Add mkdocs-material dependency to requirements.txt

.gitignore

@@ -1,6 +1,8 @@
 # Python
 .venv
 __pycache__/
+*.egg-info/
+dist/
 
 # Ignore sample JSON files
 *.sample.json
@@ -12,7 +14,13 @@ prototype_*.py
 *.secret
 *.client_secret
 
+# Environment files
+*.env
+
 # Certificate files
 *.pem
 *.key
 *.crt
+
+# Harvester output
+reference

.vscode/settings.json

@@ -2,5 +2,6 @@
     "debug.autoExpandLazyVariables": "off",
     "debug.inlineValues": "off",
     "debugpy.debugJustMyCode": true,
-    "debugpy.showPythonInlineValues": false
+    "debugpy.showPythonInlineValues": false,
+    "python.terminal.useEnvFile": true
 }

BUILD.md

@@ -0,0 +1,21 @@
+# Build Instructions
+
+Run the following command to build the project:
+
+```shell
+python -m build
+```
+
+This will create distribution files in the `dist/` directory.
+
+Install the built package using pip:
+
+```shell
+pip install --no-index dist/docs_harvester-0.1.0-py3-none-any.whl
+```
+
+Install in editable mode for development:
+
+```shell
+pip install -e .
+```

DEVOPS.md

@@ -0,0 +1,9 @@
+# DevOps Notes
+
+## DevOps OAuth2 Flow
+
+Type: **oauth2**  
+Flow: **accessCode**  
+Authorization URL: `https://app.vssps.visualstudio.com/oauth2/authorize&response_type=Assertion`  
+Token URL: `https://app.vssps.visualstudio.com/oauth2/token?client_assertion_type=urn:ietf:params:oauth:client-assertion-type:jwt-bearer&grant_type=urn:ietf:params:oauth:grant-type:jwt-bearer`  
+Scopes: `vso.code`

devops/__init__.py

@@ -0,0 +1 @@
+# devops package

devops/azure.py

@@ -59,9 +59,9 @@ def get_token(
 
 def secret_credentials_auth(
         scope: str = DEVOPS_SCOPE,
-        tenant_id: str = os.environ.get("AZURE_TENANT_ID", ""),
-        client_id: str = os.environ.get("AZURE_CLIENT_ID", ""),
-        client_secret: str = os.environ.get("AZURE_CLIENT_SECRET")
+        tenant_id = os.environ.get("AZURE_TENANT_ID", ""),
+        client_id = os.environ.get("AZURE_CLIENT_ID", ""),
+        client_secret = os.environ.get("AZURE_CLIENT_SECRET")
         ) -> str:
     """
     Authenticate using client credentials. Pass credentials via environment variables,
@@ -92,8 +92,8 @@ def certificate_credentials_auth(
     # Wczytaj klucz prywatny (RSA)
     with open(pem_path, "rb") as f:
         pem = f.read()
-        key_pem  = re.search(b"-----BEGIN (?:RSA )?PRIVATE KEY-----.*?END (?:RSA )?PRIVATE KEY-----", pem, re.S).group(0)
-        cert_pem = re.search(b"-----BEGIN CERTIFICATE-----.*?END CERTIFICATE-----", pem, re.S).group(0)
+        key_pem  = re.search(b"-----BEGIN (?:RSA )?PRIVATE KEY-----.*?END (?:RSA )?PRIVATE KEY-----", pem, re.S).group(0) # type: ignore
+        cert_pem = re.search(b"-----BEGIN CERTIFICATE-----.*?END CERTIFICATE-----", pem, re.S).group(0) # type: ignore
 
         private_key = serialization.load_pem_private_key(key_pem, password=None)
         cert = x509.load_pem_x509_certificate(cert_pem)
@@ -115,7 +115,7 @@ def certificate_credentials_auth(
 
     headers = {"x5t": x5t, "kid": x5t}
 
-    assertion = jwt.encode(claims, private_key, algorithm="RS256", headers=headers)
+    assertion = jwt.encode(claims, private_key, algorithm="RS256", headers=headers) # type: ignore
 
     data = {
         "grant_type": "client_credentials",

devops/devops.py

@@ -1,17 +1,14 @@
 from __future__ import annotations
+import pathlib
 import requests
 import urllib.parse
 from uuid import UUID
 import logging
-from sk.logger import log_entity_creation
 
 DEVOPS_SCOPE = "https://app.vssps.visualstudio.com/.default"
 DEVOPS_API_VERSION = "7.1"
 
-# Get logger. It should be configured by the main application.
 log = logging.getLogger(__name__)
-# log.setLevel(logging.DEBUG)
-# log.propagate = False
 
 # Define a class decorator
 def auto_properties(mapping: dict[str,str]):
@@ -82,7 +79,6 @@ class Organization():
         self._org_url = org_url.rstrip("/") + "/" # Ensure trailing slash
         self._token = token
         self._api_version = api_version
-        log_entity_creation(log, Organization, self._org_url)
 
     def get_path(self, path: str, params: dict = {}) -> requests.Response:
         return get_url(
@@ -96,7 +92,10 @@ class Organization():
     def projects(self):
         if not hasattr(self, "_projects"):
             # Create Project objects
-            self._projects = [Project(org=self, **proj) for proj in self.get_path("_apis/projects").json().get("value", [])]
+            self._projects = []
+            for proj in self.get_path("_apis/projects").json().get("value", []):
+                p = Project(org=self, **proj)
+                self._projects.append(p)
         return self._projects
 
     def __getitem__(self, key: str) -> Project:
@@ -127,8 +126,6 @@ class Project():
         except ValueError:
             raise ValueError(f"Invalid project ID: {self._id}")
 
-        log_entity_creation(log, Project, self.id)
-
     def get_auto_properties(self):
         r = get_url(
             URL=f"{self._org._org_url}_apis/projects/{self._id}",
@@ -151,7 +148,12 @@ class Project():
     @property
     def repositories(self):
         if not hasattr(self, "_repositories"):
-            self._repositories = [Repository(org=self._org, **repo) for repo in self._org.get_path(f"{self._id}/_apis/git/repositories").json().get("value", [])]
+            self._repositories = []
+            for repo in self._org.get_path(f"{self._id}/_apis/git/repositories").json().get("value", []):
+                # Remove unnecessary nested project info
+                if "project" in repo:
+                    del repo["project"]
+                self._repositories.append(Repository(project=self, **repo))
         return self._repositories
 
     def __getitem__(self, key: str) -> Repository:
@@ -186,7 +188,6 @@ class Repository():
 
         # set other properties if provided
         self.from_args(**kwargs) # type: ignore[attr-defined]
-        log_entity_creation(log, Repository, self.id)
 
     def get_auto_properties(self):
         id = self._id if hasattr(self, "_id") else self._name # type: ignore[attr-defined]
@@ -234,7 +235,14 @@ class Item():
     def __init__(self, repository: Repository, **kwargs):
         self._repository = repository
         self.from_args(**kwargs) # type: ignore[attr-defined]
-        log_entity_creation(log, Item, self.path)
+        if "branch" in kwargs:
+            self._branch = kwargs.get("branch")
+
+    @property
+    def branch(self):
+        if hasattr(self, "_branch"):
+            return getattr(self, "_branch")
+        return None
 
     def get_auto_properties(self):
         r = self._repository._project.organization.get_path(
@@ -247,17 +255,28 @@ class Item():
         )
         self.from_json(r.json()) # type: ignore[attr-defined]
 
-    def get_content(self) -> bytes:
-        """Get the content of the item if it is a file."""
+    def get_content(self, branch: str | None = None, commit: str | None = None, tag: str | None = None) -> bytes:
+        """Get the content of the item with optional branch, commit, or tag."""
         if self.git_object_type != "blob": # type: ignore[attr-defined]
             raise ValueError("Content can only be fetched for blob items.")
 
+        params = { "path": self.path, "recursionLevel": "none" }
+        if self.branch and branch is None:
+            branch = self.branch
+
+        if branch:
+            params["version"] = branch
+            params["versionType"] = "branch"
+        elif tag:
+            params["version"] = tag
+            params["versionType"] = "tag"
+        elif commit:
+            params["version"] = commit
+            params["versionType"] = "commit"
+
         r = self._repository._project.organization.get_path(
             path=f"{self._repository._project.id}/_apis/git/repositories/{self._repository.id}/items",
-            params={
-                "path": self.path,
-                "recursionLevel": "none"
-            }
+            params=params
         )
         return r.content
 
@@ -265,22 +284,25 @@ class Item():
     def path(self):
         return self._path # type: ignore[attr-defined]
 
-    def get_child_items(self) -> list[Item]:
+    def get_child_items(self, pattern: str | None = None, recurse: bool = False) -> list[Item]:
         """Get child items if this item is a folder."""
         if self.git_object_type != "tree": # type: ignore[attr-defined]
             raise ValueError("Child items can only be fetched for folder items.")
 
         # Fetch child objects
-        objects = self._repository._project.organization.get_path(
-            path=f"{self._repository._project.id}/_apis/git/repositories/{self._repository.id}/items",
+        objects = self._repository.project.organization.get_path(
+            path=f"{self._repository.project.id}/_apis/git/repositories/{self._repository.id}/items",
             params={
                 "scopePath": self.path,
-                "recursionLevel": "oneLevel"
+                "recursionLevel": "oneLevel" if not recurse else "full"
             }
         ).json().get("value", [])
         child_items = []
         for obj in objects:
-            i = Item(repository=self._repository, path=obj.get("path"))
+            obj_path = obj.get("path")
+            if pattern and not pathlib.PurePath(obj_path).match(pattern):
+                continue
+            i = Item(repository=self._repository, path=obj_path)
             i.from_json(obj) # type: ignore[attr-defined]
             child_items.append(i)
         return child_items
@@ -291,5 +313,15 @@ class Item():
             self._children = self.get_child_items()
         return self._children
 
+    def __getitem__(self, key: str) -> Item:
+        if self.git_object_type != "tree": # type: ignore[attr-defined]
+            raise ValueError("Child items can only be accessed for folder items.")
+        if not key.startswith("/"):
+            key = pathlib.Path(self.path).joinpath(key).absolute().as_posix()
+        for child in self.children:
+            if child.path == key:
+                return child
+        raise KeyError(f"Child item with path '{key}' not found.")
+
     def __str__(self):
         return f"Item(path=\"{self._path}\" type={self.git_object_type})" # type: ignore[attr-defined]

get-token.py

@@ -12,17 +12,43 @@ Usage:
 
     Now you can use the ADO_TOKEN environment variable, for example using curl:
 
-    curl -H "Authorization: Bearer $ADO_TOKEN" https://dev.azure.com/your_organization/_apis/projects?api-version=7.1
+    curl -sH "Authorization: Bearer $ADO_TOKEN" "https://dev.azure.com/$ADO_ORGANIZATION_URL/_apis/projects?api-version=7.1"
 """
 
-from sk.azure import get_token
+from devops.azure import get_token
+from argparse import ArgumentParser
+import os
 
+args = ArgumentParser(description="Get Azure DevOps token and print it for exporting as environment variable.")
+args.add_argument("--tenant-id", type=str, default=os.getenv("AZURE_TENANT_ID"), help="Azure AD Tenant ID")
+args.add_argument("--client-id", type=str, default=os.getenv("AZURE_CLIENT_ID"), help="Azure AD Client ID")
+args.add_argument("--pem-path", type=str, default=os.getenv("AZURE_CLIENT_CERTIFICATE_PATH"), help="Path to PEM file for authentication (optional)")
+args.add_argument("--client-secret", type=str, default=os.getenv("AZURE_CLIENT_SECRET"), help="Client Secret for authentication (optional)")
+args = args.parse_args()
+
+if not args.tenant_id or not args.client_id:
+    print("Tenant ID and Client ID are required.")
+    exit(1)
+
+if args.pem_path and os.path.isfile(args.pem_path):
     token = get_token(
-    tenant_id="a7740229-47b6-45de-ad22-83721462b1bf",
-    client_id="840671c4-5dc4-40e5-aab9-7c3a07bbd652",
-    pem_path="./slawek-mba.pem"
+        tenant_id=args.tenant_id,
+        client_id=args.client_id,
+        pem_path=args.pem_path
+    )
+elif args.client_secret:
+    if not args.client_secret:
+        print("Client secret file is empty.")
+        exit(1)
+    token = get_token(
+        tenant_id=args.tenant_id,
+        client_id=args.client_id,
+        client_secret=args.client_secret
+    )
+else:
+    token = get_token(
+        tenant_id=args.tenant_id,
+        client_id=args.client_id
     )
 
-# print(f"Obtained token: {token}")
-
 print(f"export ADO_TOKEN='{token}'")

harvester.py

@@ -1,8 +1,8 @@
 #!/usr/bin/env python3
 
-from sk.devops import Organization
-org = Organization("https://dev.azure.com/mcovsandbox")
+from harvester.harvester import harvest_readmes
 
-# print(org.projects["bafe0cf1-6c97-4088-864a-ea6dc02b2727"].repositories["feac266f-84d2-41bc-839b-736925a85eaa"].items["/generate-pat.py"])
-print(org["ADO Sandbox"]["ado-auth-lab"]["/container"].url)
-print(org["ADO Sandbox"]["ado-auth-lab"]["/generate-pat.py"].url)
+if __name__ == "__main__":
+    harvest_readmes(
+        organization="mcovsandbox"
+    )

harvester/__init__.py

@@ -0,0 +1 @@
+# Harvester Package

harvester/harvester.py

@@ -0,0 +1,75 @@
+import os
+import requests
+from devops.azure import get_token
+from devops.devops import Organization, Project, Repository, Item
+import logging
+
+fmt = logging.Formatter('%(asctime)s - %(levelname)s - %(message)s')
+ch = logging.StreamHandler()
+ch.setFormatter(fmt)
+log = logging.getLogger(__name__)
+log.addHandler(ch)
+log.setLevel(logging.INFO)
+
+def sanitize_name(name: str) -> str:
+    """Sanitize a name to be filesystem-friendly."""
+    return name.lower().replace(" ", "-").replace("_", "-")
+
+def harvest_readmes(organization: str | Organization, branch: list[str | None] = ["main", "dev", None], projects: list[str] = [], output_path: str = "reference") -> None:
+    """Harvest README files from repositories."""
+    if isinstance(organization, str):
+        org = Organization("https://dev.azure.com/" + organization, token=get_token())
+    else:
+        org = organization
+
+    if projects:
+        # Target specific projects
+        target_projects = [Project(org=org, name=project_name) for project_name in projects]
+    else:
+        # Target all projects
+        target_projects = org.projects
+
+    for project in target_projects:
+        repo_index = [] # Repository index for the project.
+        log.info(f"Processing project: {project.name} with {len(project.repositories)} repositories.") # type: ignore
+        for repo in project.repositories:
+            log.info(f"...processing repository: {repo.name}")
+            readme_found = False
+            # Try each specified branch to find the README.md file
+            for branch_name in branch:
+                try:
+                    # Check if the README.md file exists
+                    readme = Item(repository=repo, path="/README.md", branch=branch_name)
+                    # Build output path and save the README content if found
+                    if readme:
+                        project_path = f"{output_path}/{sanitize_name(project.name)}" # type: ignore
+                        # Create project directory if it doesn't exist
+                        os.makedirs(project_path, exist_ok=True)
+                        # Save README content to index.md
+                        readme_content = readme.get_content(branch=branch_name)
+                        if readme_content is None or len(readme_content.strip()) == 0:
+                            continue
+                        with open(f"{project_path}/{sanitize_name(repo.name)}.md", "w") as f:
+                            f.write(readme_content.decode("utf-8"))
+                        readme_found = True
+                        break # Exit branch loop if README is found
+                except requests.exceptions.HTTPError:
+                    # Repository does not have a README.md file in the specified branch
+                    continue
+            # Register if README was not found in any branch
+            repo_index.append((repo.name, readme_found)) # README not found
+            # Log if the README was not found
+            if not readme_found:
+                log.warning(f"......README.md in repo {repo.name} is not found or empty.")
+        # Save the repository index for the project
+        with open(f"{output_path}/{sanitize_name(project.name)}/index.md", "w") as index_file: # type: ignore
+            index_file.write(f"# Repository Index for Project: {project.name}\n\n") # type: ignore
+            for repo_name, has_readme in repo_index:
+                status = "" if has_readme else " - README.md not found"
+                index_file.write(f"- [{repo_name}]({sanitize_name(repo_name)}.md){status}\n")
+
+    # Save the reference index for all projects
+    with open(f"{output_path}/index.md", "w") as ref_index_file:
+        ref_index_file.write("# Project Index\n\n")
+        for project in target_projects:
+            ref_index_file.write(f"- [{project.name}]({sanitize_name(project.name)}/index.md)\n") # type: ignore

make-venv.sh

@@ -1,6 +1,8 @@
 #! /usr/bin/env bash
 
-python3 -m venv .venv
+VERSION="${1:-3}"
+
+python${VERSION} -m venv .venv
 ./.venv/bin/pip install --upgrade pip
 ./.venv/bin/pip install -r requirements.txt
 

mkdocs.yml

@@ -0,0 +1,4 @@
+site_name: Reference Documentation
+docs_dir: reference
+theme: 
+  name: material

pyproject.toml

@@ -0,0 +1,11 @@
+[build-system]
+requires = ["setuptools", "wheel", "build"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "devops"
+version = "0.1.0"
+requires-python = ">=3.12"
+
+[tool.setuptools]
+packages = ["devops"]

requirements.txt

@@ -2,3 +2,4 @@ debugpy==1.8.17
 azure-identity==1.25.1
 requests==2.32.5
 loki-logger-handler==1.1.2
+mkdocs-material>=1.5.2

sk/__init__.py

@@ -0,0 +1 @@
+# My helper tools

sk/logger.py

@@ -1,18 +1,12 @@
 import logging
 from loki_logger_handler.loki_logger_handler import LokiLoggerHandler
-from uuid import uuid4, UUID
 
 LOG_CONSOLE = 1
 LOG_FILE = 2
 LOG_LOKI = 4
-run_id: UUID
 
 # Quick non-intrusive debug check for sk.devops logger
 def setup(name: str, handlers: int) -> logging.Logger:
-    global run_id
-    # Generate a unique run ID.
-    run_id = uuid4()
-
     logger = logging.getLogger(name)
     logger.setLevel(logging.DEBUG)
 
@@ -42,31 +36,9 @@ def setup(name: str, handlers: int) -> logging.Logger:
 
     return logger
 
-def debug(logger: logging.Logger, message: str, **kwargs):
-    global run_id
-    logger.debug(message, extra={"run_id": str(run_id), **kwargs})
-
-def info(logger: logging.Logger, message: str, **kwargs):
-    global run_id
-    logger.info(message, extra={"run_id": str(run_id), **kwargs})
-
-def warning(logger: logging.Logger, message: str, **kwargs):
-    global run_id
-    logger.warning(message, extra={"run_id": str(run_id), **kwargs})
-
-def error(logger: logging.Logger, message: str, **kwargs):
-    global run_id
-    logger.error(message, extra={"run_id": str(run_id), **kwargs})
-
-def critical(logger: logging.Logger, message: str, **kwargs):
-    global run_id
-    logger.critical(message, extra={"run_id": str(run_id), **kwargs})
-
 def log_entity_creation(logger: logging.Logger, entity_class: type, entity_key: str):
-    global run_id
     logger.debug(f'Created new "{entity_class.__name__}" object with key: "{entity_key}"',
                  extra={
                      "entity_class": entity_class.__name__,
-                     "entity_key": entity_key,
-                     "run_id": str(run_id)
+                     "entity_key": entity_key
                  })

tests.py

@@ -1,8 +1,8 @@
 #!/usr/bin/env python3
 import unittest
 import requests
-from sk.devops import Organization, Repository, Project, Item
-from sk.azure import get_token
+from devops.devops import Organization, Repository, Project, Item
+from devops.azure import get_token
 
 # Get the token outside the test class to speed up tests.
 # Each Unit test instantinates the class, so doing it here avoids repeated authentication.