[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Frontline report from the Desperate Fgrep Hacker

• From: Stefan Haustein <haustein@k...>
• To: Sean McGrath <digitome@i...>, xml-dev@x...
• Date: Fri, 11 Feb 2000 20:38:19 +0100

Sean McGrath wrote:
> 
> [John Cowan]
> >I happened to know that the element always appeared on a single
> >line of the file: the start tag, the character data, the end tag.

> SML-DEV. Bottom line: simple grepping of XML 1.0
> can get you burnt because of all the things that
> can occur in cosher XML 1.0 documents that
> a straight lexical analysis will misconstrue unless
> prefixed by an XML parser.

Made me thinking what an "xmlgrep" would need to do, and which 
additional parameters it would need: Even for SML, grep seems
dangerous, consider

<tag>I want the complete line</tag> 

vs. 

<tag>I
want the
complete 
line</tag>

"xmlgrep" could search in a given set of elements for a 
regular expression. But how would the list of elements
be specified (xpath? 2nd regexp?)? Which child elements are returned? 
All? None? configurable? Or could the problem be solved by 
preformatting XML/SML to a form where all elements get an extra line,
whereas line breaks inside text content are replaced by a space?

Best regards

Stefan

-- 
Stefan Haustein
University of Dortmund
Computer Science VIII
www-ai.cs.uni-dortmund.de

• References:
  • Re: Frontline report from the Desperate Fgrep Hacker
    • From: Sean McGrath <digitome@i...>

• Prev by Date: gold in that thar XML
• Next by Date: xml diff
• Previous by thread: Re: Frontline report from the Desperate Fgrep Hacker
• Next by thread: Re: Frontline report from the Desperate Fgrep Hacker
• Index(es):
  • Date
  • Thread