Make corrections to address issues pointed out by Ruff

mdapi/confdata/standard.py

@@ -26,7 +26,7 @@
 """
 
 # url to the database server:
-DB_FOLDER = "/var/tmp"
+DB_FOLDER = "/var/tmp"  # noqa : S108
 
 LOGGING = {
     "version": 1,

mdapi/database/__init__.py

@@ -20,11 +20,3 @@
 License and may only be used or replicated with the express permission
 of Red Hat, Inc.
 """
-
-
-def check_archive_validity():
-    """
-    This is a workaround and not a fix.
-    See https://github.com/fedora-infra/mdapi/issues/161
-    """
-    return True

mdapi/database/base.py

@@ -28,7 +28,7 @@
 
 
 def index_database(name, tempdtbs):
-    servlogr.logrobjc.info("[%s] Indexing database %s" % (name, tempdtbs))
+    servlogr.logrobjc.info(f"[{name}] Indexing database {tempdtbs}")
     if tempdtbs.endswith("primary.sqlite"):
         # TODO: Try the "with sqlite3.connect(tempdtbs) as connobjc" statement here
         connobjc = sqlite3.connect(tempdtbs)
@@ -80,24 +80,21 @@ def change_row_to_package(self, rowe):
     def obtain_all_rows(self, location, tableobj, cacheobj):
         connobjc = sqlite3.connect(location)
         sqlquery = queries.get(tableobj, DEFAULT_QUERY).format(table=tableobj)
-        for indx, rowe in enumerate(connobjc.execute(sqlquery)):
+        for indx, rowe in enumerate(connobjc.execute(sqlquery)):  # noqa : B007
             if tableobj in self.cache_dependent_tables:
                 if rowe[0] in cacheobj:
                     yield (cacheobj[rowe[0]], *rowe[1:])
                 else:
-                    servlogr.logrobjc.debug(
-                        "[%s] %s does not appear in the %s cache for %s"
-                        % (self.name, rowe[0], tableobj, location)
-                    )
-                    servlogr.logrobjc.debug("[%s] Dropping from comparison" % self.name)
+                    servlogr.logrobjc.debug(f"[{self.name}] {rowe[0]} does not appear in the {tableobj} cache for {location}")  # noqa : E501
+                    servlogr.logrobjc.debug(f"[{self.name}] Dropping from comparison")
             else:
                 yield rowe
         connobjc.close()
 
     def build_cache(self, location, tableobj, cacheobjc):
         connobjc = sqlite3.connect(location)
         sqlquery = queries.get(tableobj, DEFAULT_QUERY).format(table=tableobj)
-        for pkgId, pkgname, *args in connobjc.execute(sqlquery):
+        for pkgId, pkgname, *args in connobjc.execute(sqlquery):  # noqa : B007
             cacheobjc[pkgId] = pkgname
         connobjc.close()
 
@@ -108,7 +105,7 @@ def should_compare(self, tableobj):
         return True
 
     def main(self):
-        servlogr.logrobjc.info("[%s] Comparing %s against %s" % (self.name, self.dbsA, self.dbsB))
+        servlogr.logrobjc.info(f"[{self.name}] Comparing {self.dbsA} against {self.dbsB}")
 
         tablistA = list(self.obtain_table_names(self.dbsA))
         tablistB = list(self.obtain_table_names(self.dbsB))
@@ -121,12 +118,12 @@ def main(self):
             # We have never downloaded this before...
             # so we have nothing to compare it against. Just return and say there
             # are "no differences".
-            servlogr.logrobjc.warning(
-                "[%s] Database empty - %s cannot compare" % (self.name, self.dbsB)
-            )
+            servlogr.logrobjc.warning(f"[{self.name}] Database empty - {self.dbsB} cannot compare")
             return set()
 
-        assert len(tablistA) == len(tablistB), "Cannot compare disparate databases"
+        if len(tablistA) != len(tablistB):
+            raise ValueError("Cannot compare disparate databases")
+
         # These should be the same
         tblelist = tablistA = tablistB
 

mdapi/database/main.py

@@ -32,65 +32,61 @@
 import time
 from xml.etree import ElementTree
 
+import pyzstd
 import requests
 from fedora_messaging.api import publish
 from fedora_messaging.exceptions import ConnectionException, PublishReturned
 from mdapi_messages.messages import RepoUpdateV1
 
 from mdapi.confdata import servlogr, standard
-from mdapi.database import check_archive_validity
 from mdapi.database.base import compare_databases, index_database
 
 
 def list_branches(status="Active"):
-    urlx = "%s/api/collections?clt_status=%s" % (standard.PKGDB2_URL, status)
-    resp = requests.get(urlx, verify=standard.PKGDB2_VERIFY)
+    urlx = f"{standard.PKGDB2_URL}/api/collections?clt_status={status}"
+    resp = requests.get(urlx, verify=standard.PKGDB2_VERIFY)  # noqa : S113
     resp.raise_for_status()
     data = resp.json()
     servlogr.logrobjc.info("Branches metadata acquired.")
     return data["collections"]
 
 
 def fetch_database(name, repmdurl, location):
-    servlogr.logrobjc.info("[%s] Downloading file %s to %s" % (name, repmdurl, location))
-    resp = requests.get(repmdurl, verify=standard.DL_VERIFY)
+    servlogr.logrobjc.info(f"[{name}] Downloading file {repmdurl} to {location}")
+    resp = requests.get(repmdurl, verify=standard.DL_VERIFY)  # noqa : S113
     resp.raise_for_status()
     with open(location, "wb") as filestrm:
         filestrm.write(resp.content)
 
 
 def extract_database(name, arcvname, location):
-    servlogr.logrobjc.info("[%s] Extracting %s to %s" % (name, arcvname, location))
+    servlogr.logrobjc.info(f"[{name}] Extracting {arcvname} to {location}")
     if arcvname.endswith(".xz"):
         with lzma.open(arcvname) as inp, open(location, "wb") as otptfile:
             otptfile.write(inp.read())
     elif arcvname.endswith(".tar.gz"):
         with tarfile.open(arcvname) as tararchv:
-            tararchv.extractall(path=location, members=check_archive_validity())
+            tararchv.extractall(path=location, filter="data")
     elif arcvname.endswith(".gz"):
         with gzip.open(arcvname, "rb") as inp, open(location, "wb") as otptfile:
             otptfile.write(inp.read())
     elif arcvname.endswith(".bz2"):
         with bz2.open(arcvname) as inp, open(location, "wb") as otptfile:
             otptfile.write(inp.read())
     elif arcvname.endswith(".zst"):
-        import pyzstd
-
         with open(arcvname, "rb") as inp, open(location, "wb") as otptfile:
             pyzstd.decompress_stream(inp, otptfile)
     else:
-        servlogr.logrobjc.error("Could not extract %s" % (arcvname))
+        servlogr.logrobjc.error(f"Could not extract {arcvname}")
         raise NotImplementedError(arcvname)
 
 
 def publish_changes(name, packages, repmdurl):
-    servlogr.logrobjc.info("[%s] Publishing differences to Fedora Messaging bus" % (name))
+    servlogr.logrobjc.info(f"[{name}] Publishing differences to Fedora Messaging bus")
 
     modified = bool(packages)
     if not modified:
-        servlogr.logrobjc.warning(
-            "[%s] No real changes detected - Skipping publishing on Fedora Messaging bus" % (name)
-        )
+        servlogr.logrobjc.warning(f"[{name}] No real changes detected - Skipping publishing on Fedora Messaging bus")  # noqa : E501
         return
 
     # Just publish the suffix of the URL.  The prefix is dl.fedoraproject.org
@@ -100,7 +96,7 @@ def publish_changes(name, packages, repmdurl):
     # talking about.
 
     urlx = "/".join(repmdurl.split("/")[4:])
-    servlogr.logrobjc.info("[%s] URL %s" % (name, urlx))
+    servlogr.logrobjc.info(f"[{name}] URL {urlx}")
 
     mesgobjc = RepoUpdateV1(
         body=dict(
@@ -113,17 +109,13 @@ def publish_changes(name, packages, repmdurl):
     try:
         publish(mesgobjc)
     except PublishReturned as excp:
-        servlogr.logrobjc.error(
-            "Fedora Messaging broker rejected message %s - %s" % (mesgobjc.id, excp)
-        )
+        servlogr.logrobjc.error(f"Fedora Messaging broker rejected message {mesgobjc.id} - {excp}")
     except ConnectionException as excp:
-        servlogr.logrobjc.error(
-            "Error occurred while sending message %s - %s" % (mesgobjc.id, excp)
-        )
+        servlogr.logrobjc.error(f"Error occurred while sending message {mesgobjc.id} - {excp}")
 
 
 def install_database(name, srce, dest):
-    servlogr.logrobjc.info("[%s] Installing %s to %s" % (name, srce, dest))
+    servlogr.logrobjc.info(f"[{name}] Installing {srce} to {dest}")
     shutil.move(srce, dest)
 
 
@@ -155,10 +147,10 @@ def process_repo(repo):
     Retrieve the repo metadata at the given URL and store them using the provided name.
     """
     urlx, name = repo
-    repmdurl = "%s/repomd.xml" % urlx
-    response = requests.get(repmdurl, verify=standard.DL_VERIFY)
+    repmdurl = f"{urlx}/repomd.xml"
+    response = requests.get(repmdurl, verify=standard.DL_VERIFY)  # noqa : S113
     if not response:
-        servlogr.logrobjc.error("[%s] Failed to obtain %s - %s" % (name, repmdurl, response))
+        servlogr.logrobjc.error(f"[{name}] Failed to obtain {repmdurl} - {response}")
         return
 
     # Parse the XML document and get a list of locations and their SHAsum
@@ -167,7 +159,7 @@ def process_repo(repo):
             node.find("repo:location", standard.repomd_xml_namespace),
             node.find("repo:open-checksum", standard.repomd_xml_namespace),
         )
-        for node in ElementTree.fromstring(response.text)
+        for node in ElementTree.fromstring(response.text)  # noqa : S314
     )
 
     # Extract out the attributes that we are really interested in
@@ -189,25 +181,25 @@ def process_repo(repo):
     cacA, cacB = {}, {}
 
     if not filelist:
-        servlogr.logrobjc.warning("No SQLite database could be found in %s" % (urlx))
+        servlogr.logrobjc.warning(f"No SQLite database could be found in {urlx}")
 
     for filename, hashdata, hashtype in filelist:
-        repmdurl = "%s/%s" % (urlx, filename)
+        repmdurl = f"{urlx}/{filename}"
 
         # First, determine if the file has changed by comparing hash
         database = None
         if "primary.sqlite" in filename:
-            database = "mdapi-%s-primary.sqlite" % name
+            database = f"mdapi-{name}-primary.sqlite"
         elif "filelists.sqlite" in filename:
-            database = "mdapi-%s-filelists.sqlite" % name
+            database = f"mdapi-{name}-filelists.sqlite"
         elif "other.sqlite" in filename:
-            database = "mdapi-%s-other.sqlite" % name
+            database = f"mdapi-{name}-other.sqlite"
 
         # Have we downloaded this before?
         # Did it change?
         destfile = os.path.join(standard.DB_FOLDER, database)
         if not needs_update(destfile, hashdata, hashtype):
-            servlogr.logrobjc.info("[%s] No change detected from %s" % (name, repmdurl))
+            servlogr.logrobjc.info(f"[{name}] No change detected from {repmdurl}")
             continue
 
         # Creating temporary directories with formatted names to remove them later easily, if needed
@@ -224,9 +216,7 @@ def process_repo(repo):
                 packages = compare_databases(name, tempdtbs, destfile, cacA, cacB).main()
                 publish_changes(name, packages, repmdurl)
             else:
-                servlogr.logrobjc.warning(
-                    "[%s] Not publishing to Fedora Messaging bus - Not comparing databases" % (name)
-                )
+                servlogr.logrobjc.warning(f"[{name}] Not publishing to Fedora Messaging bus - Not comparing databases")  # noqa : E501
             install_database(name, tempdtbs, destfile)
 
 
@@ -241,32 +231,30 @@ def index_repositories():
         versdata = rels["version"]
         if versdata == "devel":
             versdata = "rawhide"
-        urlx = "%s/pub/fedora/linux/development/%s/Everything/x86_64/os/repodata" % (
-            standard.DL_SERVER,
-            versdata,
-        )
-        servlogr.logrobjc.info(
-            "Acquired repo for %s/%s of '%s' branch at %s"
-            % (rels["koji_name"], versdata, rels["status"], urlx)
-        )
+        urlx = f"{standard.DL_SERVER}/pub/fedora/linux/development/{versdata}/Everything/x86_64/os/repodata"  # noqa : E501
+        servlogr.logrobjc.info(f"Acquired repo for {rels['koji_name']}/{versdata} of '{rels['status']}' branch at {urlx}")  # noqa : E501
         repolist.append((urlx, rels["koji_name"]))
         urlx = urlx.replace("/x86_64/os/", "/source/tree/")
-        repolist.append((urlx, "src_%s" % rels["koji_name"]))
+        repolist.append((urlx, f"src_{rels['koji_name']}"))
 
     urls = {
         "Fedora Linux": [
-            "%s/pub/fedora/linux/releases/%s/Everything/x86_64/os/repodata",
-            "%s/pub/fedora/linux/updates/%s/Everything/x86_64/repodata",
-            "%s/pub/fedora/linux/updates/testing/%s/Everything/x86_64/repodata",
+            "{dlserver}/pub/fedora/linux/releases/{versname}/Everything/x86_64/os/repodata",
+            "{dlserver}/pub/fedora/linux/updates/{versname}/Everything/x86_64/repodata",
+            "{dlserver}/pub/fedora/linux/updates/testing/{versname}/Everything/x86_64/repodata",
         ],
         "Fedora EPEL": [
-            "%s/pub/epel/%s/x86_64/repodata/",
-            "%s/pub/epel/testing/%s/x86_64/repodata",
+            "{dlserver}/pub/epel/{versname}/x86_64/repodata/",
+            "{dlserver}/pub/epel/testing/{versname}/x86_64/repodata",
         ],
     }
 
     urls["Fedora"] = urls["Fedora Linux"]
-    repodict = {"fedora": ["%s", "%s-updates", "%s-updates-testing"], "epel": ["%s", "%s-testing"]}
+
+    repodict = {
+        "fedora": ["{rlid}", "{rlid}-updates", "{rlid}-updates-testing"],
+        "epel": ["{rlid}", "{rlid}-testing"]
+    }
 
     # Obtain the stable repos
     stable_releases = list_branches(status="Active")
@@ -276,22 +264,22 @@ def index_repositories():
         versdata = rels["version"]
         for jndx, urli in enumerate(urls[rels["name"]]):
             if rels["name"] in ("Fedora Linux", "Fedora"):
-                name = repodict["fedora"][jndx] % rels["koji_name"]
+                name = repodict["fedora"][jndx].format(ridx = rels["koji_name"])
             elif rels["name"] == "Fedora EPEL" and versdata == "8":
-                name = repodict["epel"][jndx] % rels["koji_name"]
+                name = repodict["epel"][jndx].format(rlid = rels["koji_name"])
                 urli = urli.replace("/x86_64/", "/Everything/x86_64/")
             elif rels["name"] == "Fedora EPEL" and versdata == "9":
-                name = repodict["epel"][jndx] % rels["koji_name"]
+                name = repodict["epel"][jndx].format(rlid = rels["koji_name"])
                 urli = urli.replace("/x86_64/", "/Everything/x86_64/")
             else:
-                name = repodict["epel"][jndx] % rels["koji_name"]
-            rurl = urli % (standard.DL_SERVER, versdata)
+                name = repodict["epel"][jndx].format(rlid = rels["koji_name"])
+            rurl = urli.format(dlserver = standard.DL_SERVER, versname = versdata)
             repolist.append((rurl, name))
             rurl = rurl.replace("/x86_64/os", "/source/tree")
-            repolist.append((rurl, "src_%s" % name))
+            repolist.append((rurl, f"src_{name}"))
 
     # Finish with the koji repo
-    repolist.append(("%s/rawhide/latest/x86_64/repodata" % standard.KOJI_REPO, "koji"))
+    repolist.append((f"{standard.KOJI_REPO}/rawhide/latest/x86_64/repodata", "koji"))
 
     # In serial
     for repo in repolist:

mdapi/main.py

@@ -47,8 +47,8 @@ def main(conffile=None):
     if conffile is not None:
         # Load the configuration file to use
         CONFIG = {}
-        with open(conffile, "r") as confobjc:
-            exec(compile(confobjc.read(), conffile, "exec"), CONFIG)
+        with open(conffile) as confobjc:
+            exec(compile(confobjc.read(), conffile, "exec"), CONFIG)  # noqa : S102
         compile_configuration(CONFIG)
 
 
@@ -70,7 +70,7 @@ def serveapp():
                 APPSERVE["logging"]["level"],
             )
         )
-        subprocess.run(startcmd.split())
+        subprocess.run(startcmd.split())  # noqa : S603
     except KeyError:
         print("Invalid configuration detected")
         return 1