Saturday 17 December 2011

Monday 28 March 2011

outlook 2003 stationary picture not showing in windows 7

I was having a problem where in outlook 2003 the pictures in my custom stationary would not display correctly in windows 7. Everything was working fine in XP but stopped working on Windows 7. I solved the problem by storing the images in the same folder as the html template and changing the html source to point to the local folder. I had previously tried the various registry changes and none of them helped. Before sorting it out I also found the pictures displayed correctly if I used word as the email editor, but this causes lots of other problems so  it was only a test. I hope this helps as it was driving me mad.

Monday 3 May 2010

A great way to map you idisk from a windows PC

This chap describes a great way to map your mobile me idisk as a drive on your windows PC. You can then use an open source bit of software to backup to it:
http://islandinthenet.com/2008/08/06/mounting-mobileme-idisk-on-windows-xp/

Tuesday 27 October 2009

Setting up Axosoft Ontime with SQL Express & Subversion

Having just gone through setting up Ontime2009 with SQL Express 2005 (for evaluation) and Subversion running on VisualSVN server there were a few blogs and hints I found very usefull:

In order to be able to access the SQL Express 2005 database from a client running on another machine you have to enable TCP-IP access on SQL Express (it is not enabled by default), this blog describes the steps:

http://blogs.msdn.com/sqlexpress/archive/2005/05/05/415084.aspx

Once this is done you should then be able to access the ontime database from any ontime client, not just the ones running on the machine with SQL Express installed!

To give Ontime access to your Subversion repository you have to enable anonymous read access on your subversion repository (as ontime browses your repository tree structure before logging in a user). Without doing this whenever you try to connect to your repository you will get an error similar to the one below:

an unhandled exception occurred:


System.NullReferenceException: Object reference not set to an instance of an object.
at Axosoft.Common.SubversionProcess.List(Boolean refresh)
at Axosoft.Common.SubversionWrapper.List(Boolean refresh) (etc...)

This blog describes the steps for VisualSVN server, although the steps with be very similar on a apache server with subversion installed seperately:

http://west-wind.com/Weblog/posts/480534.aspx

With this done you should then be able to browse your Subversion repository from the SCM tab in ontime.

This blog then describes a way to include links into Ontime with your Subversion commits:

http://www.webapper.com/blog/index.php/2009/03/26/axosoft-ontime-to-subversion-svn-hooks-via-tortoisesvn/

However I preffered this approach as it did not invlove modifying primary key generation in the ontime SQL database, you just have to be a little bit more formal in your commits (prefix id with a 'd', 'f' or 't'):

http://brandonbehrens.blogspot.com/2009/04/integrating-subversion-with-axosoft.html


When you have done this you just need to add the ontime properties into your existing subversion repositories. The only thing I found that wasn't obvious was you have to check out a local copy of the entire repository from the root (branches, tags & trunk... the whole lot all in one go!) then for the root folder use the properties window from Tortiose to set up everything as described in the blog above. For each property you must also check the 'Apply property recursively' box so that the properties are added to all your existing files and folders (if you are using the properties box on the repository directly, NOT on a working copy, this option is greyed out). Any new ones you create will automatically have these properties added but the existing ones will not unless you ask them to have:




Once you have done this for each property you then just have to commit the whole lot back again. The next time you check something in or browse the repository you will see the following in Tortoise, Subclipse, or whichever subversion client you are using:












 


 
The nice thing about using these new features within Eclipse is that when you click on the 'OnTimeID' link it opens up OnTime in a web browser inside Eclipse which makes cross-referencing very quick & easy.

Will post some more as I have more problems...

Tuesday 1 September 2009

Migrating a local repository created with TortoiseSVN to a Subversion server hosted repository

If you have created a repository using TortoiseSVN which you access through the URL 'file:///C:/...' but want to migrate this repository to one which is managed and accessed through subversion & Apache, then the following is what worked for me. I originally started off with just TortioseSVN without a Subversion server as it was really easy to setup and generally an excellent way to get to grips with using a repository. Recently as I have been using it more and more I started to run into limitaions with network drives and accessing the data from other computers (it's made very clear in the TortoiseSVN help file that this would be the case, so it was not a suprise). I decided that it was time to bite the bullet and move everything across to a subversion server which I would access through Apache.
At first I thought there would not be an easy way to get all the data in my TortoiseSVN repository into a new Subversion repository but it turned out it was very straightforward indeed (for me at least)!

The first thing to do is to get your Subversion server up and running with Apache, which I have described here. As TortoiseSVN creates repositories in exactly the same format as Subversion, you can use all the subversion administrator commands on the TortoiseSVN repository as though it were any other Subversion repository:

1/ If your TortoiseSVN repository is not stored on the machine which is running Subversion and Apache make a copy of the repository and save that onto the Subversion machine. All of the following steps need to be done on the machine running Subversion and Apache.

2/ Open up a terminal window (start->run->cmd) and change the directory to the directory which has your TortoiseSVN repository inside it (but not to the respository directory itself). In my case my TortioseSVN  repository directory was strored in the root of my C:\ drive so the path in my terminal window looked like 'C:\'

3/ At the prompt in the terminal window type the following then press enter:
svnadmin dump TortoiseRepository > dumpfile
If your TortoiseSVN repository is not called 'TortoiseRepository' just change the name above to whatever you have called it. If the dump works you should see lots of '*Dumped revision' messages. The 'dumpfile' now contains an entire export of your TortoiseSVN repository and all the revisions, comments etc.

4/ Now change the directory in your terminal window to the directory which has your new repository inside (but not to the respository directory itself). In my case I stored my new subversion repository in the root of my C:/ drive again, so the path in my terminal window looked like 'C:\'. Also, make sure you copy the 'dumpfile' (this could be very large depending on the size of the repository you have dumped to it) which will be stored in the path from '2/' above into this directory.

5/ At the prompt in the terminal window type the following then press enter:
svnadmin load Repository < dumpfile
If your new subversion repository is not called 'Repository' just change the name above to whatever you have called it. If the load works you should get lots of 'adding path' messages. Your new repository now contains all the data which used to live in your old TortoiseSVN repository.

6/ Your old TortoiseSVN repository and the 'dumpfile' can now be deleted or kept as a backup. If you want to access your new repository from the machine which is running Subverion and Apache you can use the folllowing URL in Tortoise 'http://localhost/svn'.

7/ Allthough you are not using the repository you created in TortioseSVN any more, you can of course still use TortoiseSVN as your client to access the data on your Subverison server. I personaly use Tortoise as the client where ever possible as I find it so easy to use and the help file is so good.

If you have any problems have a look at Repository Administration section of the subversion book which goes into this process in some detail. The bit I realy needed to know when starting the above was that you could use the svnadmin comands on the Tortoise created repository. Once you know this the subversion book leads you through it. I hope this helps...

Monday 31 August 2009

How To configure SVN 1.6 on a machine allready running Apache 2.2 under Windows (XP/Server 2003):

I have previously been using VisualSVN or just TorioseSVN in standalone mode but wanted to be able to do other things with Apache on the same machine than just host subversion (VisualSVN limitation) or upgrade TortoiseSVN without worrying about corrupting up my repository (standalone mode)! The following is what worked for me, with some pointers taken from the Subversion install Readme file,the subversion book and a fair bit of head scratching:


2/ this should automatically add 'C:\Program Files\Subversion\bin' to the SYSTEM PATH, check by:

(right click 'my computer' -> properties -> Advanced -> Enviroment Variables -> System Variables -> path)

it should be in there somewhere, if it was the last thing you installed it should be at the end!

3/ open up command prompt (start -> run -> cmd) and at the C:\> prompt type:

svnadmin create c:/Repository/

This will create a new repository in the directory c:\Repository.If you want it in a different location just subsitute 'c:\Repository' for the path you want.

**WARNING** as subverision was developed in Unix/Linux when you type 'svnadmin create' the slashes in the directoty location have to go the wrong way to what you are used to in Windows (treat it like a URL rather than a directory path)!!

4/ Copy 'mod_dav_svn.so' and 'mod_authz_svn.so'

from -> C:\Program Files\Subversion\bin

to -> C:\Program Files\Apache Group\Apache2\modules (or where ever Apache is installed)


5/ Make a backup of and then edit the Apache configuration file and make the following changes: (C:\Program Files\Apache Group\Apache2\conf\httpd.conf)


A. Uncomment the following two lines:
#LoadModule dav_fs_module modules/mod_dav_fs.so

#LoadModule dav_module modules/mod_dav.so


B. Add the following two lines to the end of the LoadModule section:
LoadModule dav_svn_module modules/mod_dav_svn.so

LoadModule authz_svn_module modules/mod_authz_svn.so


C. Add the following to end of the file. Note: This Location directive is a minimal example with no authentication directives. For other options, especially authentication options, see the Subversion INSTALL file, the Subversion Book, or the TortoiseSVN Manual.

DAV svn

SVNPath C:/Repository/ (or where ever you created the repository in step 3)

Note the forward slashes again in the path!

6/ Restart the PC & check that Apache has started properly (if not double check all of the above)

7/ On the machine tha Apache is running on navigate in a web browser to: http://localhost/svn/if everything is setup correctly you should see this picture :


8/ Now to import the directory structure into subversion:

A. create a directory called temp
B. Inside 'temp' create a directory called 'Project' (or whatever you want your project to be called NO SPACES THOUGH!)
C. Inside 'Project' create three directories called 'branches', 'tags' & 'trunk'
D. right click on 'temp' and then from the TortoiseSVN menu(http://tortoisesvn.net/downloads) select import. (you can do all of this from a command prompt but I find TortoiseSVN so easy to use there is not much point!)
E. in the URL enter: http://localhost/svn/ (provided you are doing this on the machine which is running Apache)
F. in the message you can put something like 'creating directory structure to hold project'
G. Then click 'OK'
9/ Subversion is now set-up and ready to use (BUT WITHOUT ANY SECURITY ENABLED!) see subverion book & TortoiseSVN for the rest.



Thursday 20 August 2009

Emergence for Sat Navs & avoiding traffic

I have been thinking if there is any way that we could make sat navs more intelligent and get them to provide usefull data for other vehicles to use. Now that you can combine your mobile phone & sat nav into one device it is very easy for the sat nav to send info back to a central server that anyone can access. What would be the point? Well, the sat nav will know what road you are on and what the average speed should be on that road. If your speed then drops to say 1/3 of what it should be, the sat nav could send this info to a central server. If quite a few sat navs were doing this you have immediately got an up to date traffic service (including for all the alternative routes as well). Your clever little sat nav could then retrieve this data and plan you an alternative route (and maybe even tell the central server what that is). With a few clever algorithms to server could keep track of which alternative routes people are taking and make sure that everyone is not redirected the same way. Could this be the future for Sat Navs & traffic...maybe

Tuesday 4 August 2009

Code for New Blogger Tag Cloud / Label Cloud - phydeaux3

This code is very clearly described and works a treat! You just need to make sure that your labels are only single words:

Code for New Blogger Tag Cloud / Label Cloud - phydeaux3

Sunday 19 July 2009

Mobile Devices & The Internet

The possibilities really are endless now. The iPhone combined with networking sites have just opened the flood gates. The way people use computers and the Internet is going to change very very rapidly now. It's extremely exciting to be working in a time when a lot of industries are going to be turned on their heads. The question is not going to be what new developments are going to come out in the next few years, but what is going to come out in the next few days or months.

http://www.buzzmachine.com/what-would-google-do/


http://www.stevenberlinjohnson.com/


http://www.techcrunch.com/

Tuesday 14 July 2009

Project Planning software

Today I have been mostly using openproj for project planning. It seems to compare well with Microsoft project and can do all the basic stuff just as well. There seems to be some general limitations with project planning software though that would benefit from looking at how object orientated software is written. If you could have the idea of funtions in planning software (re-usable groups of tasks with standard inputs and outputs), this would change the funtionality considerably. It would also make adding in or modifying a group of tasks that is used more than once very efficient. Maybe openproj, being open source, would be the ideal way to try this out... Will think about this more

http://openproj.org/