Allow excluding certain usernames from changelog (#128)

* Allow excluding certain usernames from changelog

Also expand known bot list to include `github-actions[bot]` and `changeset-bot`.

* Document ignoring multiple usernames

* Use sets/lists, depending on whether we need duplicates

* Make work with --all

* Add test for --ignore-contributor

* Apply suggestion from @bsipocz

Co-authored-by: Brigitta Sipőcz <b.sipocz@gmail.com>

* Allow wildcards

* Sort contributors to each PR

* Reformat

* Fix spelling of commenter

* Sort order: PR author, then others

* Move minimum supported Python to 3.10

* Apply suggestion from @choldgraf

---------

Co-authored-by: Chris Holdgraf <choldgraf@gmail.com>
Co-authored-by: Brigitta Sipőcz <b.sipocz@gmail.com>

Author: 3 people

.github/workflows/tests.yaml

@@ -36,7 +36,7 @@ jobs:
         include:
           # Only test oldest supported and latest python version to reduce
           # GitHub API calls, as they can get rate limited
-          - python-version: 3.9
+          - python-version: "3.10"
           - python-version: 3.x
 
     steps:

docs/use.md

@@ -134,6 +134,19 @@ To include Issues and Pull Requests that were _opened_ in a time period, use the
 
 (use:token)=
 
+## Remove bots from the changelog
+
+`github-activity` ships with a known list of bot usernames, but your project may use ones not on our list.
+To ignore additional usernames from the changelog, use the `--ignore-contributor` flag:
+
+```
+github-activity ... --ignore-contributor robot-one --ignore-contributor 'robot-two*'
+```
+
+Wildcards are matched as per [filename matching semantics](https://docs.python.org/3/library/fnmatch.html#fnmatch.fnmatch).
+
+If this is a generic bot username, consider contributing it back to [our list](https://github.com/executablebooks/github-activity/blob/main/github_activity/github_activity.py#L73).
+
 ## Use a GitHub API token
 
 `github-activity` uses the GitHub API to pull information about a repository's activity.

github_activity/cli.py

@@ -21,6 +21,7 @@
     "include-opened": False,
     "strip-brackets": False,
     "all": False,
+    "ignore-contributor": [],
 }
 
 parser = argparse.ArgumentParser(description=DESCRIPTION)
@@ -130,6 +131,11 @@
     action="store_true",
     help=("""Whether to include all the GitHub tags"""),
 )
+parser.add_argument(
+    "--ignore-contributor",
+    action="append",
+    help="Do not include this GitHub username as a contributor in the changelog",
+)
 
 # Hidden argument so that target can be optionally passed as a positional argument
 parser.add_argument(
@@ -214,6 +220,7 @@ def main():
         include_opened=bool(args.include_opened),
         strip_brackets=bool(args.strip_brackets),
         branch=args.branch,
+        ignored_contributors=args.ignore_contributor,
     )
 
     if args.all:

github_activity/github_activity.py

@@ -1,5 +1,7 @@
 """Use the GraphQL api to grab issues/PRs that match a query."""
+import dataclasses
 import datetime
+import fnmatch
 import os
 import re
 import shlex
@@ -69,25 +71,26 @@
 }
 
 # exclude known bots from contributor lists
-# TODO: configurable? Everybody's got their own bots.
+# Also see 'ignore-contributor' flag/configuration option.
 BOT_USERS = {
-    "codecov",
-    "codecov-io",
-    "dependabot",
-    "github-actions",
-    "henchbot",
-    "jupyterlab-dev-mode",
-    "lgtm-com",
-    "meeseeksmachine",
-    "names",
-    "now",
-    "pre-commit-ci",
-    "renovate",
-    "review-notebook-app",
-    "support",
-    "stale",
-    "todo",
-    "welcome",
+    "changeset-bot*",
+    "codecov*",
+    "codecov-io*",
+    "dependabot*",
+    "github-actions*",
+    "henchbot*",
+    "jupyterlab-dev-mode*",
+    "lgtm-com*",
+    "meeseeksmachine*",
+    "names*",
+    "now*",
+    "pre-commit-ci*",
+    "renovate*",
+    "review-notebook-app*",
+    "support*",
+    "stale*",
+    "todo*",
+    "welcome*",
 }
 
 
@@ -225,6 +228,7 @@ def generate_all_activity_md(
     include_opened=False,
     strip_brackets=False,
     branch=None,
+    ignored_contributors: list[str] = None,
 ):
     """Generate a full markdown changelog of GitHub activity of a repo based on release tags.
 
@@ -259,6 +263,8 @@ def generate_all_activity_md(
         E.g., [MRG], [DOC], etc.
     branch : string | None
         The branch or reference name to filter pull requests by.
+    ignored_contributors : list
+        List of usernames not to include in the changelog.
 
     Returns
     -------
@@ -322,6 +328,7 @@ def filter(datum):
             include_opened=include_opened,
             strip_brackets=strip_brackets,
             branch=branch,
+            ignored_contributors=ignored_contributors,
         )
 
         if not md:
@@ -337,6 +344,26 @@ def filter(datum):
     return output
 
 
+@dataclasses.dataclass(slots=True)
+class ContributorSet:
+    """This class represents a sorted set of PR contributor usernames.
+
+    The sorting is special, in that the author is placed first.
+    """
+
+    author: str = ""
+    other: set = dataclasses.field(default_factory=set)
+
+    def add(self, contributor):
+        self.other.add(contributor)
+
+    def __iter__(self):
+        if self.author:
+            yield self.author
+        for item in sorted(self.other - {self.author}):
+            yield item
+
+
 def generate_activity_md(
     target,
     since=None,
@@ -349,6 +376,7 @@ def generate_activity_md(
     strip_brackets=False,
     heading_level=1,
     branch=None,
+    ignored_contributors: list[str] = None,
 ):
     """Generate a markdown changelog of GitHub activity within a date window.
 
@@ -418,30 +446,41 @@ def generate_activity_md(
     comment_response_cutoff = 6  # Comments on a single issue
     comment_others_cutoff = 2  # Comments on issues somebody else has authored
     comment_helpers = []
-    all_contributors = []
+    all_contributors = set()
     # add column for participants in each issue (not just original author)
     data["contributors"] = [[]] * len(data)
+
+    def ignored_user(username):
+        return any(fnmatch.fnmatch(username, bot) for bot in BOT_USERS) or any(
+            fnmatch.fnmatch(username, user) for user in ignored_contributors
+        )
+
+    def filter_ignored(userlist):
+        return {user for user in userlist if not ignored_user(user)}
+
     for ix, row in data.iterrows():
-        item_commentors = []
-        item_contributors = []
+        # Track contributors to this PR
+        item_contributors = ContributorSet()
+
+        # This is a list, since we *want* duplicates in here—they
+        # indicate number of times a contributor commented
+        item_commenters = []
 
         # contributor order:
         # - author
         # - committers
         # - merger
         # - reviewers
 
-        item_contributors.append(row.author)
+        item_contributors.author = row.author
 
         if row.kind == "pr":
-            for committer in row.committers:
-                if committer not in row.committers and committer not in BOT_USERS:
-                    item_contributors.append(committer)
+            for committer in filter_ignored(row.committers):
+                item_contributors.add(committer)
             if row.mergedBy and row.mergedBy != row.author:
-                item_contributors.append(row.mergedBy)
-            for reviewer in row.reviewers:
-                if reviewer not in item_contributors:
-                    item_contributors.append(reviewer)
+                item_contributors.add(row.mergedBy)
+            for reviewer in filter_ignored(row.reviewers):
+                item_contributors.add(reviewer)
 
         for icomment in row["comments"]["edges"]:
             comment_author = icomment["node"]["author"]
@@ -451,36 +490,37 @@ def generate_activity_md(
                 continue
 
             comment_author = comment_author["login"]
-            if comment_author in BOT_USERS:
+            if ignored_user(comment_author):
                 # ignore bots
                 continue
 
-            # Add to list of commentors on items they didn't author
+            # Add to list of commenters on items they didn't author
             if comment_author != row["author"]:
                 comment_helpers.append(comment_author)
 
-            # Add to list of commentors for this item so we can see how many times they commented
-            item_commentors.append(comment_author)
+            # Add to list of commenters for this item so we can see how many times they commented
+            item_commenters.append(comment_author)
 
             # count all comments on a PR as a contributor
-            if comment_author not in item_contributors:
-                item_contributors.append(comment_author)
+            item_contributors.add(comment_author)
 
-        # Count any commentors that had enough comments on the issue to be a contributor
-        item_commentors_counts = pd.Series(item_commentors).value_counts()
-        item_commentors_counts = item_commentors_counts[
-            item_commentors_counts >= comment_response_cutoff
+        # Count any commenters that had enough comments on the issue to be a contributor
+        item_commenters_counts = pd.Series(item_commenters).value_counts()
+        item_commenters_counts = item_commenters_counts[
+            item_commenters_counts >= comment_response_cutoff
         ].index.tolist()
-        for person in item_commentors_counts:
-            all_contributors.append(person)
+        for person in item_commenters_counts:
+            all_contributors.add(person)
 
         # record contributor list (ordered, unique)
-        data.at[ix, "contributors"] = item_contributors
+        data.at[ix, "contributors"] = list(item_contributors)
 
     comment_contributor_counts = pd.Series(comment_helpers).value_counts()
-    all_contributors += comment_contributor_counts[
-        comment_contributor_counts >= comment_others_cutoff
-    ].index.tolist()
+    all_contributors |= set(
+        comment_contributor_counts[
+            comment_contributor_counts >= comment_others_cutoff
+        ].index.tolist()
+    )
 
     # Filter the PRs by branch (or ref) if given
     if branch is not None:
@@ -517,7 +557,7 @@ def generate_activity_md(
     closed_prs = closed_prs.query("state != 'CLOSED'")
 
     # Add any contributors to a merged PR to our contributors list
-    all_contributors += closed_prs["contributors"].explode().unique().tolist()
+    all_contributors |= set(closed_prs["contributors"].explode().unique().tolist())
 
     # Define categories for a few labels
     if tags is None:
@@ -615,7 +655,7 @@ def generate_activity_md(
                     [
                         f"[@{user}](https://github.com/{user})"
                         for user in irowdata.contributors
-                        if user not in BOT_USERS
+                        if not ignored_user(user)
                     ]
                 )
                 this_md = f"- {ititle} [#{irowdata['number']}]({irowdata['url']}) ({contributor_list})"
@@ -663,7 +703,7 @@ def generate_activity_md(
 
     # Add a list of author contributions
     all_contributors = sorted(
-        set(all_contributors) - BOT_USERS, key=lambda a: str(a).lower()
+        filter_ignored(all_contributors), key=lambda a: str(a).lower()
     )
     all_contributor_links = []
     for iauthor in all_contributors:

tests/test_cli.py

@@ -117,3 +117,28 @@ def test_cli_all(tmpdir, file_regression):
     md = path_output.read_text()
     index = md.index("## v0.2.0")
     file_regression.check(md[index:], extension=".md")
+
+
+def test_cli_ignore_user(tmpdir):
+    """Test that a full changelog is created"""
+    path_tmp = Path(tmpdir)
+    path_output = path_tmp.joinpath("out.md")
+    cmd = f"github-activity executablebooks/github-activity --ignore-contributor choldgraf -s v1.0.2 -o {path_output}"
+    run(cmd.split(), check=True)
+    md = path_output.read_text()
+    assert not "@choldgraf" in md
+
+
+def test_contributor_sorting(tmpdir, file_regression):
+    """Test that PR author appears first, then rest of contributors, sorted"""
+    path_tmp = Path(tmpdir)
+    path_output = path_tmp.joinpath("out.md")
+
+    org, repo = ("jupyter-book", "mystmd")
+
+    cmd = (
+        f"github-activity {org}/{repo} -s mystmd@1.5.1 -u mystmd@1.6.0 -o {path_output}"
+    )
+    run(cmd.split(), check=True)
+    md = path_output.read_text()
+    file_regression.check(md, extension=".md")

tests/test_cli/test_contributor_sorting.md

@@ -0,0 +1,32 @@
+# mystmd@1.5.1...mystmd@1.6.0
+
+([full changelog](https://github.com/jupyter-book/mystmd/compare/mystmd@1.5.1...mystmd@1.6.0))
+
+## Enhancements made
+
+- 🎯 Render static HTML pages to expected server path [#2178](https://github.com/jupyter-book/mystmd/pull/2178) ([@stefanv](https://github.com/stefanv), [@agoose77](https://github.com/agoose77), [@bsipocz](https://github.com/bsipocz), [@choldgraf](https://github.com/choldgraf), [@rowanc1](https://github.com/rowanc1))
+- 🔗 Fix URLs in table of contents directive [#2140](https://github.com/jupyter-book/mystmd/pull/2140) ([@brianhawthorne](https://github.com/brianhawthorne), [@rowanc1](https://github.com/rowanc1), [@stefanv](https://github.com/stefanv))
+
+## Bugs fixed
+
+- 🏷️ Add NPM binary name to whitelabelling [#2175](https://github.com/jupyter-book/mystmd/pull/2175) ([@agoose77](https://github.com/agoose77), [@rowanc1](https://github.com/rowanc1), [@stefanv](https://github.com/stefanv))
+- Add `ipynb` format option in validators [#2159](https://github.com/jupyter-book/mystmd/pull/2159) ([@kp992](https://github.com/kp992), [@agoose77](https://github.com/agoose77))
+
+## Documentation improvements
+
+- 📖 Remove out of date readme note [#2155](https://github.com/jupyter-book/mystmd/pull/2155) ([@rowanc1](https://github.com/rowanc1), [@choldgraf](https://github.com/choldgraf))
+- 📖 A few miscellaneous documentation updates [#2154](https://github.com/jupyter-book/mystmd/pull/2154) ([@rowanc1](https://github.com/rowanc1))
+
+## Other merged PRs
+
+- 🚀 Release [#2180](https://github.com/jupyter-book/mystmd/pull/2180) ([@rowanc1](https://github.com/rowanc1))
+- 🚀 Release [#2130](https://github.com/jupyter-book/mystmd/pull/2130) ([@rowanc1](https://github.com/rowanc1))
+
+## Contributors to this release
+
+The following people contributed discussions, new ideas, code and documentation contributions, and review.
+See [our definition of contributors](https://github-activity.readthedocs.io/en/latest/#how-does-this-tool-define-contributions-in-the-reports).
+
+([GitHub contributors page for this release](https://github.com/jupyter-book/mystmd/graphs/contributors?from=2025-07-05&to=2025-07-21&type=c))
+
+@agoose77 ([activity](https://github.com/search?q=repo%3Ajupyter-book%2Fmystmd+involves%3Aagoose77+updated%3A2025-07-05..2025-07-21&type=Issues)) | @brian-rose ([activity](https://github.com/search?q=repo%3Ajupyter-book%2Fmystmd+involves%3Abrian-rose+updated%3A2025-07-05..2025-07-21&type=Issues)) | @brianhawthorne ([activity](https://github.com/search?q=repo%3Ajupyter-book%2Fmystmd+involves%3Abrianhawthorne+updated%3A2025-07-05..2025-07-21&type=Issues)) | @bsipocz ([activity](https://github.com/search?q=repo%3Ajupyter-book%2Fmystmd+involves%3Absipocz+updated%3A2025-07-05..2025-07-21&type=Issues)) | @choldgraf ([activity](https://github.com/search?q=repo%3Ajupyter-book%2Fmystmd+involves%3Acholdgraf+updated%3A2025-07-05..2025-07-21&type=Issues)) | @kp992 ([activity](https://github.com/search?q=repo%3Ajupyter-book%2Fmystmd+involves%3Akp992+updated%3A2025-07-05..2025-07-21&type=Issues)) | @rowanc1 ([activity](https://github.com/search?q=repo%3Ajupyter-book%2Fmystmd+involves%3Arowanc1+updated%3A2025-07-05..2025-07-21&type=Issues)) | @stefanv ([activity](https://github.com/search?q=repo%3Ajupyter-book%2Fmystmd+involves%3Astefanv+updated%3A2025-07-05..2025-07-21&type=Issues))