Re: [xml] Remove whitespaces from text nodes

[Georges-André SILBER <gasilber luxia fr>]

OK, but in this case it really depends on your input XML format and what you consider "useless".

If you only have "locally" useless whitespaces like here:

<txt>     Some text </txt>

and you want to get <txt>Some text</txt> you can still use the function below and "strip" every text node 
with a C function (I don't think that a standard C function exists for that).

But, if you have mixed content like:

<p>  Some text<b><i> in bold italics </i></b>and continuing</p>

it is tricky to define "useless" whitespaces in a recursive descent because the decision is not local (here, 
clearly, the whitespaces in the <i> elements should not be removed).

So, depending on your definition of "useless", the difficulty of the answer can range from very simple to 
very complicated.

Best regards,

Georges-André SILBER
LUXIA

Le 16 févr. 2012 à 09:32, spam spam spam spam free fr a écrit :

> I think this function removes blank nodes.
> That's not exactly what I want.
> I want to strip useless whitespaces from text nodes.
> These nodes aren't considered as blank nodes because they contains also visible characters.
>
> ----- Mail original -----
> De: "Georges-André SILBER" <gasilber luxia fr>
> À: "spam spam spam spam" <spam spam spam spam free fr>
> Cc: xml gnome org
> Envoyé: Jeudi 16 Février 2012 09:14:17
> Objet: Re: [xml] Remove whitespaces from text nodes
>
> Hi,
>
> I wrote a small function for this purpose some time ago.
> I didn't test it with the last versions of libxml2 nor did I ensure that this code if correct but it gives 
> you the idea of a method that you can use to remove blank nodes.
>
> The usage is for instance:
>
> doc = xmlReadFile (xmlfile, NULL, 0);
> if (doc == NULL)
>   {
>       /* Deal with error... */
>      return 1;
>    }
> glbRemoveBlankNodes (xmlDocGetRootElement(doc));
>
> Hope this helps,
>
> Best regards,
>
> Georges-André SILBER
> LUXIA
>
> int
> glbRemoveBlankNodes (xmlNodePtr n)
> {
>  xmlNodePtr cur;
>  xmlNodePtr next;
>
>  if (n == NULL)
>    return 0;
>
>  cur = n->children;
>  while (cur)
>    {
>      next = cur->next;      
>      if (xmlIsBlankNode (cur))
>       {
>         xmlUnlinkNode (cur);
>         xmlFreeNode (cur);
>       }
>      else
>       glbRemoveBlankNodes (cur);
>      cur = next;
>    }
>
>  return 0;
> }
>
>
> Le 16 févr. 2012 à 08:57, spam spam spam spam free fr a écrit :
>
> > Yes you are right.
> > But I am not sure my function will do a good job.
> > I know 2 whitespaces : " ", "\t", ... But I am not sure that I know all of them.
> > My function will probably forgot to strip some whitespaces...
> > This is the reason why I would like to use an already defined function.
> >
> > Is there a function which do this job?
> >
> > ----- Mail original -----
> > De: "Liam R E Quin" <liam holoweb net>
> > À: "spam spam spam spam" <spam spam spam spam free fr>
> > Cc: xml gnome org
> > Envoyé: Jeudi 16 Février 2012 08:40:31
> > Objet: Re: [xml] Remove whitespaces from text nodes
> >
> > On Thu, 2012-02-16 at 08:28 +0100, spam spam spam spam free fr wrote:
> > > [...].
> > > Anyway, there seems to have no other solution with libxml2 only.
> >
> > The spaces are part of the text of the document, so it's not likely that
> > a conformant XML parser will strip them for you.
> >
> > You could of course remove the spaces in C after parsing, just as if you
> > decided to remove every occurrence of an upper-case "B" from the input.
> >
> > That's just standard C string processing.
> >
> > -- 
> > Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
> > Pictures from old books: http://fromoldbooks.org/
> >
> > _______________________________________________
> > xml mailing list, project page  http://xmlsoft.org/
> > xml gnome org
> > http://mail.gnome.org/mailman/listinfo/xml
>
> _______________________________________________
> xml mailing list, project page  http://xmlsoft.org/
> xml gnome org
> http://mail.gnome.org/mailman/listinfo/xml