fix: removed weird descpritions

convert.py

@@ -56,7 +56,14 @@ def _clean_html(text):
     # remove extra spaces at the beginning of the text
     lines = [line.strip() for line in text.split("\n")]
 
-    return "\n".join(lines)
+    output = "\n".join(lines)
+
+    output = output.strip()
+
+    if output == "null":
+        return ""
+
+    return output
 
 
 def _clean_image(image):
@@ -107,11 +114,8 @@ def run():
             "text": _clean_html(description),
         }
 
-        # Skipping rows with empty texts
-        if not obj["text"]:
-            continue
-
-        rows.append(obj)
+        if obj["text"]:
+            rows.append(obj)
 
     # Add a label to each row
     _categories = list(categories)

data/all/.~lock.train.csv#

@@ -0,0 +1 @@
+,marcio,marcio-AX370-Gaming,10.04.2024 19:56,file:///home/marcio/.config/libreoffice/4;

data/all/test.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9f4467feb81a1d13d44286edba2c20fd336dff55d159bc118b8b97f9a3b6010
-size 168228486
+oid sha256:fb60f014952b4e1ae53fb3788e3e2b3a76e927ebf65931d87f2f8bcf5d22ad73
+size 168573267

data/all/train.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:906253e12209cfa45dad42d4465e010fda158a2eed65f9632611eacd1f0a0253
-size 674478545
+oid sha256:b8a9cf578e8a092d61ed556cebecdf96f3a2a13c8773792e6facbf03f36d8239
+size 674112358

data/top4/test.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6905b392ca173cc15fcc75c3242bd4d3ebb5689c1366835e40674dfb7e828a74
-size 86414336
+oid sha256:8fb2c8e2218bd3e976e147bbefc3bfa6b5575bb29966910a7d835bd1e5cd5d6a
+size 85898679

data/top4/train.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f0633bceac2f5775eb0d2fb456cc62edd203af9125d01d85e0ea9ff7d729bc7a
-size 258077981
+oid sha256:f52847721502d0f473e6eac9e7704cb13ddf325ab9c36a79ec811c19e2614f56
+size 258202157

test_convert.py

@@ -0,0 +1,24 @@
+"""
+Module to test some previously found bugs in the convert.py file.
+"""
+
+from unittest import TestCase
+
+
+from convert import _clean_html
+
+
+class TestConvert(TestCase):
+    """
+    Class to test the convert.py file.
+    """
+
+    def test_clean_description(self):
+        """
+        Test some weird description which contains a "null" value after parsed,
+        which causes problem when parsed by a CSV parser.
+        """
+        description = 'null\n<div class=\'nifad\'><a href="http://www.pheedo.com/click.phdo?x=a8a276f7a37a42499f8b1c69f9acf191&u=182589199"><img src="http://www.pheedo.com/img.phdo?x=a8a276f7a37a42499f8b1c69f9acf191&u=182589199" border="0"/></a></div>'
+        parsed = _clean_html(description)
+
+        self.assertEqual(parsed, "")