feat: 'debug data' now enumerates combinable files also

4 files changed, 89 insertions, 43 deletions

diff --git a/CHANGES.rst b/CHANGES.rst
index 5aaa0c4a..65f0c21a 100644
--- a/CHANGES.rst
+++ b/CHANGES.rst
@@ -28,6 +28,9 @@ Unreleased
 - Fix: The HTML report now will not overwrite a .gitignore file that already
   exists in the HTML output directory (follow-on for `issue 1244`_).
 
+- Debug: The `coverage debug data` command will now sniff out combinable data
+  files, and report on all of them.
+
 
 .. _changes_612:
 
diff --git a/coverage/cmdline.py b/coverage/cmdline.py
index e996ffff..f3a466e3 100644
--- a/coverage/cmdline.py
+++ b/coverage/cmdline.py
@@ -17,7 +17,7 @@ import coverage
 from coverage import Coverage
 from coverage import env
 from coverage.collector import CTracer
-from coverage.data import line_counts
+from coverage.data import CoverageData, combinable_files, line_counts
 from coverage.debug import info_formatter, info_header, short_stack
 from coverage.exceptions import BaseCoverageException, ExceptionDuringRun, NoSource
 from coverage.execfile import PyRunner
@@ -601,8 +601,8 @@ class CoverageScript:
         elif options.action == "combine":
             if options.append:
                 self.coverage.load()
-            data_dirs = args or None
-            self.coverage.combine(data_dirs, strict=True, keep=bool(options.keep))
+            data_paths = args or None
+            self.coverage.combine(data_paths, strict=True, keep=bool(options.keep))
             self.coverage.save()
             return OK
 
@@ -786,23 +786,12 @@ class CoverageScript:
                 for line in info_formatter(sys_info):
                     print(f" {line}")
             elif info == 'data':
-                self.coverage.load()
-                data = self.coverage.get_data()
                 print(info_header("data"))
-                print(f"path: {data.data_filename()}")
-                if data:
-                    print(f"has_arcs: {data.has_arcs()!r}")
-                    summary = line_counts(data, fullpath=True)
-                    filenames = human_sorted(summary.keys())
-                    print(f"\n{len(filenames)} files:")
-                    for f in filenames:
-                        line = f"{f}: {summary[f]} lines"
-                        plugin = data.file_tracer(f)
-                        if plugin:
-                            line += f" [{plugin}]"
-                        print(line)
-                else:
-                    print("No data collected")
+                data_file = self.coverage.config.data_file
+                self.do_debug_data_file(data_file)
+                for filename in combinable_files(data_file):
+                    print("-----")
+                    self.do_debug_data_file(filename)
             elif info == 'config':
                 print(info_header("config"))
                 config_info = self.coverage.config.__dict__.items()
@@ -817,6 +806,27 @@ class CoverageScript:
 
         return OK
 
+    def do_debug_data_file(self, filename):
+        """Implementation of 'coverage debug data'."""
+        data = CoverageData(filename)
+        filename = data.data_filename()
+        print(f"path: {filename}")
+        if not os.path.exists(filename):
+            print("No data collected: file doesn't exist")
+            return
+        data.read()
+        print(f"has_arcs: {data.has_arcs()!r}")
+        summary = line_counts(data, fullpath=True)
+        filenames = human_sorted(summary.keys())
+        nfiles = len(filenames)
+        print(f"{nfiles} file{'' if nfiles == 1 else 's'}:")
+        for f in filenames:
+            line = f"{f}: {summary[f]} line{'' if summary[f] == 1 else 's'}"
+            plugin = data.file_tracer(f)
+            if plugin:
+                line += f" [{plugin}]"
+            print(line)
+
 
 def unshell_list(s):
     """Turn a command-line argument into a list."""
diff --git a/coverage/data.py b/coverage/data.py
index 68ba7ec3..e7c94b4f 100644
--- a/coverage/data.py
+++ b/coverage/data.py
@@ -53,11 +53,36 @@ def add_data_to_hash(data, filename, hasher):
     hasher.update(data.file_tracer(filename))
 
 
+def combinable_files(data_file, data_paths=None):
+    """Make a list of data files to be combined.
+
+    `data_file` is a path to a data file.  `data_paths` is a list of files or
+    directories of files.
+
+    Returns a list of absolute file paths.
+    """
+    data_dir, local = os.path.split(os.path.abspath(data_file))
+
+    data_paths = data_paths or [data_dir]
+    files_to_combine = []
+    for p in data_paths:
+        if os.path.isfile(p):
+            files_to_combine.append(os.path.abspath(p))
+        elif os.path.isdir(p):
+            pattern = os.path.join(os.path.abspath(p), f"{local}.*")
+            files_to_combine.extend(glob.glob(pattern))
+        else:
+            raise CoverageException(f"Couldn't combine from non-existent path '{p}'")
+    return files_to_combine
+
+
 def combine_parallel_data(
     data, aliases=None, data_paths=None, strict=False, keep=False, message=None,
 ):
     """Combine a number of data files together.
 
+    `data` is a CoverageData.
+
     Treat `data.filename` as a file prefix, and combine the data from all
     of the data files starting with that prefix plus a dot.
 
@@ -79,21 +104,7 @@ def combine_parallel_data(
     raised.
 
     """
-    # Because of the os.path.abspath in the constructor, data_dir will
-    # never be an empty string.
-    data_dir, local = os.path.split(data.base_filename())
-    localdot = local + '.*'
-
-    data_paths = data_paths or [data_dir]
-    files_to_combine = []
-    for p in data_paths:
-        if os.path.isfile(p):
-            files_to_combine.append(os.path.abspath(p))
-        elif os.path.isdir(p):
-            pattern = os.path.join(os.path.abspath(p), localdot)
-            files_to_combine.extend(glob.glob(pattern))
-        else:
-            raise CoverageException(f"Couldn't combine from non-existent path '{p}'")
+    files_to_combine = combinable_files(data.base_filename(), data_paths)
 
     if strict and not files_to_combine:
         raise CoverageException("No data to combine")
diff --git a/tests/test_cmdline.py b/tests/test_cmdline.py
index 1d87c19a..112e5d68 100644
--- a/tests/test_cmdline.py
+++ b/tests/test_cmdline.py
@@ -817,24 +817,46 @@ class CmdLineWithFilesTest(BaseCmdLineTest):
         data.write()
 
         self.command_line("debug data")
-        assert self.stdout() == textwrap.dedent("""\
+        assert self.stdout() == textwrap.dedent(f"""\
             -- data ------------------------------------------------------
-            path: FILENAME
+            path: {data.data_filename()}
             has_arcs: False
-
             2 files:
             file1.py: 17 lines [a_plugin]
             file2.py: 23 lines
-            """).replace("FILENAME", data.data_filename())
+            """)
 
-    def test_debug_data_with_no_data(self):
+    def test_debug_data_with_no_data_file(self):
         data = CoverageData()
         self.command_line("debug data")
-        assert self.stdout() == textwrap.dedent("""\
+        assert self.stdout() == textwrap.dedent(f"""\
             -- data ------------------------------------------------------
-            path: FILENAME
-            No data collected
-            """).replace("FILENAME", data.data_filename())
+            path: {data.data_filename()}
+            No data collected: file doesn't exist
+            """)
+
+    def test_debug_combinable_data(self):
+        data1 = CoverageData()
+        data1.add_lines({"file1.py": range(1, 18), "file2.py": [1]})
+        data1.write()
+        data2 = CoverageData(suffix="123")
+        data2.add_lines({"file2.py": range(1, 10)})
+        data2.write()
+
+        self.command_line("debug data")
+        assert self.stdout() == textwrap.dedent(f"""\
+            -- data ------------------------------------------------------
+            path: {data1.data_filename()}
+            has_arcs: False
+            2 files:
+            file1.py: 17 lines
+            file2.py: 1 line
+            -----
+            path: {data2.data_filename()}
+            has_arcs: False
+            1 file:
+            file2.py: 9 lines
+            """)
 
 
 class CmdLineStdoutTest(BaseCmdLineTest):