Write down the version numbers for those two packages - you only need the v2008xxxxxxxx bit.Ģ) Check out the relevant sources using the CVS wizard. Select the Features tab and sort it by Feature ID by clicking the column header. This can be done by selecting Help -> About Eclipse on the menu and then clicking the Installation Details button. No worries, it is actually surprisingly easy, if a little long winded.īefore you start this it is probably a good idea to back-up your eclipse directory as it may well break things horribly!ġ) You need to find out which versions of the packages belowĪre used in Ganymede. > matches this patch and how to build it myself (I never did that before, but > Or maybe simply point me out to where I can find the correct sources that > helios, is it easy for you to build it for galileo ? > another issue prevents me to move to helios). > I'd like to try out the patch too, but I'm on galileo (because unfortunately You built it for ganymede and helios, is it easy for you to build it for galileo ? Or maybe simply point me out to where I can find the correct sources that matches this patch and how to build it myself (I never did that before, but I can try) I'd like to try out the patch too, but I'm on galileo (because unfortunately another issue prevents me to move to helios). > I hope this proves to be of some help to others. > version from CVS (and applied the patch, obviously) and made them publicly > bothered to build the components from source, I have built the latest stable > works as expected, as well as helping others with this issue who can't be > So, in an attempt to try and help move this along by proving that the patch > I felt bad after my last comment on this bug. Is the intent to fix this issue? Given that a big chunk of Spring development (and other frameworks such as Mule) is driven by custom XSD's for wiring up components, this is a high-value feature in the daily lives of many developers.Īfter 2 years, I was hoping that eclipse's XML editor would be all over this. > In my oppinion, a change of this magnitude is best addressed early in a new > (see bug 266683), the problem space has changed a bit and will likely require a > With the recent change to provide a global content model cache in WTP 3.2 M7 > support for this particular substitution groups scenario. > such a core mechanism - the content model - outweighed the benefits of getting However, at the time, my judgement was that the risks of changing > issue as soon as it was reported, found the cause and attached a potential > Folks, just a bit of clarification on the status here: I have looked into this A shared resource set will avoid redundant resource loading. It seems to me that, in addition to fixing this loss of function for substitution groups,using a shared resource set could also speed up things when an XML document uses multiple schemas that potentially reference each other through includes/imports/redefines. Thus the solution appears to require some sort of state sharing to save and share a common resource set for all XSD models used by the CM document manager for the current editor instance. The XSD infoset model will only populate the substitution groups information properly when all schemas that refer to each other are loaded in the same resource set. This works fine for most schemas but breaks down when substitution groups are used. Building on this assumption, the XSD EMF infoset model for each schema is loaded in its own EMF resource set. The CM document framework seems to assume that the CM documents created for each referenced schema are independent of each other. This problem appears to be caused by the way the XSD models that back the CM documents are being loaded. The commercial Oxygen XML editor handles it without a problem. IntelliJ had the same symptom and fixed it in a recent release: The problem is due to the use of substitutionGroups in the XML schema.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |