stlplace

Oracle
SQL: Inner join and out join

Function, (stored) procedure and package

What is a trigger in Oracle

“…Oracle lets you define procedures called triggers that run implicitly when an INSERT, UPDATE, or DELETE statement is issued against the associated table or, in some cases, against a view, or when database system actions occur. These procedures can be written in PL/SQL or Java and stored in the database, or they can be written as C callouts.

Triggers are similar to stored procedures. A trigger stored in the database can include SQL and PL/SQL or Java statements to run as a unit and can invoke stored procedures. However, procedures and triggers differ in the way that they are invoked. A procedure is explicitly run by a user, application, or trigger. Triggers are implicitly fired by Oracle when a triggering event occurs, no matter which user is connected or which application is being used…”

A shorter version at orafaq forum

Longer version at orafaq.com/wiki

7. Compare and contrast TRUNCATE and DELETE for a table.
Both the truncate and delete command have the desired outcome of getting rid of all the rows in a table. The difference between the two is that the truncate command is a DDL operation and just moves the high water mark and produces a now rollback. The delete command, on the other hand, is a DML operation, which will produce a rollback and thus take longer to complete.

8. Give the reasoning behind using an index.
Faster access to data blocks in a table.

11. Give some examples of the types of database contraints you may find in Oracle and indicate their purpose.
A Primary or Unique Key can be used to enforce uniqueness on one or more columns.
A Referential Integrity Contraint can be used to enforce a Foreign Key relationship between two tables.
A Not Null constraint – to ensure a value is entered in a column
A Value Constraint – to check a column value against a specific set of values.

DDL vs DML
Data Definition Language (DDL) statements are used to define the database structure or schema. Some examples:
CREATE – to create objects in the database
ALTER – alters the structure of the database
DROP – delete objects from the database
TRUNCATE – remove all records from a table, including all spaces allocated for the records are removed
COMMENT – add comments to the data dictionary
RENAME – rename an object

Data Manipulation Language (DML) statements are used for managing data within schema objects. Some examples:
SELECT – retrieve data from the a database
INSERT – insert data into a table
UPDATE – updates existing data within a table
DELETE – deletes all records from a table, the space for the records remain
MERGE – UPSERT operation (insert or update)
CALL – call a PL/SQL or Java subprogram
EXPLAIN PLAN – explain access path to data
LOCK TABLE – control concurrency

1. Describe the difference between a procedure, function and anonymous pl/sql block. Candidate should mention use of DECLARE statement, a function must return a value while a procedure doesn?t have to.

PL/SQL developer vs. Toad

Reporting
SQL Server Reporting Services (Microsoft); SQL Server Reporting Services (wiki)

Oracle 10g reports (PDF); Oracle Reports Tutorial

LIMS (wiki)

Java
30 Core Java questions (sys-con)

J2EE interview questions (techInterviews.com)

more j2ee interview questions

Java: hashMap and hashTable

Hashtable is synchronized, whereas HashMap is not. This makes HashMap better for non-threaded applications, as unsynchronized Objects typically perform better than synchronized ones.

Difference between Get and Post method on web (link one)

Lazy initialization, dependency injection

3 Things in OO:
data encapsulation, polymorphism, dynamic dispatch (when a method is invoked on an object, the object itself determines what code gets executed by looking up the method at run time in a table associated with the object. This feature distinguishes an object from an abstract data type (or module), which has a fixed (static) implementation of the operations for all instances. It is a programming methodology that gives modular component development while at the same time being very efficient.)

How to use static method? (Include class name)

Aggregation

Throwable interface (exception)

What is reflection

Interface vs. abstract class
non-static method
final: immutable, e.g. String
finalized method

tree-reflection

Read the rest

drop down box
Clear a list for selectionOneMenu every time, got burned by this in a “page number” drop down for Richfaces data scroller. And some other places.

Use validator
keep in mind it will overwrite the required = “true”, if the user does not select a valid value from drop down, it will give error message during Save method.

Assign 0 to “–select–” ?
Yes or no. Add 0 means 0 is a selectable option, so if one wants to enforce user to select something using required=”true”, it will pass this check. Thus comes in the validator, in which one can say “0″ is not valid.

Page number for data scroller
1) t:saveState if we can not have session scope for dataBean;
2) the drop down tip above;
3) When “adding or deleting rows” and it cause the total number of pages changes, make sure set the correct page number in the “action” (save) methods.

The following was written more than a year ago…

======
Got opportunity to work on those two technologies recently due to job change. Had some gotchas as I transform from C++, CAD programming to this new program paradigm.

Data problem vs. coding error
In CAD world, sometimes there is problem with the data (CAD file) itself. But most of the time I focus my effort debugging the coding error in my own code, not trying to debugging the data itself. In Oracle/Database world, I found it more often that the data itself could complicate problem. Because the relational database has relationship between table, sometimes we can not add/delete/update table at will due to dependencies (data integrity). This also bring about the second topic I want to discuss.

C++ vs. Java
Java is not as flexible as C++. For instance, how to pass data back from a function, so far I found one way: using the return data. In C++, I used the pointer and reference quite a bit. The wiring of inheritance and interface (implementation) could also got a bit complicated: basically one needs to draw the relationships on a paper to better understand all those.

Eclipse vs. Visual Studio
Eclipse (the free Java IDE) is quite powerful, actually. It’s also highly configurable, with so many plug-ins, e.g., this subclipse integrates Eclipse with Subversion. I remember I could not do such things inside Visual Studio (for Perforce).

Read the rest

Stanford commencement speech (2005)

Macintosh 1984 commercial

Steve Jobs and Bill Gates in 1983 Apple event

My other post on Steve Jobs.

Walt Mossberg on Steve Jobs.

Dirty flag on a HTML form is much trickier than I thought, that is, it is tricker when I started to programming for dirty flag.

Saw this javascript snippet. It works great except in my case, I was generating dynamic drop down (selection menu), and in a lot of cases it behaves as opposite to expected (i.e., it shows the form is dirty when it is not, and vice versa). Found out there are couple things came into play.

1) The default selection (type “0″), or select “nothing”, I forgot to set the value “0″ (string type).

2) Second, I forgot to clear some of the derivative selections when I make selection the primary selection menu (basically it will dynamically generate other drop downs based on the primary selection).

3) Third, allow type “0″ has other effect, I had to remove required=”true” validation from the selection menu, and validate it manually. Also, after I made this optional (removed required=”true”), the actionListener will get triggered when I make selection type “0″, previously it was not entering it, and caused other rendering issues.

4) Last but not least, I added a “validator” to the primary selection menu (basically it will give error if I select “0″ type), and it will not trigger the actionListener again. Madness. So basically I have to do the validation somewhere else.

It looks like it. Obviously my observation could be quite limited. But here is the background, in the good old days, Apple ranks app mostly by the download number in a given amount of time, and the ratings/reviews. So basically the popular the app, the higher the rank. This process could reinforce itself. Apple realized this approach probably has its limitation, one example being a fairly popular app Facebook (by usage) is not ranked the 1st. So Apple changed its algorithm a bit, it not only ranks by the download number, it also monitors the usage. By “usage” I assume Apple measures mostly by network activity, my reasoning being if an app just does its thing locally, Apple has no way to know it.

Also keep in mind the rating and reviews also play a role in the ranking, my understanding is if all things equal, higher rating or favorable reviews will push the rank higher.

More recently, it seems Apple started to give usage more weight. This is just from my own observation, I have 2 little apps, both are standalone (no network features), and one does fairly well last few month, until recent, it totally went down cliff. I did not know why until I did keyword search, in the past, it ranked 2 or 3, not it fell out of top 25. I tried to rescue such as free download for a limited time. After the free period, download number drops back, and this thing fell out top 25 again

I think from product lifecycle management (PLM) point of view, this app thing is not as good as traditional software revenue model

(Update 06-24-11) Last night I found “More iPhone 3 development” (by Dave mark and Jeff Lamarche) chapter 14 deveoted full chapter to multithread including timer.

(Original) As of iOS 4, Apple made blocks and Grand Central Dispatch (GCD) available. Previously it was available on Mac.

Some pre-requisite or recommend readings
WWDC 2010 video session 206: Introduction to Blocks and Grand Central Dispatch GCD (you need to be a registered Apple developer to download this).

Or if you read the nice intro “guide to blocks and grand central dispatch by Cocoa Samurai (Colin Wheeler).

The problem I attempt to solve
I have couple buttons waiting for the user to touch, and the buttons can not wait there forever, they wait for 2 seconds if a user does not touch any of them, and a new set of button will display at that time.

I thought about this problem for a while, thought about using NSNotifications which is essentially a callback mechanism (see “love to be notified” by Cocoanetics for more info). I need more, essentially I need a second thread which is a timer, to notify the main thread (which does display and handles touch) when 2 seconds is up.

I found fieryrobot’s Watchdog timer in GCD which meets my basic need for timer. But I need to return from timer to the main thread. Quote Fieryrobot:

…If you needed to return results to your main thread, you can just use dispatch_async to execute the code on the main queue (dispatch_get_main_queue() as we’ve seen in previous posts…

I am sure there is ways using blocks (callbacks) to do that, but I don’t know on top of my head. So I used Notification instead. Here is what I did in ViewDidLoad
[[NSNotificationCenter defaultCenter] addObserver:self
selector:@selector(twoSecondsNotification:)
name:@"TwoSecondsReached"
object:nil];

In the timer, I post notification when it’s time to go back to main queue (thread).
// Hey, let's actually do something when the timer fires!
dispatch_source_set_event_handler(_timer, ^{

// pass control back to main queue
dispatch_async(dispatch_get_main_queue(), ^{
[[NSNotificationCenter defaultCenter] postNotificationName:@"TwoSecondsReached" object:nil userInfo:nil];
});

last but not least, the “TwoSecondsReached” function
- (void)twoSecondsNotification:(NSNotification *) notification
{
// clean up the timer here
if(timer != nil)
{
[timer invalidate];
[timer release];
}

// do things you want to do here.
}

One more thing
In my code, I initialize timer and release timer in different methods, and I found it’s easy to lose tack of timers and have multiple timers. One phenomena I saw is multiple timers fire at the same time, and make the app go crazy. I am thinking use a singleton pattern for the timer at this time.

I prepared the following summary for 360 iDev conference this fall at Denver. The talk did not make into the agenda, but I think it may be interesting to some western iOS (mobile) developers. After all, the market has great potentials.

Some background
iPhone officially came to China late 2009, but iPhone 1G came to China shortly after the US launch in Summer 2007. The first Apple store opened in Beijing July 2008. According to Steve Jobs Sept 2010 music event and Apple Q4 2010 earning report, greater China is the fastest growing market. Chinese costumers are snapping up iPod, iPhone, iPad and Macs quickly. How about apps? In this talk, we will discuss the China iOS app development companies (some have their apps featured by Apple), the current status of China iOS apps and the potential of iOS app markets (both consumer and enterprises).

Xujiahui

Some notable China based iOS development companies
139.me (Its Colorful Aquarium app was featured by Apple).
Rye Studio (featured article on San Jose Mercury News here).

Read the rest

I was working on a new app and this app has something to do with timing. Basically I need to record the time between a button being displayed and the time a user touches the button (note they are in two different functions).

NSDate and memory leak
To accomplish this, I am using NSDate class, I declared NSDate * start in the class,
and then in one function, I was doing
start = [NSDate date];

and in the second function, I was doing
NSDate *methodFinish = [NSDate date];
NSTimeInterval intervalTime = [methodFinish timeIntervalSinceDate:start];

But the code above crashes immediately. After some research I believe this has something to do with memory, if we copy (retain) the NSDate as suggested by this thread in iphonedevsdk.com. It worked.

NSDate* now = [NSDate date]; // unretained
start = [now copy]; // retained

Sleep

Read the rest

I was following John Muchow’s tutorial “Distribute Ad Hoc Applications over the air OTA“. It worked as expected until the last step, when I click the link on my iPad 1, it says “can not download the app…”.

Did google on this, found one relevant discussion on stackoverflow. I was using my web hosting company web interface uploading the file, so I switched to ftp, changed to binary mode. Same results.

I realized my iPad 1 was still at iOS 4.2.1, while my Xcode was building 4.3 target. So I changed the target to 4.2 (in Xcode => Project Setting => build target), and repeated relevant steps in the tutorial (mainly to create the .ipa and .plist). Things worked after I made the change.

Btw, I found Xcode 3.2.6 build-in capability to share and distribute (submit to app store) is cool.