed under the terms of the Modified BSD License.
 
 from .base import Preprocessor
 from IPython.utils.traitlets import Unicode
 
 
 class RevealHelpPreprocessor(Preprocessor):
 
     url_prefix = Unicode('reveal.js', config=True,
                          help="""The URL prefix for reveal.js.
                          This can be a a relative URL for a local copy of reveal.js,
                          or point to a CDN.
                          
                          For speaker notes to work, a local reveal.js prefix must be used.
                          """
     )
 
     def preprocess(self, nb, resources):
         """
         Called once to 'preprocess' contents of the notebook.
 
         Parameters
         ----------
         nb : NotebookNode
             Notebook being converted
         resources : dictionary
             Additional resources used in the conversion process.  Allows
             preprocessors to pass variables into the Jinja engine.
         """
 
         for index, cell in enumerate(nb.cells):
 
             #Make sure the cell has slideshow metadata.
             cell.metadata.slide_type = cell.get('metadata', {}).get('slideshow', {}).get('slide_type', '-')
 
             # Get the slide type. If type is start, subslide, or slide,
             # end the last subslide/slide.
             if cell.metadata.slide_type in ['slide']:
                 nb.cells[index - 1].metadata.slide_helper = 'slide_end'
             if cell.metadata.slide_type in ['subslide']:
                 nb.cells[index - 1].metadata.slide_helper = 'subslide_end'
             # Prevent the rendering of "do nothing" cells before fragments
             # Group fragments passing frag_number to the data-fragment-index
             if cell.metadata.slide_type in ['fragment']:
                 nb.cells[index].metadata.frag_number = index
                 i = 1
                 while i < len(nb.cells) - index:
-                    nb.cells[index + i].metadata.frag_helper = 'fragment_end'
-                    nb.cells[index + i].metadata.frag_number = index
-                    i += 1
+                    # We need to break the loop when a new slide or subslide is
+                    # found to avoid the propagation of the data-fragment-index
+                    # across multiple slides/subslides
+                    if nb.cells[index + i].metadata.slideshow.slide_type in ['slide', 'subslide']:
+                        break
+                    else:
+                        nb.cells[index + i].metadata.frag_helper = 'fragment_end'
+                        nb.cells[index + i].metadata.frag_number = index
+                        i += 1
             # Restart the slide_helper when the cell status is changed
             # to other types.
             if cell.metadata.slide_type in ['-', 'skip', 'notes', 'fragment']:
                 nb.cells[index - 1].metadata.slide_helper = '-'
 
         if not isinstance(resources['reveal'], dict):
             resources['reveal'] = {}
         resources['reveal']['url_prefix'] = self.url_prefix
         return nb, resources