-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Philip Neustrom, on 2008-12-23 16:41, wrote:
Excerpts from Paul Ivanov's message of Tue Dec 23 16:24:34 -0800 2008:
Philip wrote:
There is export.py which will produce an XML dump, but I figured a
pg_dump is probably good?
I decided against threadjacking the map overhaul conversation, but I've
been meaning to ask about this:
1. Has anyone tried to make a corresponding import.py to go from XML
back to a DB?
2. what are your strategies for mirroring a live wiki when doing
development?
I run off a patched up sqlite branch locally when I play around doing
bugfixes, but I wanted to backup my wiki in its entirety and couldn't
think of a good way to do this (images borked in the db dump, though
that was probably some encoding issue). Anyone have a good way of
mirroring everything on another machine?
thanks,
Paul
There's no import.py as of yet, as far as I know. Making it wouldn't be
difficult, but I'm not sure it'd have much utility now: database dump /
import is generally the way to go and the export.py stuff
(http://www.projectsycamore.org/Export) was thrown together for folks on Wiki
Spot who wanted dumps of their wikis (e.g. might need to only be the globally
visable pages).
I use pg_dump for postgres with --format=c
sudo -u postgres pg_dump --format=c wikispot
--file=/srv/sql/postgresql/wikispot.db.out
is the command I usually do. The format=c means you'll need to use
pg_restore to import the dump, but it's compressed.
The images shouldn't be borked on a dump - definitely sounds like an encoding
issue.
If you're using sqlite it's insanely simple: just copy the single database
file (woo sqlite). share/data/wiki.db is the file.
--philip
I got excited when I first spotted export.py, because surely, there's
an import.py somewhere. I thought I could run sqlite locally to do
development, and transfer all of the data from mysql/pg using export.py
and the would-be import.py.
Up to now, I've been running off sqlite to do development but on a
default (empty) wiki. I wanted to test if I could replicate my
production wiki which runs off mysql, so I set up mysql locally,
imported the the dump, recreated the views, and got stuck when i got
this trace in browser.
Traceback (most recent call last):
File /home/pi/syc/Sycamore/request.py, line 747, in run
Page(query, self).send_page(count_hit=1)
File /home/pi/syc/Sycamore/Page.py, line 832, in send_page
polite_msg=polite_msg)
File /home/pi/syc/Sycamore/wikiutil.py, line 1614, in send_title
request.write(request.theme.header(d))
File /home/pi/syc/Sycamore/theme/eggheadbeta.py, line 679, in header
'banner_html': self.banner(d),
File /home/pi/syc/Sycamore/theme/eggheadbeta.py, line 68, in banner
wikiutil.init_logo_sizes(self.request)
File /home/pi/syc/Sycamore/wikiutil.py, line 408, in init_logo_sizes
logo = Image.open(logo_file)
File /usr/lib/python2.5/site-packages/PIL/Image.py, line 1917, in open
raise IOError(cannot identify image file)
IOError: cannot identify image file
Anyway, I'll play around with the encoding, and update the wiki
instructions if I figure something out.
Paul
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFJUedLe+cmRQ8+KPcRAkHuAJ9WIRdPEk7jsA2q9WaSchR28o397wCgj0HV
VmSbQPLZQvmo9OzJMGLxUw4=
=Bkp/
-END PGP SIGNATURE-
___
Sycamore-Dev mailing list
sycamore-...@wikispot.org
http://www.projectsycamore.org/
https://tools.cernio.com/mailman/listinfo/sycamore-dev
