Replace deprecated cgi module with multipart for parsing multipart PO…

…ST requests (#28)

* Replace deprecated cgi module with multipart

* Update to supported Python versions

Drop Python 3.7 and 3.8 from CI, as both had test failures due to
other packages dropping support for these now EOL versions.

Add Python 3.11 and 3.12 to CI.

.github/workflows/ci.yaml

@@ -6,7 +6,7 @@ jobs:
     strategy:
       max-parallel: 3
       matrix:
-        python-version: ["3.7", "3.8", "3.9", "3.10"]
+        python-version: ["3.9", "3.10", "3.11", "3.12"]
 
     steps:
       - name: checkout
@@ -20,7 +20,7 @@ jobs:
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
-          pip install -U black pytest pytest-cov
+          pip install -U black pytest pytest-cov setuptools[core]
           python setup.py -q install
 
       - name: Style Check

cdxj_indexer/postquery.py

@@ -1,17 +1,19 @@
-from warcio.utils import to_native_str
+import base64
+import json
+import sys
 
 from urllib.parse import unquote_plus, urlencode
 from io import BytesIO
 
+from multipart import MultipartParser
+from warcio.utils import to_native_str
+
 from cdxj_indexer.amf import amf_parse
 
-import base64
-import cgi
-import json
-import sys
 
 MAX_QUERY_LENGTH = 4096
 
+
 # ============================================================================
 def append_method_query_from_req_resp(req, resp):
     len_ = req.http_headers.get_header("Content-Length")
@@ -93,25 +95,16 @@ def handle_binary(query_data):
             query = handle_binary(query_data)
 
     elif mime.startswith("multipart/"):
-        env = {
-            "REQUEST_METHOD": "POST",
-            "CONTENT_TYPE": mime,
-            "CONTENT_LENGTH": len(query_data),
-        }
-
-        args = dict(fp=BytesIO(query_data), environ=env, keep_blank_values=True)
-
-        args["encoding"] = "utf-8"
-
         try:
-            data = cgi.FieldStorage(**args)
-        except ValueError:
+            boundary = mime.split("boundary=")[1]
+            parser = MultipartParser(BytesIO(query_data), boundary, charset="utf8")
+        except (ValueError, IndexError):
             # Content-Type multipart/form-data may lack "boundary" info
             query = handle_binary(query_data)
         else:
             values = []
-            for item in data.list:
-                values.append((item.name, item.value))
+            for part in parser:
+                values.append((part.name, part.value))
 
             query = urlencode(values, True)
 

setup.py

@@ -51,6 +51,7 @@ def run_tests(self):
         # temp fix for requests
         "idna<3.0",
         "py3amf",
+        "multipart",
     ],
     zip_safe=True,
     entry_points="""