UrXML

UrXML is yet another XML parser/pretty-printer based on RSXP package by CK Kashyap.

UrXML is vim-friendly and can be used to pretty-print the XML-fragment of non-XML file.

Additionaly, the tool is able to translate the usual XML into Ur/Web dialect of XML (uses several hardcoded rules)

Installation

Haskell platform is required.

$ git clone https://github.com/grwlf/urxml
$ cd urxml
$ cabal configure
$ cabal build
$ cabal install

Usage

$ urxml --help
XML converter/indenter

Usage: urxml [-w|--text-width INT] [-m|--right-margin INT] [-i|--start-indent INT] [-s|--tab-stop INT] [-e|--expand-tab] [-1|--skip-first] [-S|--skip-schema] [-u|--ur-attrs] [FILE]
  Ur/Web XML indenter (Ur/Web dialect of XML supports {}-style attributes).

Available options:
  -h,--help                Show this help text
  -w,--text-width INT      Recommended text width (not strict)
  -m,--right-margin INT    Right margin
  -i,--start-indent INT    Indent to start from
  -s,--tab-stop INT        Tab stop
  -e,--expand-tab          Expand tab
  -1,--skip-first          Don't indent 1st line
  -S,--skip-schema         Don't print DOCTYPE and stuff
  -u,--ur-attrs            Convert tag attributes to the Ur/Web format
  FILE                     File or `stdin' to read from the stdin

Examples

Simple pretty-printing:

$ urxml test.xml 
<xml>
    <a>Under a</a>
    <b>
        <c a='b'>
            under c1 asd asd asd asd asd as dasd asd as dasdasd asd asdas dasd asdas das das
            dasdas dasdasdas das dasdasdasd
        </c>
        <c c='d'>under c2</c>
        <c number=33 width="33" height="44">under c3</c>
        <c src="images/asdasdad.gif" urcode={main
      {}}>under c4</c>
    </b>
</xml>

Pretty-print with indentation:

$ urxml -i 5 test.xml 
     <xml>
         <a>Under a</a>
         <b>
             <c a='b'>
                 under c1 asd asd asd asd asd as dasd asd as dasdasd asd asdas dasd asdas das das
                 dasdas dasdasdas das dasdasdasd
             </c>
             <c c='d'>under c2</c>
             <c number=33 width="33" height="44">under c3</c>
             <c src="images/asdasdad.gif" urcode={main {}}>under c4</c>
         </b>
     </xml>

Pretty-print with indentation, don't indent first line:

$ urxml -i 5 -1 test.xml 
<xml>
         <a>Under a</a>
         <b>
             <c a='b'>
                 under c1 asd asd asd asd asd as dasd asd as dasdasd asd asdas dasd asdas das das
                 dasdas dasdasdas das dasdasdasd
             </c>
             <c c='d'>under c2</c>
             <c number=33 width="33" height="44">under c3</c>
             <c src="images/asdasdad.gif" urcode={main {}}>under c4</c>
         </b>
     </xml>

and so on.

VIM integration

UrXML is the easy way of formatting XML inside non-XML documents. Just paste the following lines into your ~/.vimrc and format the XML by visually selecting the fragment (note, that it should be a correct part of XML with matching tags) and pressing the '!' key, as set in the last line of the fragment.

" ~/.vimrc fragment
function! ProgramFilter(vt, ...)
  normal `<
  let p = getpos ('.')

  let [qr, qt] = [getreg('"'), getregtype('"')]
  let [oai, ocin, osi, oinde] = [&ai, &cin, &si, &inde]
  setl noai nocin nosi inde=

  let [sm, em] = ['[<'[a:0], ']>'[a:0]]
  exe 'norm!`' . sm . a:vt . '`' . em . 'x'

  let tw = &tw
  setl tw=0

  let cmd = 'urxml -i ' . (p[2]-1) . ' -1 --tab-stop 2 --ur-attrs stdin'

  let out = system(cmd, @")
  let out = substitute(out, '\n\n$', '', '')
  exe "norm!i\<c-r>=out\r"

  let &tw = tw
  let [&ai, &cin, &si, &inde] = [oai, ocin, osi, oinde]
  call setreg('"', qr, qt)
endfunction
vnoremap <silent> ! :<c-u>call ProgramFilter(visualmode(), 1)<cr>

RSXP's README

The original project is located at

https://github.com/ckkashyap/really-simple-xml-parser

A really simple xml parser in Haskell using Parsec. I wrote this primarily to learn Parsec and I intend to use it in my automations where I need to parse XML's returned from bug tracking systems and other systems.

I am not an XML fan - in fact, I hate it.

I'd like to acknowledge the help I got from the wonderful Haskell community for this. I'd particularly like to call out the help provided by the following folks -

1. Antoine Latter aslatter@gmail.com
2. Simon Hengel sol@typeful.net