R help - Feb 2007 - gsub: replacing a.*a if no occurence of b in .*

I am trying to read a number of XML files using xmlTreeParse(). Unfortunately,
some of them are malformed in a way that makes R crash. The problem is that
closing tags are sometimes repeated like this:

<tag>value1</tag><tag>value2</tag>some
garbage</tag></tag><tag>value3</tag>

I want to preprocess the contents of the XML file using gsub() before feeding
them to xmlTreeParse() to clean them up, but I can't figure out how to do
it.
What I need is something that transforms the example above into:

<tag>value1</tag><tag>value2</tag><tag>value3</tag>

Some kind of "</tag>.*</tag>" that only matches if there
is no "<tag>" in ".*".

Thanks in advance for you ideas,

Uli

Ulrich Keller <ulrich.keller at emacs.lu> writes:
> I am trying to read a number of XML files using xmlTreeParse().
Unfortunately,
> some of them are malformed in a way that makes R crash. The problem is that
> closing tags are sometimes repeated like this:
>
> <tag>value1</tag><tag>value2</tag>some
garbage</tag></tag><tag>value3</tag>
>
> I want to preprocess the contents of the XML file using gsub() before
feeding
> them to xmlTreeParse() to clean them up, but I can't figure out how to
do it.
> What I need is something that transforms the example above into:
>
>
<tag>value1</tag><tag>value2</tag><tag>value3</tag>
>
> Some kind of "</tag>.*</tag>" that only matches if
there is no "<tag>" in ".*".
>
> Thanks in advance for you ideas,
Hmm, there are things you just cannot do with RE's, and I suspect that
this is one of them. Something involving explicit splitting of the
strings might work, though. How's this for size?
> trim <-
   
function(x)paste(sub("</tag>.*","</tag>",x),collapse="<tag>")
> sapply(strsplit(x,"<tag>"),trim)
[1]
"<tag>value1</tag><tag>value2</tag><tag>value3</tag>"


-- 
   O__  ---- Peter Dalgaard             ?ster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark          Ph:  (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)                  FAX: (+45) 35327907

I assume <tag> is known.

This removes any occurrence </tag>.*</tag> where .* does not
contain <tag> or </tag>.

The regular expression, re, matches </tag>, then does a greedy
match (?U) for anything followed by </tag> but uses a zero
width lookahead subexpression (?=...) for the second </tag>
so that it it can be rematched again.  gsubfn in package
gsubfn is like the usual gsub except that instead of
replacing the match with a string it passes the match
to function f and then replaces the match with the output
of f.  See the gsubfn home page:
  http://code.google.com/p/gsubfn/
and vignette.


library(gsubfn)

text <-
paste("<tag>value1</tag><tag>value2</tag>some",
"garbage</tag></tag><tag>value3</tag>")

re <- "</tag>((?U).*(?=</tag>))"
f <- function(x) if (regexpr("<tag>", x) > 0) x else
""

gsubfn(re, f, text, backref = 0, perl = TRUE)


On 2/24/07, Ulrich Keller <ulrich.keller at emacs.lu>
wrote:
> I am trying to read a number of XML files using xmlTreeParse().
Unfortunately,
> some of them are malformed in a way that makes R crash. The problem is that
> closing tags are sometimes repeated like this:
>
> <tag>value1</tag><tag>value2</tag>some
garbage</tag></tag><tag>value3</tag>
>
> I want to preprocess the contents of the XML file using gsub() before
feeding
> them to xmlTreeParse() to clean them up, but I can't figure out how to
do it.
> What I need is something that transforms the example above into:
>
>
<tag>value1</tag><tag>value2</tag><tag>value3</tag>
>
> Some kind of "</tag>.*</tag>" that only matches if
there is no "<tag>" in ".*".
>
> Thanks in advance for you ideas,
>
> Uli
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

Ulrich Keller wrote:
> I am trying to read a number of XML files using xmlTreeParse().
Unfortunately,
> some of them are malformed in a way that makes R crash. The problem is that
> closing tags are sometimes repeated like this:
> 
> <tag>value1</tag><tag>value2</tag>some
garbage</tag></tag><tag>value3</tag>
> 
> I want to preprocess the contents of the XML file using gsub() before
feeding
> them to xmlTreeParse() to clean them up, but I can't figure out how to
do it.
> What I need is something that transforms the example above into:
> 
>
<tag>value1</tag><tag>value2</tag><tag>value3</tag>
> 
> Some kind of "</tag>.*</tag>" that only matches if
there is no "<tag>" in ".*".
> 
> Thanks in advance for you ideas,

Instead of using xmlTreeParse() which really expects well-formed XML,
and assuming you cannot have the XML generation mechanism fixed, you might
try to use htmlTreeParse().
While the name suggests it is for HTML, it is really a "relaxed"
XML parser that is capable of handling malformed XML.  This typically
occurs in HTML and hence the name.
Of course, since the XML is malformed, the results will be hard to predict
as it is hard to make sense of "non-sense".


If xmlTreeParse() is actually causing R to exit (i.e. what some people
refer to as crashing), as Jeff (Horner) said, we would like to be able
to stop this. We will need the actual text/file passed to
xmlTreeParse(), version information of operating system, R and the XML
package and any locale information.  However, if by crashing you mean
generates an error, then that is expected on malformed XML inputs.

HTH,
 D.
> 
> Uli
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
-- 
Duncan Temple Lang                duncan at wald.ucdavis.edu
Department of Statistics          work:  (530) 752-4782
4210 Mathematical Sciences Bldg.  fax:   (530) 752-7099
One Shields Ave.
University of California at Davis
Davis, CA 95616, USA



-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url :
https://stat.ethz.ch/pipermail/r-help/attachments/20070224/8edba5e8/attachment.bin