Обсуждение: [SoC] Relation between project "XML improvements" and "pgxml"
Hello, on your summer of code page [1] you outline a project "XML improvements". Is there any relation (similar goals, feature overlapping, technical relation) to the "pgxml" project mentioned for some time on [2]? I have been (remotely) following Oleg Bartunov's page on GiST usage and datatypes for some time, now that I'm pondering the submission of a SoC proposal I'm wondering if the "XML improvements" project is a completely new approach, maybe even superceding the approach outlined by Oleg. [1] http://www.postgresql.org/developer/summerofcode [2] http://www.sai.msu.su/~megera/postgres/gist/ Best regards, Rob Staudinger
On 5/3/06, Robert Staudinger <robert.staudinger@gmail.com> wrote: > on your summer of code page [1] you outline a project "XML improvements". > Is there any relation (similar goals, feature overlapping, technical > relation) to the "pgxml" project mentioned for some time on [2]? No, the XML project idea submitted did not include Oleg's stuff at all. > now that I'm pondering the submission of a > SoC proposal I'm wondering if the "XML improvements" project is a > completely new approach, maybe even superceding the > approach outlined At this point in time, I don't believe there's any single best way to go regarding XML-handling in PostgreSQL. If you have a neat project idea, please propose it to us on Google's site! -- Jonah H. Harris, Database Internals Architect EnterpriseDB Corporation 732.331.1324
Yes, there is no single way, because there are different tasks. There are many papers on this theme. I'm pretty sure that first of all we need to analyze other DBMSes' experience. I'm working on it, analyzing MS, ORA and DB2 (first results of analysis will be available in several weeks). I've submitted proposal 'XMLType for PostgreSQL' to Google SoC page (my 'minimum' list from here: http://archives.postgresql.org/pgsql-hackers/2006-05/msg00044.php On 5/4/06, Jonah H. Harris <jonah.harris@gmail.com> wrote: > On 5/3/06, Robert Staudinger <robert.staudinger@gmail.com> wrote: > > on your summer of code page [1] you outline a project "XML improvements". > > Is there any relation (similar goals, feature overlapping, technical > > relation) to the "pgxml" project mentioned for some time on [2]? > > No, the XML project idea submitted did not include Oleg's stuff at all. > > > now that I'm pondering the submission of a > > SoC proposal I'm wondering if the "XML improvements" project is a > > completely new approach, maybe even superceding the > > approach outlined > > At this point in time, I don't believe there's any single best way to > go regarding XML-handling in PostgreSQL. If you have a neat project > idea, please propose it to us on Google's site! > > -- > Jonah H. Harris, Database Internals Architect > EnterpriseDB Corporation > 732.331.1324 > > ---------------------------(end of broadcast)--------------------------- > TIP 5: don't forget to increase your free space map settings > -- Best regards, Nikolay
Actually, project mentioned on Oleg's page is only in plan. I see some gap between current moment and the moment when GiST will come to power for XML support - check out my proposal (http://archives.postgresql.org/pgsql-hackers/2006-05/msg00044.php), the 'minumum' list is that gap. When we will have some basic support (ideally, according SQL:200n SQL/XML standard and based on experience taken from commercial DBMSes), we would work on index support (w/o which this project definitely won't be applicable to production purposes) - first of all, path indexes and structure indexes. This includes some labeling schema (probably prefix schema, see http://davis.wpi.edu/dsrg/vamana/WebPages/Publication.html or papers about MS' ORDPATHs). GiST and Gis will definitely help here a lot. On 5/3/06, Robert Staudinger <robert.staudinger@gmail.com> wrote: > Hello, > > on your summer of code page [1] you outline a project "XML improvements". > Is there any relation (similar goals, feature overlapping, technical > relation) to the "pgxml" project mentioned for some time on [2]? I > have been (remotely) following Oleg Bartunov's page on GiST usage and > datatypes for some time, now that I'm pondering the submission of a > SoC proposal I'm wondering if the "XML improvements" project is a > completely new approach, maybe even superceding the approach outlined > by Oleg. > > [1] http://www.postgresql.org/developer/summerofcode > [2] http://www.sai.msu.su/~megera/postgres/gist/ > > Best regards, > Rob Staudinger > > ---------------------------(end of broadcast)--------------------------- > TIP 3: Have you checked our extensive FAQ? > > http://www.postgresql.org/docs/faq > -- Best regards, Nikolay