cloud-init-dev team mailing list archive

[Scott Moser <ssmoser2+ubuntu@xxxxxxxxx>]

Scott Moser has proposed merging ~smoser/cloud-init:feature/reproducible-builds into cloud-init:master.

Commit message:
setup.py: Do not include rendered files in SOURCES.txt

cloud-init renders template files during its run of setup.py.
Those rendered files were in a temp dir and were making their
way into the SOURCES.txt file.

We do not want to include those, so here we explicitly prune them out.

The issue of SOURCES.txt containing tmp files was reproducible with:
  $ rm -Rf cloud_init.egg-info
  $ git clean --force
  $ rm -Rf ../root.d; python3 setup.py install --root=../root.d
  $ grep ^tmp cloud_init.egg-info/SOURCES.txt

The SOURCES.txt under ../root.d/ would have tmp files.



Requested reviews:
  cloud-init commiters (cloud-init-dev)

For more details, see:
https://code.launchpad.net/~smoser/cloud-init/+git/cloud-init/+merge/334991
-- 
Your team cloud-init commiters is requested to review the proposed merge of ~smoser/cloud-init:feature/reproducible-builds into cloud-init:master.

diff --git a/setup.py b/setup.py
index bf697d7..bc3f52a 100755
--- a/setup.py
+++ b/setup.py
@@ -18,11 +18,14 @@ import tempfile
 
 import setuptools
 from setuptools.command.install import install
+from setuptools.command.egg_info import egg_info
 
 from distutils.errors import DistutilsArgError
 
 import subprocess
 
+RENDERED_TMPD_PREFIX = "RENDERED_TEMPD"
+
 
 def is_f(p):
     return os.path.isfile(p)
@@ -107,7 +110,7 @@ def render_tmpl(template):
         return template
 
     topdir = os.path.dirname(sys.argv[0])
-    tmpd = tempfile.mkdtemp(dir=topdir)
+    tmpd = tempfile.mkdtemp(dir=topdir, prefix=RENDERED_TMPD_PREFIX)
     atexit.register(shutil.rmtree, tmpd)
     bname = os.path.basename(template).rstrip(tmpl_ext)
     fpath = os.path.join(tmpd, bname)
@@ -156,6 +159,25 @@ elif os.path.isfile('/etc/redhat-release'):
     USR_LIB_EXEC = "usr/libexec"
 
 
+class MyEggInfo(egg_info):
+    """This makes sure to not include the rendered files in SOURCES.txt."""
+
+    def find_sources(self):
+        ret = egg_info.find_sources(self)
+        # update the self.filelist.
+        self.filelist.exclude_pattern(RENDERED_TMPD_PREFIX + ".*",
+                                      is_regex=True)
+        # but since mfname is already written we have to update it also.
+        mfname = os.path.join(self.egg_info, "SOURCES.txt")
+        if os.path.exists(mfname):
+            with open(mfname) as fp:
+                files = [f for f in fp
+                         if not f.startswith(RENDERED_TMPD_PREFIX)]
+            with open(mfname, "w") as fp:
+                fp.write(''.join(files))
+        return ret
+
+
 # TODO: Is there a better way to do this??
 class InitsysInstallData(install):
     init_system = None
@@ -229,6 +251,7 @@ if os.uname()[0] != 'FreeBSD':
 # adding on the right init system configuration files
 cmdclass = {
     'install': InitsysInstallData,
+    'egg_info': MyEggInfo,
 }
 
 requirements = read_requires()