[xml] Search on xmlsoft.org (WAS: decoding xml entities)

["William M. Brack" <wbrack mmm com hk>]

> > Cory Nelson said:
> > > Is there a function in libxml2 that will take a string and decode
> > > all the
> > > entities (< etc) in it?
>
> William Brack said:
> > Well, if I go to http://xmlsoft.org, and go from the main page to
> > the Search Engine (by clicking on "Search), then put in the word
> > "entities", I get a list of functions to consider for this task.
> >
> > Out of that list, xmlStringDecodeEntities certainly looks promising.
> >
> > Is there some reason you were not able to go through these same
> > steps?
> >
> > Bill
Cory Nelson said:
> I did that, and the function didn't (and still doesn't) come up.
>
> Treating people like an idiot because the search engine doesn't work
> that
> great is really nice.  Thanks.

Hmmm.. we seem to be awfully defensive, don't we?  If I wanted to
treat you "like an idiot", I can assure you I would not have
bothered to respond at all to your query.  Rather, I am very
concerned to try to help make the available tools as useful as
possible.  If there is any problem with them (as in this case, where
your search didn't find what you needed) I try to elicit more
information as to why they failed so that they can be fixed or
improved.  If I had responded with only a link to the function, it
is doubtful whether you would have elaborated on the problem.

Chris Ryland provided some further information: the search worked
when restricted to API's, but didn't appear in the more general
"all" search.  Peter Jacobi offered a positive suggestion that we
modify the ranking algorithm on the search so as to increase the
value of API links.  I will certainly consider both of these points,
and would welcome any other input.

The actual "search engine" is a php script.  As with all the rest of
this project, the source for that script is in the library source
distribution (libxml2/doc/search.php).  From my quick reading before
preparing this response, it looks as if the script only displays the
first 75 hits (in ranked order).  The database which is used by that
script is created and maintained by a python script
(libxml2/doc/index.py).  Patches to correct or enhance these scripts
are always welcome!

Despite your sarcasm, I assume the information I provided was
helpful to you.  I'm certain that the other responses on the list
were helful to me.  Truly a "win-win" situation :-)

Bill