Question about java.text.Normalizer

[Julien Martin]

Hello,

I am facing an issue with my java.text.Normalizer which is working fine in one part of the application and not so well in another part of the application.

Here is how I use the Normaliser:

public static String normaliserURL(String string) {
        log.debug("normaliserURL");
        return Normalizer.normalize(string, Form.NFD).replaceAll("\\p{InCombiningDiacriticalMarks}+", "");
}

It works well when I use the facelets function from facelets code as shown here:

<pretty:link mappingId="sculptureAction" styleClass="miniature-link">
<f:param value="#{view.locale.language}"/>
<f:param value="#{jbm:normaliserURL(sculpture.sculpturei18nMap[view.locale.language].titre)}" />
<f:param value="#{sculpture.sculptureID}" />
<h:graphicImage value="#{initParam['com.jeanbaptistemartin.static.url']}/#{sculpture.photoMiniature}" styleClass="miniature-image" alt="#{sculpture.sculpturei18nMap[view.locale.language].titre}, #{sculpture.sculpturei18nMap[view.locale.language].matiere}, #{jbm:renvoyerAnnee(sculpture.annee)}."/>
</pretty:link>

(I get this: "The+fall", notice the "+" sign.)

However, when I try to use it as follows from java code:

log.info("'" + FaceletsFunctions.normaliserURL(service.findByID(sculptureID).getSculpturei18nMap().get(abonneLocale).getTitre()) + "'");

It does not work and I get a space as shown here:

(see: "The fall", notice the space...)

Can anyone please help?

Thanks in advance,

Julien.

[Jesper de Jong]

Julien Martin wrote:

public static String normaliserURL(String string) {
    log.debug("normaliserURL");
    return Normalizer.normalize(string, Form.NFD).replaceAll("\\p{InCombiningDiacriticalMarks}+", "");
}

This is an interesting piece of code! It replaces characters with accents etc. with characters without accents. Right? So if you have something like "héllô", it will return "hello". People are sometimes asking for something like this on the forums, this looks like a good solution.

But how did you know about the regular expression "\\p{InCombiningDiacriticalMarks}+"? The character class InCombiningDiacriticalMarks is not documented in the API documentation of class java.util.regex.Pattern.

About your problem:

Sorry, I don't have a direct solution for you. But there are certain URL / HTTP parameter encodings in which spaces are replaced by + signs. Maybe your string that contains the + sign is encoded using such an encoding somewhere else in your program.

[Julien Martin]

Hello Jesper,
Thanks for your reply.
Yes it looks like the spaces are replaced by "+" signs by the pretty link (see facelets/html code included in my first message).
Do you think I should replace spaces by "+" signs right from regex without waiting for my pretty link to do it?
Does anyone see a solution?
Julien.