Developer Notes

Made a space for my "tech notes"

2008-12-22T23:53:00.003-05:00

I'm moving my tech stuff to tech0x20.com (Tech Space)

I originally created three sites on my thomaspowell.com domain:

linux-notes - I started my blogging experiment here.
vim - My "adventures" in vim, not a very broad audience, but I liked keeping public notes on my favorite programming editor here.
developernotes - I realized that I was researching and wanting to publicly note things that did not apply to the other two blogs.

Moreover, my role and experience has changed in general. I'm no longer a full-time developer. I haven't been that for a year or more. I still research programming and system configuration, but I just as often research technology, trends, and concepts. I had been using the "developer notes" blog for this, but I was starting to branch out into a new blog when it hit me. I *really* needed to kick the blogger habit. Therefore, I have started a new home with a shiny wordpress theme at tech0x20.com.

Installing JSPWiki on an XAMPP for Windows install

2008-12-22T07:39:00.011-05:00

I've been experimenting with XAMPP as a way to setup a temporary and transportable Apache install with PHP, MySQL, and a database admin tool (phpMyAdmin). Before starting, be sure to have Sun J2EE 5.x SDK (or higher) installed.

As mentioned on PortableApps.com, by following the "Installation" without the installer instructions, XAMPP will run from a removable drive (I'm using an SD card, myself--quite a bit slower than other options, but VERY cheap).

First, I downloaded the ZIP package from from the xampp for windows page:

Open the xampp-win32-*.*.*.zip file that was just downloaded and extract to C:\ (change to flash drive drive letter here if necessary).

Run [drive]:\xampp\xampp_start.exe to test starting up the web server and mysql.
Run [drive]:\xampp\xampp_stop.exe to shut the web server and mysql down again.

Next download the Apache Tomcat plug-in from the XAMPP for Windows Add-Ons area.

Extract this .zip file into to the c:\xampp folder (again, substitute c: for appropriate drive letter if necessary). You should get a "Confile File Overwrite" prompt when you do this--the Tomcat files should overwrite the original xampp files, so select [Yes to All].

After the Tomcat files have been extracted, run [drive]:\xampp\setup_xampp.bat to setup Tomcat.

Test the installation by running

[drive]:\xampp\xampp_start.exe to start xampp, then
[drive]:\xampp\tomcat_start.bat to start Tomcat.
Browse to http://localhost:8080, which should be something like the following:

Finally download JSPWiki (Current stable release):

Extract the JSPWiki.war file from the JSPWiki download and copy to the [drive]:\xampp\tomcat\webapps folder.

Run [drive]:\xampp\tomcat_stop.bat and then [drive]:\xampp\tomcat_start.bat to restart Tomcat. Web application JSPWiki should get loaded as part of the restart. I had to restart a second time (probably started too soon after shutting down.)

Browse to http://localhost:8080/JSPWiki to view the JSPWiki page. You will be prompted to begin installation:

For the installation options, I used:

Application name - JSPWiki
BaseURL - http://localhost:8080/JSPWiki/
Page Storage - /p/web/www-data/jspwiki/ (default) -- note that this will be the path off of your root drive wherever xampp and Tomcat are running from. This is where the wiki pages will be stored.
All other options - default values

Finally:

Press [Configure!]
[drive]:\xampp\tomcat_stop.bat to stop Tomcat.
[drive]:\xampp\tomcat_start.bat to start Tomcat.
Browse to http://localhost:8080/JSPWiki/ to start using your JSPWiki. By default, versioning is turned on, which is accomplished by storing every version of the file in an "OLD" folder just off the main page storage folder.

Troubleshooting JSPWiki Installation

Insight DNS issues #insight

2008-12-15T20:34:00.004-05:00

Half of my DNS requests are timing out, and DHCP release/renew works about 50% of the time...

thomas.powell

Initial Question/Comment: Half my DNS requests are failing, DHCP renew fails half the time.

8:29:33 PM System

Ashley H has joined this session!

8:29:33 PM System

Connected with Ashley H

8:29:33 PM System

Thank you for contacting Insight Communications Customer Netchat Support, please allow me a moment to read your question and I will be more than happy to assist you.

8:29:53 PM Ashley H

We are currently experiencing dns issues in your area that may cause slow browsing speeds or intermittent issues loading pages.

...Hey Insight, Power Down and Power Up. It's easy!

Added @ 8:50 PM:
Thanks to @DanielLight I'm using OpenDNS now.

Of course, now I can't connect to the server addresses half the time. Thanks, Insight!

Every time you make an Excel spreadsheet this big, a kitten meets its maker

2008-12-15T13:25:00.003-05:00

I'm not really that elitist about non-programmers using applications in an advanced way the programmers could find a better way to do. I've never had to port functionality from an Excel spreadsheet that was 2 MB in size prior to data being added--at least not more than once.

However, I am a bit disturbed that Excel is the new Access:

The total number of available columns in Excel
Old Limit: 256 (2^8)
New Limit: 16k (2^14)
The total number of available rows in Excel
Old Limit: 64k (2^16)
New Limit: 1M (2^20)
Number of unique colours allowed a single workbook
Old Limit: 56 (indexed colour)
New Limit: 4.3 billion (32-bit colour)

Therefore, Excel is now Microsoft's new small business data warehouse. (Apologies to DW people.)

You could also store and represent a very large 32-bit color bitmap inside an Excel spreadsheet. Also supported are Base-64 Encoded BLOBs :-) :

The total number of characters that can display in a cell
Old Limit: 1k (when the text is formatted)
New Limit: 32k or as many as will fit in the cell (regardless of formatting)
The number of characters per cell that Excel can print
Old Limit: 1k
New Limit: 32k

I'd hate to have to debug a 8k character formula with 64 levels of nesting:

The maximum length of formulas (in characters)
Old Limit: 1k characters
New Limit: 8k characters
The number of levels of nesting that Excel allows in formulas
Old Limit: 7
New Limit: 64

A 255 argument function. The mind shudders. Even if you did not use a computer science major to create the program, such an unwieldly program might benefit from the education and experience of a computer science major. Maybe there is a better way to do this, by applying some mid-level software engineering techniques instead of stream of consciousness coding.

Maximum number of arguments to a function
Old Limit: 30
New Limit: 255

TwitterThreads.com - what I've been looking for. Almost.

2008-12-15T09:43:00.002-05:00

I just searched on "twitter" and "threads" in google. What I found was TwitterThreads.com. Simple, huh?

It's not 100% what I had envisioned: My vision was probably something more like expandable threading by author sorted by date of most recent update. The TwitterThreads version looks like it shows all posts that you'd normally see if you were following someone, but limited to a single day's posts.

This is pretty much what I wanted to build, with a few additional features... Other things I thought of were [+]/[-] Expand/Collapse functionality, an option to configure the number of days shown, and read/unread functionality of some sort.

There is also a mobile TwitterThreads site. On my ancient Treo 650, this is a beautifully elegant and simple interface. It may become my new twitter interface on my phone.

Something both interfaces are missing are direct links to "reply-to" and "favorite" individual tweets. I was also a little disappointed that the timestamp link on each tweet went to the person's home instead of direct linking to the tweet on Twitter. This makes the lack of reply-to and favorite functionality more of an issue for me. If the timestamp had linked to the individual tweet, I could reply-to or favorite through the Twitter web interface.

Overall, I am still thrilled to see that someone has implemented this idea. Like any programmer geek, I would have liked to be the first, but I wouldn't really have made the time to throw something like this together.

Proof that I'm not the only one with problems with the Lenovo T-61 laptop

2008-12-05T11:27:00.008-05:00

I had virtually no problems with my IBM T60 laptop, but my T61 has had several problems.

The most recent problem that I've had is that I went into hibernate mode last night, and this morning, my PC would not turn on:

I dock my PC, the dock indicator light comes on, but the battery and AC power indicators on the laptop do not. The power button does not respond.
I remove my PC from the dock and plug in the AC directly, still the battery and AC power indicators on the laptop do not come on.
I remove the battery and plug in the AC directly. Still nothing.
I finally panic enough to call our internal help desk (early call might get turnaround by lunch time) and the tech offers one last suggestion... pull the battery out (AC too) and let power state drain, then re-insert battery. Crisis averted.

Lenovo support thread for T61 power on problem

I've also had problems with the LCD screen being locked in an "off" state on [Max Battery] mode (Presentation mode works well, though).

When I had the PC in [Max Battery] performance, the machine would not come out of hibernate at least 50% of the time.

I had a DAILY blue screen on Windows XP when I was docked (wired connection to LAN) and wireless was turned on. This was remedied by having my laptop upgraded to a newer version of Symantec Security software than was widely deployed in the organization.

Added 2008.12.05:
If my laptop was in its docking station when I tried to hibernate, it would restart immediately. I have now gotten into the habit of undocking first, then hibernating. According to another blog, this is due to a Wake On LAN setting in Windows Device manager and BIOS.

cnet user review: Blue Screen of Death w/ Bad Customer Service

Added 2008.12.09:
T61 Problems?

Added 2008.12.12:
I ended up in the weird power state again today, but this time I was prepared to pull my battery, wait 10 seconds, and replace. Voilà! It works!

Finding distances between two zip codes in PHP

2008-11-29T14:40:00.002-05:00

I found this script to calculate distances between two zip codes from PHP (requires MySQL). According to this Idealog post, the calculation relies on a text file of Zip code lon/lats from CFDynamics, available on the downloads page.

I have yet to try it; however, I have a specific implementation that I was needing it for. Does anyone else have a better way to align searcher's proximity to location-based data? I'd like to classify by city, but I'm sure that is a lot more of a gray area than allowing the searcher to specify a specific radius in which to search or letting the resource specify maximum distance to travel.

Added 2008.11.30:
I love Twitter. A few hours after posting this, I received a suggestion that I could get longitude/latitude from Google Maps.

On code.google.com, I found the following question: I need to convert addresses to latitude/longitude pairs. Can I do that with the Maps API?

Yes, this process is called "geocoding." The Google Maps API provides two methods for performing geocoding. If you wish to geocode from within your Google Maps API application you can do so using the GClientGeocoder object. Alternatively you can send geocoding requests directly to the HTTP geocoder.

Gmail terminal theme--for your inner nerd

2008-11-20T22:14:00.003-05:00

I don't know if I like it for nostalgia and geekiness, or just plain geekiness. There's only one thing I'd change with the terminal theme... make it truly like the BBS experiences of 15+ years ago. I want ANSI animation and color ANSI art--especially for the Gmail logo.

Added: Even more importantly, the terminal theme works great if you get LISTSERV mailings that have tables rendered in all ASCII characters or code.

I suspect that the delicious bookmark plug-in for Firefox 3 is not keeping me logged for two weeks.

2008-11-15T09:32:00.006-05:00

Not only that, but whenever I'm required to log-in to bookmark a page, the plug-in does not continue on to actually creating the bookmark.

This morning I was forced to log in. I believe the loss of information is coinciding with Firefox updates. Anyone else have this experience?

Home PC:
2008.11.15, 9:15 AM
2008.11.29, 2:29 PM
2008.12.14. 6:16 PM

Work PC:
2008.11.26, 10:01 AM
2008.12.11, 5:08 PM

Updated 2008.11.29: So far, no glitches. However, I still dislike the fact that the log-in prompt prevents tagging and you have to have to select the tag icon again after logging in.

Updated 2008.12.11: Looks like everything's fine, just that I'm overly sensitive to having to login.

Autotext lines in Word (using 2003 here)

2008-11-10T15:03:00.003-05:00

Sorry if this is pretty trivial for the true power users of Word...

I don't know if Word 2007 works this way or not, but I've discovered a couple additional characters in Word that auto-correct to divider lines. (Other than 'equals', 'minus', and 'underscore'). The 'hash' divider was of the most interest to me, as I'm trying to write a white paper and need distinctive, yet compact, format markings. (The quicker the better, as well.) Just enter 3 successive '#' (hash), '~' (tilde), '-' (minus sign), '=' (equals), '_' (underscore), or '*' (asterisk) characters and immediately press enter. The text will be replaced by a dividing line.

(This trick does something similar OpenOffice 2.4.)

Reading List by Amazon

2008-11-09T08:22:00.004-05:00

It looks like LinkedIn has found a way to make itself relevant for day-to-day use... The Amazon Readling List provides a way for you to post what you're currently reading, want to read, or have read. You can also see the contents of other people's reading lists--in your network, in your industry, or all recent updates. Your reading list will automatically appear on your LinkedIn Home and Profile pages.

So far, I only have one book listed. I'll probably start adding to this reading list from my Safari bookshelf list, and all the various other lists that I've been compiling over the last couple of years.

You can get to my LinkedIn profile at http://www.linkedin.com/in/twilliampowell

Reading List Application full view:

On your LinkedIn Home:

On your LinkedIn Profile:

Adding from featured applications page.

99 bottles of beer and hello world programming examples

2008-11-03T22:43:00.002-05:00

I'm a sucker for different programming languages, especially in trivial and whimsical implementations...

99 Bottles of Beer in 1200+ programming languages

Esoteric Programming Languages wiki - contains Hello World samples.

The hello world collection (420+ languages)

Hello World project

List of hello world programs on Wikibooks

Found some Gtk/Ruby signal stuff.

2008-11-03T21:04:00.001-05:00

(For gtk2+ruby, I used the Windows installer package)

For some reason, I'm trying to learn a portable language/toolkit.

First, I had to find the correct signal to connect to. Looks like focus_out_event is it.

currentMileage.signal_connect("focus_out_event") {
mileagePerDay.text=((targetMileage.text.to_f - currentMileage.text.to_f)
/(Time.parse(targetDate.text)-Time.now)*3600*24).to_s
}

signal_connect vs signal_connect_after... apparently, improper usage of signal_connect (for "focus_out_event") will generate the following warning which the signal occurs
Gtk-WARNING **: GtkEntry - did not receive focus-out-event

From the description, what has happened is a timeout with a now invalid pointer (because of the signal's redirect). Using signal_connect_after() ensures that the default handler has run first.

currentMileage.signal_connect_after("focus_out_event") {
mileagePerDay.text=((targetMileage.text.to_f - currentMileage.text.to_f)
/(Time.parse(targetDate.text)-Time.now)*3600*24).to_s
}

Next up: Encapsulate some of this stuff. In this exercise, I realized that I forgot some of the basics, like how variable scope is defined in Ruby: http://www.informit.com/articles/article.aspx?p=18225&seqNum=2.

Some additional Ruby links...
Rubyist reference
RubyDocs reference to global variables
6 Ways to Run Shell Commands in Ruby

I remember the good old days of domain squatting

2008-10-30T10:09:00.001-04:00

At least you had to pony up a few bucks per domain name back then. Now there's Twitter squatting. It's fairly predictable, really... Every single letter TwitID is taken. Common surnames... taken. Common words... taken. Apparently, you don't even need to use a legitimate or at least legitimately unique e-mail address to do it.

Classic Blogger and PHP?

2008-10-30T08:42:00.001-04:00

A question via e-mail:

I stumbled upon your blog: http://developernotes.thomaspowell.com
What I noticed - and it's something that I wanted also - is that your blogpages are all ending in .php, even your individual blogpost pages (http://developernotes.thomaspowell.com/2008/10/remember-milk-is-on-twitter.php)

How do you do that in settings? I can only get the index page to end as PHP. The individual posts remain in HTML.

Could you share this nugget of wisdom with me?

(Note: I've only done this publishing to another host via FTP)

In Blogger dashboard, click the [Settings] link for your blog.

Under the [Settings] Tab -> [Publishing] Tab, edit the Blog Filename field. (index.html -> index.php)

There is also a setting for the archive file:
Under the [Settings] Tab -> [Archive] Tab, edit the Archive filename field. (archive.html -> archive.php)

One more thing to note... all the old *.html files will still be left on the server, so if you want them removed, you'll have to clean them yourself.

(You may also want to redirect old links to the new extension).

Is a Yahoo developer platform worth the time investment to learn?

2008-10-29T10:16:00.001-04:00

Yahoo Introduces Social Developer Platform

"3 elements":

Yahoo Social Platform, a set of REST-based social application programming interfaces (APIs) for utilizing social data related to Profiles, Connections, Updates, Contacts, and Status.
Yahoo Query Language, a new Web service API for accessing other Web services using a SQL-style query language, rather than a lower-level programming code. Yahoo describes it as "a command-line version of Pipes," Yahoo's visual programming system for mashing up and remixing Web data, like RSS feeds.
Yahoo Application Platform, a set of software and services to build applications that run on Yahoo. It includes a browser-based development environment, various APIs and Web services, a distribution and discovery infrastructure, and a runtime and rendering environment.

Yahoo Open Strategy (Y!OS) 1.0 Developer Release announcement and List of Y!OS Documentation.

Is this new hope for Yahoo, or another potential dead-end for an industry laggard? I remember Yahoo killing its Facebook wannabe, which, with TechCrunch keeping score, makes Yahoo 0-4 On Social Networking. Yahoo does not equal AOL, but I can't help but think of AOL's shutting down of AOL Journals, Pictures, Hometown, Bluestring, and XDrive. Is it worth investing the time to figure out something that will potentially be on the chopping block in the future?

Remember the Milk is on Twitter!

2008-10-20T15:30:00.003-04:00

This may be old news to many RTM/Twitter users, but it was a pleasant surprise to me today.

As I was perusing the Remember the Milk keyboard shortcuts, I noticed that at the bottom, under "Services", Twitter was listed. So I clicked on the link, and the "Services / Remember the Milk for Twitter page" came up, where I could enter my Twitter id. I was presented with a verification code for the next step.

From there, I typed "follow rtm" and then "d rtm [verification code]".

Now that all that is set up, I can interact with Twitter through direct msg:

"d rtm pick up the milk"

"d rtm call jimmy at 5pm tomorrow"

"d rtm !complete call jimmy"

All the instructions and some command examples are at: http://www.rememberthemilk.com/services/twitter/

Of course, I'm still learning how to even use the RTM Date Formats.

Added:
Of course, since I can txt updates to Twitter (40404), I can add tasks for today by texting
"d rtm pick up the milk today"

I don't need a laptop or notepad to record my "action items" anymore. This is awesome.

Added 8/23:
Thanks to @louisvilllesoup, I looked into sending tasks from e-mail to RTM inbox as well... this is a pretty robust feature set.

A much simpler setup is importing a list via e-mail, which enables adding a list in bulk to a specific list (e.g., personal) by specifying the list in the subject line and list items on individual lines in the message body. A downside to this method is that signature lines and legal disclaimers get added (per line) to your to do list. I'll need to research if there is a way around this.

Trying to dig up IT relevant podcasts

2008-10-20T09:50:00.002-04:00

Any suggestions?

InfoWorld Podcasts: These look promising...

Infoworld Daily - includes a general tech news segment which is repeated in the NetworkWorld 360 podcast as well.
The Virtualization Report
Storage Sprawl
The SOA Report
...a few discontinued ones are out there, too.

Enterprise 2.0 podcasts - looks like recordings from a conference? I'll have to download and check them out tonight. Not as promising as the Infoworld ones. I see that a lone spammer has managed to post spam comments in "response" to a few of the links.

NetworkWorld podcasts: Again, some promise here.

NetworkWorld's Twisted Pair (this sounds more like entertainment, but possibly relevant news)
NetworkWorld Panorama
NetworkWorld 360 - includes a general tech news segment.
Voices from the IT Roadmap
NetworkWorld's Newsmaker of the Week
Forrester Fundamentals
Network Downtime (entertainment)
Converging on Microsoft
Cisco News and Reviews
JavaWorld's Java Technology Insider
LinuxCast

Same old grind / a new day / gotta work / for the pay / Burma-shave

2008-10-20T08:12:00.002-04:00

Today's xkcd comic has Twitter all Twitterpated.

On comments and blogs

2008-10-17T15:59:00.002-04:00

A tweet by @JasonFalls reminded of this topic.

Have you ever been in a large meeting where someone asks a question that is uncomfortably inappropriate? The perceived anonymity of the internet only seems to only embolden such tendencies.

Copywrite, Ink.: Allowing Anonymous: Communicators Divided - The business and communications justifications for allowing anonymous, allowing moderated comments, or allowing no comments at all. I would tend to agree that considering "[allowing] no comments at all" myopic is a bit harsh, for two reasons:

While the average lone miscreant is relatively harmless, many organizations might (justifiably) consider themselves potential targets of coordinated attacks.
The more popular the blog, the faster the comments degrade into flame wars. I avoid the comments on digg.com and several news sites because I've seen mildly provacative devolve into vitriolic hatred and ignorance--sometimes over something as benign as a story about a local basketball game.

I do, however, like this final point:

However, and I cannot stress this enough, I do advise communicators and public relations professionals to never make anonymous comments or, if they do, they need to be prepared to answer for such posts in a world where no communication is really private. Not anymore.

Of course, my feelings might be partially influenced by the Unedited Voice of a Person:

Do comments make it a blog? Do the lack of comments make it not a blog? Well actually, my opinion is different from many, but it still is my opinion that it does not follow that a blog must have comments, in fact, to the extent that comments interfere with the natural expression of the unedited voice of an individual, comments may act to make something not a blog.

Joel Spolsky draws from this to make the point that you get a few insights, followed by a spew of noise/filth that no one would say out loud if they had to take ownership of their words.

Finally, xkcd illustrates:

Especially on YouTube:

In a time when I'm aggravated at Insight, Verizon cheers me so...

2008-10-17T09:51:00.002-04:00

...of course, only in a schadenfreude kind of way.

Part of my reluctance to change carriers for phone and internet service is the whole "devil you know" dilemma. I've never really considered Verizon a viable option for my needs, and I know very few people in the area that are on Verizon.

Verizon plays fast and loose with the wrong 1,200 e-mail addresses is an almost juvenile example of Verizon's customer service failings... just a mere mass e-mail with a cc: version bcc: used for the mailing list, plus some exchange server glitches. There's also the equally juvenile Run-amok Verizon robo-caller torments 1,400 customers incident.

When I think Verizon, I think of: Verizon CEO thinks it's unreasonable to expect your cellphone to work at home or Verizon doesn't know dollars from cents... Oh yeah, and 792 Octillion Dollars.

See? I'm not insane...

2008-10-16T21:52:00.002-04:00

Less than an hour later, no changes to my setup or running programs.

Firefox 3

Then things fall apart again:
IE 7

40 minutes later:

My current Insight cable speed tests.

2008-10-16T21:11:00.001-04:00

(No, I'm not just trying to squeeze maximum throughput here--I just want basic pages to load properly--the slow speeds are yet another indicator that there is a problem.)

Under IE 7:

Under Firefox 3:

These are actually excellent results for me right now... I've been dropping to 88 Kps down/22 Kps up. Aggravating thing is, at certain times of the day, I get over 3000 Kbs up and over 300 Kbs down (sometimes 7000 up, 700 down). Supposedly, since my line is "good" (according to the tech at the call center) and behavior just as bad without router plugged in, it hasto be the modem. I don't really have time to drive to Insight office to get replacement and then hook up. Besides, I told the tech that connection has been inconsistent since the storm. I guess it's cheaper for me to come have the modem replaced than to have someone check the line these days.

At least when I used to have bad internet service before, I'd at least get decent/on-site service. I guess I need to "power down/power up" 30 times, try turning the cable modem upside down, and letting the water run before getting a tech out here.

Added:
Dear Insight Broadband, Please Treat Your Customers with Respect
Insight want to save? you money

Update:
So far so good with the new modem:

This should tide me over until my cell phone contract runs out, at least.

Follow Cost on Twitter

2008-10-15T16:09:00.001-04:00

My follow cost for Twitter

All these silly little Top 10 lists.

2008-10-14T16:47:00.001-04:00

Okay, so I just love punditry sometimes...

10 Things That Won't Make Windows 7 A Success (and 1 That Will)

8. Death By A Thousand Versions Of The Same Product. How many versions of Vista are there again? Okay, if you say so. We only need 3 versions of Windows 7: a home version, a business version, and a mobile version. Three products, and three SKUs. That's it. Give us any more and we'll send you back into timeout, Microsoft.

10. Annoying The User. UAC was the direct approach. Ask the end user at every turn possible a question they don't care about, don't know the answer to, for something they are going to do any way. Or there's the indirect approach, the sum of a lot annoying little problems: slower file copy, a slower computer using experience, a lack of drivers, stability issues, confusing product editions, etc. Either approach works great at annoying the end user. Doing both guarantees it will happen.

Also, some really cool graphic design going on here on a couple of these slideshows:

Random Articles found through NetworkWorld

2008-10-14T15:40:00.001-04:00

Sending a message to TSA will cost you time - the lesson here? Don't try to be clever in what you take through security. Evan Roth has designed personalized plates that allow the bags screeners for TSA to see a "message" that you've had etched into a steel plate. Of course, these plates obstruct the view of your baggage, pretty much guaranteeing that you'll be searched.

2008 I.T. Industry Graveyard - Slide 3... The IT department. I don't know about this. This prediction is along the same lines as the prediction of the death of the desktop computer 10 years ago and the "this time it's different" about the dot-com boom, the housing market, or [insert your future prediction here]. Until the barriers to replacing the legacy systems are met, someone will always be making the connections to the old data through SOA, or worse yet, changing the old applications for minor changes in the business environment.

A contrasting (if not differing) opinion from ComputerWorld.com blogs: Vista R.I.P.

Problem 1 from Project Euler . net ("set"-based solution)

2008-10-13T20:57:00.001-04:00

This time, I fill a hash map for step values of 3 and of 5. No noticeable difference from my first try, until you start trying to find results for much larger than stopping at 999, at which point amount of memory consumed/allocated makes this method way more inefficient.

x=0

h = {}
3.step(999, 3) {
|i|
h[i] = 1
}
5.step(999, 5) {
|i|
h[i] = 1
}

h.each { |k,v|
x=x+k
}
print x,"\n"

A fledgling blog on the progress of technology

2008-10-13T13:42:00.001-04:00

I think I'm generally more interested in the small-time blogs than the well-established blog. Well-established blogs with a solid following can succumb to group-think in the guise of introspection.

Sometimes, a newer blogging voice can present a different perspective on existing topics:

Found from a @implu tweet Retire your desktop: The Future:
- Greenspan's perception of the Information Technology bonus (in terms of productivity)
- Wall Street's irrational exuberance leading to the DotCom Bubble.
- Gen Y's unintentional leadership of humanity into freedom.

Problem 1 from Project Euler . net

2008-10-12T22:44:00.001-04:00

Trying to reinforce the syntax in ruby, I did Problem 1 from Project Euler . net:

If we list all the natural numbers below 10 that are multiples of 3 or 5, we get 3, 5, 6 and 9. The sum of these multiples is 23.
Find the sum of all the multiples of 3 or 5 below 1000.

x=0
1.upto(999) {
|i|
x=x+i if (i % 5 == 0 || i % 3 == 0)
}
print x

I'd like to do this by merging two sets of numbers next.

My first powershell

2008-10-12T22:39:00.001-04:00

A foreach loop that calls ping with a specific buffer size, and sends output to a ping*.txt file corresponding to the buffer size:

foreach ($number in 32, 256, 1024) { ping -n100 -l$number yahoo.com > ping$number.txt }

For when I really can't get my brain fired up on a given day

2008-10-10T15:25:00.001-04:00

Project Euler for programming problems.: "Project Euler is a series of challenging mathematical/computer programming problems that will require more than just mathematical insights to solve. Although mathematics will help you arrive at elegant and efficient methods, the use of a computer and programming skills will be required to solve most problems."

I was looking up V=IR, P=IV, etc... and found this site: Equation Sheet site. It has algebra, calculus, statistics, and physics, etc... equations.

A simple AJAX TODO List and links to the code

PC World's opinions on products and websites.

2008-10-06T16:43:00.001-04:00

100 Incredibly Useful and Interesting Web Sites

The Bottom 10: The Web's Most Useless Sites - Digg made it! Wow. Some pretty ugly sites otherwise as well.

The 10 Most Overrated Products -- a little too much credit to Microsoft, especially with the underrated Zune vs. the overrated iPod. Part of the greatness of the iPod and iTunes is the market saturation of the product. My ripped CDs, my iTunes-purchased AAC files, and my podcasts co-exist seemlessly. The wifi sharing feature of the Zune is made relatively useless by the lack of adoption.

Cleaning up a brand new PC

2008-10-06T16:13:00.001-04:00

I've found a new blog to scour: DigitalInspiration.

Bought a New Computer? Now Remove all the Pre-Installed Trial Software - this article points to the program PC Decrapifier and the full list of programs that the program can remove.

A related post of theirs, Spring Cleaning Utilities for your PC : Reclaim Hard Disk Space, provides more clean-up tips and utilities.

Send Tweets to Twitter via e-mail

2008-10-06T15:50:00.001-04:00

Send Twitter Updates via e-mail

Interesting concept:

Create a dummy blog
Set-up a private mail-to-blogger address.
Use twitterfeed.com and your OpenID to associate the rss.xml feed for the blog to your Twitter account

To post to Twitter, all you have to do is e-mail your Twitter updates to the mail-to-blogger address.

Google Calculator cheat sheet

2008-09-26T11:09:00.001-04:00

I just found an awesome site describing some of the features in Google while looking for info on Google Calculator: Google Guide

My first find was the Quick Reference: Google Calculator (Cheat Sheet).

I like this tip from Google Guide solutions to Calculator problems:

Compute the probability of your winning the lottery if you buy 1,000 tickets each bearing five distinct independently randomly chosen integers between 1 and 99.
The following answer was contributed by Nelson Blachman (Nancy's father).

You might try [ 1 - (1 - 1/(99 choose 5))^1000 ], but this is well approximated by [ 1000/(99 choose 5) ], which is roughly 0.00001 . If each ticket costs $1, the prize had better be at least $100 million to be fair to you — and higher if you might have to share the prize with other winners.

While I was on that train of thought, I found the Lottery Mathematics entry on Wikipedia.

Also... I've always wanted to figure out the probability calculation for various poker hands.

One more interesting link: Soople.com's Visual Interface to Google's calculator

Way to go, Beshear.

2008-09-25T16:55:00.001-04:00

Kentucky is getting a lot of love on the Tech Blogs...

Kentucky tries to seize gambling site domain names - on ArsTechnica

State of Kentucky Seizes Control of 141 Domain Names - on SlashDot

Kentucky judge moves to seize gambling sites' domains - on ValleyWag

Kentucky Governor Seizes Online Gambling Domain Names - on TechDirt

Not only is the state of Kentucky being laughed at for this massive overstepping of jurisdiction, but every single other facet of Kentucky living and business is being dragged through the mud. Just look at the comments. Never mind that... just reading the articles is enough.

The trouble with Twitter

2008-09-10T08:46:00.000-04:00

The Trouble with Twitter

The trouble with social networking... (see The Social Brain Hypothesis, page 184, column 3)

British anthropologist Robin Dunbar [...] noted in 1992 that humans—like other primates—can handle only 150 relationships. If we try to add many more connections, our little brains get overloaded.

The conclusion from a ballpark example...

Thus Twitter has a real value of $12.26 per user. Compare that with Facebook, which has a perceived value of $300 a user—or at least it did last year, when Microsoft purchased its 1.6% stake for $240 million and the site had 50 million users.

Even with this example, there is the difficulty of inserting the ads... In messages? On the front end interface?

My thoughts on Yammer

2008-09-09T05:30:00.002-04:00

I commented on @joewheeler's take on Yammer (Yammer.com):

I'm not entirely sure that there is as huge of a barrier to adoption from the corporate side. Office Communicator and Windows Messenger before that offered private network IM capability. Additionally, the Millenials and like-minded young adults are starting to filter into the workplace as interns and entry-level employees... Watching job websites, I often see articles about attracting the younger workforce. Microblogging is one way.

The true barrier is *not* the old school management thought (although it is a realistic barrier). The true barrier is old school employee thought. Microblogging escapes the mindset of even much of the 30-something crowd.

Added:
Another possible barrier to Yammer could be seen in a private network microblogging installation. Ultimately, this is Yammer's barrier to growth. However, much like AOL IM vs Office Communicator, grassroots adoption of microblogging will occur long before critical mass is achieved to provide a business justification for paying for or even just installing such software on a server and client machines.

The company e-mail validation on Yammer counteracts such software installs as well; however, a departing emploee would still have his or her access to internal company updates after departure, as long as the employee's e-mail has been validated. In Twitter, you can make updates private or block people from following you, but having to do this in Yammer would probably defeat the purpose it was created for.

Additionally:
Yammer, much like AIM, has the potential to put company information outside of the company’s hands. Of course, every user does need to sign up and verify through a corporate e-mail address. Moreover, those who have left the company can be requested for removal (one would have to wonder the validation steps here.) Nevertheless, there is no guarantee or auditing of the controls that would keep a company’s data private.

A recommended book for me to read

2008-09-05T14:14:00.001-04:00

CODE: The Hidden Language of Computer Hardware and Software

ISBN: 0-7356-0505-X; Microsoft Press; September 29, 1999; 400 pages (hardcover).
ISBN: 0-7356-1131-9; Microsoft Press; October 11, 2000; 400 pages (softcover).

The growth of Wikipedia

2008-09-04T13:44:00.000-04:00

http://en.wikipedia.org/wiki/Wikipedia:Size_of_Wikipedia

I love these graphs. That's all. An increase of 2,000 articles per day on Wikipedia. Even if you managed to read 3,000 articles per day, you'd spend 2 years just reading the current set, and in that time, Wikipedia would probably have doubled in size.

Firefox vs. Chrome

2008-09-02T16:34:00.000-04:00

I'm currently trying out the Chrome browser from Google. Very nice so far, however, pros and cons for each so far are:

Firefox:
Pros--

TwitterFox
Delicious
Ubiquity
ScribeFire

Cons--

Memory hog.

Chrome:
Pros--

Pretty (in a minimalist sort of way)
Fast
Relatively lightweight (in a 2008 application sense, not in an absolute program sense)
Special about: pages

Cons--

Looks like some Facebook functionality does not work. I think mostly involving Facebook apps.
Fails to render secure sites that are lost in the IE vs. Netscape 4.x days. Wouldn't be a huge downside if I didn't actually know of one that I use regularly.

Google browser officially to debut on Windows tomorrow

2008-09-01T18:34:00.001-04:00

Google "launches" browser with a cartoon

It would seem to be a natural progression that if you're using Google services for 90% of your web usage (Google, Maps, Gmail, Blogger, Sites, Docs, ...) then a Google browser would make things run "even smoother."

This is something that Internet Explorer has had in a way for some time. ...and Firefox has 20% without being the default install in any set up (unless some Linux distros opt for Firefox over the native WM browser). 20% of web users made the choice to use Firefox. IE 6's share is 25%, so I would venture that that 25% did not *choose* IE. In fact, 0.13 percentage points of IE traffic is pre-6.0.

Google will have a much more effective platform through which to push their browser.

Free tools article at Lifehacker

2008-09-01T14:12:00.001-04:00

Free: Free Replacements for Paid Tools

I never really considered GParted as a replacement for PartitionMagic.

Great article on life-hacker about Windows Performance tweak myths.

2008-09-01T14:06:00.001-04:00

Mythbusting: Debunking Common Windows Performance Tweaking Myths

Registry cleaners, memory optimizers, and Vista settings... oh my.

First pieces of code in PowerShell

2008-09-01T12:56:00.001-04:00

new-object:

$sw = new-object System.IO.StreamWriter("c:\ts.txt")

get-member:

$sw | get-member -memberType methods

For some reason, this method of invoking object methods is reminiscent of both Ruby and Smalltalk for me.

From the PowerShell team blog

2008-08-28T09:45:00.001-04:00

PowerShell Team Blog blog entry on the Linux Magazine

Bash vs. Vista PowerShell from Linux Magazine

PowerShell Script Center

Cannot hibernate Windows XP machine "Insuffienct Resources Exist to Complete the API"

2008-08-27T21:14:00.001-04:00

For me, this was happening on a Lenovo T61 with 2GB of RAM... and for every time that I tried to hibernate.

Microsoft Knowledge Base Article (KB909095) : The computer occasionally does not hibernate and you receive an "Insufficient System Resources Exist to Complete the API" error message in Windows XP with Service Pack 2, in Windows XP Tablet PC Edition 2005, or in Windows XP Media Center Edition 2005.

This problem typically occurs when the computer uses 1 gigabyte (GB) or more of RAM.

Hotfix for KB909095

Update 9:41PM: Hotfix worked great.

Lotus revived? On Linux? To compete with Microsoft?

2008-08-26T21:03:00.001-04:00

IBM targets Microsoft with desktop Linux initiative

This really made me do a double-take:

IBM announced a new partnership with Red Hat, Novell, and Canonical to offer "Microsoft-free" personal computers with IBM's Lotus Notes and Lotus Symphony software.

Lotus... as in Lotus 1-2-3 vs. Quattro Pro... partnering with Canonical... as in Ubuntu Canonical. The mind reels. Over and over and over...

Massive local mashup of news, events, flickr, Craigslist, Twitter, etc...

2008-08-26T20:47:00.001-04:00

Alas, my town (Louisville, KY) isn't one of the city choice at the moment, but this could be the next evolution of web 2.0, social networking, mashups, however many buzzwords you can rattle off.

Fwix | Local

This is definitely something that I want to keep an eye on. There's serious potential here.

ScribeFire

2008-08-26T08:27:00.000-04:00

I'm testing out ScribeFire, after finding out that the blogger bar for Firefox does not work in Linux. So far, this has been a pretty exciting experience. The setup wizard works great with Blogger... all you need is your blog URL and Blogger login to set up.

Will I ever use the Blogger.com interface for posting again?

Some promising links on growing an internal wiki.

2008-08-19T17:25:00.000-04:00

From the site, Grow Your Wiki.

21 Days of Wiki Adoption

Top 10 Organizational Wiki Tips (and how to use them)

Of particular note from my experience is the following article:
Why
wikis should replace email for collaboration

the bigger issue here is that organizations have to look at how email is being used, and what activities would be better served by other tools, like wikis and blogs. Email has become a crutch in business communications because it’s being stretched far beyond its intended use as a communications tool. Organizations are trying to use it to collaborate and it was never designed for this, so it makes a very poor collaboration tool.

It always seems like formal documentation as a whole is abandoned in collaboration mode for a furious chain of e-mails. I would prefer collaborative documentation to not be embedding in one of 10,000 e-mails archived somewhere. In place of thread progression, how about version control in the wiki?

Part 4 in a 4-part series: Email overload killed the businessperson

Password protecting a website

2008-08-19T10:30:00.001-04:00

Password protecting a website using .htaccess - this method pretty much requires shell access, but most web hosts have a web control panel interface to do this work. It should be noted that this isn't the most fool-proof method of protecting your data, and a configuration error can inadvertently open up your directory to the world.

An example for ICDSoft:

The option for this is called "Protect" on the Control Panel.

You can browse to a directory through this interface, and click "Protect It!" next to a specific directory.

From this interface, you can add users for the directory:

symfony | Web PHP Framework

2008-08-18T22:16:00.001-04:00

Symfony an Open-Source PHP Web Framework. This would be a nice tool to use to set up a web site in.

Maybe I can finally rebuild my accompanists.org website.

Ubiquity for Mozilla web browser

2008-08-18T20:13:00.000-04:00

Mozilla labs Ubiquity -- based on the Javascript programming language -- will allow users to type commands in an input box to launch a specific action, such as sending an e-mail.

Ubiquitous Interfaces, Ubiquitous Functionality

Dutch National Supercomputer Huygens Beats Human Go Professional

2008-08-18T20:04:00.000-04:00

I never realized that Go was more complex than chess in terms of number of potential moves throughout the game.

From the Wikipedia article on Go:

The area of the board is very large (more than five times the size of a chess board). Throughout most of the game, the number of legal moves stays at around 150–250 per turn, and rarely goes below 50 (compare chess, where the average number of moves is 37). Because an exhaustive computer program for Go must calculate and compare every possible legal move in each ply (player turn), its ability to work out favorable lines of play is sharply reduced when there are a large number of possible moves.

Indispensable and portable free stuff

2008-08-15T08:49:00.000-04:00

I was listening listening to one of the old Geek Brief podcasts, and was reminded of a website with portable versions of applications that are very useful when supporting computers that aren't your own, or running from several different PCs.

Of particular note are the Development apps available.

Notepad++ is an excellent text editor based on the SciTE editor that comes with Ruby installations. It's not quite my old standby gvim, but then again, it doesn't need much customization to do its thing either.

Also included in the section is XAMPP, which includes Apache, mySQL, PHP and Perl. In addition, the FULL version package includes valuable extras like phpMyAdmin. I haven't experimented with either the lite or full package, but I will be reloading a PC soon, so I'll see then how much more smoothly this package makes setup for me.

SmartFTP is no longer free

2008-08-01T08:52:00.000-04:00

Considering my web hosts do not provide shell access, a solid FTP client was invaluable for me in making updates and changing websites, since windows "web-based" ftp client is less useful than their console client for me (web-based does not work consistently with directories above user home).

I thought I had found a solid FTP client in SmartFTP. Kent J. Chen's weblog burst my bubble. However, he did offer up an alternative product, FileZilla. Downloading it now.

Unable to hibernate in vista?

2008-08-01T08:34:00.000-04:00

I don't have Vista yet, but I've heard complaints of Hibernate not existing/working in Vista.

Hibernate not available in Vista or stopped working - from the article:

... The disk cleanup utility wiped out the file Vista uses to store a duplicate of the contents of memory when hibernating. Apparently, if that file is not present, the power features of Windows assumes hibernation is not available. Fortunately, the remedy is simple. On Windows Vista, go to Start | All Programs | Accessories and right click on the Command Prompt icon. From the menu, choose Run As Administrator. Then, click OK on the UAC prompt so the command prompt will appear with full Admin rights. Keep in mind that we assume you are logged in with an account with limited admin rights to begin with, you just need to be elevated once UAC verifies what’s going on. If you are logged in as a regular user, log out and back in as a user that is a member of the Administrator’s Group or use the built-in Administrator account. Next, use the powercfg command to re-enable hibernation by typing the following command:

powercfg /hibernate on

Go back to the power settings in Vista and you’ll notice that hibernate is now an option (if it still isn’t, try a quick reboot).

This time, when I closed the lid of my laptop, it hibernated.

Have I appropriately professed my love for Ruby yet?

2008-07-28T14:38:00.000-04:00

I used to play around with Ruby a lot. Here's another script, where I was downloading fedora ISOs and giving myself feedback as to the progress of the operation (yes, I was in cygwin also).


#!/usr/bin/ruby
require 'net/ftp'
require 'fileutils'
ftp=Net::FTP.new("ftp.linux.ncsu.edu")
ftp.login("anonymous", "me@email.com")
FileUtils.chdir("/cygdrive/c/downloads/fedora")
files=ftp.chdir("pub/fedora/linux/core/3/i386/iso")

def mecallback()
    print 'filename: ', (File.size($fileName) * 100.0 / $fileSize), "%\n"
end

$fileName="FC3-i386-disc3.iso"
$fileSize=ftp.size($fileName)
print 'filename: ', $fileName, "\n"
ftp.getbinaryfile($fileName, $fileName, 2**20) {
    print 'filename: ', (File.size($fileName) * 100.0 / $fileSize), "%\n"
}

ftp.close

Ruby is so elegant

2008-07-28T14:16:00.000-04:00

This is a script to log in to a server via Telnet and run a single script remotely.


require 'net/telnet.rb'

myserver.com = Net::Telnet::new("Host" => "myserver.com",
  "Timeout" => 30,
  "Prompt" => /[$%#] \z/n)

myserver.com.login("username", "password") { |c| print c }
myserver.com.close
myserver.com.cmd("./check_missing.sh") { |c| print c }

Regular expressions reference

2008-07-27T15:56:00.000-04:00

URI Regular expressions

E-mail regular expressions

.htaccess redirect guide

2008-07-26T21:51:00.000-04:00

My favorite tip:

Changed file extension?

This example is perfect if you've decided to switch to .php from .html pages. It will look for any .html page and redirect it to .php (ie http://www.example.com/yourpage.html and redirect it to http://www.example.com/yourpage.php). Now, be careful with this, it does mean any html page. I did this on one of my sites and had totally forgotten I had an iframe with .html content on some pages... I didn't notice for weeks that it was broken :S.
So learn from my mistake ;-) check, double check, then check again.
RedirectMatch 301 (.*)\.html$ http://www.example.com$1.php

Rip page to new window in firefox

2008-07-22T14:13:00.000-04:00

On the bookmark toolbar, right-click and select "New Bookmark..."

Under name, give your bookmark a fitting title, like, "Rip page."
Under location, enter the following code:

javascript:var x=window.open(document.URL, '_blank','width=800,height=600')

Or... just drag the following link to your bookmark toolbar.
Rip page.

Update on August 5, 2008:
I found the following works well for ripping a Google Docs page to a new window (former window closes as well.)

javascript:var%20x=window.open(document.URL,%20'_blank','width=800,height=600
,status=1,toolbar=1,location=1');window.close();

Rip page.

Optimized prime number generator

2008-07-20T17:34:00.000-04:00

Same specs as previous system, 8 hours to find all primes to ULONG_MAX.


#include <stdio.h>
#include <malloc.h>
#include <limits.h>
#include <math.h>

struct linked_list
{
    unsigned long number;
    struct linked_list *next;
};

int main(int argc, char **argv)
{
    unsigned long divisor, max_divisor, prime_test;
    unsigned long max_prime_to_store;
    struct linked_list *first=NULL, *current=NULL, *last=NULL;

    first=(struct linked_list *)malloc(sizeof(struct linked_list));
    first->number = 2L;
    first->next = NULL;
    last = first;
    printf("%lu\n", 2L);

    max_prime_to_store = 1L + sqrt(ULONG_MAX);

    for(prime_test = 3L; prime_test < ULONG_MAX; prime_test+=2L)
    {
        max_divisor=1L+sqrt(prime_test);
        for(current = first; current && current->number <= max_divisor; current = current->next)
        {
            if(prime_test % current->number == 0)
            {
                break;
            }
        }
        if(!current || current->number > max_divisor)
        {
            if(prime_test <= max_prime_to_store)
            {
                last->next = (struct linked_list *)malloc(sizeof(struct linked_list));
                last = last->next;
                last->next = NULL;
                last->number = prime_test;
            }
            printf("%lu\n", prime_test);
        }
    }
    return 0;
}

I miss C programming. Program to find prime numbers

2008-07-18T09:23:00.000-04:00

Uses unsigned long long, goes up to ULONG_LONG_MAX.

Intel Core Duo 2 processor, running Cygwin on Windows XP, and compiled using gcc 3.4.4. Stops at first million results due to time. Can reasonably get to 10 million, if you're willing to wait.


#include <stdio.h>
#include <math.h>
#include <limits.h>

int main(int argc, char **argv)
{
    unsigned long long divisor, max_divisor, prime_test;
    int break_time = 0;

    printf("%llu\n", 2LL);
    for(prime_test = 3LL; prime_test < ULONG_LONG_MAX; prime_test+=2LL)
    {
        max_divisor=1LL+sqrt(prime_test);
        for(divisor=3LL;divisor <= max_divisor; divisor+=2LL)
        {
            if(prime_test % divisor == 0)
            {
                break;
            }
        }
        if(divisor >= max_divisor)
        {
            printf("%llu\n", prime_test);
        }
        else
        {
            continue;
        }
        break_time++;
        if(break_time == 999999)
        {
            break;
        }
    }
    return 0;
}

VBScript to push files into an array based on a wildcard

2008-07-10T10:56:00.000-04:00


Option Explicit
Dim gFSO
Dim dirListing
Dim fileName

set gFSO = CreateObject("Scripting.FileSystemObject")

dirListing = ListDir("c:\??*d*")

If UBound(dirListing) = -1 then
    Wscript.Echo "No files found."
Else
    For Each fileName in dirListing
        WScript.Echo FileName
    Next
End If

'==============================================================================
' List a directory, with the last part of the directory being the path
'==============================================================================
Function ListDir (ByVal Path)
    Dim fileRegex
    Dim searchFolderName
    Dim searchFolder
    Dim searchFolderFiles
    Dim fileName
    Dim fileArray
    Dim file

    set fileRegex = New RegExp
    If gFSO.FolderExists(Path) Then ' Path is a directory, list all files in path 
        searchFolder = Path
        fileName = ""
    Else
        searchFolderName = gFSO.GetParentFolderName(Path)
        fileName = gFSO.GetFileName(Path)
        fileRegex.Global = True
        fileRegex.Pattern = "\."
        fileName = fileRegex.Replace(fileName, "\.")
        fileRegex.Pattern = "\?"
        fileName = fileRegex.Replace(fileName, ".")
        fileRegex.Pattern = "\*"
        fileName = fileRegex.Replace(fileName, ".*")
    End If
    With fileRegex
        .Pattern = fileName
        .IgnoreCase = True
        .Global = False
    End With

    ReDim fileArray(1)
    Dim fileCount : fileCount = 0

    Set searchFolder = gFSO.GetFolder(searchFolderName)
    Set searchFolderFiles = searchFolder.Files

    For Each file in searchFolderFiles
        If fileRegex.Test(file.Name) Then
            If fileCount > UBound(fileArray) Then
                ReDim Preserve fileArray(fileCount*2)
            End If
            fileArray(fileCount) = file.Path
            fileCount = fileCount + 1
        End If
    Next
    ReDim Preserve fileArray(fileCount - 1)
    ListDir = fileArray
End Function

Automatic list of labels for classic templates

2008-06-18T09:30:00.001-04:00

VBScript to convert timestamp to YYYYMMDD formatted string

2008-06-17T09:32:00.000-04:00

I started with the code on this page.

Function FormatYYYYMMDD(timeStamp)

    Dim dateMonth : dateMonth = DatePart("M", timeStamp)
    Dim dateDay : dateDay = DatePart("D", timeStamp)
    Dim dateYear : dateYear = DatePart("YYYY", timeStamp)
    Dim dateString

    dateString = dateYear

    If dateMonth < 10 Then
        dateString = dateString & "0" & dateMonth
    Else
        dateString = dateString & dateMonth
    End If

    If dateDay < 10 Then
        dateString = dateString & "0" & dateDay
    Else
        dateString = dateString & dateDay
    End If

    FormatYYYYMMDD = dateString
End Function

Microsoft Windows startup Registry Settings

2008-05-02T09:24:00.000-04:00

INFO: Run, RunOnce, RunServices, RunServicesOnce and Startup:

I believe this is pretty much a bulk of the information set by Xteq X-Setup. For those who don't want to make the purchase for a very simple edit, the following is the except from the Microsoft KB article

General order of loading:

HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\RunServicesOnce
HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\RunServices
<Logon Prompt>
HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\RunOnce Not supported in NT 3.51 and ignored by XP and 2000 in safe mode
HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\Run
HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Run
StartUp Folder
HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\RunOnce Not supported in NT 3.51 and ignored by XP and 2000 in safe mode

However,

With the exception of the HKEY_LOCAL_MACHINE\...\RunOnce key, all keys and their entries are loaded asynchronously. Therefore, all entries in the RunServices and RunServicesOnce keys can potentially run at the same time.

Entries in the HKEY_LOCAL_MACHINE\...\RunOnce key are loaded synchronously in an undefined order.

Because the HKEY_LOCAL_MACHINE\...\RunOnce key is loaded synchronously, all of its entries must finish loading before the HKEY_LOCAL_MACHINE\...\Run, HKEY_CURRENT_USER\...\Run, HKEY_CURRENT_USER\...\RunOnce, and Startup Folder entries can be loaded.

WMIC service cheat sheet.

2008-04-11T10:16:00.000-04:00

Microsoft's documentation is thorough, but I got a lot more out of the above link.

Time since last boot (Windows, VBScript)

2008-04-10T16:43:00.001-04:00

I don't remember where I found this, I was just happy to find code that did this.

 strComputer = "." ' Local computer 
set objWMIDateTime = CreateObject("WbemScripting.SWbemDateTime")
set objWMI = GetObject("winmgmts:\\" & strComputer & "\root\cimv2")
set colOS = objWMI.InstancesOf("Win32_OperatingSystem")
for each objOS in colOS
    objWMIDateTime.Value = objOS.LastBootUpTime
    Wscript.Echo "Last Boot Up Time: " & objWMIDateTime.GetVarDate & vbcrlf & _
        "Time Since Last Boot: " &  TimeSpan(objWMIDateTime.GetVarDate,Now) & _
        " (hh:mm:ss)"
next

Function TimeSpan(dt1, dt2)
    ' Function to display the difference between
    ' 2 dates in hh:mm:ss format
    If (isDate(dt1) And IsDate(dt2)) = false Then
        TimeSpan = "00:00:00"
        Exit Function
        End If

        seconds = Abs(DateDiff("S", dt1, dt2))
        minutes = seconds \ 60
        hours = minutes \ 60
        minutes = minutes mod 60
        seconds = seconds mod 60

        if len(hours) = 1 then hours = "0" & hours

        TimeSpan = hours & ":" & _
            RIGHT("00" & minutes, 2) & ":" & _
            RIGHT("00" & seconds, 2)
End Function

Getting size estimates for tables in SQL SERVER 2000

2008-03-27T17:03:00.001-04:00


-- SQL Server tables and row counts

declare tables_cursor cursor for 
select name from sysobjects where xtype='U'
declare @table_name sysname

open tables_cursor
fetch next from tables_cursor INTO @table_name

while @@fetch_status = 0
begin
    exec sp_spaceused @table_name
    fetch next from tables_cursor INTO @table_name
end
close tables_cursor
deallocate tables_cursor

Restoredb.bat

2008-03-25T10:56:00.000-04:00

Using osql and trusted connection for SQL Server 2000.


@echo off

if {%2}=={} (
   call :Usage
   goto :EOF
)
net stop mssqlserver
net start mssqlserver
osql -E -d master -Q "restore database 
%1 from disk='%2' with replace"

goto :EOF
::Subroutine area 

:Usage
   @echo Syntax: call restoredb.bat DatabaseName
FullDatabaseBackupPath

goto :EOF

"Playtime" project: Porting Didiwiki to C#.NET

2008-03-07T20:41:00.000-05:00

I'm hoping to find time to throw at least a quick port of DidiWiki to C#.NET together.

Reference 1: Creating your own Web Server using C# - This article is from 2001, so there might be a significantly better way of doing things.

Update: This CodeGuru page looks like a more original version of the same page.

Reference 2: Didiwiki Source Code that I've saved off. Original maintainer is no longer maintaining.

Importing a runningahead log.xml into SQL Server 2005 using OPENXML

2008-03-01T11:56:00.000-05:00

I don't currently have a work-based use for 2005, so I thought I'd use the only readily available collection of data that I have at my disposal... my running log. This is an export file from RunningAhead.com. I don't think I'm completely loading the XML data yet, and I don't have database tables to load to yet, either. However, this is my implementation of some OPENXML examples.

DECLARE @hdoc int
DECLARE @xmlDoc XML
SET @xmlDoc = (
  SELECT * FROM OPENROWSET (
    BULK 'C:\projects\runningahead\log.1.321C02\log.xml', SINGLE_CLOB
  ) AS xmlData
)
SELECT @xmlDoc

EXEC sp_xml_preparedocument @hdoc OUTPUT, @xmlDoc

SELECT * FROM
OPENXML(@hdoc, '/RunningAHEADLog/CourseCollection/Course', 2)
WITH
(
        ID varchar(32),
        Name varchar(255) ,
        Surface varchar(255),
        City varchar(255),
        State char(3),
        Notes varchar(255)
)

SELECT * FROM
OPENXML(@hdoc, '/RunningAHEADLog/EquipmentCollection/Shoe', 2)
WITH
(

        ID varchar(32),
        Make varchar(255) ,
        Model varchar(255),
        Serial char(20),
        InitDistance decimal(7,2),
        PurchaseDate datetime,
        Retired char(3),
        Size char(10),
        SizeSystem char(40)
)

SELECT * FROM
OPENXML(@hdoc, '/RunningAHEADLog/EventCollection/Run', 2)
WITH
(
        ID char(32),
        Date datetime,
        Time datetime,
        Type char(15),
        Weight decimal(4,1),
        WeightUnit char(12),
        AvgHR tinyint,
        MaxHR tinyint,
        Quality tinyint,
        Effort tinyint,
        CourseID char(32),
        Distance decimal(8,2),
        DistanceUnit char(12),
        Duration char(20)
)

EXEC sp_xml_removedocument @hdoc

Sql Date Time formats.

2008-02-22T13:31:00.000-05:00

Fromhttp://lists.evolt.org/archive/Week-of-Mon-20020617/115646.html

without century (yy)	with century (yyyy)	standard	outputformat
-	0 or 100 (*)	default	mon dd yyyy hh:miAM (or PM)
1	101	USA	mm/dd/yy
2	102	ANSI	yy.mm.dd
3	103	British/French	dd/mm/yy
4	104	German	dd.mm.yy
5	105	Italian	dd-mm-yy
6	106	-	dd mon yy
7	107	-	mon dd, yy
8	108	-	hh:mm:ss
-	9 or 109 (*)	default milliseconds	mon dd yyyy hh:mi:ss:mmmAM (or PM)
10	110	USA	mm-dd-yy
11	111	Japan	yy/mm/dd
12	112	ISO	yymmdd
-	13 or 113 (*)	Europe default milliseconds	dd mon yyyy hh:mm:ss:mmm(24h)
14	114	-	hh:mi:ss:mmm(24h)
20	120	-	yyyy-mm-dd hh:mi:ss(24h)
21	121	-	yyyy-mm-dd hh:mi:ss.mmm(24h)
-	126	-	yyyy-mm-dd Thh:mm:ss.mmm(no spaces)
-	130	-	dd mon yyyy hh:mi:ss:mmmAM
-	131	-	dd/mm/yy hh:mi:ss:mmmAM

(*) the default values (style 0 or 100, 9 or 109, and
13 or 113) always return the century (yyyy)

Run Command Most Recently Used List Storage

2008-02-15T09:22:00.000-05:00

At least for Windows XP, this list is stored in the registry:

My Computer\HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer\RunMRU

It looks like there are values \a through \z off of RunMRU, plus \(Default) and \MRUList. There is string value in MRUList: e.g., "nkvlsawbhczeyoijfugrtdxpmq", which determines the ordering of MRUList entries.

Miscellaneous Excel tricks that I needed today.

2008-02-11T13:29:00.000-05:00

How to hide gridlines on an Excel Worksheet.

Select a Worksheet.
Select Tools->Options...
On the [View] tab, un-check Gridlines under the Window Options section.

Row and column headers, scroll bars, and sheet tabs are also hideable through this tab.

How to Split and Freeze Panes

Freeze panes makes very nice title rows (the bottom pane does not scroll into the title row).

ILASM and ILDASM (.NET "Assembler")

2008-02-04T15:29:00.000-05:00

Reverse Engineering To Learn .NET Better

Accessible via command line using:
Start->Microsoft Visual Studio 2005->Visual Studio Tools->Visual Studio 2005 Command Prompt

C vs C# formatting.

2008-01-31T15:11:00.000-05:00


printf("%05d\n",123);


Console.WriteLine(String.Format("{0:00000}", 123))

Calculate average daily mileage required to make yearly goal

2007-09-11T11:23:00.000-04:00

This might be useful if you're programming a running site. For me, however, I wanted to write this calculation in SQL Server to expand my knowledge some.


--Calculates the average daily mileage needed to meet a target yearly mileage goal, based on current mileage.
CREATE FUNCTION averageDistanceNeeded (@targetMileage decimal(10,4), @currentMileage decimal(10,4))
RETURNS decimal(10,4)
AS
BEGIN
    Declare @daysLeft decimal(10,4)
    Declare @dt datetime

    SELECT @dt = dt from openquery
    (
        "my-server",
        'select dt = getdate()'
    )
    select @daysLeft = datediff(day, @dt, CAST (CAST (YEAR(@dt) AS VARCHAR(4)) + '-12-31' AS DATETIME))
    return CONVERT(decimal(10,4), (@targetMileage - @currentMileage) / @daysLeft)
END
GO

--Get average daily mileage needed to reach 2000 miles, based on a current mileage
-- of 960.2
select dbo.averageDistanceNeeded(2000.0, 960.2)

Further refined version of code searching

2007-09-06T10:53:00.001-04:00

I have now included views in the search code version. Both functions now return the object type the result was found in.


--Searches stored procedures, functions and views for code
-- containing the stringtosearch
CREATE FUNCTION findTextInCode (@StringToSearch varchar(100))
RETURNS @fNames TABLE (search varchar(100), name sysname, type char(2))
AS
BEGIN
   SET @StringToSearch = '%' + @StringToSearch +'%'
   INSERT @fNames
   SELECT DISTINCT @stringtosearch, SO.Name, SO.Type
   FROM sysobjects SO (NOLOCK)
   INNER JOIN syscomments SC (NOLOCK) on SO.ID = SC.ID
   AND (SO.Type = 'P' OR SO.Type LIKE '%F%' OR SO.Type = 'V')
   AND SC.Text LIKE @stringtosearch
   ORDER BY SO.Name
   RETURN
END
GO


--Searches column names from stored procedure parameters and function,
-- user tables, and view columns
CREATE FUNCTION findTextInColumnName (@StringToSearch varchar(100))
RETURNS @fColumnNames TABLE (tablename sysname, colname sysname, type char(2))
AS
BEGIN
   SET @StringToSEarch = '%' + @StringToSearch + '%'
   INSERT @fColumnNames
   SELECT DISTINCT SO.NAME, SC.NAME, SO.TYPE
   FROM syscolumns SC (NOLOCK)
INNER JOIN
   sysobjects SO (NOLOCK)
   ON SC.id = SO.id
   WHERE SC.NAME LIKE @StringToSearch
   ORDER BY SC.Name
   RETURN
END

Further refined Find Text in Column Name

2007-08-16T14:06:00.000-04:00

Now returns table (or function which returns a table) and column name.


CREATE FUNCTION Find_Text_In_Column_Name (@StringToSearch varchar(100))
RETURNS @fColumnNames TABLE (tablename sysname, colname sysname)
AS
BEGIN
   SET @StringToSEarch = '%' + @StringToSearch + '%'
   INSERT @fColumnNames
   SELECT DISTINCT SO.NAME, SC.NAME
   FROM syscolumns SC (NOLOCK)
INNER JOIN
   sysobjects SO (NOLOCK)
   ON SC.id = SO.id
   WHERE SC.NAME LIKE @StringToSearch
   ORDER BY SC.Name
   RETURN
END

Functions to find text in column names and in stored procedures and function definitions

2007-08-16T10:49:00.000-04:00

Implemented in SQL Server:
This function finds a text string in a stored procedure or function definition:


CREATE FUNCTION Find_Text_In_SP_or_FN (@StringToSearch varchar(100))
RETURNS @fNames TABLE (name sysname)
AS
BEGIN
   SET @StringToSearch = '%' + @StringToSearch +'%'
   INSERT @fNames
   SELECT DISTINCT SO.Name
   FROM sysobjects SO (NOLOCK)
   INNER JOIN syscomments SC (NOLOCK) on SO.ID = SC.ID
   AND (SO.Type = 'P' OR SO.Type LIKE '%F%')
   AND SC.Text LIKE @stringtosearch
   ORDER BY SO.Name
   RETURN
END
GO

This function finds a text string in a column name:


CREATE FUNCTION Find_Text_In_Column_Name (@StringToSearch varchar(100))
RETURNS @fColumnNames TABLE (colname sysname)
AS
BEGIN
   SET @StringToSEarch = '%' + @StringToSearch + '%'
   INSERT @fColumnNames
   SELECT DISTINCT SC.NAME
   FROM syscolumns SC (NOLOCK)
   WHERE SC.NAME LIKE @StringToSearch
   ORDER BY SC.Name
   RETURN
END
GO

Some example usage for the novices:


select * from Find_Text_In_SP_or_FN('addr')
GO
Select * from Find_Text_In_Column_Name('city')
GO
Select * from Find_Text_In_Column_Name('state')
GO

Google code search for "I'm an idiot"

2007-08-15T16:00:00.000-04:00

My new favorite toy: Google code search

Some interesting comments and such left in code: I'm an idiot.

Why the hell

wtf

Some idiot

What an idiot

Google code search even does regex. How about a string of random ? and !?

Windows XP Keyboard Navigation

2007-07-19T14:26:00.000-04:00

In Explorer:

"<ALT>VOU" - Go up one level (View, Go To, Up one level

Stored procedure to search functions and stored procedures.

2007-07-12T10:09:00.000-04:00


-- Altered slightly from http://www.knowdotnet.com/articles/storedprocfinds.html
CREATE PROCEDURE Find_Text_In_SP_or_FN (@StringToSearch varchar(100))
AS
   SET @StringToSearch = '%' + @StringToSearch +'%'
   SELECT DISTINCT SO.Name
   FROM sysobjects SO (NOLOCK)
   INNER JOIN syscomments SC (NOLOCK) on SO.ID = SC.ID
   AND (SO.Type = 'P' OR SO.Type LIKE '%F%')
   AND SC.Text LIKE @stringtosearch
   ORDER BY SO.Name
GO

Turn pseudo-enums expressed as #defines into strings.

2007-06-11T00:29:00.000-04:00

I have an include file that has 50+ #defines with constants (0,1,2,3,...,etc). I have a dump program to dump the data structure that these #defines are used in into a .csv file. I used this little script to dump case statements out for the #defines, with readable strings instead of arbitrary constant values.


cat $1 | awk '/#define/{
   print "\t\tcase " $2 ":";
   print "\t\t\tfprintf(out_csv_fp,\"," $2 "\");";
   print "\t\t\tbreak;";
}' > out.txt

Awk script to quote all parameters for a given function name.

2007-06-11T00:14:00.000-04:00

I was trying to maintain a program that wrote 200+ lines to a single file, using a single repeated function call--many of which had all NULL parameters. There were also occasional loops involved. The data was position based, so NULL writes counted.

I was just trying to figure out:

How many times a given function was called.
At what position actual data occurred.

I created a stub function of the original offending function that logged to a file what # call this was and the data passed. Then I copied the offending functions' caller and substituted offending function calls using this script, which quotes all passed parameters (note: already quoted parameters would need a little more work):


BEGIN {
  OFS=","
}
/repeated_function_name/{

  for(i=1; i<=NF; i++)
  {
      if(i==1)
      {
          sub("[(]", "(\"", $i);
      }
      else
      {
          sub("^[ ]*","\"", $i);
      }
      if(i==NF)
      {
          sub("[)]", "\")", $i);
      }
      else
      {
          sub("[ ]*$", "\"", $i);
      }
  }

  print $0
  next;
}
{ print $0; }

Awk to split c functions into multiple files.

2007-05-29T16:19:00.000-04:00

This, at present, is overly simplistic and makes many stupid assumptions:

That the return type is short.
That there are no short global variables declared.
That everything is well-formatted so that the return type is always first column, and that there is a space between both function name and return type (as in, only spaces--no comments) and function name and open parenthesis of the arg list.


cat $1 | awk '
BEGIN {
    functionNameFile="";
}
/^short/ {
    /* assumes return type of short */
    /* assumes space between return type, function name, and parm list */
    functionNameFile=$2.".c";
    print $0 > functionNameFile;
}
/^\}/ {
    print $0 >> functionNameFile;
    functionNameFile="";
    next;
}
{
    if(functionNameFile    != "") {
        print $0 >> functionNameFile;
    }
}
'

Windows Vista problems resticting GNU GCC apps to 32MB

2007-05-29T12:52:00.000-04:00

Disturbing to see this come about:
Windows Vista problems resticting GNU GCC apps to 32MB

Buried in the footnotes is a note about a response supposedly coming from a Microsoft representative saying, "now that people are asking for this, we are planning to fix..."

(g)awk script to remove last dir from PATH

2007-04-26T09:15:00.000-04:00

This script chops off the last dir off of the PATH (still need to assign back to PATH)


echo $PATH | gawk -F':' '{
    for(i=1; i<NF; i++) {
        if(i) {
            printf (":");
        } 
        printf("%s", $i);
    }
}'

Writing an onLoad function that includes a delay.

2007-04-17T09:59:00.000-04:00

This code was written out of desperation when I could not use an embedded APPLET's isActive() function to check if the APPLET was loaded. Instead, I waited for 500 milliseconds. Hopefully, there are better uses for this code.


<SCRIPT>

function OnLoadFunction(milliseconds_time)
{
 setTimeout("delayedOnLoadFunction()", milliseconds_time);
}

function delayedOnLoadFunction()
{
 /* Function body */
}
</SCRIPT>

<body onLoad="OnLoadFunction(500)">

Insert archive parameter into HTML applet tag.

2007-03-29T09:12:00.000-04:00

Under Linux:
This code, named replace_applet.sh, adds a archive parameter (classes.jar) to all applet tags for *.htm files in a directory.


ls *.htm | while read x 
do
 cat $x | sed -e '
/<APPLET/ {
 s/<APPLET /<APPLET archive="classes.jar"/
}' > $x.$$
 cp $x.$$ $x
 rm $x.$$
done

C# tutorials, Mono and .NET - Hello World

2007-03-15T09:52:00.000-04:00

GtkSharp: Hello World
Hello World in Different Styles - mainly console output.
The Hello World Application Windows Forms example from C# and the .NET Framework

Java: Preventing a child window's close operation from exiting the app.

2007-02-16T22:23:00.000-05:00

I included a second JFrame in NetBeans, which I used as a child window. When the child window was closed, all windows were closed. I finally figured out that the default close operation needed to be either HIDE_ON_CLOSE or DISPOSE_ON_CLOSE. The default in Java is HIDE_ON_CLOSE, but in NetBeans is EXIT_ON_CLOSE.

I first changed this property in the main window code that invokes the child window, but then I looked at the Properties page for the child window. "defaultCloseOperation" is the first property. "DISPOSE" can be selected here.

From Sun Java Documentation

setDefaultCloseOperation

public void setDefaultCloseOperation(int operation)

Sets the operation that will happen by default when the user initiates a "close" on this frame. You must specify one of the following choices:

DO_NOTHING_ON_CLOSE (defined in WindowConstants): Don't do anything; require the program to handle the operation in the windowClosing method of a registered WindowListener object.
HIDE_ON_CLOSE (defined in WindowConstants): Automatically hide the frame after invoking any registered WindowListener objects.
DISPOSE_ON_CLOSE (defined in WindowConstants): Automatically hide and dispose the frame after invoking any registered WindowListener objects.
EXIT_ON_CLOSE (defined in JFrame): Exit the application using the System exit method. Use this only in applications.

The value is set to HIDE_ON_CLOSE by default.

Link to Random Image Password Generation

2007-02-14T22:13:00.000-05:00

Captchas.net

Reducing redundant code by wrapping JNI functions.

2007-02-10T18:26:00.000-05:00

My HelloWorldNative.c, with the "nativePrintObject" function reduced by wrapping the object and integer field retrieval:


JNIEXPORT void JNICALL Java_helloworld_Main_nativePrintObject(JNIEnv *env, jobject ths, jobject wsv)
{
 jint ipAddress;
 jstring jPageVisited;
 char *pageVisited;
 jstring jRefererURL;
 char *refererURL;
 jobject ipObject;

 ipAddress = getIntegerField(env, wsv, "ipAddress");
 printf("  wsv.ipAddress = %d.%d.%d.%d\n",
     ( ipAddress & 0xff000000 ) >> 24,
     ( ipAddress &amp; 0x00ff0000 ) >> 16,
     ( ipAddress & 0x0000ff00 ) >> 8,
     ipAddress &amp;amp;amp; 0x000000ff);

 if(!(jPageVisited = getStringField(env, wsv, "pageVisited"))) {
     return;
 }

 pageVisited = (*env)->GetStringUTFChars(env, jPageVisited, 0);

 if(!pageVisited) {
     return;
 }
 printf("  wsv.pageVisted = %s\n", pageVisited);
 (*env)->ReleaseStringUTFChars(env, jPageVisited, pageVisited);

 if(!(jRefererURL = getStringField(env, wsv, "refererURL"))) {
     return;
 }

 if(!(refererURL = (*env)->GetStringUTFChars(env, jRefererURL, 0))) {
     return;
 }
 printf("  wsv.refererURL = %s\n", refererURL);
 (*env)->ReleaseStringUTFChars(env, jRefererURL, refererURL);

 if(!(ipObject = getObjectField(env, wsv, "ipAddr", "helloworld", "IPAddress"))) {
     return;
 }

 printf("  wsv.ipAddr = %d.%d.%d.%d\n",
     getIntegerField(env, ipObject, "byte1"),
     getIntegerField(env, ipObject, "byte2"),
     getIntegerField(env, ipObject, "byte3"),
     getIntegerField(env, ipObject, "byte4")
 );
}

jniUtils.h, the header for the JNI wrapper:


/*
* File:   jniUtils.h
* Author: thomas
*
* Created on February 10, 2007, 5:30 PM
*/
#include <jni.h>

#ifndef _jniUtils_H
#define _jniUtils_H

#ifdef __cplusplus
extern "C" {
#endif
jint getIntegerField(JNIEnv *env, jobject ths, char *fieldName);
void putIntegerField(JNIEnv *env, jobject ths, char *fieldName, jint ji);
jstring getStringField(JNIEnv *env, jobject ths, char *fieldName);
jobject getObjectField(JNIEnv *env, jobject ths, char *fieldName,
  char *objectPackage, char *objectClass);

#ifdef __cplusplus
}
#endif


#endif /* _jniUtils_H */

jniUtils.c, the wrapper for the JNI C interface. Note that not all types have been implemented. Also, error checking may not be performed in all scenarios. See that the "signature" for objects is broken up into package and class name.


#include <stdlib.h>
#include <string.h>
#include "jniUtils.h"

jint getIntegerField(JNIEnv *env, jobject ths, char *fieldName)
{
   jclass c;
   jfieldID fid;
  
   c = (*env)->GetObjectClass(env, ths);
   fid = (*env)->GetFieldID(env, c, fieldName, "I");
   return (*env)->GetIntField(env, ths, fid);
}

void putIntegerField(JNIEnv *env, jobject ths, char *fieldName, jint ji)
{
   jclass c;
   jfieldID fid;
  
   c = (*env)->GetObjectClass(env, ths);
   fid = (*env)->GetFieldID(env, c, fieldName, "I");
   (*env)->SetIntField(env, ths, fid, ji);
}

jstring getStringField(JNIEnv *env, jobject ths, char *fieldName)
{
   jclass c;
   jfieldID fid;
  
   return (jstring) getObjectField(env, ths, fieldName, "java/lang", "String");
}

jobject getObjectField(JNIEnv *env, jobject ths, char *fieldName,
   char *objectPackage, char *objectClass)
{
   char *objectSignature;
   jclass c;
   jfieldID fid;
   jobject objectField = 0;
  
   objectSignature = (char *) malloc(strlen(objectPackage) + strlen(objectClass) + 10);
  
   strcpy(objectSignature, "L");
   strcat(objectSignature, objectPackage);
   strcat(objectSignature, "/");
   strcat(objectSignature, objectClass);
   strcat(objectSignature, ";");
   c = (*env)->GetObjectClass(env, ths);
   fid = (*env)->GetFieldID(env, c, fieldName, objectSignature);
  
   if(!fid) {
       return 0;
   }
   objectField = (*env)->GetObjectField(env, ths, fid);
  
   free(objectSignature);
  
   return objectField;
}

Passing an object including java Objects to a C library.

2007-02-10T16:51:00.000-05:00

Very similar to including Java Strings:
The passed object, WebSiteVisit:

package helloworld;

/**
*
* @author thomas
*/
public class WebSiteVisit {
 /** URL of the web page that linked user to this one */
 public String refererURL;
 public String pageVisited;
 public int ipAddress;
 IPAddress ipAddr;

 /** Creates a new instance of WebSiteVisit */
 public WebSiteVisit() {
 }

 public WebSiteVisit(String referer, String page, int ip) {
     this.refererURL = referer;
     this.ipAddress = ip;
     this.pageVisited = page;
     this.ipAddr = new IPAddress(ip);
 }
}

The embedded object, IPAddress:

package helloworld;

/**
*
* @author thomas
*/
public class IPAddress {
  public int byte1, byte2, byte3, byte4;
  /** Creates a new instance of IPAddress */
  public IPAddress() {
  }
  public IPAddress(int address) {
      byte1 = (address & 0xff000000) >> 24;
      byte2 = (address & 0x00ff0000) >> 16;
      byte3 = (address & 0x0000ff00) >> 8;
      byte4 =  address & 0x000000ff;
  }

}

The C source to the dynamic library, including the IPAddress object access.

JNIEXPORT void JNICALL Java_helloworld_Main_nativePrintNumber
 (JNIEnv *env, jobject obj, jint ji)
{
   jfieldID fid;
   jint version;
   jclass c;
   c = (*env)->GetObjectClass(env, obj);
   fid = (*env)->GetFieldID(env, c, "version_number", "I");
   version = (*env)->GetIntField(env, obj, fid);
   (*env)->SetIntField(env, obj, fid, ji*ji);
   printf("::: %d, %d\n", fid, version);
   hello_number(ji);
}
JNIEXPORT void JNICALL Java_helloworld_Main_nativePrintObject
 (JNIEnv *env, jobject ths, jobject wsv)
{
   jfieldID fid;
   jint ipAddress;
   jstring jPageVisited;
   char *pageVisited;
   jstring jRefererURL;
   char *refererURL;
   jclass c;
   jobject ipObject;
   jclass ipObjectClass;
   jint byte1, byte2, byte3, byte4;
  
   c = (*env)->GetObjectClass(env, wsv);

   .
   .
   .
  
   fid = (*env)->GetFieldID(env, c, "ipAddr", "Lhelloworld/IPAddress;");
   if(!fid) {
       return;
   }
   ipObject = (*env)->GetObjectField(env, wsv, fid);
   ipObjectClass = (*env)->GetObjectClass(env, ipObject);
  
   fid = (*env)->GetFieldID(env, ipObjectClass, "byte1", "I");
   byte1 = (*env)->GetIntField(env, ipObject, fid);
  
   fid = (*env)->GetFieldID(env, ipObjectClass, "byte2", "I");
   byte2 = (*env)->GetIntField(env, ipObject, fid);

   fid = (*env)->GetFieldID(env, ipObjectClass, "byte3", "I");
   byte3 = (*env)->GetIntField(env, ipObject, fid);
  
   fid = (*env)->GetFieldID(env, ipObjectClass, "byte4", "I");
   byte4 = (*env)->GetIntField(env, ipObject, fid);

   printf("wsv.ipAddr = %d.%d.%d.%d\ns", byte1, byte2, byte3, byte4);
}

Developer Notes

Made a space for my "tech notes"

Installing JSPWiki on an XAMPP for Windows install

Insight DNS issues #insight

Every time you make an Excel spreadsheet this big, a kitten meets its maker

TwitterThreads.com - what I've been looking for. Almost.

Proof that I'm not the only one with problems with the Lenovo T-61 laptop

Finding distances between two zip codes in PHP

Gmail terminal theme--for your inner nerd

I suspect that the delicious bookmark plug-in for Firefox 3 is not keeping me logged for two weeks.

Autotext lines in Word (using 2003 here)

Reading List by Amazon

99 bottles of beer and hello world programming examples

Found some Gtk/Ruby signal stuff.

I remember the good old days of domain squatting

Classic Blogger and PHP?

Is a Yahoo developer platform worth the time investment to learn?

Remember the Milk is on Twitter!

Trying to dig up IT relevant podcasts

Same old grind / a new day / gotta work / for the pay / Burma-shave

On comments and blogs

In a time when I'm aggravated at Insight, Verizon cheers me so...

See? I'm not insane...

My current Insight cable speed tests.

Follow Cost on Twitter

All these silly little Top 10 lists.

Random Articles found through NetworkWorld

Problem 1 from Project Euler . net ("set"-based solution)

A fledgling blog on the progress of technology

Problem 1 from Project Euler . net

My first powershell

For when I really can't get my brain fired up on a given day

PC World's opinions on products and websites.

Cleaning up a brand new PC

Send Tweets to Twitter via e-mail

Google Calculator cheat sheet

Way to go, Beshear.

The trouble with Twitter

My thoughts on Yammer

A recommended book for me to read

The growth of Wikipedia

Firefox vs. Chrome

Google browser officially to debut on Windows tomorrow

Free tools article at Lifehacker

Great article on life-hacker about Windows Performance tweak myths.

First pieces of code in PowerShell

From the PowerShell team blog

Cannot hibernate Windows XP machine "Insuffienct Resources Exist to Complete the API"

Lotus revived? On Linux? To compete with Microsoft?

Massive local mashup of news, events, flickr, Craigslist, Twitter, etc...

ScribeFire

Some promising links on growing an internal wiki.

Password protecting a website

symfony | Web PHP Framework

Ubiquity for Mozilla web browser

Dutch National Supercomputer Huygens Beats Human Go Professional

Indispensable and portable free stuff

SmartFTP is no longer free

Unable to hibernate in vista?

Have I appropriately professed my love for Ruby yet?

Ruby is so elegant

Regular expressions reference

.htaccess redirect guide

Rip page to new window in firefox

Optimized prime number generator

I miss C programming. Program to find prime numbers

VBScript to push files into an array based on a wildcard

Tag Cloud for ftp blogger blogs

Automatic list of labels for classic templates

VBScript to convert timestamp to YYYYMMDD formatted string

Microsoft Windows startup Registry Settings

WMIC service cheat sheet.

Time since last boot (Windows, VBScript)

Getting size estimates for tables in SQL SERVER 2000

Restoredb.bat

"Playtime" project: Porting Didiwiki to C#.NET

Importing a runningahead log.xml into SQL Server 2005 using OPENXML

Sql Date Time formats.

Run Command Most Recently Used List Storage

Miscellaneous Excel tricks that I needed today.