Replace MarkupPy by ElementTree for html conversion

HISTORY.md

@@ -1,5 +1,12 @@
 # History
 
+## Unreleased
+
+### Changes
+
+- The html export format does not depend on MarkupPy any longer, therefore the
+  tablib[html] install target was removed also.
+
 ## 3.5.0 (2023-06-11)
 
 ### Improvements

docs/formats.rst

@@ -101,9 +101,6 @@ The ``html`` format is currently export-only. The exports produce an HTML page
 with the data in a ``<table>``. If headers have been set, they will be used as
 table headers.
 
-This format is optional, install Tablib with ``pip install "tablib[html]"`` to
-make the format available.
-
 jira
 ====
 

pyproject.toml

@@ -34,7 +34,6 @@ dynamic = ["version"]
 
 [project.optional-dependencies]
 all = [
-    "markuppy",
     "odfpy",
     "openpyxl>=2.6.0",
     "pandas",
@@ -44,7 +43,6 @@ all = [
     "xlwt",
 ]
 cli = ["tabulate"]
-html = ["markuppy"]
 ods = ["odfpy"]
 pandas = ["pandas"]
 xls = ["xlrd", "xlwt"]

src/tablib/formats/__init__.py

@@ -15,7 +15,6 @@
 uninstalled_format_messages = {
     "cli": {"package_name": "tabulate package", "extras_name": "cli"},
     "df": {"package_name": "pandas package", "extras_name": "pandas"},
-    "html": {"package_name": "MarkupPy package", "extras_name": "html"},
     "ods": {"package_name": "odfpy package", "extras_name": "ods"},
     "xls": {"package_name": "xlrd and xlwt packages", "extras_name": "xls"},
     "xlsx": {"package_name": "openpyxl package", "extras_name": "xlsx"},
@@ -101,8 +100,7 @@ def register_builtins(self):
         if find_spec('odf'):
             self.register('ods', 'tablib.formats._ods.ODSFormat')
         self.register('dbf', 'tablib.formats._dbf.DBFFormat')
-        if find_spec('MarkupPy'):
-            self.register('html', 'tablib.formats._html.HTMLFormat')
+        self.register('html', 'tablib.formats._html.HTMLFormat')
         self.register('jira', 'tablib.formats._jira.JIRAFormat')
         self.register('latex', 'tablib.formats._latex.LATEXFormat')
         if find_spec('pandas'):

src/tablib/formats/_html.py

@@ -1,10 +1,6 @@
 """ Tablib - HTML export support.
 """
-
-import codecs
-from io import BytesIO
-
-from MarkupPy import markup
+from xml.etree import ElementTree as ET
 
 
 class HTMLFormat:
@@ -17,48 +13,38 @@ class HTMLFormat:
     def export_set(cls, dataset):
         """HTML representation of a Dataset."""
 
-        stream = BytesIO()
-
-        page = markup.page()
-        page.table.open()
-
+        table = ET.Element('table')
         if dataset.headers is not None:
-            new_header = [item if item is not None else '' for item in dataset.headers]
-
-            page.thead.open()
-            headers = markup.oneliner.th(new_header)
-            page.tr(headers)
-            page.thead.close()
-
-        page.tbody.open()
+            head = ET.Element('thead')
+            tr = ET.Element('tr')
+            for header in dataset.headers:
+                th = ET.Element('th')
+                th.text = str(header) if header is not None else ''
+                tr.append(th)
+            head.append(tr)
+            table.append(head)
+
+        body = ET.Element('tbody')
         for row in dataset:
-            new_row = [item if item is not None else '' for item in row]
+            tr = ET.Element('tr')
+            for item in row:
+                td = ET.Element('td')
+                td.text = str(item) if item is not None else ''
+                tr.append(td)
+            body.append(tr)
+        table.append(body)
 
-            html_row = markup.oneliner.td(new_row)
-            page.tr(html_row)
-        page.tbody.close()
-
-        page.table.close()
-
-        # Allow unicode characters in output
-        wrapper = codecs.getwriter("utf8")(stream)
-        wrapper.writelines(str(page))
-
-        return stream.getvalue().decode('utf-8')
+        return ET.tostring(table, method='html', encoding='unicode')
 
     @classmethod
     def export_book(cls, databook):
         """HTML representation of a Databook."""
 
-        stream = BytesIO()
-
-        # Allow unicode characters in output
-        wrapper = codecs.getwriter("utf8")(stream)
-
+        result = ''
         for i, dset in enumerate(databook._datasets):
-            title = (dset.title if dset.title else 'Set %s' % (i))
-            wrapper.write(f'<{cls.BOOK_ENDINGS}>{title}</{cls.BOOK_ENDINGS}>\n')
-            wrapper.write(dset.html)
-            wrapper.write('\n')
+            title = dset.title if dset.title else f'Set {i}'
+            result += f'<{cls.BOOK_ENDINGS}>{title}</{cls.BOOK_ENDINGS}>\n'
+            result += dset.html
+            result += '\n'
 
-        return stream.getvalue().decode('utf-8')
+        return result

tests/requirements.txt

@@ -1,6 +1,5 @@
 pytest
 pytest-cov
-MarkupPy
 odfpy
 openpyxl>=2.6.0
 pyyaml

tests/test_tablib.py

@@ -14,7 +14,6 @@
 from uuid import uuid4
 
 import pytest
-from MarkupPy import markup
 from openpyxl.reader.excel import load_workbook
 
 import tablib
@@ -624,47 +623,49 @@ def test_row_has_tags(self):
 
 
 class HTMLTests(BaseTestCase):
-    def test_html_export(self):
+    founders_html = (
+        "<table>"
+        "<thead>"
+        "<tr><th>first_name</th><th>last_name</th><th>gpa</th></tr>"
+        "</thead>"
+        "<tbody>"
+        "<tr><td>John</td><td>Adams</td><td>90</td></tr>"
+        "<tr><td>George</td><td>Washington</td><td>67</td></tr>"
+        "<tr><td>Thomas</td><td>Jefferson</td><td>50</td></tr>"
+        "</tbody>"
+        "</table>"
+    )
+
+    def test_html_dataset_export(self):
         """HTML export"""
-
-        html = markup.page()
-        html.table.open()
-        html.thead.open()
-
-        html.tr(markup.oneliner.th(self.founders.headers))
-        html.thead.close()
-
-        html.tbody.open()
-        for founder in self.founders:
-            html.tr(markup.oneliner.td(founder))
-        html.tbody.close()
-
-        html.table.close()
-        html = str(html)
-
-        self.assertEqual(html, self.founders.html)
+        self.assertEqual(self.founders_html, self.founders.html.replace('\n', ''))
 
     def test_html_export_none_value(self):
         """HTML export"""
 
-        html = markup.page()
-        html.table.open()
-        html.thead.open()
-
-        html.tr(markup.oneliner.th(['foo', '', 'bar']))
-        html.thead.close()
-
-        html.tbody.open()
-        html.tr(markup.oneliner.td(['foo', '', 'bar']))
-        html.tbody.close()
-
-        html.table.close()
-        html = str(html)
-
         headers = ['foo', None, 'bar']
-        d = tablib.Dataset(['foo', None, 'bar'], headers=headers)
+        d = tablib.Dataset(['foø', None, 'bar'], headers=headers)
+        expected = (
+            "<table>"
+            "<thead>"
+            "<tr><th>foo</th><th></th><th>bar</th></tr>"
+            "</thead>"
+            "<tbody>"
+            "<tr><td>foø</td><td></td><td>bar</td></tr>"
+            "</tbody>"
+            "</table>"
+        )
+        self.assertEqual(expected, d.html.replace('\n', ''))
 
-        self.assertEqual(html, d.html)
+    def test_html_databook_export(self):
+        book = tablib.Databook()
+        book.add_sheet(self.founders)
+        book.add_sheet(self.founders)
+        self.maxDiff = None
+        self.assertEqual(
+            book.html.replace('\n', ''),
+            f"<h3>Founders</h3>{self.founders_html}<h3>Founders</h3>{self.founders_html}"
+        )
 
 
 class RSTTests(BaseTestCase):