replace default: with ns1, default2: with ns2: etc.

[bmix]

First things first: This is such a great plugin! Wow! Really! I read it on the closed issues (#1), that you did it for a friend, so you may not even appreciate, how nice and convenient this added XPath functionality is! :-) So, thanks a lot, man!

This is not really an important issue, but convention in the XML world seems to be to use the nsN prefix for namespaces, which are anonymous.

So, default:foo would become ns1:foo and default2:bar would become ns2:bar.

I have no backing on this via some specs, it's just from experience by working with XML IDEs and reading some articles on the web. It may be more familiar to XML users.

[keith-hall]

Hi, thanks for the feedback! Glad to hear that this plugin is useful for you :)

I agree that nsx notation probably makes more sense / will be more familiar to people, thanks for the suggestion. From a cursory glance, I believe the following places will need some tweaks to facilitate this:

• sublime-xpath/xpath.py

  Line 648 in f4916d1

  defaultNamespacePrefix = settings.get('default_namespace_prefix', 'default')

• sublime-xpath/xpath.sublime-settings

  Line 22 in 118b2df

  "default_namespace_prefix": "default",

• sublime-xpath/lxml_parser.py

  Lines 281 to 298 in f4916d1

  	def unique_namespace_prefixes(namespaces, replaceNoneWith = 'default', start = 1):
  	"""Given an ordered dictionary of unique namespace prefixes and their URIs in document order, create a dictionary with unique namespace prefixes and their mappings."""
  	unique = collections.OrderedDict()
  	for key in namespaces.keys():
  	if len(namespaces[key]) == 1:
  	try_key = key or replaceNoneWith
  	unique[try_key] = (namespaces[key][0], key)
  	else: # find next available number. we can't just append the number, because it is possible that the new numbered prefix already exists
  	index = start - 1
  	for item in namespaces[key]: # for each item that has the same prefix but a different namespace
  	while True:
  	index += 1 # try with the next index
  	try_key = (key or replaceNoneWith) + str(index)
  	if try_key not in unique.keys() and try_key not in namespaces.keys():
  	break # the key we are trying is new
  	unique[try_key] = (item, key)
  	return unique

  (to always append a number, as opposed to just having ns:foo for example. It may make sense to add an additional setting for this, but maybe I'm just customization crazy!)

• sublime-xpath/tests.py

  Line 97 in e59195e

  def xpath_tests():

  (because default is hard-coded in the tests... that'd surely fail if someone tries to run the tests after changing the setting - oops, I didn't think of that when I wrote this!)

Would you be interested in working on this and submitting a PR, @bmix ? If not, I'll try to find some spare time to work on it in the near future.

[bmix]

Thanks for hints! I will, maybe, slowly, investigate this. I am deep in a website project right now, and never did any Sublime plugin development, so I have to learn a little more about it. However, as this is not an important issue, save your time. ;-) There is no need to act on your side.

Oh, btw, the only "official" reference I found regarding the nsN syntax was this: https://www.w3.org/XML/2008/xsdl-exx/ns1

And I was wrong: The numbering goes: ns0, ns1, ns2, etc.

but maybe I'm just customization crazy!

Yes, that may be in this case! ;-)