Hello -
I am downloading en.wikipedia and trying to get both the cur and old tables on my system.
I was able to do an import into a 3.* version of MySQL, but I do not have administrative control of that machine, so I could get the table space set up to be big enough.
I have total control over this machine. It is a Mac OS X 10.3 running MySQL 4.1.0-alpha-debug. The SQL files that I was able to download gave me a syntax error before any data makes it into the tables. I am re-downloading to see if that helps.
Does anyone have any suggestions? Are there known issues with importing into 4.1.* MySQL?
I see that there were, last January, discussions about being able to be a replication client of the databases, but I do not any follow-up on these discussions, so it is not clear that one can become a replication client at this point.
Also, is there a history of the downloadable files? In other words, if I look at today's file, I can see that the md5 digest for http://download.wikipedia.org/archives/en/20031010_old_table.sql.bz2 is 6178dc8bbb25c9788b04cd5cb692a70b, but given a copy of 20030922_old_table.sql.bz2, I cannot seem to tell what its md5 digest was supposed to be. It would be helpful to be able to look at a list of the data files that were put up in the past and see what their digest was.
thanx - ray
On Sat, Oct 11, 2003 at 03:15:19PM -0700, ray@ganymede.org wrote:
I have total control over this machine. It is a Mac OS X 10.3 running MySQL 4.1.0-alpha-debug. The SQL files that I was able to download gave me a syntax error before any data makes it into the tables. I am re-downloading to see if that helps.
What message did it give exactly?
Regards,
JeLuF
On Saturday, Oct 11, 2003, at 15:15 US/Pacific, ray@ganymede.org wrote:
Does anyone have any suggestions? Are there known issues with importing into 4.1.* MySQL?
I've never tried it with 4.1... try on 4.0 and compare. (Also knowing exactly what the syntax erorrs are would help.)
I see that there were, last January, discussions about being able to be a replication client of the databases, but I do not any follow-up on these discussions, so it is not clear that one can become a replication client at this point.
Not yet, no. A key issue is how to deal with the user tables.
Also, is there a history of the downloadable files?
You want the md5sums-YYYYMMDD files in http://download.wikipedia.org/archives/
-- brion vibber (brion @ pobox.com)
Brion Vibber wrote:
On Saturday, Oct 11, 2003, at 15:15 US/Pacific, ray@ganymede.org wrote:
I see that there were, last January, discussions about being able to be a replication client of the databases, but I do not any follow-up on these discussions, so it is not clear that one can become a replication client at this point.
Not yet, no. A key issue is how to deal with the user tables.
Which reminds me, are we waiting for consensus on merging the user accounts of all the wikipedias into a single one, or is it just that noone bothered to implement it?
Magnus
On Sunday, Oct 12, 2003, at 01:46 US/Pacific, Magnus Manske wrote:
Which reminds me, are we waiting for consensus on merging the user accounts of all the wikipedias into a single one, or is it just that noone bothered to implement it?
Seems to me it was mainly an implementation issue; also how to deal with the existing duplicate account names.
-- brion vibber (brion @ pobox.com)
wikitech-l@lists.wikimedia.org