ai-cr 0.5.0__py3-none-any.whl → 0.6.0__py3-none-any.whl

ai_code_review/__main__.py

@@ -1,4 +1,4 @@
-from .cli import main
-
-if __name__ == "__main__":
-    main()
+from .cli import main
+
+if __name__ == "__main__":
+    main()

ai_code_review/bootstrap.py

@@ -59,4 +59,4 @@ def bootstrap():
     except Exception as e:
         logging.error(f"Unexpected configuration error: {e}")
         raise SystemExit(3)
-    mc.logging.LoggingConfig.STRIP_REQUEST_LINES = [100, 15]
+    mc.logging.LoggingConfig.STRIP_REQUEST_LINES = [300, 15]

ai_code_review/cli.py

@@ -2,35 +2,33 @@ import asyncio
 import logging
 import sys
 import os
-import shutil
+import textwrap
+import tempfile
 import requests
 
 import microcore as mc
-import async_typer
 import typer
-from ai_code_review.utils import parse_refs_pair
 from git import Repo
 
-from .core import review
+from .core import review, get_diff, filter_diff
 from .report_struct import Report
 from .constants import ENV_CONFIG_FILE
 from .bootstrap import bootstrap
 from .project_config import ProjectConfig
-from .utils import is_app_command_invocation
+from .utils import no_subcommand, parse_refs_pair
 
-
-app = async_typer.AsyncTyper(pretty_exceptions_show_locals=False)
-default_command_app = async_typer.AsyncTyper(pretty_exceptions_show_locals=False)
-if sys.platform == "win32":
-    asyncio.set_event_loop_policy(asyncio.WindowsSelectorEventLoopPolicy())
+app = typer.Typer(pretty_exceptions_show_locals=False)
+app_no_subcommand = typer.Typer(pretty_exceptions_show_locals=False)
 
 
 def main():
-    if is_app_command_invocation(app):
-        app()
-    else:
+    if sys.platform == "win32":
+        asyncio.set_event_loop_policy(asyncio.WindowsSelectorEventLoopPolicy())
+    if no_subcommand(app):
         bootstrap()
-        default_command_app()
+        app_no_subcommand()
+    else:
+        app()
 
 
 @app.callback(invoke_without_command=True)
@@ -39,27 +37,7 @@ def cli(ctx: typer.Context):
         bootstrap()
 
 
-@default_command_app.async_command(name="review", help="Perform code review")
-@app.async_command(name="review", help="Perform code review")
-async def cmd_review(
-    refs: str = typer.Argument(
-        default=None,
-        help="Git refs to review, [what]..[against] e.g. 'HEAD..HEAD~1'"
-    ),
-    what: str = typer.Option(None, "--what", "-w", help="Git ref to review"),
-    against: str = typer.Option(
-        None,
-        "--against", "-vs", "--vs",
-        help="Git ref to compare against"
-    ),
-    filters: str = typer.Option(
-        "", "--filter", "-f", "--filters",
-        help="""
-        filter reviewed files by glob / fnmatch pattern(s),
-        e.g. 'src/**/*.py', may be comma-separated
-        """,
-    )
-):
+def args_to_target(refs, what, against) -> tuple[str | None, str | None]:
     _what, _against = parse_refs_pair(refs)
     if _what:
         if what:
@@ -75,34 +53,103 @@ async def cmd_review(
             )
     else:
         _against = against
-    await review(what=_what, against=_against, filters=filters)
+    return _what, _against
 
 
-@app.async_command(help="Configure LLM for local usage interactively")
-async def setup():
-    mc.interactive_setup(ENV_CONFIG_FILE)
+def arg_refs() -> typer.Argument:
+    return typer.Argument(
+        default=None,
+        help="Git refs to review, [what]..[against] e.g. 'HEAD..HEAD~1'"
+    )
 
 
-@app.async_command()
-async def render(format: str = Report.Format.MARKDOWN):
-    print(Report.load().render(format=format))
+def arg_what() -> typer.Option:
+    return typer.Option(None, "--what", "-w", help="Git ref to review")
 
 
-@app.async_command(help="Review remote code")
-async def remote(url=typer.Option(), branch=typer.Option()):
-    if os.path.exists("reviewed-repo"):
-        shutil.rmtree("reviewed-repo")
-    Repo.clone_from(url, branch=branch, to_path="reviewed-repo")
-    prev_dir = os.getcwd()
-    try:
-        os.chdir("reviewed-repo")
-        await review()
-    finally:
-        os.chdir(prev_dir)
+def arg_filters() -> typer.Option:
+    return typer.Option(
+        "", "--filter", "-f", "--filters",
+        help="""
+            filter reviewed files by glob / fnmatch pattern(s),
+            e.g. 'src/**/*.py', may be comma-separated
+            """,
+    )
 
 
-@app.async_command(help="Leave a GitHub PR comment with the review.")
-async def github_comment(
+def arg_out() -> typer.Option:
+    return typer.Option(
+        None,
+        "--out", "-o", "--output",
+        help="Output folder for the code review report"
+    )
+
+
+def arg_against() -> typer.Option:
+    return typer.Option(
+        None,
+        "--against", "-vs", "--vs",
+        help="Git ref to compare against"
+    )
+
+
+@app_no_subcommand.command(name="review", help="Perform code review")
+@app.command(name="review", help="Perform code review")
+def cmd_review(
+    refs: str = arg_refs(),
+    what: str = arg_what(),
+    against: str = arg_against(),
+    filters: str = arg_filters(),
+    merge_base: bool = typer.Option(default=True, help="Use merge base for comparison"),
+    out: str = arg_out()
+):
+    _what, _against = args_to_target(refs, what, against)
+    asyncio.run(review(
+        what=_what,
+        against=_against,
+        filters=filters,
+        use_merge_base=merge_base,
+        out_folder=out,
+    ))
+
+
+@app.command(help="Configure LLM for local usage interactively")
+def setup():
+    mc.interactive_setup(ENV_CONFIG_FILE)
+
+
+@app.command()
+def render(format: str = Report.Format.MARKDOWN):
+    print(Report.load().render(format=format))
+
+
+@app.command(help="Review remote code")
+def remote(
+    url: str = typer.Argument(..., help="Git repository URL"),
+    refs: str = arg_refs(),
+    what: str = arg_what(),
+    against: str = arg_against(),
+    filters: str = arg_filters(),
+    merge_base: bool = typer.Option(default=True, help="Use merge base for comparison"),
+    out: str = arg_out()
+):
+    _what, _against = args_to_target(refs, what, against)
+    with tempfile.TemporaryDirectory() as temp_dir:
+        logging.info(f"Cloning [{mc.ui.green(url)}] to {mc.utils.file_link(temp_dir)} ...")
+        repo = Repo.clone_from(url, branch=_what, to_path=temp_dir)
+        asyncio.run(review(
+            repo=repo,
+            what=_what,
+            against=_against,
+            filters=filters,
+            use_merge_base=merge_base,
+            out_folder=out or '.',
+        ))
+        repo.close()
+
+
+@app.command(help="Leave a GitHub PR comment with the review.")
+def github_comment(
     token: str = typer.Option(
         os.environ.get("GITHUB_TOKEN", ""), help="GitHub token (or set GITHUB_TOKEN env var)"
     ),
@@ -110,7 +157,7 @@ async def github_comment(
     """
     Leaves a comment with the review on the current GitHub pull request.
     """
-    file = "code-review-report.txt"
+    file = "code-review-report.md"
     if not os.path.exists(file):
         print(f"Review file not found: {file}")
         raise typer.Exit(4)
@@ -155,3 +202,35 @@ async def github_comment(
     else:
         logging.error(f"Failed to post comment: {resp.status_code} {resp.reason}\n{resp.text}")
         raise typer.Exit(5)
+
+
+@app.command(help="List files in the diff. Might be useful to check what will be reviewed.")
+def files(
+    refs: str = arg_refs(),
+    what: str = arg_what(),
+    against: str = arg_against(),
+    filters: str = arg_filters(),
+    merge_base: bool = typer.Option(default=True, help="Use merge base for comparison"),
+    diff: bool = typer.Option(default=False, help="Show diff content")
+):
+    _what, _against = args_to_target(refs, what, against)
+    repo = Repo(".")
+    patch_set = get_diff(repo=repo, what=_what, against=_against, use_merge_base=merge_base)
+    patch_set = filter_diff(patch_set, filters)
+    print(
+        f"Changed files: "
+        f"{mc.ui.green(_what or 'INDEX')} vs "
+        f"{mc.ui.yellow(_against or repo.remotes.origin.refs.HEAD.reference.name)}"
+        f"{' filtered by '+mc.ui.cyan(filters) if filters else ''}"
+    )
+
+    for patch in patch_set:
+        if patch.is_added_file:
+            color = mc.ui.green
+        elif patch.is_removed_file:
+            color = mc.ui.red
+        else:
+            color = mc.ui.blue
+        print(f"- {color(patch.path)}")
+        if diff:
+            print(mc.ui.gray(textwrap.indent(str(patch), "  ")))

ai_code_review/core.py

@@ -1,6 +1,8 @@
 import fnmatch
 import logging
+from os import PathLike
 from typing import Iterable
+from pathlib import Path
 
 import microcore as mc
 from git import Repo
@@ -9,22 +11,76 @@ from unidiff.constants import DEV_NULL
 
 from .project_config import ProjectConfig
 from .report_struct import Report
+from .constants import JSON_REPORT_FILE_NAME
+
+
+def is_binary_file(repo: Repo, file_path: str) -> bool:
+    """
+    Check if a file is binary by attempting to read it as text.
+    Returns True if the file is binary, False otherwise.
+    """
+    try:
+        # Attempt to read the file content from the repository tree
+        content = repo.tree()[file_path].data_stream.read()
+        # Try decoding as UTF-8; if it fails, it's likely binary
+        content.decode("utf-8")
+        return False
+    except (UnicodeDecodeError, KeyError):
+        return True
+    except Exception as e:
+        logging.warning(f"Error checking if file {file_path} is binary: {e}")
+        return True  # Conservatively treat errors as binary to avoid issues
 
 
 def get_diff(
     repo: Repo = None,
     what: str = None,
-    against: str = None
+    against: str = None,
+    use_merge_base: bool = True,
 ) -> PatchSet | list[PatchedFile]:
     repo = repo or Repo(".")
     if not against:
         against = repo.remotes.origin.refs.HEAD.reference.name  # origin/main
     if not what:
         what = None  # working copy
-    logging.info(f"Reviewing {mc.ui.green(what or 'working copy')} vs {mc.ui.yellow(against)}")
+    if use_merge_base:
+        if what is None:
+            try:
+                current_ref = repo.active_branch.name
+            except TypeError:
+                # In detached HEAD state, use HEAD directly
+                current_ref = "HEAD"
+                logging.info(
+                    "Detected detached HEAD state, using HEAD as current reference"
+                )
+        else:
+            current_ref = what
+        merge_base = repo.merge_base(current_ref or repo.active_branch.name, against)[0]
+        against = merge_base.hexsha
+        logging.info(
+            f"Using merge base: {mc.ui.cyan(merge_base.hexsha[:8])} ({merge_base.summary})"
+        )
+    logging.info(
+        f"Making diff: {mc.ui.green(what or 'INDEX')} vs {mc.ui.yellow(against)}"
+    )
     diff_content = repo.git.diff(against, what)
     diff = PatchSet.from_string(diff_content)
-    return diff
+    diff = PatchSet.from_string(diff_content)
+
+    # Filter out binary files
+    non_binary_diff = PatchSet([])
+    for patched_file in diff:
+        # Check if the file is binary using the source or target file path
+        file_path = (
+            patched_file.target_file
+            if patched_file.target_file != DEV_NULL
+            else patched_file.source_file
+        )
+        if file_path == DEV_NULL or is_binary_file(repo, file_path.lstrip("b/")):
+            logging.info(f"Skipping binary file: {patched_file.path}")
+            continue
+        non_binary_diff.append(patched_file)
+    return non_binary_diff
 
 
 def filter_diff(
@@ -33,7 +89,6 @@ def filter_diff(
     """
     Filter the diff files by the given fnmatch filters.
     """
-    print([f.path for f in patch_set])
     assert isinstance(filters, (list, str))
     if not isinstance(filters, list):
         filters = [f.strip() for f in filters.split(",") if f.strip()]
@@ -44,7 +99,6 @@ def filter_diff(
         for file in patch_set
         if any(fnmatch.fnmatch(file.path, pattern) for pattern in filters)
     ]
-    print([f.path for f in files])
     return files
 
 
@@ -61,22 +115,32 @@ def file_lines(repo: Repo, file: str, max_tokens: int = None) -> str:
 
 
 def make_cr_summary(cfg: ProjectConfig, report: Report, diff):
-    return mc.prompt(
-        cfg.summary_prompt,
-        diff=mc.tokenizing.fit_to_token_size(diff, cfg.max_code_tokens)[0],
-        issues=report.issues,
-        **cfg.prompt_vars,
-    ).to_llm() if cfg.summary_prompt else ""
+    return (
+        mc.prompt(
+            cfg.summary_prompt,
+            diff=mc.tokenizing.fit_to_token_size(diff, cfg.max_code_tokens)[0],
+            issues=report.issues,
+            **cfg.prompt_vars,
+        ).to_llm()
+        if cfg.summary_prompt
+        else ""
+    )
 
 
 async def review(
+    repo: Repo = None,
     what: str = None,
     against: str = None,
-    filters: str | list[str] = ""
+    filters: str | list[str] = "",
+    use_merge_base: bool = True,
+    out_folder: str | PathLike | None = None,
 ):
     cfg = ProjectConfig.load()
-    repo = Repo(".")
-    diff = get_diff(repo=repo, what=what, against=against)
+    repo = repo or Repo(".")
+    out_folder = Path(out_folder or repo.working_tree_dir)
+    diff = get_diff(
+        repo=repo, what=what, against=against, use_merge_base=use_merge_base
+    )
     diff = filter_diff(diff, filters)
     if not diff:
         logging.error("Nothing to review")
@@ -114,12 +178,14 @@ async def review(
                 if lines[file]:
                     f_lines = [""] + lines[file].splitlines()
                     i["affected_code"] = "\n".join(
-                        f_lines[i["start_line"]: i["end_line"]+1]
+                        f_lines[i["start_line"]: i["end_line"] + 1]
                     )
     exec(cfg.post_process, {"mc": mc, **locals()})
+    out_folder.mkdir(parents=True, exist_ok=True)
     report = Report(issues=issues, number_of_processed_files=len(diff))
     report.summary = make_cr_summary(cfg, report, diff)
-    report.save()
+    report.save(file_name=out_folder / JSON_REPORT_FILE_NAME)
     report_text = report.render(cfg, Report.Format.MARKDOWN)
     print(mc.ui.yellow(report_text))
-    open("code-review-report.txt", "w", encoding="utf-8").write(report_text)
+    text_report_path = out_folder / "code-review-report.md"
+    text_report_path.write_text(report_text, encoding="utf-8")

ai_code_review/utils.py

@@ -92,11 +92,11 @@ def is_running_in_github_action():
     return os.getenv("GITHUB_ACTIONS") == "true"
 
 
-def is_app_command_invocation(app: typer.Typer) -> bool:
+def no_subcommand(app: typer.Typer) -> bool:
     """
     Checks if the current script is being invoked as a command in a target Typer application.
     """
-    return (
+    return not (
         (first_arg := next((a for a in sys.argv[1:] if not a.startswith('-')), None))
         and first_arg in (
             cmd.name or cmd.callback.__name__.replace('_', '-')
@@ -106,7 +106,7 @@ def is_app_command_invocation(app: typer.Typer) -> bool:
     )
 
 
-def parse_refs_pair(refs: str):
+def parse_refs_pair(refs: str) -> tuple[str | None, str | None]:
     SEPARATOR = '..'
     if not refs:
         return None, None

{ai_cr-0.5.0.dist-info → ai_cr-0.6.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: ai-cr
-Version: 0.5.0
+Version: 0.6.0
 Summary: LLM-agnostic GitHub AI Code Review Tool with integration to GitHub actions
 License: MIT
 Keywords: static code analysis,code review,code quality,ai,coding,assistant,llm,github,automation,devops,developer tools,github actions,workflows,git
@@ -15,13 +15,12 @@ Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Software Development
-Requires-Dist: GitPython (==3.1.44)
-Requires-Dist: ai-microcore (==4.0.0.dev18)
-Requires-Dist: anthropic (==0.52.2)
-Requires-Dist: async-typer (==0.1.8)
-Requires-Dist: google-generativeai (==0.8.5)
-Requires-Dist: typer (==0.9.4)
-Requires-Dist: unidiff (==0.7.5)
+Requires-Dist: GitPython (>=3.1.44,<4.0.0)
+Requires-Dist: ai-microcore (==4.0.0.dev19)
+Requires-Dist: anthropic (>=0.52.2,<0.53.0)
+Requires-Dist: google-generativeai (>=0.8.5,<0.9.0)
+Requires-Dist: typer (>=0.16.0,<0.17.0)
+Requires-Dist: unidiff (>=0.7.5,<0.8.0)
 Project-URL: Homepage, https://github.com/Nayjest/github-ai-code-review
 Project-URL: Repository, https://github.com/Nayjest/github-ai-code-review
 Description-Content-Type: text/markdown
@@ -30,6 +29,7 @@ Description-Content-Type: text/markdown
 <a href="https://pypi.org/project/ai-code-review/" target="_blank"><img src="https://badge.fury.io/py/ai-code-review.svg" alt="PYPI Release"></a>
 <a href="https://github.com/Nayjest/ai-code-review/actions/workflows/code-style.yml" target="_blank"><img src="https://github.com/Nayjest/ai-code-review/actions/workflows/code-style.yml/badge.svg" alt="Pylint"></a>
 <a href="https://github.com/Nayjest/ai-code-review/actions/workflows/tests.yml" target="_blank"><img src="https://github.com/Nayjest/ai-code-review/actions/workflows/tests.yml/badge.svg" alt="Tests"></a>
+<img src="https://github.com/Nayjest/ai-code-review/blob/main/coverage.svg" alt="Code Coverage">
 <a href="https://github.com/Nayjest/ai-code-review/blob/main/LICENSE" target="_blank"><img src="https://img.shields.io/static/v1?label=license&message=MIT&color=d08aff" alt="License"></a>
 </p>
 
@@ -91,7 +91,8 @@ jobs:
 > ⚠️ Make sure to add `LLM_API_KEY` to your repository’s GitHub secrets.
 
 💪 Done!  
-PRs to your repository will now receive AI code reviews automatically. ✨
+PRs to your repository will now receive AI code reviews automatically. ✨  
+See [GitHub Setup Guide](https://github.com/Nayjest/ai-code-review/blob/main/documentation/github_setup.md) for more details.
 
 ### 2. Run Code Analysis Locally
 

ai_cr-0.6.0.dist-info/RECORD

@@ -0,0 +1,15 @@
+ai_code_review/.ai-code-review.toml,sha256=z-3iliCvkkpdAr9l_yUyM-IbgRm9hbAA3CIV6ocuurY,10378
+ai_code_review/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+ai_code_review/__main__.py,sha256=EClCwCzb6h6YBpt0hrnG4h0mlNhNePyg_xBNNSVm1os,65
+ai_code_review/bootstrap.py,sha256=jqioR_UtTsn5nXezmjMLU3aB8tzlVz73ZRBk33ud5F4,2336
+ai_code_review/cli.py,sha256=1wrF-s6U_f8BjwCl70G7IXIBEhmMUbQDvij5AMDxQ5M,7318
+ai_code_review/constants.py,sha256=K9mNxTq9seTG3aVm__3r1lXb5oCOQjH24Cl3hfX9FsE,281
+ai_code_review/core.py,sha256=BJNs4ZER2-bMulXY2apY6B6hI0nvRCOrLqsJ7L8Bizc,6693
+ai_code_review/project_config.py,sha256=RDbplmncALKw0zgqSG8POofi300z0DPvtF33wt7_1Sk,3651
+ai_code_review/report_struct.py,sha256=N-EnNMwBY9LyJ9sdFHpUzn2fwBvxo5TZYYiJagBl8Po,3488
+ai_code_review/utils.py,sha256=vlzU3M89qK6_mVkBMnppZaOFsXddVsIBVdfmbN3cxDY,2939
+ai_cr-0.6.0.dist-info/entry_points.txt,sha256=u0N5NroPYEGqmDGaGqFmiijJ5swzpe7EyKBupnkp1FY,49
+ai_cr-0.6.0.dist-info/LICENSE,sha256=XATf3zv-CppUSJqI18KLhwnPEomUXEl5WbBzFyb9OSU,1096
+ai_cr-0.6.0.dist-info/METADATA,sha256=poIlyrRUJQEKwRN7JIgKGOjgL1DR291tw6jibTi4EMs,5835
+ai_cr-0.6.0.dist-info/WHEEL,sha256=fGIA9gx4Qxk2KDKeNJCbOEwSrmLtjWCwzBz351GyrPQ,88
+ai_cr-0.6.0.dist-info/RECORD,,

ai_cr-0.5.0.dist-info/RECORD

@@ -1,15 +0,0 @@
-ai_code_review/.ai-code-review.toml,sha256=z-3iliCvkkpdAr9l_yUyM-IbgRm9hbAA3CIV6ocuurY,10378
-ai_code_review/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-ai_code_review/__main__.py,sha256=MSmt_5Xg84uHqzTN38JwgseJK8rsJn_11A8WD99VtEo,61
-ai_code_review/bootstrap.py,sha256=b9BlhRgliwsjmWRCCZcsTivsOw9AbXOf8_AZYjQTx-c,2336
-ai_code_review/cli.py,sha256=L1utFMHs4D9HVcuJ1VC96PEvP6q26UHsfCz_Lq49VMg,4893
-ai_code_review/constants.py,sha256=K9mNxTq9seTG3aVm__3r1lXb5oCOQjH24Cl3hfX9FsE,281
-ai_code_review/core.py,sha256=WMGU2zwZTJ1P6PkUId-RJP7KHQbzB31GKSnI5wsedcA,4177
-ai_code_review/project_config.py,sha256=RDbplmncALKw0zgqSG8POofi300z0DPvtF33wt7_1Sk,3651
-ai_code_review/report_struct.py,sha256=N-EnNMwBY9LyJ9sdFHpUzn2fwBvxo5TZYYiJagBl8Po,3488
-ai_code_review/utils.py,sha256=UzFp2ShNzQvsydxJ98bdr_j2rBUHcm62pPtl6hBuZD4,2914
-ai_cr-0.5.0.dist-info/entry_points.txt,sha256=u0N5NroPYEGqmDGaGqFmiijJ5swzpe7EyKBupnkp1FY,49
-ai_cr-0.5.0.dist-info/LICENSE,sha256=XATf3zv-CppUSJqI18KLhwnPEomUXEl5WbBzFyb9OSU,1096
-ai_cr-0.5.0.dist-info/METADATA,sha256=k3FaB6nbVV0xbgZYPg_VQ6zMauB65Xy6CLWI4qFOMF0,5609
-ai_cr-0.5.0.dist-info/WHEEL,sha256=fGIA9gx4Qxk2KDKeNJCbOEwSrmLtjWCwzBz351GyrPQ,88
-ai_cr-0.5.0.dist-info/RECORD,,