From: Sebastian B. <sb...@jb...> - 2004-09-14 08:00:22
|
Hi, I don't know the answer to this, but some trial and error revealed this: - The problem of mixing up special characters does not only occur with distinct values. It occurs wherever the result contains atomic strings. E.g. the following query also exhibits it (assume that "sometag" exists and contains umlauts): fn:data(//sometag) - The data is still all right when it is returned vom RpcConnection. However, it reemerges mixed up on the client side. I suspect that something is going wrong with the encoding in the XML RPC libraries. There is a line return responseProcessor.processResponse (response, requestProcessor.getEncoding()); in XmlRpcWorker#execute. Hope this helps somebody (sorry Tom :-( ) Sebastian On Monday 13 September 2004 15:12, Tom Wilczak wrote: > Hi, > I started using eXist in the last 3 months in support of several > content management systems. > Difficult for me to comment on that topic as I have not used other > databses in support of that type of work. > > However, the archival data that I am managing contains several of > the more unusual Latin-1 characters. > I am sure that you all are familiar with several of these. (umlaut, > accents, etc.) > > By and large exist supports these completely transparently. > However, when I query this text using the distinct values function, > the characters get rather garbled and are returned as completely > different entities. > > I did look into the > org.exist.xquery.functions.FunDistinctValues.java file, and its > supporting superclasses/interfaces, but could not exactly determine > where this function is going off. > > I was not able to find any previous posts on this topic, (somewhat > surprising), and am wondering if anyone else has encountered this > problem in the past. > I suspect that one of the underlying data types or > iterartors/readers does not properly implement support for > characters longer than one byte( a bit odd given java's support for > unicode, and has me scratching my head ). > Any suggestions from someone more familiar with the source than > myself? > > I have looked into workarounds (ie: selecting the underlying tree, > and using xsl to restrict the results) however, when the data is > queried in this manner, the requests are significantly slower than > when using the function. > > Thanks in advance, > Tom > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170 > Project Admins to receive an Apple iPod Mini FREE for your > judgement on who ports your project to Linux PPC the best. > Sponsored by IBM. Deadline: Sept. 13. Go here: > http://sf.net/ppc_contest.php > _______________________________________________ > Exist-open mailing list > Exi...@li... > https://lists.sourceforge.net/lists/listinfo/exist-open |