More code repository improvements

src/zenml/cli/code_repository.py

@@ -162,6 +162,32 @@ def register_code_repository(
         cli_utils.declare(f"Successfully registered code repository `{name}`.")
 
 
+@code_repository.command("describe", help="Describe a code repository.")
+@click.argument(
+    "name_id_or_prefix",
+    type=str,
+    required=True,
+)
+def describe_code_repository(name_id_or_prefix: str) -> None:
+    """Describe a code repository.
+
+    Args:
+        name_id_or_prefix: Name, ID or prefix of the code repository.
+    """
+    client = Client()
+    try:
+        code_repository = client.get_code_repository(
+            name_id_or_prefix=name_id_or_prefix,
+        )
+    except KeyError as err:
+        cli_utils.error(str(err))
+    else:
+        cli_utils.print_pydantic_model(
+            title=f"Code repository '{code_repository.name}'",
+            model=code_repository,
+        )
+
+
 @code_repository.command("list", help="List all connected code repositories.")
 @list_options(CodeRepositoryFilter)
 def list_code_repositories(**kwargs: Any) -> None:

src/zenml/client.py

@@ -34,7 +34,7 @@
     Union,
     cast,
 )
-from uuid import UUID, uuid4
+from uuid import UUID
 
 from pydantic import ConfigDict, SecretStr
 
@@ -4980,12 +4980,7 @@ def _validate_code_repository_config(
             )
         )
         try:
-            # This does a login to verify the credentials
-            code_repo_class(id=uuid4(), config=config)
-
-            # Explicitly access the config for pydantic validation, in case
-            # the login for some reason did not do that.
-            _ = code_repo_class.config
+            code_repo_class.validate_config(config)
         except Exception as e:
             raise RuntimeError(
                 "Failed to validate code repository config."

src/zenml/code_repositories/base_code_repository.py

@@ -15,7 +15,7 @@
 
 from abc import ABC, abstractmethod
 from typing import TYPE_CHECKING, Any, Dict, Optional, Set, Type
-from uuid import UUID
+from uuid import UUID, uuid4
 
 from zenml.config.secret_reference_mixin import SecretReferenceMixin
 from zenml.logger import get_logger
@@ -82,6 +82,22 @@ def from_model(cls, model: CodeRepositoryResponse) -> "BaseCodeRepository":
         )
         return class_(id=model.id, config=model.config)
 
+    @classmethod
+    def validate_config(cls, config: Dict[str, Any]) -> None:
+        """Validate the code repository config.
+
+        This method should check that the config/credentials are valid and
+        the configured repository exists.
+
+        Args:
+            config: The configuration.
+        """
+        # The initialization calls the login to verify the credentials
+        code_repo = cls(id=uuid4(), config=config)
+
+        # Explicitly access the config for pydantic validation
+        _ = code_repo.config
+
     @property
     def id(self) -> UUID:
         """ID of the code repository.

src/zenml/code_repositories/git/local_git_repository_context.py

@@ -19,6 +19,10 @@
 from zenml.code_repositories import (
     LocalRepositoryContext,
 )
+from zenml.constants import (
+    ENV_ZENML_CODE_REPOSITORY_IGNORE_UNTRACKED_FILES,
+    handle_bool_env_var,
+)
 from zenml.logger import get_logger
 
 if TYPE_CHECKING:
@@ -70,11 +74,13 @@
             from git.exc import InvalidGitRepositoryError
             from git.repo.base import Repo
         except ImportError:
+            logger.debug("Failed to import git library.")
             return None
 
         try:
             git_repo = Repo(path=path, search_parent_directories=True)
         except InvalidGitRepositoryError:
+            logger.debug("No git repository exists at path %s.", path)
             return None
 
         remote_name = None
@@ -124,13 +130,19 @@
     def is_dirty(self) -> bool:
         """Whether the git repo is dirty.
 
-        A repository counts as dirty if it has any untracked or uncommitted
-        changes.
+        By defauly, a repository counts as dirty if it has any untracked or
+        uncommitted changes. Users can use an environment variable to ignore
+        untracked files.
 
         Returns:
             True if the git repo is dirty, False otherwise.
         """
-        return self.git_repo.is_dirty(untracked_files=True)
+        ignore_untracked_files = handle_bool_env_var(
+            ENV_ZENML_CODE_REPOSITORY_IGNORE_UNTRACKED_FILES, default=False
+        )
+        return self.git_repo.is_dirty(
+            untracked_files=not ignore_untracked_files
+        )
 
     @property
     def has_local_changes(self) -> bool:

src/zenml/constants.py

@@ -175,6 +175,9 @@ def handle_int_env_var(var: str, default: int = 0) -> int:
 ENV_ZENML_PIPELINE_RUN_API_TOKEN_EXPIRATION = (
     "ZENML_PIPELINE_API_TOKEN_EXPIRATION"
 )
+ENV_ZENML_CODE_REPOSITORY_IGNORE_UNTRACKED_FILES = (
+    "ZENML_CODE_REPOSITORY_IGNORE_UNTRACKED_FILES"
+)
 
 # Materializer environment variables
 ENV_ZENML_MATERIALIZER_ALLOW_NON_ASCII_JSON_DUMPS = (

src/zenml/integrations/github/code_repositories/github_code_repository.py

@@ -15,7 +15,8 @@
 
 import os
 import re
-from typing import List, Optional
+from typing import Any, Dict, List, Optional
+from uuid import uuid4
 
 import requests
 from github import Consts, Github, GithubException
@@ -62,6 +63,20 @@ class GitHubCodeRepositoryConfig(BaseCodeRepositoryConfig):
 class GitHubCodeRepository(BaseCodeRepository):
     """GitHub code repository."""
 
+    @classmethod
+    def validate_config(cls, config: Dict[str, Any]) -> None:
+        """Validate the code repository config.
+
+        This method should check that the config/credentials are valid and
+        the configured repository exists.
+
+        Args:
+            config: The configuration.
+        """
+        code_repo = cls(id=uuid4(), config=config)
+        # Try to access the project to make sure it exists
+        _ = code_repo.github_repo
+
     @property
     def config(self) -> GitHubCodeRepositoryConfig:
         """Returns the `GitHubCodeRepositoryConfig` config.

src/zenml/integrations/gitlab/code_repositories/gitlab_code_repository.py

@@ -15,7 +15,8 @@
 
 import os
 import re
-from typing import Optional
+from typing import Any, Dict, Optional
+from uuid import uuid4
 
 from gitlab import Gitlab
 from gitlab.v4.objects import Project
@@ -63,6 +64,20 @@ class GitLabCodeRepositoryConfig(BaseCodeRepositoryConfig):
 class GitLabCodeRepository(BaseCodeRepository):
     """GitLab code repository."""
 
+    @classmethod
+    def validate_config(cls, config: Dict[str, Any]) -> None:
+        """Validate the code repository config.
+
+        This method should check that the config/credentials are valid and
+        the configured repository exists.
+
+        Args:
+            config: The configuration.
+        """
+        code_repo = cls(id=uuid4(), config=config)
+        # Try to access the project to make sure it exists
+        _ = code_repo.gitlab_project
+
     @property
     def config(self) -> GitLabCodeRepositoryConfig:
         """Returns the `GitLabCodeRepositoryConfig` config.

src/zenml/pipelines/pipeline_definition.py

@@ -695,8 +695,8 @@ def _create_deployment(
             deployment=deployment, stack=stack
         )
 
-        local_repo_context = (
-            code_repository_utils.find_active_code_repository()
+        local_repo_context = code_repository_utils.find_active_code_repository(
+            log=True
         )
         code_repository = build_utils.verify_local_repository_context(
             deployment=deployment, local_repo_context=local_repo_context

src/zenml/utils/code_repository_utils.py

@@ -85,12 +85,14 @@ def set_custom_local_repository(
 
 def find_active_code_repository(
     path: Optional[str] = None,
+    log: bool = False,
 ) -> Optional["LocalRepositoryContext"]:
     """Find the active code repository for a given path.
 
     Args:
         path: Path at which to look for the code repository. If not given, the
             source root will be used.
+        log: Log a message if there was no matching code repository.
 
     Returns:
         The local repository context active at that path or None.
@@ -106,7 +108,8 @@ def find_active_code_repository(
         return _CODE_REPOSITORY_CACHE[path]
 
     local_context: Optional["LocalRepositoryContext"] = None
-    for model in depaginate(list_method=Client().list_code_repositories):
+    code_repositories = depaginate(list_method=Client().list_code_repositories)
+    for model in code_repositories:
         try:
             repo = BaseCodeRepository.from_model(model)
         except ImportError:
@@ -125,6 +128,12 @@ def find_active_code_repository(
         local_context = repo.get_local_context(path)
         if local_context:
             break
+    else:
+        if code_repositories and log:
+            # There are registered code repositories, but none was matching the
+            # current path -> We log the path to help in debugging issues
+            # related to the source root.
+            logger.info("No matching code repository found at path %s.", path)
 
     _CODE_REPOSITORY_CACHE[path] = local_context
     return local_context