From 74c225fa458bed2e1d2fd11cbaa3af05dbf1fbc7 Mon Sep 17 00:00:00 2001
From: Ayan Sinha Mahapatra <ayansmahapatra@gmail.com>
Date: Tue, 8 Aug 2023 17:21:57 +0530
Subject: [PATCH] Add support for custom repository URLs

Adds support for getting packages from custom repository URLs,
in the case of maven packages.

Signed-off-by: Ayan Sinha Mahapatra <ayansmahapatra@gmail.com>
---
 minecode/visitors/maven.py | 21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

diff --git a/minecode/visitors/maven.py b/minecode/visitors/maven.py
index 023f672a..293a6bb8 100644
--- a/minecode/visitors/maven.py
+++ b/minecode/visitors/maven.py
@@ -61,6 +61,9 @@
     logger.setLevel(logging.DEBUG)
 
 
+MAVEN_BASE_URL = 'https://repo1.maven.org/maven2'
+
+
 class GzipFileWithTrailing(gzip.GzipFile):
     """
     A subclass of gzip.GzipFile supporting files with trailing garbage. Ignore
@@ -104,7 +107,7 @@ def get_seeds(self):
         # also has a npm mirrors: https://maven-eu.nuxeo.org/nexus/#view-repositories;npmjs~browsestorage
 
 
-def get_pom_text(namespace, name, version, qualifiers={}):
+def get_pom_text(namespace, name, version, qualifiers={}, base_url=MAVEN_BASE_URL):
     """
     Return the contents of the POM file of the package described by the purl
     field arguments in a string.
@@ -116,7 +119,8 @@ def get_pom_text(namespace, name, version, qualifiers={}):
         namespace=namespace,
         name=name,
         version=version,
-        qualifiers=qualifiers
+        qualifiers=qualifiers,
+        base_url=base_url,
     )
     # Get and parse POM info
     pom_url = urls['api_data_url']
@@ -279,11 +283,17 @@ def map_maven_package(package_url, package_content):
     db_package = None
     error = ''
 
+    if "repository_url" in package_url.qualifiers:
+        base_url = package_url.qualifiers["repository_url"]
+    else:
+        base_url = MAVEN_BASE_URL
+
     pom_text = get_pom_text(
         namespace=package_url.namespace,
         name=package_url.name,
         version=package_url.version,
-        qualifiers=package_url.qualifiers
+        qualifiers=package_url.qualifiers,
+        base_url=base_url,
     )
     if not pom_text:
         msg = f'Package does not exist on maven: {package_url}'
@@ -302,11 +312,14 @@ def map_maven_package(package_url, package_content):
         ancestor_pom_texts=ancestor_pom_texts,
         package=package
     )
+    
+
     urls = get_urls(
         namespace=package_url.namespace,
         name=package_url.name,
         version=package_url.version,
-        qualifiers=package_url.qualifiers
+        qualifiers=package_url.qualifiers,
+        base_url=base_url,
     )
     # In the case of looking up a maven package with qualifiers of
     # `classifiers=sources`, the purl of the package created from the pom does