<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
OK, I've found the answer... Apologies for inundating your
mailboxes.<br>
Here it is, for reference, in an old post (21 Oct 2016) by Andrew
Hardie:<br>
<br>
<blockquote type="cite">
<pre wrap="">- this page will give you a form for each text-metadata field that was installed with the datatype "classification"
- each of these forms will list all the category handles that exist within the given classification
- by default, category handles are mapped to a "description" that is the same as the category handle itself
- BUT you can use the forms here to change the category descriptions to something more user-friendly
- since category handles are limited to short codes with no spacing or punctuation this is often useful
- if you do this, then the "descriptions" will show up in a whole lot of different places in the user interface instead of the category handles, including:
-- restricted query form
-- concordance header for a restricted query
-- distribution display
-- text metadata page.</pre>
</blockquote>
So basically, the underscores can be removed, and more exotic
characters added, if required, by editing the "Category description"
fields in the "Manage text categories" form.<br>
Best,<br>
Graham.<br>
<br>
<br>
<div class="moz-cite-prefix">Le 21/03/2019 à 11:13, Graham Ranger --
UAPV a écrit :<br>
</div>
<blockquote type="cite"
cite="mid:feabacc7-4915-676b-9558-3f4b43e642af@univ-avignon.fr">
<meta http-equiv="content-type" content="text/html; charset=utf-8">
Hello to all,<br>
I would like some help formatting metadata for a corpus.<br>
I understand that the "text id" field has to use only ASCII
alphnumeric characters plus de underscore. However, from my
experiments, this constraint appears to apply to all fields.<br>
And so, while the metadata for the BE 2006 corpus, on the cqpweb
interface at Lancaster, appears as "Press, Entire text, A. Press:
Reportage" I would only be able to display this sort of
information as "Press, Entire_text", "A_Press_Reportage", etc. I
have played with the "Free text" "Classification" opposition, but
that makes no difference. When my text is formated with spaces, or
punctuation, it simply does not show in the metadata.<br>
I'm doing this with a separate text file including metadata, but
the other possibility, i.e. including metadata as attributes
inside the corpus xml has not proved any more satisfactory.<br>
Many thanks in advance for any help.<br>
Best,<br>
Graham.<br>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
CWB mailing list
<a class="moz-txt-link-abbreviated" href="mailto:CWB@sslmit.unibo.it">CWB@sslmit.unibo.it</a>
<a class="moz-txt-link-freetext" href="http://liste.sslmit.unibo.it/mailman/listinfo/cwb">http://liste.sslmit.unibo.it/mailman/listinfo/cwb</a>
</pre>
</blockquote>
<br>
</body>
</html>