JCU Research Portal - Nigels Blog

Plone email gateway proposal

nigel — 2006-10-15T17:36+00:00

To encourage people to blog more with plone/quills what I think we need is an email gateway, to at least get the content to the site. I've tried to research what other blogging agents do, and it basically appears that a special email account per blog is set up. My proposal is as follows:

Set up a single blog email address for each plone instance (research, verg, dart, hpc,...)
Have each instance check its box at a given periodicy
Each user on the site can set preferences specifying the default blog and categories for email blogs
When an email arrives it is mapped, using the From address to a user, then to a blog and category. Overrides should be allowed, perhaps in the body?

If the message is not signed (PGP, S/MIME) then the blog sits in the drafts to be published. (normal quills interface) Send a reply email with a link to the drafts folder.
If the email is sigend it is published immediately.

If it doesn't map to a user drop the email

AutoJar - jar file combiner

nigel — 2006-10-09T20:10+00:00

I thought I had blogged this eariler, but apparently not.
AutoJar Combine all class files associated with an application into one jar.

Cleaning up after cancelled AT, and still using the parent node

nigel — 2006-07-24T20:41+00:00

An AT BaseContent derived product does not clean up after itself if I go Add, then press Cancel. I'm left with an object "named" [...]. As explained here, you need to register your type with the portal_factor so that cancellations are cleaned up. However, doing this means you can't use the self.getParentNode() method until the object is saved. So defaults and the like will not work if they derive their values from the parent. There is a methid, self.getFolderWhenPortalFactory(), which overcomes this problem. If you are not using the portal_factory then it will just return the parent. If you are using the portal_factory then it will return the folder where the object was created!

Archetype lifecycle

nigel — 2006-07-24T18:14+00:00

This is from my traces which I ran against a BaseContent type, using these docs as a guide.

User	Plone
Create object
	manage_afterAdd
(Edit page displayed) Save
	manage_beforeDelete manage_afterAdd at_post_create_script
(Saved item shown) Click edit
	(None)
(Edit page displayed) Save
	at_post_edit_script
(Saved item shown) Click delete
	manage_beforeDelete

Some Archetype notes

nigel — 2006-07-24T00:12+00:00

I've been playing around with writing my own plone product from scratch for a totally unrelated purpose, but in the midst of this I've made a few important starter hints for useing Archetypes.

BaseFolder and BaseContent are the types to inherit from
To set defaults from the parent of a content, set the
default_method = 'get_default_value_x'
def get_default_value_x(self):
self.getParentNode()...
Or, overrider the initializeArchetype method
Set the meta_type static variable for the object
To restrict types on a BaseFolder allowed_content_types = ( 'TypeA', 'TypeB')
And on save is manage_afterEdit (I have not tested this myself yet)

Debian hanging on boot at udev

nigel — 2006-07-17T15:28+00:00

It seems it is libnss-ldap which is causing udev to hang. Removing it does fix the issue, but we still need ldap, so I don't know what the full answer is yet.

Bug#375215: libnss-ldap hangs udev at startup

SSH timeout. Connection reset by peer

nigel — 2006-06-14T22:32+00:00

If you get messages like : Read from remote host xxx.xxx.xxx: Connection reset by peer in your idle SSH sessions, and you are behind a NAT, then it probably has to do with this.

To solve, put this line in your /etc/ssh/ssh_config:
ServerAliveInterval 60

Optimize your ext2/ext3 filesystem

nigel — 2006-06-05T18:25+00:00

This is a repost from here: http://blog.drinsama.de/erich/en/linux/2006060401-optimize-your-ext23-fs

...use tune2fs -l /dev/hda5 | grep features to get a list of your filesystem features. If you have "dir_index" in there then you're all set. Ignore any "needs_recovery" - that is good for mounted filesystems (it just means "not cleanly unmounted")
To add dir_index, do the following.
Unmount the filesystem. If this is your primary filesystem, use a bootable CD like Knoppix or some Linux install CD.
Make sure the filesystem is really not mounted. Just for safety of your data. Run e2fsck /dev/hda5 - this should take less than one second if your filesystem is clean and is needed anyway otherwise...
Execute tune2fs -O +dir_index /dev/hda5 to enable the feature for your filesystem
Execute e2fsck -Df /dev/hda5. This will force a filesystem check with directory optimization (!) on your filesystem.
Mount your system (or reboot, if you were using a live CD)

Getting RSS feeds from private blogs.

nigel — 2006-05-30T14:42+00:00

If you want to be able to access somes RSS feed, which is in a private blog, using software other than an integrated Firefox or Safari reader, then you need to slightly alter the RSS URL to tell plone to accept HTTP auth. To do this, append the following to the RSS feed:

?disable_cookie_login__=1

So, if my blog was private, it's RSS feed would be:

https://plone.jcu.edu.au/research/Members/nigel/nigels-blog/feed.rdf?disable_cookie_login__=1

Network Manager doesn't find devices

nigel — 2006-05-23T22:19+00:00

After upgrading network-manager from 0.6.2-1 to 0.6.2-3 none of my network devices show up. This is due to a change in the way devices are enumerated. If your device is listed in /etc/network/interfaces then it will not be managed by Network Manager. So deleting all references to non lo devices solved the problem.

Livesearch in Plone

nigel — 2006-05-23T18:57+00:00

The situation where live search returns one set of results, and normal search doesn't is caused by the fact live search appends * (wildcard) to the search, while normal search doesn't.

Speeding up blogs

nigel — 2006-05-07T12:45+00:00

I've noticed that it can take a long time to display some large blogs when using the plone viewer. This seems to be able to be alleviated if you turn off showing the category pictures in the blog itself.

The weigh in

nigel — 2006-05-05T22:30+00:00

The results are in for the bike weigh in:
Mat's: 9kg
Tristan's: 11kg
Nigel's: 13kg
Ian's: 13kg
Russell's: 14kg

Now how much does it cost per kilo?

Beagle - Evolution

nigel — 2006-05-02T13:53+00:00

Beagle has this nasty problem for me. When beagled starts this happens:

Debug: Getting calendar changes for file:///home/nigel/.evolution/calendar/local/system
Error: Unhandled exception thrown.  Exiting immediately.
Error: GLib.GException: Unknown error
in <0x00187> Evolution.Cal:GetChanges (string,Evolution.CalComponent[]&,Evolution.CalComponent[]&,string[]&)
in [0x0003f] (at /tmp/buildd/beagle-0.2.4/beagled/EvolutionDataServerQueryable/CalContainer.cs:92) Beagle.Daemon.EvolutionDataServerQueryable.CalContainer:IndexChanges ()

I'm not sure how to actually fix it, but a work around was found here, which basically equates to: apt-get remove beagle-backend-evolution