On Thu, Jun 12, 2008 at 10:47 AM, Sandro Böhme <[EMAIL PROTECTED]> wrote: > Hi Florian, > > thanks for your feedback! You are probably right. This seems to be the only > solution.
have you considered cloning an existing workspace to a new one using BundleDbPM? cheers stefan > > Best regards, > > Sandro > > Florian Holeczek schrieb: >> >> Hi Sandro, >> >> what about the brute force solution... iterating over all nodes, >> retrieving their properties content into the corresponding Java >> objects and doing your search in these? >> >> Regards, >> Florian >> >> Ursprüngliche Nachricht vom 11.06.2008 um 18:57: >>> >>> Hello, >> >>> currently we are trying to export the content from one repository to >>> another (in this case for migrating to the BundlePersistanceManager). The >>> problem is, that our repository contains special characters like #x0d, #x0a, >>> #x04. As of JCR-674 [1] the export seems to work but the import does not >>> decode the base64 encoded values [2]. Our initial thought was, to just >>> remove some of these chars. But it doesn't seem to >>> be possible to create a search query to find these chars. It would just >>> be good for us to know if this is correct, or if we missed something. We >>> are depending on preserving the uuids across the repositories so the only >>> way seems to be JCR export/import. >>> At the moment we are in a situation where we cannot migrate to the other >>> PM and cannot use the JCR export/import in general with Jackrabbit. Does >>> somebody know an other way of fixing such content in the repository or >>> exporting/importing it? Any hint is much appreciated. >> >>> Best regards, >> >>> Sandro >> >>> [1] - String properties with invalid XML characters export as invalid XML >>> https://issues.apache.org/jira/browse/JCR-674 >> >>> [2] - JCR-1228 Support xs:base64Binary values in system view import >>> https://issues.apache.org/jira/browse/JCR-1228 > >
