DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG· RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT <http://issues.apache.org/bugzilla/show_bug.cgi?id=38183>. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND· INSERTED IN THE BUG DATABASE.
http://issues.apache.org/bugzilla/show_bug.cgi?id=38183 ------- Additional Comments From [EMAIL PROTECTED] 2006-01-10 13:06 ------- (In reply to comment #4) > Thanks! a few comments: > > > The "internal codecs" as I called them are moved to a new source > > directory (sources-internal-codec) > > I will probably not do this, I will move them all into a common > package instead. Ok, but that will make it more difficult to work with in an IDE. Now I can just remove the sources-internal-codec from the source dir list and work with Batik in a GNU-Classpath-compatible way. Sure, it's easy to keep the things apart in the Ant build but having a clean separation helps a lot, especially in showing the dependencies. I guess we're back to the same argument we had with XML Graphics Commons. > >- I've not reimplemented every single feature in the various image > > transcoders for the Image I/O API as is available for the internal > > codecs. Examples are Gamma/Chromaticity handling and background > > color in PNG. > > Hmm, gamma handling _is_ important for PNG (especially on decode). > This is why we moved from using the JDK decoder to our own local one. > Fortunately this is not the real problem area with the Batik > codecs - so I will see what can be done. I guessed as much. Thanks. > On Tiff it isn't the most important part. It would be nice for > some use cases. > > > - I've experienced problems handling metadata and encoding parameters > > when writing JPEG images through Image I/O. The current code works > > for most use cases, though. > > This is a little more troubling, what sorts of problems? The main problem is that the JPEG codec does not support setting the bitmap resolution through the standard metdata format. Image I/O has a good concept for handling metadata but the implementation is a catastrophe. Other problems mostly involved IIOMetadata.setFromTree/mergeFromTree(). Settings get lost when merging and stuff like that. A lot of trial and error was necessary to get this far and I've found almost no helpful resources on the net. > > - The TIFFTranscoder now supports specifying the compression > > method through a Transcoding hint (type String). > > Sounds good! > > > - I've changed the code that prepares an XML parser to use JAXP in > > order to remove the mandatory dependency on Apache Xerces. I hope > > I did this the right way. > > This also looks good. > > > You can forcibly disable the compilation of the Sun-dependant > > internal codecs if you create a file (called "build-local.properties") > > Is there any reason you didn't use 'build.properties' file? > This file is only used for 'local' mods anyways (otherwise you would > put them in the ant file wouldn't you?). I like to separate build.properties from build-local.properties, because that latter is normally in the ignores list, so people can override their build settings without accidentally committing their special settings to the repository. A build.properties can help a lot for specifying default behaviour and serving as a template for a build-local.properties. See FOP where this is done very cleanly. BTW, I'm very surprised about the amount of attention this patch already got. Looks like I picked something important. :-) -- Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
