Ensure text in table preview is valid

audbcards/core/dataset.py

@@ -13,9 +13,8 @@
 import audeer
 import audformat
 
+from audbcards.core import utils
 from audbcards.core.config import config
-from audbcards.core.utils import format_schemes
-from audbcards.core.utils import limit_presented_samples
 
 
 class _Dataset:
@@ -350,7 +349,7 @@ def schemes_summary(self) -> str:
         e.g. ``'speaker: [age, gender, language]'``.
 
         """
-        return format_schemes(self.header.schemes)
+        return utils.format_schemes(self.header.schemes)
 
     @functools.cached_property
     def schemes_table(self) -> typing.List[typing.List[str]]:
@@ -405,7 +404,9 @@ def tables_preview(self) -> typing.Dict[str, typing.List[typing.List[str]]]:
         Shows the header
         and the first 5 lines for each table
         as a list of lists.
-        All table values are converted to strings.
+        All table values are converted to strings,
+        stripped from HTML tags or newlines,
+        and limited to a maximum length of 100 characters.
 
         Returns:
             dictionary with table IDs as keys
@@ -434,9 +435,11 @@ def tables_preview(self) -> typing.Dict[str, typing.List[typing.List[str]]]:
                 verbose=False,
             )
             df = df.reset_index()
-            preview[table] = [df.columns.tolist()] + df.head(5).astype(
-                "string"
-            ).values.tolist()
+            header = [df.columns.tolist()]
+            body = df.head(5).astype("string").values.tolist()
+            # Remove unwanted chars and limit length of each entry
+            body = [[utils.parse_text(column) for column in row] for row in body]
+            preview[table] = header + body
         return preview
 
     @functools.cached_property
@@ -562,7 +565,7 @@ def _scheme_to_list(self, scheme_id):
                     label[:-1] + r"\_" if label.endswith("_") else label
                     for label in labels
                 ]
-                labels = limit_presented_samples(
+                labels = utils.limit_presented_samples(
                     labels,
                     15,
                     replacement_text="[...]",

audbcards/core/utils.py

@@ -1,6 +1,8 @@
+import re
 import typing
 
 import matplotlib.pyplot as plt
+import pandas as pd
 
 import audeer
 import audformat
@@ -99,6 +101,30 @@ def limit_presented_samples(
     return samples
 
 
+def parse_text(text: str) -> str:
+    """Remove unsupported characters and restrict length.
+
+    Args:
+        text: input text
+
+    Returns:
+        parsed text
+
+    """
+    # Missing text
+    if pd.isna(text):
+        return text
+    # Remove newlines
+    text = text.replace("\n", "\\n")
+    # Remove HTML tags
+    text = re.sub("<[^<]+?>", "", text)
+    # Limit length
+    max_characters_per_entry = 100
+    if len(text) > max_characters_per_entry:
+        text = text[: max_characters_per_entry - 3] + "..."
+    return text
+
+
 def set_plot_margins(
     *,
     left=0,

tests/test_utils.py

@@ -65,7 +65,19 @@ def test_format_schemes(scheme_names, scheme_dtypes, labels, expected):
     ],
 )
 def test_limit_presented_samples(sample, limit, replacement_text, expected):
-    limited_sample = audbcards.core.dataset.limit_presented_samples(
+    limited_sample = audbcards.core.utils.limit_presented_samples(
         sample, limit, replacement_text
     )
     assert limited_sample == expected
+
+
+@pytest.mark.parametrize(
+    "text, expected",
+    [
+        ("abc\ndef", "abc\\ndef"),
+        ("a" * 101, "a" * 97 + "..."),
+        ('<a href="http://www.google.de">text link</a>', "text link"),
+    ],
+)
+def test_parse_text(text, expected):
+    assert audbcards.core.utils.parse_text(text) == expected