Re: UTF8 IPTC comments in V17.5

Hello,

I am using LT v17.5 and Visual Studio 2010 (C# with framework 4.0).

I can manipulate an image and write IPTC comments to it, but I have problems with the encoding of those comments. I cannot find an example showing how to write IPTC comments in UTF8 (or Unicode).

The following code shows how I write the comments:

++++++++++++++++++++++++++++++
// Loading IPTC values
RasterCollection comments = new RasterCollection();
comments.Add(new RasterCommentMetadata(RasterCommentMetadataType.IptcObjectName, Encoding.ASCII.GetBytes(foto.Code)));
comments.Add(new RasterCommentMetadata(RasterCommentMetadataType.IptcCaption, Encoding.ASCII.GetBytes(foto.Caption)));
codecs.WriteComments(destFileName, 1, comments);
++++++++++++++++++++++++++++++

As most of my comments are not represented in ASCII, the text in the IPTC field is not read properly. How should I write the comments in UTF8 in order that a common image viewer can read those comments properly?

This is an example of Caption text: "Museo de las Artesanías"

Thank you for your help.
Xavier.

If you want specifically UTF8 encoding you using use the Encoding.UTF8 encoder.

Replace Encoding.ASCII.GetBytes( ) above with the Encoding.UTF8.GetBytes( ) call. This should give you what you're looking for.

In my test with ASCII I got this when decoded with UTF8:
Museo de las Artesan?as

With UTF8 encoding and decoding with UTF8 I get the appropriate string.

Hello Danny,

I already did that. When I use directly UTF8 I get the following in my viewer (IrfanView):
Museo de las Artesanías

I was wondering whether I have to put an escape sequence for a common viewer to interpret the text in UTF8. Does it make any sense?

Thanks for your help,
Xavier.

The problem you're seeing is that IrfanView and Windows decode the tags as ASCII. The tag's are proper in UTF8 if you use:

RasterCommentMetadata cm = codecs.ReadComment( fileName, 1 , RasterCommentMetadataType.IptcObjectName );
Console.WriteLine( Encoding.UTF8.GetString( cm.GetData() ) );

You will see the tag is properly saved at UTF8. The problem is the viewer and the nature of ASCII encoding.

Hi Danny,

I understand what you say. However, how a common viewer knows which specific encoding system has been used to codify the IPTC comments for a given image? As there exist a lot of possibilities, there should be a way of telling to a viewer that an image has coded its IPTC comments using an specific enconding: Something like a "IptcCodedCharacterSet".

I have opened a LT generated image using PhotoShopc CS5 and it also has problems with the coding.

What am I missing? Thank you for your help.

Xavier.

Common viewers are going to decode them as ASCII as that is the encoding IPTC is built for. Yes it can accept UTF8, but that is not what common viewers are going to read.

http://leadtools.com/help/leadtools/v175/dh/L/Leadtools~Leadtools.RasterCommentMetadataDataType.html

As you can see in the above link we do not register IPTC tags as UTF8. With leadtools you can however build a viewer which can recognize tags written in UTF8, but common viewers will not recognize these tags.