From b780c80b92d2386ce5c6bb01360a046de8d3c91c Mon Sep 17 00:00:00 2001
From: Rohan Devasthale <rdevasth@redhat.com>
Date: Wed, 2 Oct 2024 12:01:56 -0400
Subject: [PATCH] Initial commit for poc

---
 requirements.txt          |  1 +
 src/fromager/candidate.py |  3 +++
 src/fromager/resolver.py  | 15 +++++++++++++++
 3 files changed, 19 insertions(+)

diff --git a/requirements.txt b/requirements.txt
index 90212171..ce6607e5 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -18,3 +18,4 @@ tomlkit
 tqdm
 virtualenv
 wheel
+mousebender
diff --git a/src/fromager/candidate.py b/src/fromager/candidate.py
index e7a824a3..4bd3d74d 100644
--- a/src/fromager/candidate.py
+++ b/src/fromager/candidate.py
@@ -25,6 +25,7 @@ def __init__(
         name: str,
         version: Version,
         url: str,
+        metadata_hash: tuple[str, str] | None,
         extras: typing.Iterable[str] | None = None,
         is_sdist: bool | None = None,
         build_tag: BuildTag = (),
@@ -35,9 +36,11 @@ def __init__(
         self.extras = extras
         self.is_sdist = is_sdist
         self.build_tag = build_tag
+        self.metadata_hash = metadata_hash
 
         self._metadata: Metadata | None = None
         self._dependencies: list[Requirement] | None = None
+        self.metadata_url: str | None = None
 
     def __repr__(self) -> str:
         if not self.extras:
diff --git a/src/fromager/resolver.py b/src/fromager/resolver.py
index b8c3ca56..15f04540 100644
--- a/src/fromager/resolver.py
+++ b/src/fromager/resolver.py
@@ -11,6 +11,7 @@
 from operator import attrgetter
 from platform import python_version
 from urllib.parse import urljoin, urlparse
+import mousebender
 
 import github
 import html5lib
@@ -105,7 +106,19 @@ def get_project_from_pypi(
     simple_index_url = sdist_server_url.rstrip("/") + "/" + project + "/"
     logger.debug("%s: getting available versions from %s", project, simple_index_url)
     data = session.get(simple_index_url).content
+    metadata_response =  session.get(simple_index_url, headers={"accept": mousebender.simple.ACCEPT_SUPPORTED})
     doc = html5lib.parse(data, namespaceHTMLElements=False)
+    parsed_metadata_doc = mousebender.simple.parse_project_details(metadata_response.content, metadata_response.headers["content-type"], project)
+    files = parsed_metadata_doc["files"]
+    
+    for dct in files:
+        if dct.get("yanked", False):
+            continue
+        hash_val = dct.get("data-dist-info-metadata")
+        if hash_val:
+            _metadata_hash = hash_val
+            _metadata_url = dct.get("url") + ".metadata"
+
     for i in doc.findall(".//a"):
         candidate_url = urljoin(simple_index_url, i.attrib["href"])
         py_req = i.attrib.get("data-requires-python")
@@ -174,10 +187,12 @@ def get_project_from_pypi(
             name,
             version,
             url=candidate_url,
+            metadata_hash=_metadata_hash,
             extras=extras,
             is_sdist=is_sdist,
             build_tag=build_tag,
         )
+        c.metadata_url = _metadata_url
         if DEBUG_RESOLVER:
             logger.debug(
                 "%s: candidate %s (%s) %s", project, filename, c, candidate_url