server2/TODO-merge | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+)
New commits: commit 907618ddf1e6812d04ab0cb773d46c44ed3f3626 Author: Matt Domsch Matt_Domsch@dell.com Date: Thu Aug 11 08:32:50 2011 -0500
add TODO-merge to track work needing done to finish merging
diff --git a/server2/TODO-merge b/server2/TODO-merge new file mode 100644 index 0000000..7bcc138 --- /dev/null +++ b/server2/TODO-merge @@ -0,0 +1,30 @@ +<mdomsch> ok, the big stuff I see still needs doing... +<mdomsch> 1) in model/*, most of the object methods haven't been converted from SO yet +<mdomsch> this includes the hard-coded methods to implement cascaded deletes +<mdomsch> 1b) making sure FAS integration is chosen from the config.ini file, rather than hard-coded. +<mdomsch> e.g. from development.ini, it should not use FAS +<mdomsch> 2) setting up python-migrate (sqlalchemy-migrate) to be able to update the schema over time cleanly. I spent most of yesterday making sure that, for sqlite at least, the SQL to create the tables using SA matches what the old SO had done. +* Gomex has quit (Remote host closed the connection) +<mdomsch> I need to be able to modify the schema going forward, most of the MM1.4 changes I had made require schema changes +<mdomsch> 3) port the helper scripts from server/ into server2/. such as update-master-directory-list. Where it makes sense, move functions into the app rather than in separate helpers, leave the helpers just for invocation +<mdomsch> 4) port the MM1.4 changes into the tg2merge branch. I see ~100 patches made to the server/ directory +<pingou> I saw that too, after I was done porting 1.3 :/ +<mdomsch> pingou: you've been a great help so far, and I really appreciate all the effort you've put into it. I couldn't have made it even this far without you. THANK YOU! +<mdomsch> today I have to do "real work"; hopefully tomorrow afternoon I can get back into this again +<pingou> mdomsch: no problem, the hardest part is still on you: "making sure it works as it should" +<pingou> :) +<mdomsch> :-) +<pingou> mdomsch: and it was fun to do :) +<mdomsch> that caught me off guard last night - if there's a fas user 'manager', and if we had lockouts after x many failed attempts, they'd be locked out :-) +<pingou> mdomsch: two things +<pingou> 1) right now it seems to not start (at least for me, might be a change in the db) +<pingou> 2) there was a page (text) that didn't quite work +<mdomsch> what db are you using? +<pingou> sqlite, but I didn't regenerate it +<pingou> mdomsch: working now, was a change in the db schema I guess +<mdomsch> I spent several hours comparing SQL schemas yesterday and fixing it up to match what SO generated +<pingou> mdomsch: /rsync_acl is the still buggy function +<mdomsch> there are still a few slight differences (TIMESTAMP vs DATETIME fields) +<pingou> I have a script laying somewhere to compare to db scheme ^^ +<mdomsch> ok, don't sweat that too much - it's not used a lot. I see the xmlrpc stuff needs to be handled too - that's more important +<pingou> but might be a mysql specific script