From f5aeec87cca786dc9654c77fd87a47e4c7b055f1 Mon Sep 17 00:00:00 2001
From: behnazh-w <behnaz.hassanshahi@oracle.com>
Date: Wed, 2 Aug 2023 15:54:34 +1000
Subject: [PATCH] fix: handle cloning issues when repo is in an unexpected
 state

Signed-off-by: behnazh-w <behnaz.hassanshahi@oracle.com>
---
 src/macaron/slsa_analyzer/analyzer.py         |  1 +
 .../slsa_analyzer/git_service/gitlab.py       | 35 ++++++++++++++++---
 2 files changed, 31 insertions(+), 5 deletions(-)

diff --git a/src/macaron/slsa_analyzer/analyzer.py b/src/macaron/slsa_analyzer/analyzer.py
index 25d239fa5..8fa18b546 100644
--- a/src/macaron/slsa_analyzer/analyzer.py
+++ b/src/macaron/slsa_analyzer/analyzer.py
@@ -700,6 +700,7 @@ def _prepare_repo(
         try:
             git_service.check_out_repo(git_obj, branch_name, digest, not is_remote)
         except RepoCheckOutError as error:
+            logger.error("Failed to check out repository at %s", resolved_local_path)
             logger.error(error)
             return None
 
diff --git a/src/macaron/slsa_analyzer/git_service/gitlab.py b/src/macaron/slsa_analyzer/git_service/gitlab.py
index 2c5abb756..17597e026 100644
--- a/src/macaron/slsa_analyzer/git_service/gitlab.py
+++ b/src/macaron/slsa_analyzer/git_service/gitlab.py
@@ -23,6 +23,7 @@
 from abc import abstractmethod
 from urllib.parse import ParseResult, urlunparse
 
+from git import GitError
 from pydriller.git import Git
 
 from macaron.errors import CloneError, ConfigurationError, RepoCheckOutError
@@ -129,7 +130,7 @@ def clone_repo(self, clone_dir: str, url: str) -> None:
         # Therefore, we don't need to catch and handle the CloneError exceptions here.
         repo = git_url.clone_remote_repo(clone_dir, clone_url)
 
-        # If ``git_url.clone_remote_repo`` returns an Repo instance, this means that the repository is freshly cloned
+        # If ``git_url.clone_remote_repo`` returns a Repo instance, this means that the repository is freshly cloned
         # with the token embedded URL. We will set its value back to the original non-token URL.
         # If ``git_url.clone_remote_repo`` returns None, it means that the repository already exists so we don't need
         # to do anything.
@@ -139,7 +140,16 @@ def clone_repo(self, clone_dir: str, url: str) -> None:
             except ValueError as error:
                 raise CloneError("Cannot find the remote origin for this repository.") from error
 
-            origin_remote.set_url(url)
+            try:
+                # Even though the documentation of ``set_url`` function does not explicitly mention
+                # ``ValueError`` or ``GitError`` as raised errors, these errors might be raised based
+                # on the implementation.
+                origin_remote.set_url(url)
+            except (ValueError, GitError) as error:
+                raise CloneError(
+                    "Failed to set the remote origin URL because this repository is in an unexpected state."
+                    f" Consider removing the cloned repository at {clone_dir}."
+                ) from error
 
     def check_out_repo(self, git_obj: Git, branch: str, digest: str, offline_mode: bool) -> Git:
         """Checkout the branch and commit specified by the user of a repository.
@@ -168,7 +178,7 @@ def check_out_repo(self, git_obj: Git, branch: str, digest: str, offline_mode: b
 
         Raises
         ------
-        RepoError
+        RepoCheckOutError
             If there is error while checkout the specific branch and digest.
         """
         remote_origin_url = git_url.get_remote_origin_of_local_repo(git_obj)
@@ -183,11 +193,26 @@ def check_out_repo(self, git_obj: Git, branch: str, digest: str, offline_mode: b
         except CloneError as error:
             raise RepoCheckOutError("Cannot parse the remote origin URL of this repository.") from error
 
-        origin_remote.set_url(reconstructed_url, remote_origin_url)
+        try:
+            # Even though the documentation of ``set_url`` function does not explicitly mention
+            # ``ValueError`` or ``GitError`` as raised errors, these errors might be raised based
+            # on the implementation.
+            origin_remote.set_url(reconstructed_url, remote_origin_url)
+        except (ValueError, GitError) as error:
+            raise RepoCheckOutError(
+                "Failed to set the remote origin URL because this repository is in an unexpected state."
+                " Consider removing the cloned repository."
+            ) from error
 
         check_out_status = git_url.check_out_repo_target(git_obj, branch, digest, offline_mode)
 
-        origin_remote.set_url(remote_origin_url, reconstructed_url)
+        try:
+            origin_remote.set_url(remote_origin_url, reconstructed_url)
+        except (ValueError, GitError) as error:
+            raise RepoCheckOutError(
+                "Failed to set the remote origin URL because this repository is in an unexpected state."
+                " Consider removing the cloned repository."
+            ) from error
 
         if not check_out_status:
             raise RepoCheckOutError(