Merge pull request #797 from AnswerDotAI/html2ft_space

only strip newlines when parsing strings

Author: jph00

fasthtml/components.py

@@ -207,14 +207,14 @@ def _f(*c, target_id=None, **kwargs): return ft_hx(tag, *c, target_id=target_id,
 def html2ft(html, attr1st=False):
     """Convert HTML to an `ft` expression"""
     rev_map = {'class': 'cls', 'for': 'fr'}
-    
+
     def _parse(elm, lvl=0, indent=4):
-        if isinstance(elm, str): return repr(elm.strip()) if elm.strip() else ''
+        if isinstance(elm, str): return repr(elm.strip("\n")) if elm.strip() else ''
         if isinstance(elm, list): return '\n'.join(_parse(o, lvl) for o in elm)
         tag_name = elm.name.capitalize().replace("-", "_")
         if tag_name=='[document]': return _parse(list(elm.children), lvl)
         cts = elm.contents
-        cs = [repr(c.strip()) if isinstance(c, str) else _parse(c, lvl+1)
+        cs = [repr(c.strip("\n")) if isinstance(c, str) else _parse(c, lvl+1)
               for c in cts if str(c).strip()]
         attrs, exotic_attrs  = [], {}
         for key, value in sorted(elm.attrs.items(), key=lambda x: x[0]=='class'):

nbs/api/01_components.ipynb

@@ -1120,14 +1120,14 @@
     "def html2ft(html, attr1st=False):\n",
     "    \"\"\"Convert HTML to an `ft` expression\"\"\"\n",
     "    rev_map = {'class': 'cls', 'for': 'fr'}\n",
-    "    \n",
+    "\n",
     "    def _parse(elm, lvl=0, indent=4):\n",
-    "        if isinstance(elm, str): return repr(elm.strip()) if elm.strip() else ''\n",
+    "        if isinstance(elm, str): return repr(elm.strip(\"\\n\")) if elm.strip() else ''\n",
     "        if isinstance(elm, list): return '\\n'.join(_parse(o, lvl) for o in elm)\n",
     "        tag_name = elm.name.capitalize().replace(\"-\", \"_\")\n",
     "        if tag_name=='[document]': return _parse(list(elm.children), lvl)\n",
     "        cts = elm.contents\n",
-    "        cs = [repr(c.strip()) if isinstance(c, str) else _parse(c, lvl+1)\n",
+    "        cs = [repr(c.strip(\"\\n\")) if isinstance(c, str) else _parse(c, lvl+1)\n",
     "              for c in cts if str(c).strip()]\n",
     "        attrs, exotic_attrs  = [], {}\n",
     "        for key, value in sorted(elm.attrs.items(), key=lambda x: x[0]=='class'):\n",
@@ -1299,7 +1299,9 @@
     "#|hide\n",
     "def test_html2ft(html: str, attr1st=False):\n",
     "    # html -> ft -> html\n",
-    "    assert html == to_xml(eval(html2ft(html, attr1st))).strip()"
+    "    soup1 = BeautifulSoup(html, 'html.parser')\n",
+    "    soup2 = BeautifulSoup(to_xml(eval(html2ft(html, attr1st))).strip(), 'html.parser')\n",
+    "    assert soup1.prettify() == soup2.prettify()\n"
    ]
   },
   {
@@ -1313,6 +1315,7 @@
     "test_html2ft('<input value=\"Profit\" name=\"title\" id=\"title\" class=\"char\">')\n",
     "test_html2ft('<div id=\"foo\"></div>')\n",
     "test_html2ft('<div id=\"foo\">hi</div>')\n",
+    "test_html2ft('<div>Howdy <a href=\"https://answer.ai\">answer</a> how are you?</div>')\n",
     "test_html2ft('<div x-show=\"open\" x-transition:enter=\"transition duration-300\" x-transition:enter-start=\"opacity-0 scale-90\">Hello 👋</div>')\n",
     "test_html2ft('<div x-transition:enter.scale.80 x-transition:leave.scale.90>hello</div>')"
    ]
@@ -1361,13 +1364,7 @@
    "source": []
   }
  ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "python3",
-   "language": "python",
-   "name": "python3"
-  }
- },
+ "metadata": {},
  "nbformat": 4,
  "nbformat_minor": 5
 }