Mercurial > hg > LGDataverses

annotate scripts/migration/scrub_email_usernames.sql @ 14:be7787c36e58 default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
new: nofity LGSercies for deleted files
author Zoe Hong <zhong@mpiwg-berlin.mpg.de>
date Mon, 02 Nov 2015 16:41:23 +0100
parents a50cf11e5178
children
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
10
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1 -- first, find users with e-mails as usernames
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
2 select id, username, email from vdcuser where username like '%@%'
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
3 --and username != email;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
4
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
5 -- then find which those which would create duplicates after truncating
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
6 -- (verify that adding 1 would be OK; if not, you may need to update some individually)
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
7 select u1.id, u1.username, u2.id, u2.username from vdcuser u1, vdcuser u2
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
8 where u1.id != u2.id
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
9 and u1.username like '%@%'
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
10 and split_part (u1.username, '@', 1) = u2.username
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
11
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
12 -- for those usernames, truncate and add 1, so no duplicates
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
13 update vdcuser set username = split_part (username, '@', 1) ||'1'
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
14 where id in (
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
15 select u1.id from vdcuser u1, vdcuser u2
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
16 where u1.id != u2.id
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
17 and u1.username like '%@%'
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
18 and split_part (u1.username, '@', 1) = u2.username
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
19 )
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
20
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
21 -- now truncate the rest
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
22 update vdcuser set username = split_part (username, '@', 1) where username like '%@%'
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
23
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
24 -- confirm no duplicates
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
25 select id, username, email from vdcuser where username in (
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
26 select username from vdcuser
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
27 group by username having count(*) > 1
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
28 )