record package metadata that describes co-installability

Specifically all entries in the Conflicts header are saved in the
conflict table, all entries in the Provides header are saved in the
provide table (to cover conflicts with virtual packages) and packages
using dpkg-divert in preinst get a magic "_dpkg-divert" entry in their
conflict table. With this metadata it should be possible to compute
undeclared file conflicts.

1 files changed, 12 insertions, 2 deletions

diff --git a/importpkg.py b/importpkg.py
index aeccda5..f3868ff 100755
--- a/importpkg.py
+++ b/importpkg.py
@@ -58,14 +58,24 @@ def process_package(filelike, hash_functions):
                 raise ValueError("unexpected control.tar.gz")
             state = "control"
             tf = tarfile.open(fileobj=af, mode="r|gz")
+            meta = None
+            use_dpkg_divert = False
             for elem in tf:
+                if elem.name == "./preinst":
+                    if "dpkg-divert" in tf.extractfile(elem).read():
+                        use_dpkg_divert = True
+                    continue
                 if elem.name != "./control":
                     continue
                 if state != "control":
                     raise ValueError("duplicate control file")
                 state = "control_file"
-                yield process_control(tf.extractfile(elem).read())
-                break
+                meta = process_control(tf.extractfile(elem).read())
+            if meta is None:
+                raise ValueError("control file not found")
+            if use_dpkg_divert:
+                meta["conflicts"].add("_dpkg-divert")
+            yield meta
             continue
         elif name == "data.tar.gz":
             tf = tarfile.open(fileobj=af, mode="r|gz")