770293 - updates from https://bugzilla.mozilla.org/show_bug.cgi?id=769808

Status: RESOLVED FIXED

(Testing :: Mozbase, defect)

Description

[Jeff Hammel]

The sync to m-c bug
https://bugzilla.mozilla.org/show_bug.cgi?id=769808 illustrates
several nits in the mozbase codebase that should be fixed:

* mozbase/manifestdestiny/README.md should have a MPL license

* update docstring for
  https://github.com/mozilla/mozbase/blob/master/manifestdestiny/manifestparser/manifestparser.py#L696
  ; should be

+        run-if.os = win linux
+        skip-if.os = mac

*
  https://github.com/mozilla/mozbase/blob/master/manifestdestiny/tests/test.py
  should have a she-bang

* testing/mozbase/mozhttpd/README.md should have an MPL license

* print_help and exit are redundant in mozhttpd.py since we already
  error:
  https://github.com/mozilla/mozbase/blob/master/mozhttpd/mozhttpd/mozhttpd.py#L247

* bring back the shebang for
  https://github.com/mozilla/mozbase/blob/master/mozhttpd/tests/api.py
  and
  https://github.com/mozilla/mozbase/blob/master/mozhttpd/tests/filelisting.py

* mozinfo/README.md should have a MPL license block

* mozinstall/README.md shoud have a MPL license block

* mozlog/README.md should have a MPL license block

* mozprocess/README.md should have a license block

*
  https://github.com/mozilla/mozbase/blob/master/mozprocess/mozprocess/wpk.py
  has a test in the main block. Its somewhat silly.  We did kill the
  shebang for this file, so we could bring back the shebang, though
  tbh I'd prefer to just kill the test.  If we don't kill the test, we
  can add it to the test suite as a unittest marked only for running
  on windows

*
  https://github.com/mozilla/mozbase/blob/master/mozprocess/tests/mozprocess1.py
  should have the shebang restored

*
  https://github.com/mozilla/mozbase/blob/master/mozprocess/tests/mozprocess2.py
  should have the shebang restored

* mozprofile/README.md should have a MPL license block

* remove the useless main section: https://github.com/mozilla/mozbase/blob/master/mozprofile/mozprofile/prefs.py#L212

* mozrunner/README.md should have a MPL license block

*
  https://github.com/mozilla/mozbase/blob/master/mozrunner/mozrunner/utils.py
  should have the shebang restored

* test-manifest.ini should have a MPL license block

* test.py should have a MPL license block

Comment 1

[Jeff Hammel]

Actually, talking to :sid0 on IRC perhaps we don't need to license the readmes?

:gerv, do these README files need licenses?

Comment 2

[Gervase Markham [:gerv]]

For maximum clarity, everything needs a copyright license. In practice, READMEs often don't have one.

Gerv

Comment 3

[Jeff Hammel]

(In reply to Gervase Markham [:gerv] from comment #2)
> For maximum clarity, everything needs a copyright license. In practice,
> READMEs often don't have one.
> 
> Gerv

So it sounds like we do want the licenses in these files

Comment 4

[Jeff Hammel]

Attachment: fixes

Comment 5

[William Lachance (:wlach)]

(In reply to Jeff Hammel [:jhammel] from comment #3)
> (In reply to Gervase Markham [:gerv] from comment #2)
> > For maximum clarity, everything needs a copyright license. In practice,
> > READMEs often don't have one.
> > 
> > Gerv
> 
> So it sounds like we do want the licenses in these files

I think that's only the case if "maximum clarity" is a more important requirement than other considerations. I don't mean to be a pain, but I really don't want to have to look at the stuff first thing when I just want to read some documentation. If it's consistent with other practices at Mozilla, my vote would be to leave the licensing stuff out of these files.

Comment 6

[Andrew Halberstadt [:ahal]]

Comment on attachment 638519 [details] [diff] [review]
fixes

Review of attachment 638519 [details] [diff] [review]:
-----------------------------------------------------------------

Looks good. Personally I'm indifferent about the READMEs, so r+ from me. If you tend to agree with wlach feel free to remove them.

Comment 7

[Jeff Hammel]

(In reply to William Lachance (:wlach) from comment #5)
> (In reply to Jeff Hammel [:jhammel] from comment #3)
> > (In reply to Gervase Markham [:gerv] from comment #2)
> > > For maximum clarity, everything needs a copyright license. In practice,
> > > READMEs often don't have one.
> > > 
> > > Gerv
> > 
> > So it sounds like we do want the licenses in these files
> 
> I think that's only the case if "maximum clarity" is a more important
> requirement than other considerations. I don't mean to be a pain, but I
> really don't want to have to look at the stuff first thing when I just want
> to read some documentation. If it's consistent with other practices at
> Mozilla, my vote would be to leave the licensing stuff out of these files.

FWIW, I would also prefer not to have licenses in the READMEs for the same reason.  However, I am not a lawyer and can't make this call.  :gerv, could you clarify?  I'm inclined to not land until we have clarification here.

A situation I do want to avoid is having the files in m-c differ from that in the mozbase repository.  So if the files in m-c require license blocks, I am inclined having them in the mozbase repo as well.  This makes syncing easier

Comment 8

[Jeff Hammel]

(In reply to Andrew Halberstadt [:ahal] from comment #6)
> Comment on attachment 638519 [details] [diff] [review]
> fixes
> 
> Review of attachment 638519 [details] [diff] [review]:
> -----------------------------------------------------------------
> 
> Looks good. Personally I'm indifferent about the READMEs, so r+ from me. If
> you tend to agree with wlach feel free to remove them.

Thanks. I do agree with wlach but will wait for further legal clarification

Comment 9

[Gervase Markham [:gerv]]

I won't lose sleep if you leave them out :-)

Gerv

Comment 10

[Jeff Hammel]

pushed minus the licenses for READMEs: https://github.com/mozilla/mozbase/commit/984e2bcc02636b4a773ccd61f2f229f4e0359161