sys.exit(1) from makemigrations if no changes found

359 views
Skip to first unread message

Tim Heap

unread,
Oct 28, 2014, 9:22:42 PM10/28/14
to django-d...@googlegroups.com
Hi all,

I have created a ticket for this (https://code.djangoproject.com/ticket/23728) but I would like some input before I work on it. I will copy the content of the ticket below for ease of reading:

It would be very useful for continuous deployment, testing, commit hooks, and other applications if django-admin makemigrations signaled via an exit code if any migrations were found. Commits in projects could be rejected if migrations were outstanding, continuous deployment systems could fail the build on outstanding migrations, and potentially other uses. No more would hasty commits break things when developers forgot to make migrations!

Changes to the code to make this happen are easy enough, but I am unsure how the command should behave. The grep unix utility is a example to copy. Under normal operation, grep always exits 0 unless an error happens, regardless of whether it found any matches. Invoking grep with the -q/--quiet flag causes grep to be silent, not printing anything, as well as exiting 0 if matches are found and 1 if nothing is found.

I am proposing django-admin makemigrations should exit with 1 (or anything non-zero) if no migrations to make were found, or exit 0 if migrations to make were found. As the command is instructed to make migrations, not making any is the error case.

I am unsure how this new functionality should be selected by the user when invoking makemigrations. The options I see are:
  1. Enable this always. This is very simple to implement and easy to understand. Good unixy tools use error codes extensively to signal errors. This may be surprising behaviour when compared to grep though, and breaks backwards compatibility in a minor way.
  2. Enable this when the --dry-run flag is enabled. Now this flag can be used to check for migrations that need to be created both visually via the printed text, and composed in shell commands.
  3. Add a new flag -e/--exit (or similar). The sole purpose of this flag would be to exit with 1 when no migrations were found. This could be combined with --dry-run to just check for migrations that need to be made.
  4. Add a new flag -q/--quiet that copies the behaviour of greps -q/--quiet flag: silences output and exits with 1 when no migrations were found. This duplicates functionality though, as logging can be silenced using -v0 already.
My personal preference is for option 2. I was surprised when enabling --dry-run did not signal its result via the exit code. 3 would be the cleanest and most composable option, as 4 could be emulated using -ev0.

I will implement this change using 2, unless other people have opinions on the matter.

Regards,
Tim

Andrew Godwin

unread,
Oct 28, 2014, 10:04:31 PM10/28/14
to django-d...@googlegroups.com
I'm actually fine with option 1 - always exiting with a status code if no migrations are found. Since the status code does nothing useful at the moment, I don't see any backwards compatibility issues, and as long as it's a suitably small patch, it should be fine.

(As for being surprising compared to grep, there are many other commands with different exit codes one could draw parallels to; I'm not sure being consistent with a very different utility is a worthwhile cause).

Andrew

--
You received this message because you are subscribed to the Google Groups "Django developers (Contributions to Django itself)" group.
To unsubscribe from this group and stop receiving emails from it, send an email to django-develop...@googlegroups.com.
To post to this group, send email to django-d...@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers.
To view this discussion on the web visit https://groups.google.com/d/msgid/django-developers/92203dcb-a775-4c17-a831-97d01ce8af3c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Marc Tamlyn

unread,
Oct 29, 2014, 3:44:31 AM10/29/14
to django-d...@googlegroups.com

I agree number 1 is fine. For most general interaction with the command users won't notice.

Marc

Tim Heap

unread,
Oct 29, 2014, 7:29:34 PM10/29/14
to django-d...@googlegroups.com
The backwards compatibility issue I was thinking of is for people using `makemigrations` in a script currently, expecting it to exit with 0 unless something very wrong has happened. For example, if `makemigrations` is run in a bash script with `set -e`, and it does not find any migrations, that script will then exit. Alternately, if someone is using a one-liner like `./manage.py makemigrations && ./manage.py migrate && another command` the makemigrations command will cause the whole command to abort early, even though no 'error' as such has happened. Backwards compatibility aside, this would be awkward to work around in scripts without simply ignoring the exit code of `makemigrations` completely, which then ignores legitimate errors.

I have made a patch that calls `sys.exit(1)` when no changes are found, and made a pull request at https://github.com/django/django/pull/3441

Tim

Markus Holtermann

unread,
Oct 29, 2014, 7:30:50 PM10/29/14
to django-d...@googlegroups.com
I like the general idea, but I slightly disagree with "As the command is instructed to make migrations, not making any is the error case.". Yes, grep works that way but I thing this are two different use cases. I think makemigrations on a project without any changes in the models' database representation shouldn't exit with RC>0. I'd prefer to have the return code either only set on dry-run or explicitly when using -e/--exit. I'm in favor of 2. or 3.

/Markus

Tim Heap

unread,
Oct 29, 2014, 7:45:47 PM10/29/14
to django-d...@googlegroups.com
I also agree that users should opt-in to enabling the error, using option 2 or 3, but I will implement what ever the popular option is.

My argument regarding "As the command is instructed to make migrations, not making any is the error case." was more about when to return the error: either when changes are found, or when changes are not found. As the command is attempting to make a migration, not finding any changes is the exceptional case, and should return an error code (if enabled via flags, or however it will eventually work).

Daryl

unread,
Oct 29, 2014, 9:21:11 PM10/29/14
to django-d...@googlegroups.com
I'm +1 on being able to continue using the line
./manage.py makemigrations && ./manage.py migrate

I can't see many (any?) situation where someone *wouldn't* run
makemigrations & migrate as one logical operation, whether by typing
the commands or running a script.
What would the workflow be where you would make migrations but not apply them?

D

Tim Heap

unread,
Oct 29, 2014, 9:47:17 PM10/29/14
to django-d...@googlegroups.com
If makemigrations exits with something other than 0, it will not have created any migrations to run anyway, so running the makemigrations & migrate pair here would not be affected. However, a useful one liner that I have been using when developing is:
./manage.py makemigrations && ./manage.py migrate && ./manage.py runserver
If the exit code was always used, chains like this would become much more awkward to use, which is why my preference is for option 2 or 3.

Making migrations and not applying them is useful if you want to review or edit the migrations before applying them - perhaps to split the migration in to two logical chunks for two different commits, or to append some manual migration code to the automatically generated migration. As part of continuous integration I also want to run:
./manage.py makemigrations --dry-run --exit || fail-build "Outstanding database changes with no migrations"
This way, the build fails if there are outstanding migrations. I am not interested in creating or running migrations in this instance, just checking that everything is up to date.
Message has been deleted

Shai Berger

unread,
Oct 30, 2014, 8:06:06 PM10/30/14
to django-d...@googlegroups.com
On Thursday 30 October 2014 15:22:01 Dan Poirier wrote:
> Wouldn't enabling this by default cause a problem for any automated
> deploys?

The discussion is about the makemigrations command, not migrate. The OP wants
to use it to find automatically if there are model changes not covered by
migrations. While it is valid to do so in deployment, I don't think it is very
common practice.

Shai.

Luke Plant

unread,
Oct 31, 2014, 8:57:50 AM10/31/14
to django-d...@googlegroups.com
My preference for this would be option 3., on the grounds that:

* I'd rule out 1 because of the slight backwards incompatibility (which
can affect people - I remember a deploy script failing because of a
changed exit code for a Mercurial command). In particular, currently
there are other conditions when you get a non-zero exit code - for the
case of some crash due to incorrect setup or other runtime error. This
is a distinct case from "this command errored because there were no
migrations to make".

* I'd rule out 2 because it really makes --dry-run behave differently
from normal, which breaks expectations

Basically, having a flag like "--exit" is nice and explicit.

Luke
> 1. Enable this always. This is very simple to implement and easy to
> understand. Good unixy tools use error codes extensively to signal
> errors. This may be surprising behaviour when compared to grep
> though, and breaks backwards compatibility in a minor way.
> 2. Enable this when the --dry-run flag is enabled. Now this flag can be
> used to check for migrations that need to be created both visually
> via the printed text, and composed in shell commands.
> 3. Add a new flag -e/--exit (or similar). The sole purpose of this flag
> would be to exit with 1 when no migrations were found. This could be
> combined with --dry-run to just check for migrations that need to be
> made.
> 4. Add a new flag -q/--quiet that copies the behaviour of greps
> -q/--quiet flag: silences output and exits with 1 when no migrations
> were found. This duplicates functionality though, as logging can be
> silenced using -v0 already.
>
> My personal preference is for option 2. I was surprised when enabling
> --dry-run did not signal its result via the exit code. 3 would be the
> cleanest and most composable option, as 4 could be emulated using -ev0.
>
> I will implement this change using 2, unless other people have opinions
> on the matter.
>
> Regards,
> Tim
>
> --
> You received this message because you are subscribed to the Google
> Groups "Django developers (Contributions to Django itself)" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to django-develop...@googlegroups.com
> <mailto:django-develop...@googlegroups.com>.
> To post to this group, send email to django-d...@googlegroups.com
> <mailto:django-d...@googlegroups.com>.
> <https://groups.google.com/d/msgid/django-developers/92203dcb-a775-4c17-a831-97d01ce8af3c%40googlegroups.com?utm_medium=email&utm_source=footer>.
> For more options, visit https://groups.google.com/d/optout.


--
"I was sad because I had no shoes, until I met a man who had no
feet. So I said, "Got any shoes you're not using?" (Steven Wright)

Luke Plant || http://lukeplant.me.uk/

Tim Heap

unread,
Nov 10, 2014, 11:47:55 PM11/10/14
to django-d...@googlegroups.com
The preference seems to be for option 3, adding a new flag --exit. I have implemented this and updated the pull request at https://github.com/django/django/pull/3441.
Reply all
Reply to author
Forward
0 new messages