[Boost-users] PropertyTree's XML parsers survey

Subject: [Boost-users] PropertyTree's XML parsers survey
From: Tan, Tom (Shanghai) (TTan_at_[hidden])
Date: 2009-05-18 04:49:27

>From: "Robert Ramey" <ramey_at_[hidden]>
>Subject: Re: [Boost-users] PropertyTree's XML parsers survey

>The serialization library has been using the spirit library
>for parsing XML for years. This is for both utf-8 and wide characters
>
>Robert Ramey

I just played with the examples demo_xml_load along with boost 1.39. As
I mainly use xml as configuration files. I found some features I really
like are missing:

- often I need to edit the xml file manually with 3rd tools(gedit, kate
notepad++ etc) other than the app itself. When saving back with those
editors,  they may add some BOM bytes(0xFFFE for UTF16LE, 0xFEFF for
UTF16BE, 0xefbbbf for utf8) and replace the line-breakers(0x0a, 0x0a0d)
with another fashion(e.g 0x0d0a -> 0x0a) the system default.  In such
cases, I wish boost.serialize would detect and ignore them as the real
data the xml carries do NOT change. 

- I often embed some comment nodes inside the xml to facilitate the
manual edit. E.g, enumeration of accepted value some fields. I wish
boost.serialize ignores them when parsing, and embbded them when
saving(is this possible).

- I wish only the hierarchy not the order or nodes matters. Ex.
<a>
<first = "1st">
<second = "2nd">
</a> 
Would be equivalent to
<a>
<second = "2nd">
<first = "1st">
</a>
It seems not to be support yet.

On the whole, my impression is boost.serialize imposes some restrictions
(orders for instance) that make sense to plain text files, but limits
the buil-in flexibility that coming with xml.