On 3/02/19 10:16 PM, Chris Angelico wrote:
On Sun, Feb 3, 2019 at 8:09 PM DL Neil <pythonl...@danceswithmice.info> wrote:
On 3/02/19 9:45 PM, Chris Angelico wrote:
Which is why I always write dates in sorted format, usually eschewing
delimiters:
//CJA 20160511: Is this still happening? I don't remember seeing it in
three parts of forever.
Sure is. It is an acceptable alternative under the ISO standard.
Not sure if you're responding to the content of the comment there;
Comment? I don't see no #, ''', or """!
(am teasing)
[ccyymmdd cf ccyy-mm-dd]
Some would say it is more sensible to use when storing data because it
removes the dash/hyphen separators in exchange for implying the
fixed-format. (more bytes/characters saved if extend to include the time)
I'm not going there - recalling folk from these memory-is-cheap times
being less able to understand why we used to save 'expensive' storage
space by using yy-years (instead of ccyy) and thus 'causing' "the
millennium bug" aka Y2K!
Skipping the delimiter isn't about saving space, it's about
consistency. If I say "non-delimited sorted date", you can almost
certainly write out a character-for-character identical date - handy
if you want to search a bunch of files, for instance. Having
delimiters leaves people free to dispute whether they should be
slashes, hyphens, dots, or maybe something else.
This logic indisputable.
However, the whole purpose of an ISO standard is to remove "dispute",
locally and internationally! Thus, if not sufficiently-well stated
earlier, the standard is actually for information interchange purposes.
>> I find it much slower to decode than reading the same with embedded
>> separators!
> Sure. I mainly use it in contexts where the most important information
> is simply "that's a date", rather than actually caring what the date
> *is*.
[paras re-ordered]
+1
In case other readers are following-along-at-home, and the (above)
purpose of the standard was insufficiently obvious, I did a bit of
review 'homework':
- ISO standards are still not $free
- a good write-up from the ISO appears on the Wayback machine at
https://web.archive.org/web/20110614235056/http://www.iso.org/iso/support/faqs/faqs_widely_used_standards/widely_used_standards_other/date_and_time_format.htm
- the above makes the point about "interchange" and offers similar
examples of date-confusion to those 'here'/earlier
- mention is made of formats including/excluding delimiters
(I haven't been able to check this, but can't find any evidence that
separators other than "-" are allowed (in dates) )
- in lieu of the ISO text, those of us working over the Internet will
turn to RFC 3339
- this is a slight simplification of the ISO standard
- the ABNF appendix *requires* a dash/hyphen as (date) separator
- Markus Kuhn at Cambridge (British university) provides a readable and
thought-provoking summary at https://www.cl.cam.ac.uk/~mgk25/iso-time.html
- uses dashes/hyphens
- discusses international considerations (I've just learned that the
Chinese date notation preceded the ISO order, as did the conventions of
a number of other countries)
- supports the compact/'no debate' format "The hyphens can be omitted if
compactness of the representation is more important than human
readability" (as well as caring for my fading vision)
I wouldn't use it in a 'visible' situation though, eg a fileNM. Yes, it
is shorter, but as my eyes age (they are already older than my teeth!),
Guess your teeth better work on catching up...
Can't put them under the pressure of thinking it is a race - they could
decide to drop out!
That said, I am aware that I am not in any way a "normal person".
Using month names as per your other example is probably a fair
compromise with other humans.
In this life, one does have to make allowances...
There's normal and there's normal - like it's tomato or tomato?
I dunno. I'm the kind of normal that likes tomatoes (not to be
confused with tomatoes). Does that help?
If you like tomatoes, and tomatoes are fruit, do you (normally) chug
tomato sauce (ketchup) as if it is fruit juice?
--
Regards =dn
--
https://mail.python.org/mailman/listinfo/python-list
