copy figures as well

website/copy_notebooks.py

@@ -4,12 +4,13 @@ creates pages which wrap them and link together.
 """
 import os
 import nbformat
+import shutil
 
 PAGEFILE = """title: {title}
 slug: {slug}
 Template: page
 
-{{% notebook notebooks/{notebook_file} cells[2:] %}}
+{{% notebook notebooks/{notebook_file} cells[{cells}] %}}
 """
 
 
@@ -24,38 +25,58 @@ PAGE_DEST_DIR = abspath_from_here('content', 'pages')
 
 
 def copy_notebooks():
-    nblist = sorted(os.listdir(NB_SOURCE_DIR))
+    nblist = sorted(nb for nb in os.listdir(NB_SOURCE_DIR)
+                    if nb.endswith('.ipynb'))
     name_map = {nb: nb.rsplit('.', 1)[0] + '.html'
                 for nb in nblist}
 
+    figsource = abspath_from_here('..', 'notebooks', 'figures')
+    figdest = abspath_from_here('content', 'figures')
+
+    if os.path.exists(figdest):
+        shutil.rmtree(figdest)
+    shutil.copytree(figsource, figdest)
+
+    figurelist = os.listdir(abspath_from_here('content', 'figures'))
+    figure_map = {os.path.join('figures', fig) : os.path.join('/figures', fig)
+                  for fig in figurelist}
+
     for nb in nblist:
         base, ext = os.path.splitext(nb)
-        if ext != '.ipynb':
-            continue
         print('-', nb)
 
         content = nbformat.read(os.path.join(NB_SOURCE_DIR, nb),
                                 as_version=4)
-        title = content.cells[2].source
-        if not title.startswith('#'):
-            raise ValueError('title not found in third cell')
-        title = title.lstrip('#').strip()
 
-        # put nav below title
-        content.cells[1], content.cells[2] = content.cells[2], content.cells[1]
+        if nb == 'Index.ipynb':
+            cells = '1:'
+            title = 'Python Data Science Handbook'
+        else:
+            cells = '2:'
+            # put nav below title
+            title = content.cells[2].source
+            if not title.startswith('#') or len(title.splitlines()) > 1:
+                raise ValueError('title not found in third cell')
+            title = title.lstrip('#').strip()
+            content.cells[1], content.cells[2] = content.cells[2], content.cells[1]
 
         for cell in content.cells:
             if cell.cell_type == 'markdown':
                 for nbname, htmlname in name_map.items():
                     if nbname in cell.source:
                         cell.source = cell.source.replace(nbname, htmlname)
+                for figname, newfigname in figure_map.items():
+                    if figname in cell.source:
+                        cell.source = cell.source.replace(figname, newfigname)
+                        
         nbformat.write(content, os.path.join(NB_DEST_DIR, nb))
 
         pagefile = os.path.join(PAGE_DEST_DIR, base + '.md')
         with open(pagefile, 'w') as f:
             f.write(PAGEFILE.format(title=title,
                                     slug=base.lower(),
-                                    notebook_file=nb))
+                                    notebook_file=nb,
+                                    cells=cells))
 
 if __name__ == '__main__':
     copy_notebooks()