On 03/12/14 17:24, Dimitry Sibiryakov wrote: > 12.03.2014 14:01, Alex Peshkoff wrote: >> It's easy to add new fields to XSQLDA, but adding something to XSQLVAR >> is much harder because there is an array of this structures in the end >> of XSQLDA. From user POV this looks fine, but inside firebird library we >> get nightmare. > No, not at all. Or may be I don't see something...
For different versions we need to take into an account differetnt offsets in that array. Possible, but very unsane. > Merging changes in why.cpp which are conflicting with the patch I > prepared last week - > that was nightmare. Strange... I've applied it to one of tree copies last week and always updated from svn w/o problems. But today found a code dup in 2 places and could not commit without trivial fix. Certainly if you updated your initial version with my changed you had a lot of conflicts. But it's not something outstanding. >> Yes provided it's enough. >> Assignment of many pointers is often slower than copying memory region. > But why one would like to assign many pointers? > If to take for given that buffer is a class with unknown internal > structure, then it is > enough to assign a pointer to it if your code is going to use only buffer's > official > interface or create a new buffer of suitable structure and copy all data into > it, letting > original buffer go. Copy all data - that's what we try to avoid with new API. Certainly, in most cases that's not a trouble. But when peek performance is needed buffer as an unknown object described by interface is not good. > There is no need to force your code (engine) to work with any kind of > buffer or return > a buffer with any user's format. Engine can return its buffers in own format > and then it > is up to user - work with them as via official interface is or to convert it > into > something more suitable. > > Exactly this is done now. ------------------------------------------------------------------------------ Learn Graph Databases - Download FREE O'Reilly Book "Graph Databases" is the definitive new guide to graph databases and their applications. Written by three acclaimed leaders in the field, this first edition is now available. Download your free book today! http://p.sf.net/sfu/13534_NeoTech Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
