Hi guys,
I'm playing with the Exhibit format printer in SemanticResultFormats.
Specifically, I do an ask query like this:
{{#ask: [[Category:E22 1.Art Object]]
[[HasDeparment::+]]
[[HasClassification::+]]
[[HasObjectType::+]]
[[HasObjectName::+]]
| ?HasDeparment
| ?HasClassification
| ?HasObjectType
| ?HasObjectName
| ?P102F has title
| ?HasThumbnailImage
| format=exhibit
| views=tabular,tiles
| facets=HasDeparment,HasClassification,HasObjectType,HasObjectName
| limit=21
}}
and HasThumbnailImage is a property that points to an Image URI.
eg:
http://ssemtestdb00:8765/services/MediaBinImageURI/Default.aspx?objAcc=02.6.129&getBest=true&imgSize=3&ext=.jpg
coming out of the wiki text parser, those uris get turned into full image tags.
Which is what I want. And the table that gets generated from the ASK query
looks
like:
<tr>
<td><a href="/metwiki/index.php/02.6.129"
title="02.6.129">02.6.129</a></td>
<td>European Sculpture and Decorative Arts</td>
<td>Ceramics-Porcelain</td>
<td>normal object (accessioned)</td>
<td>Stand</td>
<td></td>
</tr>
Here is where Exhibit's javascript takes over, pulling all the data from the
table and indexing it. for faceting and sorting and whatnot, I guess.
However, as it does this, it explicitely replaces the img tag with just the
text
of the uri.
in
extensions/SemanticResultFormats/Exhibit/exhibit/scripts/data/importers/html-table-importer.js,
around line 150:
var img, imgs = table.getElementsByTagName("img");
while( img = imgs[0] ){ // replace any images with their respective URLs
donundeen: I don't want to do that!!!
img.parentNode.replaceChild( document.createTextNode( img.src ),
img );
}
However, I can't just remove this chunk of code, because the bit that comes
after will extract the text anyways, and not get any from the image tag:
var items = [], td, raw;
for( row = 1; tr = trs[row]; row++ ) {
var item = {};
var tds = tr.getElementsByTagName("td");
for( col = 0; td = tds[col]; col++ ) {
var raw = textOf( td ); // this part removes anything that's not text.
hence, no image tags get through. But I want to see images...
data = columnData[col].valueParser( raw, td, row, col );
if( data == null || raw === "" ) {
continue;
}
if( typeof data == 'object' && !(data instanceof Array) ) {
for( var property in data ) {
item[property] = data[property];
}
} else {
item[fields[col]] = data;
}
}
if( type )
item.type = type;
items.push( item );
parsed.items = items;
}
Is there a way to accomplish what I want? I just want to see the images in the
tabular and Tiles output formats. I don't care about using the value of the
image tag (whatever that would mean) for faceting or sorting.
Does exhibit have a setting to help with this? Or a newer version or something?
I'm using:
MediaWiki 1.15.1
PHP 5.2.9 (apache2handler)
MySQL 5.1.33-community
Semantic MediaWiki (Version 1.5.0)
Semantic Forms (Version 1.9) Forms for adding and editing semantic data
Or if you guys can recommend a better forum for this issue (an exhibit-focused
forum, maybe?) I'd appreciate it.
Thanks!------------------------------------------------------------------------------
This SF.net email is sponsored by
Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev
_______________________________________________
Semediawiki-devel mailing list
Semediawiki-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/semediawiki-devel
