Optimize logic for GitLab URL parsing

pyproject.toml

@@ -23,7 +23,7 @@ description = "A package to perform IP Analysis for GitHub and GitLab projects"
 readme = "README.md"
 license = { text = "EPL-2.0" }
 requires-python = ">=3.9"
-dependencies = ["chardet==5.2.0", "Jinja2==3.1.6", "python-gitlab==6.1.0", "PyGithub==2.6.1"]
+dependencies = ["chardet==5.2.0", "Jinja2==3.1.6", "PyGithub==2.6.1", "python-gitlab==6.1.0", "requests==2.32.4"]
 classifiers = [
     "Programming Language :: Python :: 3",
     "License :: OSI Approved :: Eclipse Public License 2.0 (EPL-2.0)",

src/eclipse/ipa/glab/remote.py

@@ -20,6 +20,7 @@ from datetime import datetime
 from pathlib import Path
 from subprocess import PIPE, Popen
 from time import sleep
+from urllib.parse import urlparse
 
 import gitlab
 
@@ -39,8 +40,9 @@ def get_dependency_locations(gl, config, repositories):
         line_count = 0
         for repo in repositories:
             # From URL to ID (required by python-gitlab)
-            repo = repo.replace(config.get('General', 'GitlabURL', fallback='https://gitlab.eclipse.org') + '/', '')
-            repo = repo.replace('.git', '')
+            repo = urlparse(repo).path.lstrip('/').removesuffix('.git')
+            # Remove any trailing slashes
+            repo = repo.rstrip('/')
 
             dependency_locations[repo] = {}
             line_count = line_count + 1
@@ -63,8 +65,9 @@ def get_dependency_locations(gl, config, repositories):
                         tokens = line.strip().split(';')
                         proj_id = tokens[0]
                         # Get rid of base URL, if included (required by python-gitlab)
-                        proj_id = proj_id.replace(config.get('General', 'GitlabURL',
-                                                             fallback='https://gitlab.eclipse.org') + '/', '')
+                        proj_id = urlparse(proj_id).path.lstrip('/').removesuffix('.git')
+                        # Remove any trailing slashes
+                        proj_id = proj_id.rstrip('/')
                         try:
                             dependency_locations[proj_id][tokens[2]].append(tokens[1])
                         except KeyError:
@@ -88,8 +91,9 @@ def get_dependency_locations(gl, config, repositories):
                         line_count = line_count + 1
                         proj_id = line.strip()
                         # Get rid of base URL, if included (required by python-gitlab)
-                        proj_id = proj_id.replace(config.get('General', 'GitlabURL',
-                                                             fallback='https://gitlab.eclipse.org') + '/', '')
+                        proj_id = urlparse(proj_id).path.lstrip('/').removesuffix('.git')
+                        # Remove any trailing slashes
+                        proj_id = proj_id.rstrip('/')
                         dependency_locations[proj_id] = {}
             print("Read " + str(line_count) + " project(s) from " + input_file)
             logger.info("Read " + str(line_count) + " project(s) from " + input_file)
@@ -101,8 +105,10 @@ def get_dependency_locations(gl, config, repositories):
     elif config.has_option('Groups', 'BaseGroupID'):
         # Get rid of base URL, if included in BaseGroupID (required by python-gitlab)
         base_group_id = config.get('Groups', 'BaseGroupID')
-        base_group_id = base_group_id.replace(config.get('General', 'GitlabURL',
-                                             fallback='https://gitlab.eclipse.org') + '/', '')
+        base_group_id = urlparse(base_group_id).path.lstrip('/')
+        # Remove any trailing slashes
+        base_group_id = base_group_id.rstrip('/')
+
         try:
             # Set base group ID to work
             base_group = gl.groups.get(base_group_id, lazy=True)
@@ -132,7 +138,11 @@ def get_dependency_locations(gl, config, repositories):
             dependency_locations[proj.path_with_namespace] = {}
     # Work with a single project ID
     elif config.has_option('Projects', 'SingleProjectID'):
-        dependency_locations[config.get('Projects', 'SingleProjectID')] = {}
+        # Get rid of base URL, if included (required by python-gitlab)
+        repo = urlparse(config.get('Projects', 'SingleProjectID')).path.lstrip('/').removesuffix('.git')
+        # Remove any trailing slashes
+        repo = repo.rstrip('/')
+        dependency_locations[repo] = {}
     else:
         # No valid option provided, exit
         print("Insufficient parameters provided. Exiting...")