Re: [NTG-context] Ctx & XML

11 Dec 2014

      On 12/11/2014 8:33 AM, Procházka Lukáš Ing. - Pontex s. r. o. wrote:
...
Hello,
On Thu, 11 Dec 2014 02:24:24 +0100, Hans Hagen  wrote:
...
On 12/10/2014 11:08 PM, Lukáš Procházka wrote:
I assume that you also want the content of elements, not just the
attributes.
Yes...
...
...
Or how to achieve this in a best way under Ctx: XML file -> Lua
table/tree?
It all depends on what kin dof table you want (if the content is an
indexed table and attributes keys then you have the problem that the
element cannot be a key too as it can clash) .. anyway, it's no big deal
to provide a stupid table, so I can provide something because after all
the loaded xml file is a table already.
It would be great.
...
local x = xml.load("file.xml")
local t = xml.totable(x)
where x is an index/hash mix plus _tag, _type and _namespace fields as
well (after all I assume that you need to know that too).
(btw, I bet that you can do what you want with the existing
functionality as well, so there is not much benefit in such a table)
Probably in the next upload,
.... Means - some functionality (for me to test or to play with) might
be available tomorrow morning?
...
Hans
BTW, I prepared a really-simple-XML-test which shows XML -> Lua
table/tree conversion, which I'd like to achieve (and which (approx.)
provides LuaXml lib).
---- XML
<a aaa="AAA">
   <b bbb="B1"/>
   <b bbb="B2">
     c
   </b>
</a>
----
---- Lua representation
{ [0] = "a", -- Element name
   { [0] = "b", -- Element name
     -- No nested members
     bbb = "B1", -- Atts
   },
   { [0] = "b", -- Element name
     { "c", }, -- Nested member
     bbb = "B2", -- Atts
   },
   aaa = "AAA", -- Atts
}
----
that already makes assumptions that it's just data ... the context xml 
parser is a roundtrip one so a bit more control is provided:

local str = [[
<?xml version="1.0" ?>
<a one="1">
     <!-- rubish -->
   <b two="1"/>
   <b two="2">
     c > d
   </b>
</a>
]]

I've added some flags:

inspect(xml.totable(xml.convert(str)))

table={
  {
   "xml version=\"1.0\" ",
   ["_type"]="instruction",
  },
  "\
",
  {
   "\
     ",
   {
    " rubish ",
    ["_type"]="comment",
   },
   "\
   ",
   {
    ["_tag"]="b",
    ["_type"]="_element",
    ["two"]="1",
   },
   "\
   ",
   {
    "\
     c > d\
   ",
    ["_tag"]="b",
    ["_type"]="_element",
    ["two"]="2",
   },
   "\
",
   ["_tag"]="a",
   ["_type"]="_element",
   ["one"]="1",
  },
  ["_type"]="root",
}

inspect(xml.totable(xml.convert(str),true))

table={
  {
   "xml version=\"1.0\"",
   ["_type"]="instruction",
  },
  {
   {
    "rubish",
    ["_type"]="comment",
   },
   {
    ["_tag"]="b",
    ["_type"]="_element",
    ["two"]="1",
   },
   {
    "c > d",
    ["_tag"]="b",
    ["_type"]="_element",
    ["two"]="2",
   },
   ["_tag"]="a",
   ["_type"]="_element",
   ["one"]="1",
  },
  ["_type"]="root",
}

inspect(xml.totable(xml.convert(str),true,true))

table={
  {
   [0]="a",
   {
    [0]="b",
    ["two"]="1",
   },
   {
    [0]="b",
    "c > d",
    ["two"]="2",
   },
   ["one"]="1",
  },
}

The last one is yours. Of course in all cases you loose info that is 
present in the original representation (where you can also access the
nodes via expressions). Don't expect additional helpers for this 
'totable' variant.

Hans

-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
     tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com
                                              | www.pragma-pod.nl
-----------------------------------------------------------------