Add a little script to check for missing packages in comps

It would be nice to make 'package is listed in group, but does
not exist' a fatal error (at least optionally) for dnf etc, but
we can't really do that at present because it's *not* been a
fatal error for so long that our comps have tons of stale
entries for packages that no longer exist, and also entries that
aren't properly archified for packages which only exist on some
arches. This script pokes through the comps file for current
Rawhide and identifies all (I hope) pkgreqs which specify a
package that is not actually available in current Rawhide, per
arch.

Signed-off-by: Adam Williamson <awilliam@redhat.com>

check-missing

@@ -0,0 +1,49 @@
+#!/bin/python3
+
+# This script looks for the 'latest' comps-fXX.xml.in file, assumes it's
+# for Rawhide, and looks through its `pkgreq` lines for ones that specify
+# packages that do not currently exist in Rawhide. It is arch-aware. It
+# expects to be run on a Fedora system with network access, as it will
+# try to query the 'rawhide' dnf repo to get lists of currently-existing
+# packages.
+
+import glob
+import subprocess
+import xml.etree.ElementTree as ET
+
+ARCHES = ('aarch64', 'armv7hl', 'i686', 'ppc64le', 's390x', 'x86_64')
+
+# gather package lists. this eats lots of RAM. I don't care.
+pkgs = {}
+for arch in ARCHES:
+    pkgtext = subprocess.run(('dnf', '--forcearch={}'.format(arch), '--disablerepo=*', '--enablerepo=rawhide', 'repoquery', '--qf=%{NAME}'), capture_output=True, text=True).stdout
+    pkgs[arch] = pkgtext.splitlines()
+
+# find the *latest* comps file (assume it's rawhide)
+compsfiles = glob.glob('comps-f*.xml.in')
+latest = sorted(compsfiles, key=lambda x: int(''.join(c for c in x if c.isdigit())))[-1]
+# find package reqs in comps
+root = ET.parse(latest).getroot()
+pkgreqs = root.findall('.//packagereq')
+
+# check!
+for pkgreq in pkgreqs:
+    # list of arches the package is missing on
+    missing = []
+    # arches the package is listed for (if no 'arch' key, it's listed for all)
+    reqarches = pkgreq.get('arch', '').replace('armhfp', 'armv7hl').replace('i386', 'i686').replace('ppc64,','')
+    if reqarches:
+        reqarches = reqarches.split(',')
+    else:
+        reqarches = ARCHES
+
+    # do the actual check, append arch to 'missing' if it's not there
+    for arch in reqarches:
+        if arch in pkgs and pkgreq.text not in pkgs[arch]:
+            missing.append(arch)
+
+    # print the result
+    if missing == list(ARCHES):
+        print('Package {} not found for any arch'.format(pkgreq.text))
+    elif missing:
+        print('Package {} not found for arches {}'.format(pkgreq.text, ', '.join(missing)))