we have hundreds of 90% new dell,HP,IBM servers and HP,lenovo,dell,samsung,fujitsu,toshiba,sony,Acer,Asus laptops imported from USA, stored in our Hongkong warehouse.
Welcome to contact us!
http://www.net-silkroad.com
Name: Bruce Chan
Tel: +86-13480134956
Fax: +86-13714409176
Mobile: +86-13823510166
E-mail: 13823510166@139.com
Add: Building 4,Seg Industry Park, Zhenxing Road, Shenzhen City, China
MSN: brucechan2000@hotmail.com
Skype: bruce.chan2000

No matter how many times I type ‘exit’, like millions of times before, it just gets back to what you see in the screen shot! Sometimes I really hate Valve man!

Here is what I have in my startup script:
“screen -m -S tf2 ./srcds_run -game tf +hostport 27015 -replay -autoupdate -tickrate 66 -steambin /home/tf2/hlds/steam -maxplayers 24 +map ctf_2fort”

Does anyone have any ideas? I have tried a few things which you can find here

The location information is based on information held by the IANA - http://www.iana.org/ - who are the global coordinator of the IP network, as well as the DNS root. Users are assigned IP addresses by Internet service providers (ISPs). ISPs obtain allocations of IP addresses from a local Internet registry (LIR) or National Internet Registry (NIR), or from their appropriate Regional Internet Registry (RIR), which is split into five regions - AfriNIC:Africa Region, APNIC:Asia/Pacific Region, ARIN:North America Region, LACNIC:Latin America and some Caribbean Islands, RIPE NCC:Europe, the Middle East, and Central Asia – and is also added too by secondary sources such as user submitted information or ”guesstimates from adjacent Class C range[2] and/or gleaned from network hops”.
Basically, servers by default do not have geolocation information, and even those that do, may not be accurate if they’re based on averaging network hops from other people (going back to the location of this blog, I think WordPress has submitted the location of its servers). If the only way you could tell where your house was based on the time other people took to get there, then you’d have to account for the fact that not everyone travels at the same speed. If the hop is 5ms from server A, 5ms from Server B and 5ms from Server C, then it should be simple to triangulate to the position, providing the speed of the medium is the same. But they’re not.

Now, all of this research has been brought on by a post on a friends church Facebook page, by a person who was carrying out a vulnerability scan of what they said was a child porn site. I haven’t verified that’s what it is, or how he knows that’s what it is, but as we’re just interested with ip addresses the content isn’t really important to this blog post.

Pointing out the obvious first, the person posting and the church are based in the UK, while the site the person is looking for has “dallas” in the title, and – based on the RIR – is based in the US (ip geolocation may not always be accurate, but as Continents come under different registries, telling that part is). After passing through the suspect ip the route takes another dozen hops.

His local area network is hop one of nineteen, while the server he believes to be in the church is hop two of nineteen, which for a residential set up is usually an ip associated with your ISP. If you go back to the GPS analogy, when you plan a car journey from your house then the first line will be your house (origination), and the second the street you live on (your route to the outside world). Where you go from there depends on your ultimate destination, but you need those two parts. Based on that screenshot, he was leaving his house and going down someone elses road, and he said he didn’t recognise it!

In theory, this tracert could be completely accurate…if he was in the church and therefore using their server. The routing would then go from his machine, to the church server, to the outside internet. But he said he was in his own house – two miles away – plus the ip addresses allocated by the church wired and wireless routers don’t match the numbers shown, so we can rule that out.

So that leaves something more sinister. His internet connection has been hijacked, and instead of going to his ISP, it’s going to this unknown ip address, which he then types it into a locate ip address site, and get the church as the result. Given he’s doing something that requires a bit of IT know how, detecting something so weird should probably have alarmed him; instead of going to the internet via his normal route he’s connecting directly to a church two miles away. Yet instead he then shares this as proof of wrong doing that he wants all too see.
Problem is – as discussed above – ip geolocation is not exact, and results vary from address to address. In this case the site he tried said the church, while http://iplocation.truevue.org/ says it’s in Yorkshire, onehttp://www.geobytes.com/IpLocator.htm?GetLocation gives Newcastle, while  http://www.iplocation.net/ puts it in the middle of the Irish sea! For all we know, it could be a proxy server and his information was being shared with people in Russia, but as he was too busy believing it was the church he didn’t seem to consider the dangerous consequences of this deviation.
Sadly for him, the only proof it shows is he doesn’t know as much as he thinks he does. Sadly for everyone else, he’ll probably continue doing things like this without stopping to consider the consequences of his misplaced accusations.

As a father I support the crusade against paedophiles.
But as a father I wouldn’t want my kids – or anyone elses kids – to have to grow up without a parent because they’ve been accused based on flimsy evidence like this, which is obviously wrong to those who know what to look for, but can still tarnish innocent peoples lives.
In the play, Goody Proctor is saved from death because she’s pregnant, but other characters who where falsely accused did not get off so lightly.

To edit the settings of the “User Map” follow this steps:

1) From the main screen choose Map / Download

2) Choose the last entry of the list of maps, the “User” map.

3) Open the menu and tap on “edit map provider”

4) This are the default settings for the OpenStreetMap-Mapnik-Server

You can see some fields to edit. The main setting is the URL of the map server. The URL contains some placeholders {0}..{3}. The amount of placeholders depends on the amount of sub-servers the map server has (load balanced).

The settings

1. URL
   • for Server Count > 1: The placeholders are replaced with the server-number = {0}, zoom = {1}, tile-x = {2} and tile-y = {3}
   • for Server Count = 1: The placeholders are replaced with zoom = {0}, tile-x = {1} and tile-y = {2}
2. Parameter Type
   • You should always choose OSM
3. Server Count
   • Some map providers use more than one server to balance the traffic and reduce the server-stress
   • If this option should not be used, set it to 1 and replace the placeholder for the servers with nothing or a static value.
4. Max Zoom
   • Type in the max zoom level the map provider allows. Some map servers provide map tiles until zoom level 16 only, escpecially such with included hill-shading. Usually this value is 18.
5. File Extension
   • Type in the file-extension (png or jpg) of the format the map tiles are returned from the server.

Examples

Here is an example for the settings of a finnish map server.

And thus it looks like in the map:

More examples

Some links for potential map tile servers:

http://www.neongeo.com/wiki/doku.php?id=map_servers   (replace [z] with {0}, [x] with {1} and [y] with {2})

As test values for zoom, x and y you can use: zoom=10, x=542 and y=297 (Oslo, Norway) e.g. http://opencache.statkart.no/gatekeeper/gk/gk.open_gmaps?layers=topo2&zoom=10&x=542&y=297

Leave a reply if you found a useful map provider which could be interesting for others.

English: The original Piggly Wiggly Store, Memphis, Tennessee. The first self service grocery store, opened 1916. Français : Le premier supermarché Piggly Wiggly ouvert en 1916 à Memphis, Tennessee (Photo credit: Wikipedia)

It’s five o’clock in the afternoon, and you just got off work. You have gone to the grocery store to get the one item you needed to finish dinner. After picking up your cart you see an immense stack of Coke strategically placed like the White House, and realized you needed a case of soda because it’s been a long day and you are thirsty. Then you turn the corner and you see all of these beautiful vegetables displayed and realized that you could make a great salad with just a couple more items. Rounding another corner you see dessert  peering through a little glass case at you, and of course that Angel Food cake would be great with some strawberriesand whipped cream. So, back to produce and on to the one thing you needed. Passing the meat counter you see fresh Porterhouse steaks displayed on sale, and “Shazam” you have tomorrow night’s main course. On each end cap is displayed amazing Little Debbie treats and Tortilla Chips.  Turning you go over to the aisle where the salsa is located because you really wanted a different brand. Now you have four more items and you are headed for that one thing… What was it? Oh yeah!  Get milk… Then that takes you past the ice cream, and then another display with sale wines. Excellent… Oh, now you see one of those “Chip Clips” hanging next to the wine. Turning around you see the new Sports Illustrated displayed on the magazine rack. Then before you get to the register you see a display of beef jerky and realize how hungry you have become. With the saltiness of the jerky you are going  to want a soda in the cooler, next to the jerky,  to wash that down. By the time you get through the register, the milk you came for just cost you 83.55, Does  it dawn on you then that you have just been cleverly outsmarted by technical genius of a grocery store team that you thought was well below your pay grade? No, but you do remember to pick up flowers for your wife on the way out so she won’t yell at you for getting the Little Debbie’s.

English: Little Debbie Zebra Cakes, made by McKee Foods. (Photo credit: Wikipedia)

As a commissioned, or tipped, employee I would like to suggest that you need to take the “grocery store” approach to the services you provide. The key to a great haircut, meal, or outfit sale is to give your guest what they want while adding nuances along the way that will enhance their experience. We want them to walk away from that occasion in time knowing this was the best opportunity ever, and that they couldn’t duplicate it anywhere else with anyone else for that price.

When you introduce your customer to the menu they need to see all of the things that made your establishment special – your “Signature’s” so to speak. If it is a suit, they couldn’t have done it just right without one more really nice tie. In addition, the really needed the French Cuff shirts to set off the ensemble.Your job is to be aware of everything available to them with each option and then enhance their experience with such an option. Here’s the tough part – ready? You have to lay that suggestion in so softly they don’t notice they were just sold on it. They key is to never look hungry – like you need the sale, and never question in your mind whether they need something; – assume they do.

If the majority of the people that walk in have a set tip percentage in their mind then it is to your benefit to deliver the best experience at the optimum price point.

English: unrecognized dessert. (Photo credit: Wikipedia)

Here’s a trick for a server – It’s a bit of a Ninja maneuver – So, don’t let  just anyone try this. When you approach your table, before you get their drink order, introduce them to some of your favorite items. These items will be a little higher price point than the lower end items everyone else sells. Here’s what happens next. You get their drink order and are heading to the back to get their drinks. The time you are away works for you. Their mouth is now watering for the specialty Filet as opposed to the chopped steak they would normally be looking at. What just happened was you diverted their eyes from the inexpensive items and focused them with a laser fine gaze open the mouth watering top dollar steak. The time you have gone away has turned into time for that steak to melt into their braincells. Try it, and if you do it just right you will have them eating out of the palm of your hand, actually the best part of the menu the majority of the time.

The key to making money as a commissioned professional is to fully educate yourself with the items that are available to you and how to pair them with the accessories that are available with each item. You will find that the more tools you add to your toolbox the more prepared you are to to do specialty work in your job.

Always remember, most anyone can wait a table, or sell an outfit, but it takes a special person to create an experience.

Cheers!

Related articles

• Hungry? Don’t Go Grocery Shopping!
• Angel Food Cake
• Grocery Shopping While Hungry Is A Terrible Idea

• Backups. Added a line in the script for the PPM enhanced db
• Started the paperwork to mark the old certs as superseded. Here’s hoping I did that right!
• General mucking about getting integrated with the Novetta portal. The IE browser plugin doesn’t work. But Firefox does. Switching over.

1:30 – 4:00 FP

• Starting to walk through the OpenAL source code
• Meeting with Shawn this afternoon or Thursday
  • Discussed various FP concepts. Of the concepts, iRevolution seems to be the best fit with the IRAD dept. There was also some discussion about building a data visualization capability.

1) Before create the new server in the server view in Eclipse, you need to make sure Eclipse can access the Tomcat files:

cd /usr/share/tomcat7
sudo ln -s /var/lib/tomcat7/conf conf
sudo ln -s /etc/tomcat7/policy.d/03catalina.policy conf/catalina.policy
sudo ln -s /var/log/tomcat7 log
sudo chmod -R 777 /usr/share/tomcat7/conf

Make sure all the files in /usr/share/tomcat7/conf at least have the read permission for all users. During my installation, I had to put the command:

cd /usr/share/tomcat7/conf
sudo chmod 644 tomcat-users.xml

2) Then, you may create the new server in the server view in Eclipse and follow the instruction to choose the tomcat7 as the server.

If you encounter problem during the server creation, you need to remove the previous installed but not successful Tomcat7 server, and then re-creation it follow the creation steps above.

To remove the previous installed server, you can:

a) in Eclipse, go to:
Window->Preferences->Server->RunTime Environment

Click the the Apache Tomcat v7.0 in the right, and select remove
b) or go the directory:

cd ~/workspace/.metadata/.plugins/org.eclipse.core.runtime/.settings

then remove/rename the file:

rm org.eclipse.jst.server.tomcat.core.prefs

Then, you may re-create the new server.

Step 0: Using spare parts or asking a geeky friend to find some.  I’m willing to bet that most people who read this are no more than two degrees of separation away from personally knowing a technophile with spare audio, video, or computer parts.  Technophiles are always upgrading their set-ups, which means that they have random electronic hardware that is taking up space in a basement or storage bin since the last upgrade.  Hint: Technophiles are typically happy to pass-along and help to set up used hardware for other aspiring technophiles.  There are many strains of technophilia – I’m an audiophile at heart.  Somewhere, there is a wood-paneled mixing board from the 70′s that I gave to a friend who gave it to a friend who scrapped it for spare parts.  Here’s a look into my ‘modern’ spare parts bin:

• I probably have more audio cables and adapters than your local Radio Shack store, a mixing board with two fried channels (thanks to “audio engineer” Steve for overloading the mix, circa 2007), a used amplifier that would blow the cones out of most home audio speakers, a shoebox full of audio connectors and adapters, and a couple of computer sound cards.  Sure, I might be hoarding a collection of audio fossils that have little- or no monetary value, but… I have spare components that bridge decades of audio technology and, I swear, will come in handy later!
• For video: a couple of old Roku players (with a bit of integration) = streaming TV video from a home server.  One fossil of a TV uses a RF modulator box so that the Roku can plug into a lonely coaxial input (archaic, but less expensive than buying a new TV).  I have no desire to stream 1080p/HD videos anyway – I’m clearly not a videophile.
• Spare computer parts are a plus.  Extra ethernet and video cables often come in handy, and it never hurts to have a backup DVD-ROM or hard drive.

Step 1: Building a server or finding one that is mostly built.  Home media servers don’t need to be cutting-edge technology, and they don’t need to cost thousands of dollars.  If you have a screwdriver and a bit of know-how (or a geeky friend), you can repurpose a desktop computer or a second-hand server.  As long as the system has enough processing power to get the job done, it is a step in the right direction.

• While shopping for a home media server, I stayed away from the big-name computer stores and websites.  After patiently waiting for a deal on my local Craigslist site, and with a little negotiation (down to $100), I was able to get a used server with a DVD-ROM, Intel dual-core processor, 2GB RAM, 30GB IDE hard drive for the operating system, on-board video and ethernet connections, and a handful of USB ports.  I installed a spare sound card and a new 1TB SATA hard drive to get it ready for playing audio and storing a growing media library.  The cooling fans run constantly and sound like an aircraft taking off, but at least I won’t have to worry about overheating (I was planning on keeping the server in a cabinet anyway).  For under $200, the basic server hardware is in place.
• I’d like to think myself a computer programmer, but I’m not.  Still, I chose a Linux operating system in order to avoid the massive size, viruses, and general crankiness of a Windows platform.  I know enough about Linux to run some basic terminal commands but not enough to be comfortable setting up a server using only command lines.  I decided to install Ubuntu 12.04 – it is a popular platform that provides a nice mix of intuitive graphical user interface and customizability, and the Ubuntu user forums are full of solutions for common problems.  Note: Linux has more programming hurdles than Windows and does require some initiative, but I’ve never encountered a problem that I couldn’t fix after reading through the forums.

Step 2: Setting up the home network.  Moving beyond the nuts and bolts of hardware setup, integrating and managing a home network does take some effort.  For my purposes, I needed the server to be accessible to- and share folders with Windows-based computers over WiFi.  Because every network is a little different, I’ll focus only on the tweaks made to the server itself.

• File system:  Because I’ll be accessing the storage drive from a Windows computer, as a precaution, I set the file system of the media storage drive to NTFS – a native file system for the Windows interface that allows larger file sizes than the FAT32 file system (a consideration for large video files).  So far, neither Ubuntu nor Windows has had any problems reading or writing from the media drive.  I used the gparted package when setting-up the new drive.
• Samba File Server:  The Samba package allowed me to share folders with a Windows computer on my home network.  Setting it up required me to learn a little bit about manually editing system files on the server to set up profiles and shares, but it wasn’t too overwhelming.  With Samba, I can access my media folders from a remote computer running Windows – an absolute necessity!
• xRDP:  xRDP allowed me to remotely access the server from a Windows computer.  This was important so that I didn’t have to constantly sit at the server box when updating the server system.  There is a little bit of lag for screen refresh, but the interface works well overall and I am able to control the computer remotely.  The only major hurdle that I’ve found: login sessions appear to be independent of one another, even when using the same login account… this means that if I start playing music while physically sitting at the server, and then log in remotely to play music, I end up with two overlapping audio streams instead of being able to skip songs or otherwise interact with the session in progress.  More on that in a future post…

Overall, the server took about a week (playing around in the evening hours) to setup and configure basic functionality.  I’m now adding audio and video media with some success of streaming to a Roku player and Windows computer.  My next “Thrifty Technophilia” post will cover the software that makes that all happen.  Until then… feel free to post questions or points to ponder!

SELECT VirtualServerName = SERVERPROPERTY('ServerName')
    , ActiveNode = SERVERPROPERTY('ComputerNamePhysicalNetBIOS') 
    , PassiveNode = CASE (SELECT COUNT(*) FROM sys.dm_os_cluster_nodes WHERE NodeName <> SERVERPROPERTY('ComputerNamePhysicalNetBIOS'))
        WHEN 0 THEN 'Stand-Alone'
        WHEN 1 THEN (SELECT NodeName FROM sys.dm_os_cluster_nodes WHERE NodeName <> SERVERPROPERTY('ComputerNamePhysicalNetBIOS'))
        ELSE Cast((SELECT COUNT(*) FROM sys.dm_os_cluster_nodes WHERE NodeName <> SERVERPROPERTY('ComputerNamePhysicalNetBIOS')) as VarChar(2)) + ' Passive Nodes'
        END

SELECT PassiveNodes = NodeName 
FROM sys.dm_os_cluster_nodes 
WHERE NodeName <> SERVERPROPERTY('ComputerNamePhysicalNetBIOS') 

SELECT ClusterDrives = DriveName 
FROM sys.dm_io_cluster_shared_drives
ORDER BY DriveName

Create a new batch script with Notepad. Enter the following code, changing to include the name of your VM.

cd "C:\Program Files\Oracle\VirtualBox"
VBoxHeadless --startvm name_of_your_vm_here

Save and close your file. Now click and drag it from where you saved it, over the Start Menu and drop it into the “Startup” folder. Now, that script will be executed on startup.

That’s it! Reboot and check that your VM has started running. It will be headless (ie you won’t be a VBox window) which is great for VMs you want in the background. However, you will still have a command prompt open.

sudo /etc/init.d/apache2 restart

# /etc/init.d/apache2 stop

$ sudo /etc/init.d/apache2 stop

# /etc/init.d/apache2 start

$ sudo /etc/init.d/apache2 start

Related articles

• How can i start and stop apache (kunwardharmesh.wordpress.com)

If you cannot find “servers” view, then you may have the Eclipse for Java Developers version.

To direct install the Eclipse for J2EE developers version:

1. go to the Eclipse website http://www.eclipse.org/downloads/

2. Download the right version, make sure is the J2EE version

3. Extract the File in the right directory

Our staff at Aegis, a non-profit gaming community, has been busy preparing for the new entropy launch on Minecraft 1.5.2 code, new features and balanced server-play.

This will involve an eventual map wipe and will take entropy to BETA status.  We want to get a decent player-base going and Technic is a good avenue to bring players in from.

If you want to play entropy before the new server launch, download it from the Technic launcher (and please +1 the pack up to give exposure, get more players. ** you need to register on Technic to vote, but gives access to their forums and bug reporting)

I will also be releasing a public server download for entropy in conjunction with open beta.

Bạn có thể tải mã nguồn của bài viết trước ở đây.

Ngày 18 tháng 2 năm 2003

Trong đầu tôi cứ cân nhắc mãi mớ threads treo lủng lẳng trong khi ăn món mì ống spaghetty một cách lơ đãng. Sau bữa trưa, tôi trở về phòng làm việc tìm Jerry.

“Ông C nghĩ là SocketServer sẵn sàng để dùng rồi đó, và bây giờ ông ta muốn chúng mình làm việc với ứng dụng SMSRemote.”

“Ồ, đúng nhỉ!” Tôi nói. “Thì đó là lý do có SocketServer mà – mình đã dựng xong chương trình dùng để gọi phần biên dịch SMC từ xa, gửi mã nguồn đến server và gửi ngược lại tệp tin đã biên dịch.”

Jerry nhìn tôi chờ đợi và hỏi, “mày nghĩ mình bắt đầu sao đây?”

“Tôi nghĩ là tôi cần biết người dùng sẽ sử dụng chúng ra sao cái đã,” tôi trả lời.

“Xuất sắc!” gã mỉm cười. “Bắt đầu từ cái nhìn của người dùng luôn luôn là một điều hay. Thế thì đâu là cách đơn giản nhất người dùng có thể sử dụng công cụ này?”

“Anh ta có thể yêu cầu một tệp tin nào đó được biên dịch. Lệnh ấy có thể như thế này.” tôi viết lên tường như sau:

java SMCRemoteClient myFile.sm

“Coi được đó,” Jerry nói. “Vậy mình bắt đầu sao đây?”

Mì spaghetti đã ấm trong dạ dày, và tôi cảm thấy khá vững tin sau khi làm SocketServer chạy được, thế nên tôi vớ lấy bàn phím và bắt đầu gõ:

public class SMCRemoteClient {
	public static void main(String args[]) {
		String fileName = args[0];
	}
}

“Xin lỗi!” Jerry ngắt ngang. “Mày có kiểm thử cho nó không?”

“Ý ông là sao?” tôi hỏi một cách thiếu kiên nhẫn. “Mã nguồn này thuộc dạng lẻ tẻ – sao mình phải viết kiểm thử cho nó làm chi?”

“Nếu mày không viết một kiểm thử cho nó thì làm sao mày biết là có cần hay không?” gã hỏi.

Câu hỏi ấy làm tôi khựng lại. “Tôi nghĩ điều ấy quá hiển nhiên,” sau rốt tôi nói.

“Vậy sao?” Jerry trả lời. “Tao không thấy thuyết phục cho lắm. Hãy thử một lối khác xem sao.”

Jerry vớ lấy bàn phím và xoá hết mã nguồn của tôi. Tự ái trong lòng bùng lên nhưng tôi cố dằn nó xuống. Dù gì cũng chỉ có vỏn vẹn bốn dòng code mà thôi.

“OK, mình cần những hàm nào đây?” gã hỏi.

Tôi nghĩ ngợi vài giây và nói, “mình cần lấy tên tệp tin từ dòng lệnh nhưng tôi không biết ông sẽ làm sao nếu không có phần mã nguồn ông vừa xoá mất.”

Jerry nhìn tôi với vẻ chế giễu, hắn nói, “tao biết,” và bắt đầu gõ phím.

Ðầu tiên gã viết một đoạn mã của khung làm việc kiểm thử mà giờ đã quen thuộc:

import junit.framework.*;
public class TestSMCRemoteClient extends TestCase {
	public TestSMCRemoteClient(String name) {
		super(name);
	}
}

Gã biên dịch và chạy, nắm chắc rằng kiểm thử bị hỏng vì thiếu kiểm thử, và rồi gã thêm kiểm thử sau:

public void testParseCommandLine() throws Exception {
	SMCRemoteClient c = new SMCRemoteClient();
	c.parseCommandLine(new String[] { "filename" });
	assertEquals("filename", c.filename());
}

“Được thôi,” tôi nói. “Có vẻ như ông lấy đối số của dòng lệnh bằng hàm parseCommandLine thay vì dùng main, nhưng phiền như thế làm gì?”

“Làm thế để tao có thể kiểm thử,” Jerry trả lời.

“Nhưng chẳng có gì để mà kiểm thử,” tôi cằn nhằn.

“Ðiều đó có nghĩa là viết kiểm thử thực sự rẻ.” Gã cười toe toét.

Tôi biết tôi sẽ không thắng nổi trận đấu này nên đành thở dài, vớ lấy bàn phím và viết đoạn mã sau để kiểm thử có thể đạt:

public class SMCRemoteClient {
	private String itsFilename;

	public void parseCommandLine(String[] args){
			itsFilename = args[0];
	}

	public String filename() {
		return itsFilename;
	}
}

Jerry gật đầu và nói “Tốt, nó thành công rồi”

Sau đó hắn lặng lẽ viết trường hợp kiểm thử tiếp theo.

public void testParseInvalidCommandLine() {
	SMCRemoteClient c = new SMCRemoteClient();
	boolean result = c.parseCommandLine(new String[0]);
	assertTrue("result should be false", !result);
}

Tôi lẽ ra phải biết hắn sẽ làm như thế. Hắn đã chỉ cho tôi thấy rằng việc viết kiểm thử mà tôi cho rằng không cần thiết là một ý hay.

“OK”, tôi thú nhận. “Tôi đoán việc lấy đối số của dòng lệnh ít vụn vặt hơn là tôi nghĩ. Có lẽ nó đáng để có một kiểm thử cho riêng nó.” Thế rồi tôi vớ lấy bàn phím và làm cho kiểm thử đạt.

public boolean parseCommandLine(String[] args) {
	try {
		itsFilename = args[0];
	} catch (ArrayIndexOutOfBoundsException e) {
		return false;
	}

	return true;
}

Cân nhắc kỹ lưỡng, tôi tái cấu trúc biến c và khởi tạo cho nó trong hàm setUp. Các kiểm thử đều đạt.

Trước khi Jerry có thể đề nghị trường hợp kiểm thử tiếp theo, tôi nói, “Rất có khả năng tệp tin không tồn tại. Chúng ta nên viết một kiểm thử chứng minh mình có thể xử lý trường hợp ấy được.”

“Quả vậy,” Jerry nói trong khi tóm lấy bàn phím từ tay tôi. “Nhưng để tao chỉ cho mày tao khoái làm cách đó như thế nào”

public void testFileDoesNotExist() throws Exception {
	c.setFilename("thisFileDoesNotExist");
	boolean prepared = c.prepareFile();
	assertEquals(false, prepared);
}

“Mày thấy không?” gã giảng giải. “Tao muốn đánh giá mỗi đối số của dòng lệnh trong hàm của chính nó thay vì trộn cả mớ mã phân tách và đánh giá chung với nhau.”

Trong khi đó, tôi kín đáo đảo mắt ráng ghi nhớ những điếm ấy để tham khảo sau này, tôi lấy bàn phím và thay đổi những điểm sau để làm cho kiểm thử đạt:

public void setFilename(String itsFilename) {
	this.itsFilename = itsFilename;
}

public boolean prepareFile() {
	File f = new File(itsFilename);
	if (f.exists()) {
		return true;
	} else {
		return false;
	}
}

Toàn bộ các kiểm thử đều đạt. Jerry nhìn tôi rồi ngó sang bàn phím. Hiển nhiên gã muốn “lái” nó rồi. Hôm nay dường như gã tràn đầy sáng kiến, bởi thế tôi chuyển bàn phím về phía gã.
“OK, bây giờ xem đây!” gã nói, cỗ máy trong người gã rõ ràng đang gầm rú.

public void testCountBytesInFile() throws Exception {
	File f = new File("testFile");
	FileOutputStream stream = new FileOutputStream(f);
	stream.write("some text".getBytes());
	stream.close();
	c.setFilename("testFile");
	boolean prepared = c.prepareFile();
	f.delete();
	assertTrue(prepared);
	assertEquals(9, c.getFileLength());
}

Sau khi nghiên cứu mã nguồn của gã vài giây, tôi trả lời, “Ông muốn prepareFile() để lấy độ dài của tệp tin? tại sao?”

“Tao nghĩ lát nữa mình sẽ cần chúng,” gã giải thích. “và đó là một cách hay để chứng minh mình có thể đối phó với một tệp tin hiện có.”

“Mình cần nó để làm gì kia chớ?” tôi hỏi.

“Chúng ta sẽ phải gửi nội dung của tệp tin thông qua socket đến server, phải không?” Jerry hỏi.

“Vâng.”

“Và chúng ta cần biết sẽ gửi bao nhiêu chữ,” gã kiên nhẫn giải thích.

“Hườm… có lẽ,” tôi miễn cưỡng trả lời.

“Tin tao đi,” gã mỉm cười. “xét cho cùng thì tao là người hướng đạo cơ mà.”

“OK, khỏi nói đến chuyện ấy,” tôi trả lời một cách thiếu kiên nhẫn. “Tạo sao ông lại tạo tệp tin trong kiểm thử kia chớ? sao ông không giữ tệp tin này sẵn thay vì lần nào cũng phải tạo nó ra?”

Jerry cười khẩy rồi trở nên nghiêm túc. “Tao ghét giữ lại các nguồn bên ngoài cho mấy cái kiểm thử. Bất cứ khi nào có thể được, tao để cho mấy cái kiểm thử tạo ra nguồn chúng cần. Với cách ấy, không cách nào tao bị mất nguồn cả, hoặc ngay cả trường hợp nguồn bị hỏng nữa.”

“Ồ, điều này thì quả có lý,” tôi thừa nhận, “nhưng tôi vẫn không thích thú gì với mấy thứ độ dài của tệp tin kia.”

“Nhớ đó. Mày sẽ thấy!”

Tôi lấy bàn phím và bắt đầu làm viết mã để vượt qua kiểm thử. Trong khi tôi gõ phím, tôi thấy hơi lạ vì tôi đang viết mã sản xuất trong khi thiết kế là của Jerry – nhưng những gì Jerry làm chỉ là viết những kiểm thử nhỏ. Bạn có thể thực sự xác định một thiết kế bằng cách viết những kiểm thử không?

public long getFileLength() {
	return itsFileLength;
}

public boolean prepareFile() {
	File f = new File(itsFilename);
	if (f.exists()) {
		itsFileLength = f.length();
		return true;
	} else {
		return false;
	}
}

Kiểm thử kế tiếp tạo ra một cái server giả và dùng để thử khả năng của SMCRemoteClient truy cập vào đó.

public void testConnectToSMCRemoteServer() throws Exception {
	SocketServer server = new SocketServer() {
		public void serve(Socket socket) {
			try {
				socket.close();
			} catch (IOException e) {
			}
		}
	};

	SocketService smc = new SocketService(SMCPORT, server);
	boolean connection = c.connect();
	assertTrue(connection);
}

Tôi có thể vượt qua nó mà chả gặp mấy khó khăn:

public boolean connect() {
	try {
		Socket s = new Socket("localhost", 9000);
		return true;
	} catch (IOException e) {
	}

	return false;
}

“Tuyệt!” Jerry nói. “Mình nghỉ giải lao một tí.”

“OK,” tôi trả lời, “nhưng hãy viết phần main() trước đã.”

“main() gì, dính dự gì ở đây?” gã hỏi.

“Hở? đó là main của chương trình chớ gì!”

“Thế thì sao chớ?” Jerry rụt vai. “Nó chỉ gọi parseCommandLine(), parseFile() và connect().

Còn lâu lắm mình mới kiểm tra mấy thứ đó!”

Tôi rời phòng làm việc và đi về phía phòng giải lao. Trước giờ tôi cứ nghĩ main() là hàm đầu tiên cần được viết, nhưng Jerry rất đúng. Rốt cuộc, main() chỉ là một hàm khá thiếu thú vị.

Nguồn Clean Code

Người dịch: Hoàng Ngọc Diêu | Biên tập: Phạm Anh Đới

Thợ lành nghề #9: Những thread nguy hiểm (Dịch vụ Socket 4)
Thợ lành nghề #10: Những thread lửng lơ (Vòng lặp không hạn chế)

Work seemingly wasn’t the only area of his life spiralling downwards. His girlfriend had begun to smell a rat. She’d call the restaurant demanding answers from Justin and Tina. “Is Sid there?” she’d ask frantically, and they’d have to tell her that no, he wasn’t.  It looked like Delilah wasn’t the only girl he had cheated on her with. She kicked him out of their shared house a month before his visa ran out. There was nothing else for it: Sid had to leave the restaurant, and the country.

On his last day, Sid told us all he hoped we’d be there for his leaving party. Those of us that finished work in time were in two minds, but eventually, Yeliz, Delilah and I decided to show our faces. I borrowed Ash’s coat; we went straight from work and the weather had turned freezing cold. When I popped back into the restaurant, I was sucked in to telling everyone about it; about how uncontrollably wankered Sid was, about how he was getting with Jasmine, another manager from a different restaurant, and about how weirdly he’d treated us when we arrived.

“So, I walk up to the table, Yeliz and Delilah behind me, and he’s wasted, absolutely off his tits, and I’m like, hi Sid, we thought we’d come to say goodbye, and the first thing he says is, ‘Silent! YA CUNT!’”

I was sitting on a stool opposite Amanda as she shut down the bar, setting the scene whilst eating a handful of peanuts.

She stopped what she was doing. “You cunt?” she said incredulously, her clear blue eyes widening, dimples vanishing from her cheeks. “You cunt?” Surrounding tables were beginning to turn around to watch their aristocratically beautiful waitress swear like a sailor. ”Oh my god, he called you a cunt? Well what did you say? Jesus Christ, I am furious! Did you call him a cunt back? You should have said, I’m not a cunt, I’m looking at a fucking cunt!”

And then she went back to cleaning the coffee machine.

- silentwaitress

Make sure that no project inside the Server, or you can delete existing server and create new server.

1. Right click on your server -> Properties -> then click “Switch location” and it must not [workspaces metadata].
2. Then double click your server on explorer, then the “Overview” will show.
3. check “use tomcat installation (takes control of tomcat installation)” which placed under “Server Location”.

Eclipse Apache Tomcat 404, STS 404, Spring Template project 404, Spring 404, not found

This video helps depict how guests should treat their servers when going out to eat. Videos from servers like Jay would be helpful for many restaurant foodies to see before dining out. These blogs are also fun entertainment for servers and consumers around Springfield to follow: The Oatmeal, Buzz Feed, and Tip Your Server.