Merge pull request #459 from Iamrodos/issue-93-starred-gists-warning

fix: warn and skip when --starred-gists used for different user

CHANGES.rst

@@ -1,9 +1,59 @@
 Changelog
 =========
 
-0.52.0 (2025-11-28)
+0.54.0 (2025-12-03)
 -------------------
 ------------------------
+
+Fix
+~~~
+- Send INFO/DEBUG to stdout, WARNING/ERROR to stderr. [Rodos]
+
+  Fixes #182
+
+Other
+~~~~~
+- Docs: update README testing section and add fetch vs pull explanation.
+  [Rodos]
+
+
+0.53.0 (2025-11-30)
+-------------------
+
+Fix
+~~~
+- Case-sensitive username filtering causing silent backup failures.
+  [Rodos]
+
+  GitHub's API accepts usernames in any case but returns canonical case.
+  The case-sensitive comparison in filter_repositories() filtered out all
+  repositories when user-provided case didn't match GitHub's canonical case.
+
+  Changed to case-insensitive comparison.
+
+  Fixes #198
+
+Other
+~~~~~
+- Avoid rewriting unchanged JSON files for labels, milestones, releases,
+  hooks, followers, and following. [Rodos]
+
+  This change reduces unnecessary writes when backing up metadata that changes
+  infrequently. The implementation compares existing file content before writing
+  and skips the write if the content is identical, preserving file timestamps.
+
+  Key changes:
+  - Added json_dump_if_changed() helper that compares content before writing
+  - Uses atomic writes (temp file + rename) for all metadata files
+  - NOT applied to issues/pulls (they use incremental_by_files logic)
+  - Made log messages consistent and past tense ("Saved" instead of "Saving")
+  - Added informative logging showing skip counts
+
+  Fixes #133
+
+
+0.52.0 (2025-11-28)
+-------------------
 - Skip DMCA'd repos which return a 451 response. [Rodos]
 
   Log a warning and the link to the DMCA notice. Continue backing up

README.rst

@@ -301,6 +301,8 @@ Starred gists vs starred repo behaviour
 
 The starred normal repo cloning (``--all-starred``) argument stores starred repos separately to the users own repositories. However, using ``--starred-gists`` will store starred gists within the same directory as the users own gists ``--gists``. Also, all gist repo directory names are IDs not the gist's name.
 
+Note: ``--starred-gists`` only retrieves starred gists for the authenticated user, not the target user, due to a GitHub API limitation.
+
 
 Skip existing on incomplete backups
 -----------------------------------
@@ -308,6 +310,25 @@ Skip existing on incomplete backups
 The ``--skip-existing`` argument will skip a backup if the directory already exists, even if the backup in that directory failed (perhaps due to a blocking error). This may result in unexpected missing data in a regular backup.
 
 
+Updates use fetch, not pull
+---------------------------
+
+When updating an existing repository backup, ``github-backup`` uses ``git fetch`` rather than ``git pull``. This is intentional - a backup tool should reliably download data without risk of failure. Using ``git pull`` would require handling merge conflicts, which adds complexity and could cause backups to fail unexpectedly.
+
+With fetch, **all branches and commits are downloaded** safely into remote-tracking branches. The working directory files won't change, but your backup is complete.
+
+If you look at files directly (e.g., ``cat README.md``), you'll see the old content. The new data is in the remote-tracking branches (confusingly named "remote" but stored locally). To view or use the latest files::
+
+    git show origin/main:README.md           # view a file
+    git merge origin/main                    # update working directory
+
+All branches are backed up as remote refs (``origin/main``, ``origin/feature-branch``, etc.).
+
+If you want to browse files directly without merging, consider using ``--bare`` which skips the working directory entirely - the backup is just the git data.
+
+See `#269 <https://github.com/josegonzalez/python-github-backup/issues/269>`_ for more discussion.
+
+
 Github Backup Examples
 ======================
 
@@ -357,7 +378,12 @@ A huge thanks to all the contibuters!
 Testing
 -------
 
-This project currently contains no unit tests.  To run linting::
+To run the test suite::
+
+    pip install pytest
+    pytest
+
+To run linting::
 
     pip install flake8
     flake8 --ignore=E501

bin/github-backup

@@ -16,12 +16,23 @@ from github_backup.github_backup import (
     retrieve_repositories,
 )
 
-logging.basicConfig(
+# INFO and DEBUG go to stdout, WARNING and above go to stderr
-    format="%(asctime)s.%(msecs)03d: %(message)s",
+log_format = logging.Formatter(
+    fmt="%(asctime)s.%(msecs)03d: %(message)s",
     datefmt="%Y-%m-%dT%H:%M:%S",
-    level=logging.INFO,
 )
 
+stdout_handler = logging.StreamHandler(sys.stdout)
+stdout_handler.setLevel(logging.DEBUG)
+stdout_handler.addFilter(lambda r: r.levelno < logging.WARNING)
+stdout_handler.setFormatter(log_format)
+
+stderr_handler = logging.StreamHandler(sys.stderr)
+stderr_handler.setLevel(logging.WARNING)
+stderr_handler.setFormatter(log_format)
+
+logging.basicConfig(level=logging.INFO, handlers=[stdout_handler, stderr_handler])
+
 
 def main():
     args = parse_args()

github_backup/__init__.py

@@ -1 +1 @@
-__version__ = "0.52.0"
+__version__ = "0.54.0"

github_backup/github_backup.py

@@ -1565,16 +1565,22 @@ def retrieve_repositories(args, authenticated_user):
         repos.extend(gists)
 
     if args.include_starred_gists:
-        starred_gists_template = "https://{0}/gists/starred".format(
+        if not authenticated_user.get("login") or args.user.lower() != authenticated_user["login"].lower():
-            get_github_api_host(args)
+            logger.warning(
-        )
+                "Cannot retrieve starred gists for '%s'. GitHub only allows access to the authenticated user's starred gists.",
-        starred_gists = retrieve_data(
+                args.user,
-            args, starred_gists_template, single_request=False
+            )
-        )
+        else:
-        # flag each repo as a starred gist for downstream processing
+            starred_gists_template = "https://{0}/gists/starred".format(
-        for item in starred_gists:
+                get_github_api_host(args)
-            item.update({"is_gist": True, "is_starred": True})
+            )
-        repos.extend(starred_gists)
+            starred_gists = retrieve_data(
+                args, starred_gists_template, single_request=False
+            )
+            # flag each repo as a starred gist for downstream processing
+            for item in starred_gists:
+                item.update({"is_gist": True, "is_starred": True})
+            repos.extend(starred_gists)
 
     return repos
 
@@ -1587,7 +1593,9 @@ def filter_repositories(args, unfiltered_repositories):
     repositories = []
     for r in unfiltered_repositories:
         # gists can be anonymous, so need to safely check owner
-        if r.get("owner", {}).get("login") == args.user or r.get("is_starred"):
+        # Use case-insensitive comparison to match GitHub's case-insensitive username behavior
+        owner_login = r.get("owner", {}).get("login", "")
+        if owner_login.lower() == args.user.lower() or r.get("is_starred"):
             repositories.append(r)
 
     name_regex = None
@@ -1898,11 +1906,21 @@ def backup_milestones(args, repo_cwd, repository, repos_template):
     for milestone in _milestones:
         milestones[milestone["number"]] = milestone
 
-    logger.info("Saving {0} milestones to disk".format(len(list(milestones.keys()))))
+    written_count = 0
     for number, milestone in list(milestones.items()):
         milestone_file = "{0}/{1}.json".format(milestone_cwd, number)
-        with codecs.open(milestone_file, "w", encoding="utf-8") as f:
+        if json_dump_if_changed(milestone, milestone_file):
-            json_dump(milestone, f)
+            written_count += 1
+
+    total = len(milestones)
+    if written_count == total:
+        logger.info("Saved {0} milestones to disk".format(total))
+    elif written_count == 0:
+        logger.info("{0} milestones unchanged, skipped write".format(total))
+    else:
+        logger.info("Saved {0} of {1} milestones to disk ({2} unchanged)".format(
+            written_count, total, total - written_count
+        ))
 
 
 def backup_labels(args, repo_cwd, repository, repos_template):
@@ -1955,19 +1973,17 @@ def backup_releases(args, repo_cwd, repository, repos_template, include_assets=F
             reverse=True,
         )
         releases = releases[: args.number_of_latest_releases]
-        logger.info("Saving the latest {0} releases to disk".format(len(releases)))
-    else:
-        logger.info("Saving {0} releases to disk".format(len(releases)))
 
     # for each release, store it
+    written_count = 0
     for release in releases:
         release_name = release["tag_name"]
         release_name_safe = release_name.replace("/", "__")
         output_filepath = os.path.join(
             release_cwd, "{0}.json".format(release_name_safe)
         )
-        with codecs.open(output_filepath, "w+", encoding="utf-8") as f:
+        if json_dump_if_changed(release, output_filepath):
-            json_dump(release, f)
+            written_count += 1
 
         if include_assets:
             assets = retrieve_data(args, release["assets_url"])
@@ -1984,6 +2000,17 @@ def backup_releases(args, repo_cwd, repository, repos_template, include_assets=F
                         fine=True if args.token_fine is not None else False,
                     )
 
+    # Log the results
+    total = len(releases)
+    if written_count == total:
+        logger.info("Saved {0} releases to disk".format(total))
+    elif written_count == 0:
+        logger.info("{0} releases unchanged, skipped write".format(total))
+    else:
+        logger.info("Saved {0} of {1} releases to disk ({2} unchanged)".format(
+            written_count, total, total - written_count
+        ))
+
 
 def fetch_repository(
     name,
@@ -2108,9 +2135,10 @@ def _backup_data(args, name, template, output_file, output_directory):
         mkdir_p(output_directory)
         data = retrieve_data(args, template)
 
-        logger.info("Writing {0} {1} to disk".format(len(data), name))
+        if json_dump_if_changed(data, output_file):
-        with codecs.open(output_file, "w", encoding="utf-8") as f:
+            logger.info("Saved {0} {1} to disk".format(len(data), name))
-            json_dump(data, f)
+        else:
+            logger.info("{0} {1} unchanged, skipped write".format(len(data), name))
 
 
 def json_dump(data, output_file):
@@ -2122,3 +2150,57 @@ def json_dump(data, output_file):
         indent=4,
         separators=(",", ": "),
     )
+
+
+def json_dump_if_changed(data, output_file_path):
+    """
+    Write JSON data to file only if content has changed.
+
+    Compares the serialized JSON data with the existing file content
+    and only writes if different. This prevents unnecessary file
+    modification timestamp updates and disk writes.
+
+    Uses atomic writes (temp file + rename) to prevent corruption
+    if the process is interrupted during the write.
+
+    Args:
+        data: The data to serialize as JSON
+        output_file_path: The path to the output file
+
+    Returns:
+        True if file was written (content changed or new file)
+        False if write was skipped (content unchanged)
+    """
+    # Serialize new data with consistent formatting matching json_dump()
+    new_content = json.dumps(
+        data,
+        ensure_ascii=False,
+        sort_keys=True,
+        indent=4,
+        separators=(",", ": "),
+    )
+
+    # Check if file exists and compare content
+    if os.path.exists(output_file_path):
+        try:
+            with codecs.open(output_file_path, "r", encoding="utf-8") as f:
+                existing_content = f.read()
+            if existing_content == new_content:
+                logger.debug(
+                    "Content unchanged, skipping write: {0}".format(output_file_path)
+                )
+                return False
+        except (OSError, UnicodeDecodeError) as e:
+            # If we can't read the existing file, write the new one
+            logger.debug(
+                "Error reading existing file {0}, will overwrite: {1}".format(
+                    output_file_path, e
+                )
+            )
+
+    # Write the file atomically using temp file + rename
+    temp_file = output_file_path + ".temp"
+    with codecs.open(temp_file, "w", encoding="utf-8") as f:
+        f.write(new_content)
+    os.rename(temp_file, output_file_path)  # Atomic on POSIX systems
+    return True

tests/test_json_dump_if_changed.py

@@ -0,0 +1,198 @@
+"""Tests for json_dump_if_changed functionality."""
+
+import codecs
+import json
+import os
+import tempfile
+
+import pytest
+
+from github_backup import github_backup
+
+
+class TestJsonDumpIfChanged:
+    """Test suite for json_dump_if_changed function."""
+
+    def test_writes_new_file(self):
+        """Should write file when it doesn't exist."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            output_file = os.path.join(tmpdir, "test.json")
+            test_data = {"key": "value", "number": 42}
+
+            result = github_backup.json_dump_if_changed(test_data, output_file)
+
+            assert result is True
+            assert os.path.exists(output_file)
+
+            # Verify content matches expected format
+            with codecs.open(output_file, "r", encoding="utf-8") as f:
+                content = f.read()
+                loaded = json.loads(content)
+                assert loaded == test_data
+
+    def test_skips_unchanged_file(self):
+        """Should skip write when content is identical."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            output_file = os.path.join(tmpdir, "test.json")
+            test_data = {"key": "value", "number": 42}
+
+            # First write
+            result1 = github_backup.json_dump_if_changed(test_data, output_file)
+            assert result1 is True
+
+            # Get the initial mtime
+            mtime1 = os.path.getmtime(output_file)
+
+            # Second write with same data
+            result2 = github_backup.json_dump_if_changed(test_data, output_file)
+            assert result2 is False
+
+            # File should not have been modified
+            mtime2 = os.path.getmtime(output_file)
+            assert mtime1 == mtime2
+
+    def test_writes_when_content_changed(self):
+        """Should write file when content has changed."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            output_file = os.path.join(tmpdir, "test.json")
+            test_data1 = {"key": "value1"}
+            test_data2 = {"key": "value2"}
+
+            # First write
+            result1 = github_backup.json_dump_if_changed(test_data1, output_file)
+            assert result1 is True
+
+            # Second write with different data
+            result2 = github_backup.json_dump_if_changed(test_data2, output_file)
+            assert result2 is True
+
+            # Verify new content
+            with codecs.open(output_file, "r", encoding="utf-8") as f:
+                loaded = json.load(f)
+                assert loaded == test_data2
+
+    def test_uses_consistent_formatting(self):
+        """Should use same JSON formatting as json_dump."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            output_file = os.path.join(tmpdir, "test.json")
+            test_data = {"z": "last", "a": "first", "m": "middle"}
+
+            github_backup.json_dump_if_changed(test_data, output_file)
+
+            with codecs.open(output_file, "r", encoding="utf-8") as f:
+                content = f.read()
+
+            # Check for consistent formatting:
+            # - sorted keys
+            # - 4-space indent
+            # - comma-colon-space separator
+            expected = json.dumps(
+                test_data,
+                ensure_ascii=False,
+                sort_keys=True,
+                indent=4,
+                separators=(",", ": "),
+            )
+            assert content == expected
+
+    def test_atomic_write_always_used(self):
+        """Should always use temp file and rename for atomic writes."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            output_file = os.path.join(tmpdir, "test.json")
+            test_data = {"key": "value"}
+
+            result = github_backup.json_dump_if_changed(test_data, output_file)
+
+            assert result is True
+            assert os.path.exists(output_file)
+
+            # Temp file should not exist after atomic write
+            temp_file = output_file + ".temp"
+            assert not os.path.exists(temp_file)
+
+            # Verify content
+            with codecs.open(output_file, "r", encoding="utf-8") as f:
+                loaded = json.load(f)
+                assert loaded == test_data
+
+    def test_handles_unicode_content(self):
+        """Should correctly handle Unicode content."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            output_file = os.path.join(tmpdir, "test.json")
+            test_data = {
+                "emoji": "🚀",
+                "chinese": "你好",
+                "arabic": "مرحبا",
+                "cyrillic": "Привет",
+            }
+
+            result = github_backup.json_dump_if_changed(test_data, output_file)
+            assert result is True
+
+            # Verify Unicode is preserved
+            with codecs.open(output_file, "r", encoding="utf-8") as f:
+                loaded = json.load(f)
+                assert loaded == test_data
+
+            # Second write should skip
+            result2 = github_backup.json_dump_if_changed(test_data, output_file)
+            assert result2 is False
+
+    def test_handles_complex_nested_data(self):
+        """Should handle complex nested data structures."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            output_file = os.path.join(tmpdir, "test.json")
+            test_data = {
+                "users": [
+                    {"id": 1, "name": "Alice", "tags": ["admin", "user"]},
+                    {"id": 2, "name": "Bob", "tags": ["user"]},
+                ],
+                "metadata": {"version": "1.0", "nested": {"deep": {"value": 42}}},
+            }
+
+            result = github_backup.json_dump_if_changed(test_data, output_file)
+            assert result is True
+
+            # Verify structure is preserved
+            with codecs.open(output_file, "r", encoding="utf-8") as f:
+                loaded = json.load(f)
+                assert loaded == test_data
+
+    def test_overwrites_on_unicode_decode_error(self):
+        """Should overwrite if existing file has invalid UTF-8."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            output_file = os.path.join(tmpdir, "test.json")
+            test_data = {"key": "value"}
+
+            # Write invalid UTF-8 bytes
+            with open(output_file, "wb") as f:
+                f.write(b"\xff\xfe invalid utf-8")
+
+            # Should catch UnicodeDecodeError and overwrite
+            result = github_backup.json_dump_if_changed(test_data, output_file)
+            assert result is True
+
+            # Verify new content was written
+            with codecs.open(output_file, "r", encoding="utf-8") as f:
+                loaded = json.load(f)
+                assert loaded == test_data
+
+    def test_key_order_independence(self):
+        """Should treat differently-ordered dicts as same if keys/values match."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            output_file = os.path.join(tmpdir, "test.json")
+
+            # Write first dict
+            data1 = {"z": 1, "a": 2, "m": 3}
+            github_backup.json_dump_if_changed(data1, output_file)
+
+            # Try to write same data but different order
+            data2 = {"a": 2, "m": 3, "z": 1}
+            result = github_backup.json_dump_if_changed(data2, output_file)
+
+            # Should skip because content is the same (keys are sorted)
+            assert result is False
+
+
+if __name__ == "__main__":
+    pytest.main([__file__, "-v"])