The stat module currently uses the "st_ctime" slot to hold two kinds
values which are semantically different but which are frequently
confused with one another.  It chooses which kind of value to put in
there based on platform -- Windows gets the file creation time and all
other platforms get the "ctime".  The only sane way to use this API is
then to switch on platform:

if platform.system() == "Windows":
    metadata["creation time"] = s.st_ctime
else:
    metadata["unix ctime"] = s.st_ctime

(That is an actual code snippet from the Allmydata-Tahoe project.)

Many or even most programmers incorrectly think that unix ctime is file
creation time, so instead of using the sane idiom above, they write the
following:

metadata["ctime"] = s.st_ctime

thus passing on the confusion to the users of their metadata, who may
not know on which platform this metadata was created.  This is the
situation we have found ourselves in for the Allmydata-Tahoe project --
we now have a bunch of "ctime" values stored in our filesystem and no
way to tell which kind they were.

More and more filesystems such as ZFS and Macintosh apparently offer
creation time nowadays.

I propose the following changes:

1.  Add a "st_crtime" field which gets populated on filesystems
(Windows, ZFS, Mac) which can do so.

That is hopefully not too controversial and we could proceed to do so
even if the next proposal gets bogged down:

2.  Add a "st_unixctime" field which gets populated *only* by the unix
ctime and never by any other value (even on Windows, where the unix
ctime *is* available even though nobody cares about it), and deprecate
the hopelessly ambiguous "st_ctime" field.

You may be interested in http://allmydata.org/trac/tahoe/ticket/628
("mtime" and "ctime": I don't think that word means what you think it
means.) where the Allmydata-Tahoe project is carefully unpicking the
mess we made for ourselves by confusing ctime with file-creation time.

Sounds like a good idea, perhaps you could launch a discussion on
python-dev?

Okay, I posted to python-dev:

http://mail.python.org/pipermail/python-dev/2009-June/090021.html

What was the outcome?

Well, read the thread! http://mail.python.org/pipermail/python-dev/2009-June/090021.html

Basically just a couple of +1's, and a good suggestion to name it something clearer than "crtime". 


Please fix it!

I don't think we should hijack st_ctime. All the other members of the stat object correspond exactly to what you get back from the Unix stat() call. Let's not break that. How about just leaving st_ctime as it is and adding st_creationtime, which will have the nicer semantics?

Benjamin Peterson: what do you mean "hijack ctime"? I don't think I — or anyone — has proposed anything that fits that description. Please be more specific.  My proposal in http://bugs.python.org/issue5720#msg85750 does not break anything.

Indeed, "hijacking" is a bit strong. I mean filling it with a value that is not in the underlying st_ctime field from stat() call.

Benjamin: I'm sorry, I still don't understand. Do you think my proposal would involve setting something named "ctime" to contain a value that didn't come from the underlying stat "ctime"?

You're quite right. I stupidly misread "crtime" as "ctime".

Aha! Mystery solved. I wouldn't say that you were stupid — I would say that "crtime" is way too close to "ctime", and I strongly agree with the suggestion (http://mail.python.org/pipermail/python-dev/2009-June/090026.html) on the mailing list by Greg Ewing that we name the new thing something more obvious.