preprocess_rm_cell

Author: jph00

nbprocess/_nbdev.py

@@ -40,6 +40,9 @@
          "clean_nb": "05_clean.ipynb",
          "nbdev_clean_nbs": "05_clean.ipynb",
          "default_pp_cfg": "06_mdx.ipynb",
+         "preprocess_cell": "06_mdx.ipynb",
+         "preprocess": "06_mdx.ipynb",
+         "preprocess_rm_cell": "06_mdx.ipynb",
          "InjectMeta": "06_mdx.ipynb",
          "StripAnsi": "06_mdx.ipynb",
          "InsertWarning": "06_mdx.ipynb",

nbprocess/mdx.py

@@ -1,22 +1,26 @@
 # AUTOGENERATED! DO NOT EDIT! File to edit: nbs/06_mdx.ipynb (unless otherwise specified).
 
-__all__ = ['default_pp_cfg', 'InjectMeta', 'StripAnsi', 'InsertWarning', 'RmEmptyCode', 'UpdateTags', 'HideInputLines',
-           'FilterOutput', 'CleanFlags', 'CleanMagics', 'BashIdentify', 'CleanShowDoc', 'HTMLEscape', 'RmHeaderDash',
-           'default_pps', 'mdx_exporter', 'nb2md']
+__all__ = ['default_pp_cfg', 'preprocess_cell', 'preprocess', 'preprocess_rm_cell', 'InjectMeta', 'StripAnsi',
+           'InsertWarning', 'RmEmptyCode', 'UpdateTags', 'HideInputLines', 'FilterOutput', 'CleanFlags', 'CleanMagics',
+           'BashIdentify', 'CleanShowDoc', 'HTMLEscape', 'RmHeaderDash', 'default_pps', 'mdx_exporter', 'nb2md']
 
 # Cell
 import re, uuid
-
-from .read import get_config
-from .processor import *
-from nbconvert.preprocessors import ExtractOutputPreprocessor
-
+from functools import wraps
 from fastcore.basics import *
 from fastcore.foundation import *
 from traitlets.config import Config
 from pathlib import Path
 from html.parser import HTMLParser
 
+from .read import get_config
+from .processor import *
+
+from nbconvert.preprocessors import ExtractOutputPreprocessor,Preprocessor,TagRemovePreprocessor
+from nbconvert import MarkdownExporter
+from .extract_attachments import ExtractAttachmentsPreprocessor
+from nbconvert.writers import FilesWriter
+
 # Cell
 def default_pp_cfg():
     "Default Preprocessor Config for MDX export"
@@ -26,9 +30,38 @@ def default_pp_cfg():
     c.TagRemovePreprocessor.remove_input_tags = ('remove_input', 'remove_inputs', "hide_input", "hide_inputs")
     return c
 
+# Cell
+def preprocess_cell(func):
+    "Decorator to create a `preprocess_cell` `Preprocessor` for cells"
+    @wraps(func, updated=())
+    class _C(Preprocessor):
+        def preprocess_cell(self, cell, resources, index):
+            res = func(cell)
+            if res: cell = res
+            return cell, resources
+    return _C
+
+# Cell
+def preprocess(func):
+    "Decorator to create a `preprocess` `Preprocessor` for notebooks"
+    @wraps(func, updated=())
+    class _C(Preprocessor):
+        def preprocess(self, nb, resources):
+            res = func(nb)
+            if res: nb = res
+            nb.cells = list(nb.cells)
+            return nb, resources
+    return _C
+
+# Cell
+def preprocess_rm_cell(func):
+    "Like `preprocess_cell` but remove cells where function returns `True`"
+    @preprocess
+    def _inner(nb): nb.cells = [cell for cell in nb.cells if not func(cell)]
+    return _inner
+
 # Cell
 def _mdx_exporter(pps, cfg=None, tpl_file='ob.tpl'):
-    "An mdx notebook exporter which composes preprocessors"
     cfg = cfg or default_pp_cfg()
     cfg.MarkdownExporter.preprocessors = pps or []
     tmp_dir = Path(__file__).parent/'templates/'
@@ -39,7 +72,6 @@ def _mdx_exporter(pps, cfg=None, tpl_file='ob.tpl'):
 
 # Cell
 def _run_preprocessor(pps, fname, display=False):
-    "An mdx notebook exporter which composes preprocessors"
     exp = _mdx_exporter(pps)
     result = exp.from_filename(fname)
     if display: print(result[0])
@@ -191,14 +223,11 @@ def HTMLEscape(cell):
 # Cell
 _re_hdr_dash = re.compile(r'^#+\s+.*\s+-\s*$', re.MULTILINE)
 
-@preprocess_cell
+@preprocess_rm_cell
 def RmHeaderDash(cell):
     "Remove headings that end with a dash -"
-    if cell.cell_type == 'markdown':
-        exclude = {l.strip() for l in _re_hdr_dash.findall(cell.source)}
-        if exclude:
-            lines = [l for l in cell.source.splitlines() if l not in exclude]
-            cell.source = '\n'.join(lines)
+    src = cell.source.strip()
+    return cell.cell_type == 'markdown' and src.startswith('#') and src.endswith(' -')
 
 # Cell
 def default_pps(c):