extend front matter retrieval

Author: hamelsmu

nbprocess/_modidx.py

@@ -94,8 +94,7 @@
                                       'nbprocess.processors.exec_show_docs': 'https://nbprocess.fast.ai/processors#exec_show_docs',
                                       'nbprocess.processors.clean_show_doc': 'https://nbprocess.fast.ai/processors#clean_show_doc',
                                       'nbprocess.processors.insert_warning': 'https://nbprocess.fast.ai/processors#insert_warning',
-                                      'nbprocess.processors.add_show_docs': 'https://nbprocess.fast.ai/processors#add_show_docs',
-                                      'nbprocess.processors.add_frontmatter': 'https://nbprocess.fast.ai/processors#add_frontmatter'},
+                                      'nbprocess.processors.add_show_docs': 'https://nbprocess.fast.ai/processors#add_show_docs'},
             'nbprocess.read': { 'nbprocess.read.NbCell': 'https://nbprocess.fast.ai/read#NbCell',
                                 'nbprocess.read.dict2nb': 'https://nbprocess.fast.ai/read#dict2nb',
                                 'nbprocess.read.read_nb': 'https://nbprocess.fast.ai/read#read_nb',

nbprocess/processors.py

@@ -2,7 +2,7 @@
 
 # %% auto 0
 __all__ = ['add_links', 'strip_ansi', 'hide_', 'hide_line', 'filter_stream_', 'clean_magics', 'lang_identify', 'rm_header_dash',
-           'rm_export', 'exec_show_docs', 'clean_show_doc', 'insert_warning', 'add_show_docs', 'add_frontmatter']
+           'rm_export', 'exec_show_docs', 'clean_show_doc', 'insert_warning', 'add_show_docs']
 
 # %% ../nbs/09_processors.ipynb 3
 import ast
@@ -162,24 +162,14 @@ def add_show_docs(nb):
 def _celltyp(nb, cell_type): return nb.cells.filter(lambda c: c.cell_type == cell_type)
 def _frontmatter(nb): return _celltyp(nb, 'raw').filter(lambda c: _re_fm.search(c.get('source', '')))
 
-def _title(nb): 
-    "Get the title and description from a notebook from the H1"
+def _fm(nb): 
+    "Infer the front matter from a notebook's markdown formatting"
     md_cells = _celltyp(nb, 'markdown').filter(lambda c: _re_title.search(c.get('source', '')))
     if not md_cells: return None,None
     cell = md_cells[0]
     title,desc=_re_title.match(cell.source).groups()
+    flags = re.findall('^-\s+(.*)', cell.source, flags=re.MULTILINE)
+    flags = [s.split(':') for s in flags if ':' in s]
+    flags = {k:v for k,v in flags if k and v}
     cell['source'] = None
-    return title,desc
-
-def add_frontmatter(nb):
-    "Insert front matter if it doesn't exist"
-    if _frontmatter(nb): return
-    title,desc = _title(nb)
-    code_src = nb.cells.filter(lambda x: x.cell_type == 'code').attrgot('source')
-    default_exp = first(code_src.filter().map(_re_defaultexp.search).filter())
-    default_exp = default_exp.group(1) if default_exp else None
-    if title:
-        desc = f'description: "{desc}"\n' if desc else ''
-        outfile = f'output-file: {default_exp}\n' if default_exp else ''
-        content = f'---\ntitle: {title}\n{outfile}{desc}---\n'
-        nb.cells.insert(0, NbCell(0, dict(cell_type='raw', metadata={}, source=content)))
+    return title,desc,flags

nbs/09_processors.ipynb

@@ -581,7 +581,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "faba8516-7f94-4e5c-ae33-41b27cf350b9",
+   "id": "99a8a186-47be-4b66-b1f7-14727f4932e1",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -593,19 +593,43 @@
     "def _celltyp(nb, cell_type): return nb.cells.filter(lambda c: c.cell_type == cell_type)\n",
     "def _frontmatter(nb): return _celltyp(nb, 'raw').filter(lambda c: _re_fm.search(c.get('source', '')))\n",
     "\n",
-    "def _title(nb): \n",
-    "    \"Get the title and description from a notebook from the H1\"\n",
+    "def _fm(nb): \n",
+    "    \"Infer the front matter from a notebook's markdown formatting\"\n",
     "    md_cells = _celltyp(nb, 'markdown').filter(lambda c: _re_title.search(c.get('source', '')))\n",
     "    if not md_cells: return None,None\n",
     "    cell = md_cells[0]\n",
     "    title,desc=_re_title.match(cell.source).groups()\n",
+    "    flags = re.findall('^-\\s+(.*)', cell.source, flags=re.MULTILINE)\n",
+    "    flags = [s.split(':') for s in flags if ':' in s]\n",
+    "    flags = {k:v for k,v in flags if k and v}\n",
     "    cell['source'] = None\n",
-    "    return title,desc\n",
-    "\n",
+    "    return title,desc,flags"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "77a021a4-a51a-4f18-acd5-5382ef8312c5",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#|hide\n",
+    "test_eq(_fm(read_nb('../tests/docs_test.ipynb')),\n",
+    "        ('a title', 'A description', {'key1': ' value1', 'key2': ' value2'})\n",
+    "       )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "412799d3-a241-44bd-abc7-2a3bf1334781",
+   "metadata": {},
+   "outputs": [],
+   "source": [
     "def add_frontmatter(nb):\n",
     "    \"Insert front matter if it doesn't exist\"\n",
     "    if _frontmatter(nb): return\n",
-    "    title,desc = _title(nb)\n",
+    "    title,desc,_ = _fm(nb)\n",
     "    code_src = nb.cells.filter(lambda x: x.cell_type == 'code').attrgot('source')\n",
     "    default_exp = first(code_src.filter().map(_re_defaultexp.search).filter())\n",
     "    default_exp = default_exp.group(1) if default_exp else None\n",
@@ -655,7 +679,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "8fff5b3f-f977-4301-892e-312936cb7a4e",
+   "id": "174d4398-7446-4c27-b1aa-743bcbb419d4",
    "metadata": {},
    "outputs": [],
    "source": []

tests/docs_test.ipynb

@@ -6,7 +6,9 @@
    "metadata": {},
    "source": [
     "# a title\n",
-    "> A description"
+    "> A description\n",
+    "- key1: value1\n",
+    "- key2: value2"
    ]
   },
   {