Re: [xsl] Trimming (formatting-only) leading tabs/spaces from XSLT - issues?

Subject: Re: [xsl] Trimming (formatting-only) leading tabs/spaces from XSLT - issues?
From: Philip Fearon <pgfearo@xxxxxxxxxxxxxx>
Date: Tue, 7 Jun 2011 15:26:22 +0100
This wouldn't ever be perfect, but there's a large set of rules that
could be used to determine formatting-only whitespace. The following
set is a distilled version:
>From the XML context:
1. Outside mixed-content
2. Outside where xsl:space "preserve" is in scope
3. Outside defined elements such as 'pre' and 'text'
4. If it precedes an attribute name beginning a new line
5. If it precedes an attribute value on a new line
>From text context:
4. Where the number of characters found are (within a defined margin)
is consistent with the current nesting-level - a pattern can normally
be established
5. Where irregular leading whitespace is found on consecutive lines in
a node value

The approach would be progressive where, on first load a minimal set
of 'obvious' formatting is removed, then further options provided to
the developer that steadily impose a more rigorous filtering

Phil Fearon

On Tue, Jun 7, 2011 at 2:22 PM, Dimitre Novatchev <dnovatchev@xxxxxxxxx>
> How do you know that the spaces are "formatting-only"?
> I have heard about cases (mostly with HTML) where this cannot be safely
> --
> Cheers,
> Dimitre Novatchev
> ---------------------------------------
> Truly great madness cannot be achieved without significant intelligence.
> ---------------------------------------
> To invent, you need a good imagination and a pile of junk
> -------------------------------------
> Never fight an inanimate object
> -------------------------------------
> You've achieved success in your field when you don't know whether what
> you're doing is work or play
> -------------------------------------
> Facts do not cease to exist because they are ignored.
> -------------------------------------
> I finally figured out the only reason to be alive is to enjoy it.
> On Tue, Jun 7, 2011 at 5:47 AM, Philip Fearon <philipfearon@xxxxxxxxxxx>
>> Say, (hypothetically) there was an XSLT-based system that included an
>> editor that (if used) needed to trim leading tabs/spaces from XSLT
>> when first loading a file, so as to provide a continously formatted
>> view of the XSLT code as it was edited.
>> Provided that this was done relatively reliably (i.e. without trimming
>> significant tab and space characters) and predictably, would this
>> cause significant problems in an XSLT development, test or production
>> environment?
>> So, would other editors, viewers, diff-tools, version-control systems,
>> auto-documentation systems etc. be adversely affected? If such
>> formatting characters needed to be added again, would it be best just
>> to add them when required (letting the consumer choose the formatting
>> style)?
>> I ask, because the popular consensus seems to be that trimming
>> non-XSLT code of formatting characters in this way would be a major
>> issue. This is because of established working methods and tools that
>> don't/won't/can't tolerate changes to whitespace (this seems to be
>> quite an emotive issue).Now, whitespace has added significance (an
>> understatement I think) in XML/XSLT systems. My experience (such that
>> it is) in XSLT is that working methods and tools have therefore
>> evolved to manage whitespace (removing it, comparing it, adding it, or
>> ignoring it) more effectively than text-only tools, trimming XSLT of
>> tabs/spaces therefore shouldn't be a problem and may even provide more
>> flexibility, does this hold up in the wider XSLT world?
>> Phil Fearon

Current Thread