Re: making address formatting more flexible

2011-02-09 Thread Luca Capello
Hi there!

Please do not Cc: me, I read the list.

On Tue, 08 Feb 2011 22:13:39 +0100, Johnny wrote:
 Luca Capello l...@pca.it writes:

 I do not think there is even a *used* standard within a single country,
 at least here in Switzerland I saw different layouts sometime according
 to where the sender lives (French/German/Italian part).  And none of the
 examples at the UPU website uses this ISO-3166 supposed standard:

   
 http://www.upu.int/en/activities/addressing/postal-addressing-systems-in-member-countries.html
 This is an international problem of cooperation. If there is an
 International Standard, supported by an Organisation (yes, ISO was
 intentionally capitalised), it should be adhered to, unless there is
 compelling evidence to the contrary (of outstanding benefit).

Fully agree.

 I think, in this case, it is merely a topic of custom and practice I
 think, and if any practice is to be adopted for an application with
 international usage, an internationally accepted standard should be
 adhered to, if at least to support and strengthen the collaborative
 effort, which I believe FOSS and Gnu is all about.

 If bbdb is going to use any country abbreviations, they should be
 according to ISO is my opinion. However, as far as I am aware, postcodes
 do not appear in ISO and should be locally (user) specified.

Please note that we were not discussing about country abbreviations, but
about the label for the postal code.  Now that I thought a bit more
about it, given that we use postcards we should use postcode and not
postal code as I previously suggested.

Going back to country abbreviations, FWIW BBDB-2 already adheres to the
EU standard (which as I reported is not even respected by the UPU):

  |-+-|
  | file| elisp code  |
  |-+-|
  | bbdb.el | `bbdb-format-address-continental'   |
  | bbdb-com.el | `bbdb-legal-zip-codes'  |
  | bbdb-migrate.el | ` bbdb-unmigrate-zip-codes-to-strings'  |
  | bbdb-print.el   | `bbdb-print-format-address-continental' |
  |-+-|

Thx, bye,
Gismo / Luca


pgpCmVusPv07Z.pgp
Description: PGP signature
--
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb___
bbdb-info@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bbdb-info
BBDB Home Page: http://bbdb.sourceforge.net/

Re: making address formatting more flexible

2011-02-08 Thread Johnny
Luca Capello l...@pca.it writes:

 I do not think there is even a *used* standard within a single country,
 at least here in Switzerland I saw different layouts sometime according
 to where the sender lives (French/German/Italian part).  And none of the
 examples at the UPU website uses this ISO-3166 supposed standard:

   
 http://www.upu.int/en/activities/addressing/postal-addressing-systems-in-member-countries.html
This is an international problem of cooperation. If there is an
International Standard, supported by an Organisation (yes, ISO was
intentionally capitalised), it should be adhered to, unless there is
compelling evidence to the contrary (of outstanding benefit). I think,
in this case, it is merely a topic of custom and practice I think, and
if any practice is to be adopted for an application with international
usage, an internationally accepted standard should be adhered to, if at
least to support and strengthen the collaborative effort, which I
believe FOSS and Gnu is all about.

If bbdb is going to use any country abbreviations, they should be
according to ISO is my opinion. However, as far as I am aware, postcodes
do not appear in ISO and should be locally (user) specified.
-- 
Johnny

--
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
___
bbdb-info@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bbdb-info
BBDB Home Page: http://bbdb.sourceforge.net/


Re: making address formatting more flexible

2011-02-07 Thread Luca Capello
Hi there!

On Thu, 03 Feb 2011 05:31:50 +0100, Roland Winkler wrote:
 On Wed Feb 2 2011 Johnny wrote:
 This is a good idea; the current BBDB handling seems very
 US-centered. For one, I think Postal code should replace the American
 term Zip code as default naming, as this seems to be the generic
 term.

 This becomes yet more complicated. I vaguely remember that some
 other english-speaking countries use yet different names instead of
 zip code and postal code, but I forgot where it was and what
 their alternative terminology was...

Well, I would say that by default we should use the same term used by
the UPU, which is Postcode [1], albeit quite surprisingly the main
article on Wikipedia is referred as Postal Code:

  http://www.upu.int/en/resources/postcodes/about-postcodes.html
  http://en.wikipedia.org/wiki/Postcode

FWIW, here a small survey on some websites:

  easyJet.com  Postcode/Zip
  Amazon.com   ZIP
  eBay.it  CAP (Italian for 'Postal Code')
  PayPal.com   Postal Code (localized, CH-...)
  FSF.org  ZIP/Postal Code
  FSFE.org Post Code

  If I remember correctly, Europe introduced zip codes like CH-8052,
  NL-2300RA, and SE-132 54
 I guess this stems from ISO-3166? It does seem appealing to use this
 standard, but it could take some effort to implement; e.g. would a
 lookup table be necessary?

 I am impressed what is covered by some ISO-xyz! Yet I guess what
 matters in the end is what people actually like to use in real life...

I do not think there is even a *used* standard within a single country,
at least here in Switzerland I saw different layouts sometime according
to where the sender lives (French/German/Italian part).  And none of the
examples at the UPU website uses this ISO-3166 supposed standard:

  
http://www.upu.int/en/activities/addressing/postal-addressing-systems-in-member-countries.html

  I am not yet sure what is the best way how to combine a look-up table
  based on country names with a zip-based scheme.
 Wikipedia lists a few formattings of postal codes; however this is not
 defined in ISO-3166. Maybe there's some other source for this though...

 Well, I want to ask: Is the scheme I proposed general enough to make
 most people happy? Or is it necessary to have something yet more
 complicated (and possibly less user-friendly? I wouldn't like that
 either...)

Your proposed scheme is enough customizable, so the situation will be
improved anyway WRT the actual US/EU dichotomy ;-)

Thx, bye,
Gismo / Luca


pgpVDR9Eeuzrg.pgp
Description: PGP signature
--
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb___
bbdb-info@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bbdb-info
BBDB Home Page: http://bbdb.sourceforge.net/

making address formatting more flexible

2011-02-02 Thread Roland Winkler
I am thinking about ways to make the formatting of snail mail
addresses more flexible. In my BBDB I have addresses from a few
countries in the world, and each goes with slightly different
formatting styles, in particular for city, zip code and state.

Yet currently BBDB has only two hard-coded functions
bbdb-format-address-default and bbdb-format-address-continental.
I am thinking about replacing this scheme by a generic function
bbdb-format-address that uses some kind of format strings for
addresses from different countries.

One could have the format specifiers

%s  streets (used repeatedly for each street part)
%c  city
%z  zip code
%S  state
%C  country

A problem is that not every element is always present in an address.
So if there is no country, we want to omit the delimiting newline,
too. This would require some kind of generalized format
specification.

So I thought one could have a delimiter such as `@' in the format
string. If the element referred to by a format specifier was missing,
everything between two `@' was skipped (e.g., @\n%C@ would omit
the newline, too, if there was no country).  But we do not need @
at the very beginning / very end of a format string.

So we could have for example the following format strings

%s\n@%c@, %S@ %z@\n%C USA
%s\n@%s @%c@ (%S)@\n%Cmost European countries
%s\n@%c@ %S@ %z@\n%C  Australia
%s\n@%c@ %z@ (%S)@\n%CIndia
... ...

(I guess that state names are used neither in Europe nor in
India, but if you insist you can get them...)

Also, I want to make the scheme for selecting the above address
formatting schemes more flexible. Old BBDB used the format of zip
code, where the approach was rather euro-centric. I guess this works
fine for some people. But for a more international database, it's
more reliable to use the country -- if (and only if) an address
includes the country field. If I remember correctly, Europe
introduced zip codes like CH-8052, NL-2300RA, and SE-132 54 so that
one need not spell out country names anymore. I do not know whether
people ever followed this rule. Yet it seems that this was the idea
underlying Babb-continental-zip-regexp. -- I am not yet sure what is
the best way how to combine a look-up table based on country names
with a zip-based scheme.

Comments and suggestions welcome!

Roland

--
Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
Finally, a world-class log management solution at an even better price-free!
Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
February 28th, so secure your free ArcSight Logger TODAY! 
http://p.sf.net/sfu/arcsight-sfd2d
___
bbdb-info@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bbdb-info
BBDB Home Page: http://bbdb.sourceforge.net/


Re: making address formatting more flexible

2011-02-02 Thread Leo
On 2011-02-02 23:15 +0800, Roland Winkler wrote:
 Yet currently BBDB has only two hard-coded functions
 bbdb-format-address-default and bbdb-format-address-continental.
 I am thinking about replacing this scheme by a generic function
 bbdb-format-address that uses some kind of format strings for
 addresses from different countries.

 One could have the format specifiers

 %s  streets (used repeatedly for each street part)
 %c  city
 %z  zip code
 %S  state
 %C  country

This is still awkward for Chinese addresses. In bbdb2 it was so awkward
to put address in the address field that I have to put them in the notes
field.

This is because we tend to use one line for the address and it is
specified from country/state to street/room number, for example, 香港九龍
觀塘興業街14號永興大廈6樓C1C室.

Leo

-- 
Oracle is the new evil


--
Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
Finally, a world-class log management solution at an even better price-free!
Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
February 28th, so secure your free ArcSight Logger TODAY! 
http://p.sf.net/sfu/arcsight-sfd2d
___
bbdb-info@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bbdb-info
BBDB Home Page: http://bbdb.sourceforge.net/

Re: making address formatting more flexible

2011-02-02 Thread Johnny
Roland Winkler wink...@gnu.org writes:

 I am thinking about ways to make the formatting of snail mail
 addresses more flexible. 
This is a good idea; the current BBDB handling seems very
US-centered. For one, I think Postal code should replace the American
term Zip code as default naming, as this seems to be the generic term.

 I am thinking about replacing this scheme by a generic function
 bbdb-format-address that uses some kind of format strings for
 addresses from different countries.
...
 If I remember correctly, Europe introduced zip codes like CH-8052,
 NL-2300RA, and SE-132 54
I guess this stems from ISO-3166? It does seem appealing to use this
standard, but it could take some effort to implement; e.g. would a
lookup table be necessary?

 I am not yet sure what is the best way how to combine a look-up table
 based on country names with a zip-based scheme.
Wikipedia lists a few formattings of postal codes; however this is not
defined in ISO-3166. Maybe there's some other source for this though...


-- 
Johnny

--
Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
Finally, a world-class log management solution at an even better price-free!
Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
February 28th, so secure your free ArcSight Logger TODAY! 
http://p.sf.net/sfu/arcsight-sfd2d
___
bbdb-info@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bbdb-info
BBDB Home Page: http://bbdb.sourceforge.net/


Re: making address formatting more flexible

2011-02-02 Thread Leo
On 2011-02-03 00:41 +0800, Roland Winkler wrote:
 This is still awkward for Chinese addresses. In bbdb2 it was so awkward
 to put address in the address field that I have to put them in the notes
 field.
 
 This is because we tend to use one line for the address and it is
 specified from country/state to street/room number, for example, 
  14 6 C1C .

 I am not sure I understand correctly what you are missing here. 

 It appears to me that the room number could become a second street
 part. (The street part of a BBDB address is really just a list of
 strings.)

 Also, newlines would become part of the format specification. So if
 the format specification for chinese addresses contained no
 newlines, everything would go on one line. So for the street part
 you might want to use a plain %s which would be used repeatedly
 for all street parts.

Thanks, that sounds good enough.

Leo


--
Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
Finally, a world-class log management solution at an even better price-free!
Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
February 28th, so secure your free ArcSight Logger TODAY! 
http://p.sf.net/sfu/arcsight-sfd2d
___
bbdb-info@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bbdb-info
BBDB Home Page: http://bbdb.sourceforge.net/


Re: making address formatting more flexible

2011-02-02 Thread Roland Winkler
On Thu Feb 3 2011 Leo wrote:
  It appears to me that the room number could become a second street
  part. (The street part of a BBDB address is really just a list of
  strings.)
 
  Also, newlines would become part of the format specification. So if
  the format specification for chinese addresses contained no
  newlines, everything would go on one line. So for the street part
  you might want to use a plain %s which would be used repeatedly
  for all street parts.
 
 Thanks, that sounds good enough.

OK!

Roland

--
Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
Finally, a world-class log management solution at an even better price-free!
Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
February 28th, so secure your free ArcSight Logger TODAY! 
http://p.sf.net/sfu/arcsight-sfd2d
___
bbdb-info@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bbdb-info
BBDB Home Page: http://bbdb.sourceforge.net/


Re: making address formatting more flexible

2011-02-02 Thread Roland Winkler
On Wed Feb 2 2011 Johnny wrote:
 This is a good idea; the current BBDB handling seems very
 US-centered. For one, I think Postal code should replace the American
 term Zip code as default naming, as this seems to be the generic
 term.

This becomes yet more complicated. I vaguely remember that some
other english-speaking countries use yet different names instead of
zip code and postal code, but I forgot where it was and what
their alternative terminology was...

  If I remember correctly, Europe introduced zip codes like CH-8052,
  NL-2300RA, and SE-132 54
 I guess this stems from ISO-3166? It does seem appealing to use this
 standard, but it could take some effort to implement; e.g. would a
 lookup table be necessary?

I am impressed what is covered by some ISO-xyz! Yet I guess what
matters in the end is what people actually like to use in real life...

  I am not yet sure what is the best way how to combine a look-up table
  based on country names with a zip-based scheme.
 Wikipedia lists a few formattings of postal codes; however this is not
 defined in ISO-3166. Maybe there's some other source for this though...

Well, I want to ask: Is the scheme I proposed general enough to make
most people happy? Or is it necessary to have something yet more
complicated (and possibly less user-friendly? I wouldn't like that
either...)

Roland

--
Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
Finally, a world-class log management solution at an even better price-free!
Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
February 28th, so secure your free ArcSight Logger TODAY! 
http://p.sf.net/sfu/arcsight-sfd2d
___
bbdb-info@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bbdb-info
BBDB Home Page: http://bbdb.sourceforge.net/