Added additional metadata to diff output

README.md

@@ -16,7 +16,22 @@ This friction complicates the responsible acceptance of pull requests that chang
 
 ## Example
 
-<img width="916" alt="image" src="https://user-images.githubusercontent.com/1723176/224580589-bd5e7a5f-e39f-40d3-91a2-b4bd02284100.png">
+```markdown
+### Detected 6 changes to dependencies in Poetry lockfile
+
+From base f4e6ca0f4d67d9bb3f8ab43a89ceca2d0d2be7a1 to target a86b84f85d0bb2bf2fca6d6e8c58f2ce6f9e393c:
+
+Added **pydantic** (1.10.6)
+Added **requests-mock** (1.10.0)
+Added **six** (1.16.0)
+Added **tomli** (2.0.1)
+Added **typing-extensions** (4.5.0)
+Updated **urllib3** (1.26.14 -> 1.26.15)
+
+*(5 added, 0 removed, 1 updated, 4 not changed)*
+
+<small>Generated by diff-poetry-lock 1.0.1</small>
+```
 
 ## Usage
 

diff_poetry_lock/__init__.py

@@ -0,0 +1,6 @@
+from importlib.metadata import PackageNotFoundError, version
+
+try:
+    __version__ = version("diff-poetry-lock")
+except PackageNotFoundError:
+    __version__ = "dev"

diff_poetry_lock/github.py

@@ -1,7 +1,8 @@
+import base64
+
 import requests
 from loguru import logger
 from pydantic import BaseModel, Field, parse_obj_as
-from requests import Response
 
 from diff_poetry_lock.settings import PrLookupConfigurable, Settings
 
@@ -30,6 +31,7 @@ class GithubApi:
     def __init__(self, settings: Settings) -> None:
         self.s = settings
         self.session = requests.session()
+        self._ref_hash_cache: dict[str, str] = {}
         if isinstance(self.s, PrLookupConfigurable):
             self.s.set_pr_lookup_service(self)
 
@@ -45,7 +47,7 @@ def post_comment(self, comment: str) -> None:
         logger.debug("Posting comment to PR #{}", self.s.pr_num)
         r = self.session.post(
             f"{self.s.api_url}/repos/{self.s.repository}/issues/{self.s.pr_num}/comments",
-            headers={"Authorization": f"token {self.s.token}", "Accept": "application/vnd.github+json"},
+            headers=self.api_headers(),
             json={"body": f"{MAGIC_COMMENT_IDENTIFIER}{comment}"},
             timeout=10,
         )
@@ -56,7 +58,7 @@ def update_comment(self, comment_id: int, comment: str) -> None:
         logger.debug("Updating comment {}", comment_id)
         r = self.session.patch(
             f"{self.s.api_url}/repos/{self.s.repository}/issues/comments/{comment_id}",
-            headers={"Authorization": f"token {self.s.token}", "Accept": "application/vnd.github+json"},
+            headers=self.api_headers(),
             json={"body": f"{MAGIC_COMMENT_IDENTIFIER}{comment}"},
             timeout=10,
         )
@@ -74,7 +76,7 @@ def list_comments(self) -> list[GithubComment]:
             r = self.session.get(
                 f"{self.s.api_url}/repos/{self.s.repository}/issues/{self.s.pr_num}/comments",
                 params={"per_page": 100, "page": page},
-                headers={"Authorization": f"token {self.s.token}", "Accept": "application/vnd.github+json"},
+                headers=self.api_headers(),
                 timeout=10,
             )
             r.raise_for_status()
@@ -84,28 +86,47 @@ def list_comments(self) -> list[GithubComment]:
         logger.debug("Found %d comments", len(all_comments))
         return [c for c in all_comments if c.is_diff_comment()]
 
-    def get_file(self, ref: str) -> Response:
+    def get_file(self, ref: str) -> bytes:
         logger.debug("Fetching {} from ref {}", self.s.lockfile_path, ref)
 
         r = self.session.get(
             f"{self.s.api_url}/repos/{self.s.repository}/contents/{self.s.lockfile_path}",
             params={"ref": ref},
-            headers={"Authorization": f"token {self.s.token}", "Accept": "application/vnd.github.raw"},
+            headers=self.api_headers(),
             timeout=10,
-            stream=True,
         )
         logger.debug("Response status: {}", r.status_code)
 
         if r.status_code == 404:
             raise FileNotFoundError(self.s.lockfile_path) from RepoFileRetrievalError(self.s.repository, ref)
         r.raise_for_status()
-        return r
+        file_obj = r.json()
+
+        resolved_hash = str(file_obj.get("sha", "")).strip()
+        if resolved_hash:
+            self._ref_hash_cache[ref] = resolved_hash
+            logger.debug("Cached commit hash for ref {} from contents sha", ref)
+
+        encoded_content = file_obj.get("content", "")
+        if not isinstance(encoded_content, str):
+            msg = "Invalid content returned from GitHub contents API"
+            raise TypeError(msg)
+
+        return base64.b64decode(encoded_content)
+
+    def resolve_commit_hash(self, ref: str) -> str:
+        if cached_hash := self._ref_hash_cache.get(ref):
+            logger.debug("Using cached commit hash for ref {}", ref)
+            return cached_hash
+
+        logger.warning("No cached commit hash for ref {}, falling back to ref", ref)
+        return ref
 
     def delete_comment(self, comment_id: int) -> None:
         logger.debug("Deleting comment {}", comment_id)
         r = self.session.delete(
             f"{self.s.api_url}/repos/{self.s.repository}/issues/comments/{comment_id}",
-            headers={"Authorization": f"token {self.s.token}", "Accept": "application/vnd.github+json"},
+            headers=self.api_headers(),
         )
         logger.debug("Response status: {}", r.status_code)
         r.raise_for_status()
@@ -122,7 +143,7 @@ def find_pr_for_branch(self, branch_ref: str) -> str:
         r = self.session.get(
             f"{self.s.api_url}/repos/{self.s.repository}/pulls",
             params={"head": head, "state": "open"},
-            headers={"Authorization": f"token {self.s.token}", "Accept": "application/vnd.github+json"},
+            headers=self.api_headers(),
             timeout=10,
         )
         logger.debug("Response status: {}", r.status_code)
@@ -137,6 +158,13 @@ def find_pr_for_branch(self, branch_ref: str) -> str:
         logger.debug("No open PR found for branch {}", branch)
         return ""
 
+    def api_headers(self) -> dict[str, str]:
+        return self.request_headers(self.s.token)
+
+    @staticmethod
+    def request_headers(token: str) -> dict[str, str]:
+        return {"Authorization": f"Bearer {token}", "Accept": "application/vnd.github+json"}
+
     def upsert_comment(self, existing_comment: GithubComment | None, comment: str | None) -> None:
         if existing_comment is None and comment is None:
             return

diff_poetry_lock/run_poetry.py

@@ -7,6 +7,7 @@
 from poetry.core.packages.package import Package
 from poetry.packages import Locker
 
+from diff_poetry_lock import __version__
 from diff_poetry_lock.github import GithubApi
 from diff_poetry_lock.logging_utils import configure_logging
 from diff_poetry_lock.settings import Settings, determine_and_load_settings
@@ -79,7 +80,11 @@ def post_comment(api: GithubApi, comment: str | None) -> None:
     api.upsert_comment(existing_comment, comment)
 
 
-def format_comment(packages: list[PackageSummary]) -> str | None:
+def format_comment(
+    packages: list[PackageSummary],
+    base_commit_hash: str | None = None,
+    target_commit_hash: str | None = None,
+) -> str | None:
     added = sorted([p for p in packages if p.added()], key=attrgetter("name"))
     removed = sorted([p for p in packages if p.removed()], key=attrgetter("name"))
     updated = sorted([p for p in packages if p.updated()], key=attrgetter("name"))
@@ -88,20 +93,25 @@ def format_comment(packages: list[PackageSummary]) -> str | None:
     if len(added + removed + updated) == 0:
         return None
 
-    comment = f"### Detected {len(added + removed + updated)} changes to dependencies in Poetry lockfile\n\n"
-    comment += "\n".join(p.summary_line() for p in added + removed + updated)
+    change_count = len(added + removed + updated)
+    comment = f"### Detected {change_count} changes to dependencies in Poetry lockfile\n\n"
+    if base_commit_hash and target_commit_hash:
+        comment += f"From base {base_commit_hash} to target {target_commit_hash}:\n\n"
+    summary_lines = [p.summary_line() for p in added + removed + updated]
+    comment += "\n".join(summary_lines)
     comment += (
         f"\n\n*({len(added)} added, {len(removed)} removed, {len(updated)} updated, {len(not_changed)} not changed)*"
     )
+    if __version__:
+        comment += f"\n\n<small>Generated by diff-poetry-lock {__version__}</small>\n\n"
 
     return comment
 
 
 def load_lockfile(api: GithubApi, ref: str) -> list[Package]:
-    r = api.get_file(ref)
+    file_contents = api.get_file(ref)
     with tempfile.NamedTemporaryFile(mode="wb", delete=True) as f:
-        for chunk in r.iter_content(chunk_size=1024):
-            f.write(chunk)
+        f.write(file_contents)
         f.flush()
 
         return load_packages(Path(f.name))
@@ -128,7 +138,17 @@ def do_diff(settings: Settings) -> None:
 
     logger.debug("Computing diff...")
     packages = diff(base_packages, head_packages)
-    summary = format_comment(packages)
+
+    if not any(package.changed() for package in packages):
+        summary = None
+    else:
+        base_commit_hash = api.resolve_commit_hash(settings.ref)
+        target_commit_hash = api.resolve_commit_hash(settings.base_ref)
+        summary = format_comment(
+            packages,
+            base_commit_hash=base_commit_hash,
+            target_commit_hash=target_commit_hash,
+        )
 
     if summary:
         logger.debug("Generated summary with {} characters", len(summary))