Add: plugin for finding excess space before dot in some tags

tests/plugins/test_spaces_before_dots.py

@@ -0,0 +1,103 @@
+# SPDX-License-Identifier: GPL-3.0-or-later
+# SPDX-FileCopyrightText: 2024 Greenbone AG
+
+from pathlib import Path
+
+from tests.plugins import PluginTestCase
+from troubadix.helper import CURRENT_ENCODING
+from troubadix.plugin import LinterFix
+from troubadix.plugins.spaces_before_dots import CheckSpacesBeforeDots
+
+
+class TestSpacesBeforeDots(PluginTestCase):
+
+    def test_ok(self):
+        nasl_file = Path("/some/fake/directory/test.nasl")
+        content = """
+              script_tag(name:"summary", value:"Foo Bar.");
+              script_tag(name:"solution", value:"Foo .NET.");
+              script_tag(name:"insight", value:"Foo Bar ...");
+            """
+        fake_context = self.create_file_plugin_context(
+            nasl_file=nasl_file, file_content=content
+        )
+        plugin = CheckSpacesBeforeDots(fake_context)
+        results = list(plugin.run())
+        self.assertEqual(len(results), 0)
+
+    def test_fail(self):
+        nasl_file = Path("/some/fake/directory/test.nasl")
+        content = (
+            """
+              script_tag(name:"summary", value:"Foo Bar .");
+              script_tag(name:"vuldetect", value:"Foo Bar .");
+              script_tag(name:"insight", value:"Foo Bar .");
+              script_tag(name:"impact", value:"Foo . Bar . Foo");
+            """
+            'script_tag(name:"affected", value:"Foo\n.\nBar.");'
+            'script_tag(name:"solution", value:"Foo Bar\n.");'
+        )
+        fake_context = self.create_file_plugin_context(
+            nasl_file=nasl_file, file_content=content
+        )
+        plugin = CheckSpacesBeforeDots(fake_context)
+        results = list(plugin.run())
+        self.assertEqual(len(results), 7)
+        self.assertEqual(
+            results[0].message,
+            "value of script_tag summary has at least one occurence of excess"
+            " whitespace before a dot:\n"
+            " 'script_tag(name:"
+            '"summary", value:"Foo Bar .");'
+            "'",
+        )
+
+    def test_ignore(self):
+        nasl_file = Path("/some/fake/directory/test.inc")
+        fake_context = self.create_file_plugin_context(nasl_file=nasl_file)
+        plugin = CheckSpacesBeforeDots(fake_context)
+
+        results = list(plugin.run())
+
+        self.assertEqual(len(results), 0)
+
+    def test_fix(self):
+        with self.create_directory() as tempdir:
+            path = tempdir / "file.nasl"
+            content = (
+                """
+                  script_tag(name:"summary", value:"Foo Bar .");
+                  script_tag(name:"vuldetect", value:"Foo Bar .");
+                  script_tag(name:"insight", value:"Foo .Net Bar .");
+                  script_tag(name:"impact", value:"Foo . Bar . Foo");
+                """
+                'script_tag(name:"affected", value:"Foo\n.\nBar.");'
+                'script_tag(name:"solution", value:"Foo Bar\n.");'
+            )
+            expected_modified_content = (
+                """
+                  script_tag(name:"summary", value:"Foo Bar.");
+                  script_tag(name:"vuldetect", value:"Foo Bar.");
+                  script_tag(name:"insight", value:"Foo .Net Bar.");
+                  script_tag(name:"impact", value:"Foo. Bar. Foo");
+                """
+                'script_tag(name:"affected", value:"Foo.\nBar.");'
+                'script_tag(name:"solution", value:"Foo Bar.");'
+            )
+            path.write_text(content, encoding=CURRENT_ENCODING)
+
+            fake_context = self.create_file_plugin_context(
+                nasl_file=path, file_content=content
+            )
+
+            plugin = CheckSpacesBeforeDots(fake_context)
+
+            # keep list() to consume the iterator
+            list(plugin.run())
+
+            results = list(plugin.fix())
+            self.assertEqual(len(results), 1)
+
+            self.assertIsInstance(results[0], LinterFix)
+            modified_content = path.read_text(encoding=CURRENT_ENCODING)
+            self.assertEqual(modified_content, expected_modified_content)

troubadix/plugins/__init__.py

@@ -18,6 +18,7 @@
 from typing import Iterable, List
 
 from troubadix.plugin import FilePlugin, FilesPlugin, Plugin
+from troubadix.plugins.spaces_before_dots import CheckSpacesBeforeDots
 
 from .badwords import CheckBadwords
 from .copyright_text import CheckCopyrightText
@@ -140,6 +141,7 @@
     CheckVTFilePermissions,
     CheckVTPlacement,
     CheckWrongSetGetKBCalls,
+    CheckSpacesBeforeDots,
 ]
 
 # plugins checking all files

troubadix/plugins/spaces_before_dots.py

@@ -0,0 +1,130 @@
+# SPDX-License-Identifier: GPL-3.0-or-later
+# SPDX-FileCopyrightText: 2024 Greenbone AG
+import re
+from collections.abc import Iterator
+from pathlib import Path
+
+from troubadix.helper import CURRENT_ENCODING
+from troubadix.helper.helper import is_ignore_file
+from troubadix.helper.patterns import (
+    ScriptTag,
+    get_script_tag_pattern,
+)
+from troubadix.plugin import (
+    FileContentPlugin,
+    LinterFix,
+    LinterResult,
+    LinterWarning,
+)
+
+TAGS = [
+    ScriptTag.SUMMARY,
+    ScriptTag.VULDETECT,
+    ScriptTag.INSIGHT,
+    ScriptTag.IMPACT,
+    ScriptTag.AFFECTED,
+    ScriptTag.SOLUTION,
+]
+
+# Regex pattern to match:
+# 1. A dot preceded and/or followed by any whitespace character (floating between words)
+# 2. A dot preceded by any whitespace character at the end of the string
+PATTERN = re.compile(r"\s+\.(\s|$)")
+IGNORE = [
+    # 21.04 and 22.04 are generated and should not be touched manually
+    "21.04/",
+    "22.04/",
+    # uses dots for beginning of entry in enumeration
+    "common/2008/debian/deb_246.nasl",
+    "common/2008/debian/deb_266.nasl",
+    "common/2008/freebsd/freebsd_5e92e8a2.nasl",
+    "common/2008/freebsd/freebsdsa_cpio.nasl",
+    "common/2008/freebsd/freebsdsa_cvs2.nasl",
+    "common/2009/osc_photoGallery_sql_injection.nasl",
+    "common/2009/secpod_novell_edir_mult_vuln_jul09_lin.nasl",
+    "common/2009/secpod_novell_edir_mult_vuln_jul09_win.nasl",
+    "common/2010/freebsd/freebsd_3a7c5fc4.nasl",
+    "common/2012/freebsd/freebsd_a4a809d8.nasl",
+    "common/2015/amazon/alas-2014-455.nasl",
+    "common/2015/gb_mozilla_firefox_mult_vuln01_mar15_macosx.nasl",
+    "common/2015/gb_mozilla_firefox_mult_vuln01_mar15_win.nasl",
+    "common/2015/oracle/ELSA-2009-1619.nasl",
+    "common/2015/oracle/ELSA-2011-0586.nasl",
+    "common/2016/gb_perl_privilege_escalation_vuln_win.nasl",
+    "common/2021/dropbear/gb_dropbear_ssh_filename_vuln_may20.nasl",
+    "common/2021/eclipse/gb_jetty_GHSA-v7ff-8wcx-gmc5_lin.nasl",
+    "common/2021/eclipse/gb_jetty_GHSA-v7ff-8wcx-gmc5_win.nasl",
+    "common/gsf/2009/mandriva/gb_mandriva_MDVSA_2008_140.nasl",
+    "common/gsf/2009/mandriva/gb_mandriva_MDVSA_2008_141.nasl",
+    "common/gsf/2010/mandriva/gb_mandriva_MDVA_2010_173.nasl",
+    "common/gsf/2010/mandriva/gb_mandriva_MDVSA_2010_155.nasl",
+    "common/gsf/2010/mandriva/gb_mandriva_MDVSA_2010_155_1.nasl",
+    "common/gsf/2010/mandriva/gb_mandriva_MDVSA_2010_167.nasl",
+    "common/gsf/2020/f5/gb_f5_big_ip_K11315080.nasl",
+    "common/gsf/2020/f5/gb_f5_big_iq_K11315080.nasl",
+    "common/2022/opensuse/gb_opensuse_2022_1548_1.nasl",
+    "common/2024/opensuse/gb_opensuse_2023_3247_1.nasl",
+    "common/attic/debian/deb_232_1.nasl",
+]
+
+
+class CheckSpacesBeforeDots(FileContentPlugin):
+    name = "check_spaces_before_dots"
+
+    def check_content(
+        self, nasl_file: Path, file_content: str
+    ) -> Iterator[LinterResult]:
+        """
+        This plugin checks for excess whitespace before a dot
+        in script_tags that have full sentence values
+        """
+        self.matches = []
+        if nasl_file.suffix == ".inc" or is_ignore_file(nasl_file, IGNORE):
+            return
+        for tag in TAGS:
+            pattern = get_script_tag_pattern(tag)
+            match = pattern.search(file_content)
+            if not match:
+                continue
+
+            value = match.group("value")
+            value_start = match.start("value")
+
+            for excess_match in PATTERN.finditer(value):
+                whitespace_pos = excess_match.start() + value_start
+                self.matches.append((whitespace_pos, excess_match.group()))
+                fullmatch = match.group()
+                yield LinterWarning(
+                    f"value of script_tag {match.group('name')} has at least"
+                    " one occurence of excess whitespace before a dot:"
+                    f"\n '{fullmatch}'",
+                    file=nasl_file,
+                    plugin=self.name,
+                )
+
+    def fix(self) -> Iterator[LinterResult]:
+
+        if not self.matches:
+            return
+
+        # Sort matches by position, descending order to avoid messing up indices during replacement
+        self.matches.sort(reverse=True)
+
+        file_content = self.context.file_content
+        for pos, match_str in self.matches:
+            # Replace the match by removing the excess whitespace before the dot
+            fixed_str = re.sub(r"\s+\.", ".", match_str)
+            file_content = (
+                file_content[:pos]
+                + fixed_str
+                + file_content[pos + len(match_str) :]
+            )
+
+        with open(self.context.nasl_file, "w", encoding=CURRENT_ENCODING) as f:
+            f.write(file_content)
+
+        yield LinterFix(
+            "Excess spaces were removed",
+            file=self.context.nasl_file,
+            plugin=self.name,
+        )