[report-converter] Enable multiple input for report-converter

Sometimes dynamic analysis is executed for the code base several times
with different parameters. The output of these analyses may go to
different log files which are given to report-converter one-by-one.
If there are reports for the same source file in different log output
files then the resulting .plist files will be overwritten at a
subsequent execution of report-converter.

In this patch we make it possible to give multiple log files or
directories containing log files to the report-converter. The reports
in these files and directories will be collected all and will be
converted to .plist without overriding the ones coming from a different
log file.

tools/report-converter/codechecker_report_converter/analyzers/analyzer_result.py

@@ -13,7 +13,7 @@
 from abc import ABCMeta, abstractmethod
 from collections import defaultdict
 import hashlib
-from typing import Dict, List, Optional
+from typing import Dict, Iterable, List, Optional
 
 from codechecker_report_converter.report import Report, report_file
 from codechecker_report_converter.report.hash import get_report_hash, HashType
@@ -37,7 +37,7 @@ class AnalyzerResultBase(metaclass=ABCMeta):
 
     def transform(
         self,
-        analyzer_result_file_path: str,
+        analyzer_result_file_paths: Iterable[str],
         output_dir_path: str,
         export_type: str,
         file_name: str = "{source_file}_{analyzer}_{file_hash}",
@@ -53,23 +53,27 @@ def transform(
                       export_type)
             return False
 
-        analyzer_result_file_path = os.path.abspath(analyzer_result_file_path)
-        reports = self.get_reports(analyzer_result_file_path)
-        if not reports:
-            LOG.info("No '%s' results can be found in the given code analyzer "
-                     "output.", self.TOOL_NAME)
-            return False
+        all_reports = []
 
-        self._post_process_result(reports)
+        for file_path in analyzer_result_file_paths:
+            file_path = os.path.abspath(file_path)
+            reports = self.get_reports(file_path)
+            if not reports:
+                LOG.info("No '%s' results can be found in '%s'.",
+                         self.TOOL_NAME, file_path)
 
-        for report in reports:
-            report.analyzer_result_file_path = analyzer_result_file_path
+            self._post_process_result(reports)
+
+            for report in reports:
+                report.analyzer_result_file_path = file_path
+
+                if not report.checker_name:
+                    report.checker_name = self.TOOL_NAME
 
-            if not report.checker_name:
-                report.checker_name = self.TOOL_NAME
+            all_reports.extend(reports)
 
         self._write(
-            reports, output_dir_path, parser, export_type, file_name)
+            all_reports, output_dir_path, parser, export_type, file_name)
 
         if metadata:
             self._save_metadata(metadata, output_dir_path)
@@ -79,7 +83,7 @@ def transform(
                         "and analysis command when storing the results to it. "
                         "For more information see the --help.")
 
-        return True
+        return bool(all_reports)
 
     @abstractmethod
     def get_reports(self, analyzer_result_file_path: str) -> List[Report]:

tools/report-converter/codechecker_report_converter/cli.py

@@ -9,14 +9,15 @@
 
 
 import argparse
+from collections.abc import Iterable, Sequence
 import glob
 import importlib
 import logging
 import os
 import shutil
 import sys
 
-from typing import Dict, Optional, Tuple
+from typing import Any, Dict, Iterable, Optional, Sequence, Tuple, Union
 
 
 # If we run this script in an environment where 'codechecker_report_converter'
@@ -79,7 +80,7 @@ class RawDescriptionDefaultHelpFormatter(
 
 
 def transform_output(
-    analyzer_result: str,
+    analyzer_result: Iterable[str],
     parser_type: str,
     output_dir: str,
     file_name: str,
@@ -118,15 +119,57 @@ def process_metadata(metadata) -> Tuple[Dict[str, str], Dict[str, str]]:
     return valid_values, invalid_values
 
 
+class CollectFiles(argparse.Action):
+    """
+    This report-converter tool can be given a set of files and directories as
+    positional command line arguments. This action collects all files at the
+    given locations: if an argument is a file then that file is collected, if
+    it's a directory then its content is inspected recursively.
+    """
+    def __init__(
+        self,
+        option_strings: Sequence[str],
+        dest: str,
+        nargs: Union[int, str, None],
+        **kwargs
+    ) -> None:
+        if nargs != "+":
+            raise ValueError("'nargs' option for 'input' should be '+'.")
+        super().__init__(option_strings, dest, nargs, **kwargs)
+
+    def __call__(
+        self,
+        parser: argparse.ArgumentParser,
+        namespace: argparse.Namespace,
+        values: Union[str, Sequence[Any], None],
+        option_string: Union[str, None] = None
+    ):
+        # Type of "values" can have any of the indicated types above. But in
+        # argument parser it is given by "nargs='+'", so it will be a list in
+        # this specific case.
+        assert isinstance(values, Sequence)
+
+        all_files = []
+
+        for path in values:
+            if os.path.isfile(path):
+                all_files.append(path)
+            else:
+                for root, _, files in os.walk(path):
+                    all_files.extend(os.path.join(root, f) for f in files)
+
+        setattr(namespace, 'input', all_files)
+
+
 def __add_arguments_to_parser(parser):
     """ Add arguments to the the given parser. """
     parser.add_argument('input',
-                        type=str,
-                        metavar='file',
+                        nargs='+',
+                        action=CollectFiles,
                         default=argparse.SUPPRESS,
-                        help="Code analyzer output result file which will be "
-                             "parsed and used to generate a CodeChecker "
-                             "report directory.")
+                        help="Code analyzer output result files or "
+                             "directories which will be parsed and used to "
+                             "generate a CodeChecker report directory.")
 
     parser.add_argument('-o', '--output',
                         dest="output_dir",

tools/report-converter/codechecker_report_converter/report/parser/plist.py

@@ -80,10 +80,11 @@ def __init__(self, dict_type=dict):
         # complain that too many arguments are used to invoke __init__, but
         # with newer interpreters, this is deadcode.
         if len(params) == 3 and "use_builtin_types" in params:
+            # Before 3.9 interpreter.
             # pylint: disable=E1121
             _PlistParser.__init__(self, True, dict_type)
-        # After 3.9 interpreter.
         else:
+            # After 3.9 interpreter.
             _PlistParser.__init__(self, dict_type)  # pylint: disable=E1120
 
         self.event_handler = _LXMLPlistEventHandler()

tools/report-converter/tests/functional/cmdline/test_cmdline.py

@@ -60,3 +60,41 @@ def test_metadata(self):
                 self.assertEqual(tool['name'], "golint")
                 self.assertEqual(tool['version'], analyzer_version)
                 self.assertEqual(tool['command'], analyzer_command)
+
+    def test_input_dir(self):
+        """ Execution with file and directory inputs. """
+        def count_reports(report_dir):
+            counter = 0
+
+            for plist_file in os.listdir(report_dir):
+                plist_file = os.path.join(report_dir, plist_file)
+                with open(plist_file, encoding="utf-8", errors="ignore") as f:
+                    counter += f.read().count(
+                        "issue_hash_content_of_line_in_context")
+
+            return counter
+
+        test_dir = os.path.dirname(os.path.realpath(__file__))
+        outputs_dir = os.path.join(test_dir, "test_files", "test_outputs")
+
+        with tempfile.TemporaryDirectory() as tmp_dir:
+            ret = subprocess.call(['report-converter', '-t', 'golint',
+                                   '-o', tmp_dir, outputs_dir])
+            self.assertEqual(0, ret)
+
+            self.assertEqual(
+                len(glob.glob(os.path.join(tmp_dir, '*.plist'))), 3)
+
+            self.assertEqual(count_reports(tmp_dir), 4)
+
+        test_input1 = os.path.join(outputs_dir, "subdir", "simple2.out")
+        test_input2 = os.path.join(outputs_dir, "subdir", "simple3.out")
+        with tempfile.TemporaryDirectory() as tmp_dir:
+            ret = subprocess.call(['report-converter', '-t', 'golint',
+                                   '-o', tmp_dir, test_input1, test_input2])
+            self.assertEqual(0, ret)
+
+            self.assertEqual(
+                len(glob.glob(os.path.join(tmp_dir, '*.plist'))), 2)
+
+            self.assertEqual(count_reports(tmp_dir), 3)

tools/report-converter/tests/functional/cmdline/test_files/test_outputs/simple1.out

@@ -0,0 +1 @@
+./files/simple.go:3:6: exported type T should have comment or be unexported

tools/report-converter/tests/functional/cmdline/test_files/test_outputs/subdir/simple2.out

@@ -0,0 +1,2 @@
+./files/simple.go:5:5: exported var Z should have its own declaration
+./files/b/simple.go:3:6: exported type T2 should have comment or be unexported

tools/report-converter/tests/functional/cmdline/test_files/test_outputs/subdir/simple3.out

@@ -0,0 +1 @@
+./files/b/simple.go:5:5: exported var Z2 should have its own declaration

tools/report-converter/tests/unit/analyzers/test_asan_parser.py

@@ -52,7 +52,7 @@ def tearDown(self):
     def test_asan(self):
         """ Test for the asan.plist file. """
         self.analyzer_result.transform(
-            'asan.out', self.cc_result_dir, plist.EXTENSION,
+            ['asan.out'], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
 
         with open('asan.plist', mode='rb') as pfile:

tools/report-converter/tests/unit/analyzers/test_clang_tidy_parser.py

@@ -54,7 +54,7 @@ def __check_analyzer_result(self, analyzer_result, analyzer_result_plist,
                                 source_files, expected_plist):
         """ Check the result of the analyzer transformation. """
         self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
 
         plist_file = os.path.join(self.cc_result_dir, analyzer_result_plist)
@@ -75,13 +75,13 @@ def __check_analyzer_result(self, analyzer_result, analyzer_result_plist,
     def test_empty1(self):
         """ Test for empty Messages. """
         ret = self.analyzer_result.transform(
-            'empty1.out', self.cc_result_dir, plist.EXTENSION)
+            ['empty1.out'], self.cc_result_dir, plist.EXTENSION)
         self.assertFalse(ret)
 
     def test_empty2(self):
         """ Test for empty Messages with multiple line. """
         ret = self.analyzer_result.transform(
-            'empty2.out', self.cc_result_dir, plist.EXTENSION)
+            ['empty2.out'], self.cc_result_dir, plist.EXTENSION)
         self.assertFalse(ret)
 
     def test_tidy1(self):

tools/report-converter/tests/unit/analyzers/test_coccinelle_parser.py

@@ -43,7 +43,7 @@ def test_no_cocci_output_file(self):
                                        'sample.c')
 
         ret = self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
         self.assertFalse(ret)
 
@@ -52,15 +52,15 @@ def test_transform_dir(self):
         analyzer_result = os.path.join(self.test_files)
 
         ret = self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
         self.assertFalse(ret)
 
     def test_transform_single_file(self):
         """ Test transforming single output file. """
         analyzer_result = os.path.join(self.test_files, 'sample.out')
         self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
 
         plist_file = os.path.join(self.cc_result_dir,

tools/report-converter/tests/unit/analyzers/test_cppcheck_parser.py

@@ -59,7 +59,7 @@ def test_no_plist_file(self):
                                        'divide_zero.cpp')
 
         ret = self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
         self.assertFalse(ret)
 
@@ -68,7 +68,7 @@ def test_no_plist_dir(self):
         analyzer_result = os.path.join(self.test_files, 'non_existing')
 
         ret = self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
         self.assertFalse(ret)
 
@@ -77,7 +77,7 @@ def test_transform_single_file(self):
         analyzer_result = os.path.join(
             self.test_files, 'out', 'divide_zero.plist')
         self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
 
         plist_file = os.path.join(self.cc_result_dir,
@@ -102,7 +102,7 @@ def test_transform_directory(self):
         """ Test transforming a directory of plist files. """
         analyzer_result = os.path.join(self.test_files, 'out')
         self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
 
         plist_file = os.path.join(self.cc_result_dir,

tools/report-converter/tests/unit/analyzers/test_cpplint_parser.py

@@ -42,7 +42,7 @@ def test_no_cpplint_output_file(self):
                                        'sample.cpp')
 
         ret = self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
         self.assertFalse(ret)
 
@@ -51,15 +51,15 @@ def test_transform_dir(self):
         analyzer_result = os.path.join(self.test_files)
 
         ret = self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
         self.assertFalse(ret)
 
     def test_transform_single_file(self):
         """ Test transforming single output file. """
         analyzer_result = os.path.join(self.test_files, 'sample.out')
         self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
 
         plist_file = os.path.join(self.cc_result_dir,

tools/report-converter/tests/unit/analyzers/test_eslint_parser.py

@@ -42,7 +42,7 @@ def test_no_json_file(self):
                                        'index.js')
 
         ret = self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
         self.assertFalse(ret)
 
@@ -51,15 +51,15 @@ def test_transform_dir(self):
         analyzer_result = os.path.join(self.test_files)
 
         ret = self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
         self.assertFalse(ret)
 
     def test_transform_single_file(self):
         """ Test transforming single plist file. """
         analyzer_result = os.path.join(self.test_files, 'reports.json')
         self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
 
         plist_file = os.path.join(self.cc_result_dir,

tools/report-converter/tests/unit/analyzers/test_golint_parser.py

@@ -42,7 +42,7 @@ def test_no_go_output_file(self):
                                        'simple.go')
 
         ret = self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
         self.assertFalse(ret)
 
@@ -51,15 +51,15 @@ def test_transform_dir(self):
         analyzer_result = os.path.join(self.test_files)
 
         ret = self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
         self.assertFalse(ret)
 
     def test_transform_single_file(self):
         """ Test transforming single plist file. """
         analyzer_result = os.path.join(self.test_files, 'simple.out')
         self.analyzer_result.transform(
-            analyzer_result, self.cc_result_dir, plist.EXTENSION,
+            [analyzer_result], self.cc_result_dir, plist.EXTENSION,
             file_name="{source_file}_{analyzer}")
 
         plist_file = os.path.join(self.cc_result_dir,