my $dbh;
# GOTO is gone. You're also not continually reconnecting to the DB even when already
#connected anymore, which will make your DB server admins happy
# Are you sure you really want to be doing this, though? That's still pretty nasty if it can't connect.
while(!$dbh) {
#eval is like a try{} in other languages... kind of.
# it also suppresses warnings and stuff going to stderr
# you can get the return code if you need it.
eval{$dbh = DBI->connect($dsn,$user,$pass);}
}
# Move your prepare outside of the loop. Inside the loop is defeating the point
# of using prepare() in the first place.
# also, don't use the ; at the end of your sql statement. It doesn't actually hurt anything
# but is still generally considered wrong to do
my $sth = $dbh->prepare
('select id, UNIX_TIMESTAMP(last_active) from machine_state');
for (;;)
{
$sth->execute;
while(my @row = $sth->fetchrow_array())
{
$current_time = time;
if (($row[1] != 0))
{
if ($current_time > ($row[1]+45))
{turn_off($row[0]);}
}
}
sleep(5);
}
A few other tips for you here:
Read about eval because there are some other things you may need to know about it.
prepare() should be called outside of loops. If you prepare outside of the loop and then just execute() inside of the loop you save the db server some work setting up the execution plan repeatedly and some network traffic and so on.
You should also really try to use the strict pragma and be sure to declare your variables with my (in most cases... on rare occasions you'll use our). This makes Perl behave like a proper programming language with sane variable scoping so that crazy things don't start happening as your script inevitably gets longer and more complex.
You probably don't want DBI->connect() being called repeatedly if it can't connect. If it can't connect, there's likely a problem and hammering the server and blowing cpu cycles on the client like that isn't going to help anything and can make matters worse.
The above is a good answer, but you might be interested to know that the drop-in replacement for that goto would be "continue", which just goes to the next iteration of the innermost loop.
Continue would have worked there, but really that DBI->connect() needed to be pulled out of the for() loop anyway. Once it's connected, it's connected. There's not need to call connect() again within the same script unless you close the connection for some reason.
I would put that while() loop with the DBI->connect() with a sleep() or wait() that steps up. Something like after first failure to connet wait 5 seconds, then 30 seconds, then 1 minute, then 5 minutes, then 10 minutes, then 30 minutes, then 1 hour and hold at retrying every hour until the script is manually restarted.
"word" most likely means a discrete word just like any of the words in this paragraph, although I guess it could be referring to the typing definition of "word" as in 5 characters (I think it's 5 chars). In either case it's pretty simple with Regular expressions.
I don't do them enough in java to quickly whip up an example, but basically for a very naive system you could just count the number of matches to "<beginning of line><any characters except whitespace>" or "<space><any characters except whitespace>" . There's a special char, \s, which will match space or tab whitespace so it's nice and easy to just match "whitespace" in general, and \S matches non-whitespace, assuming the "group of characters between spaces" definition of word. If it's the other, regexp has a way to count characters.
I suspect Java has a way to just count the number of matches using the Matcher class (as opposed to String.matches()), but I haven't looked at the docs for it.
In a worst case scenario it probably has a way to just dump the matches into an array, but depending on just how large this text is, you may not want to do that. In an even worse scenario, you could even do that with String.split() and split on whitespace and dump to an array. I would start with the regular expressions and seeing if the Matcher class has a method to just return the number of matches, though, because that's almost certainly the better way.
When dealing with removing words from a sentence you do want to use a regular expression. Look at \b and \B which detect word and non word boundaries.
God, I always forget about \b and \B. Note that \b doesn't count ' (apostrophe) or - (hyphen) as a word character and so Jaziek's would be 2 words without a little bit more trickery using groups between []
I don't think he meant the programming definition for "word." But doesn't String.split() in java take regex patterns as well as single characters?
It looks like String.split() will take a regex, too.
I'd still look into using the actual regex stuff unless this is like intro to java programming class and this week's chapter is on the String class since using that means dumping it into an array whereas the Matcher class has a groupCount() method (I just checked) which returns the number of matches.
\b and \B will do what he wants aside from the caveat I mentioned above. They don't match on the computer definition of word, they are like \w and \W except in java they support unicode and \w and \W don't.
Of course my solution of matching on non-space characters could also give problematic results if things like & are not supposed to count as a word.
This seems like a really simple problem so I must be having a brain fart.
I'll have input that has the command then the variable to be used.
Ex: INT v298
I need the scanner to separate the char from the int.
Do I just have my scanner scan the line, pass it to a string tokenizer which separates the two strings (1 the command, 2 the variable), then have the tokenizer also separate the char from the int, or do I just use a delimiter to ignore the v? If I do that, how do I set it up?
Would it be like so:
Scanner scan = new Scanner(System.in).useDelimiter("v");
If you're just starting out, regexes may be inappropriately complex. For some simple input parsing, you could just read the whole line and then read one character at a time from that String. For the word-counting thing, you could have one while loop which skips all whitespace one character at a time, and then another while loop which reads all non-whitespace one character at a time, incrementing the word count by 1 when it finishes. Those would be inside another loop which terminates when you reach the end of the input.
In particular, I would recommend against using Java regexes to match across line boundaries unless you are familiar with Java's regex engine. For example, Java does not normalize line breaks into \n like C does, and it has a few other oddities.
Saeris on
0
HalibutPassion FishSwimming in obscurity.Registered Userregular
Because they look like someone took a dump on your monitor. The longer or more complex the regex string, the worse it gets.
Long regex expressions can't be parsed by a human without spending an inordinate amount of time trying to figure out what it's supposed to do. It's like the worst perl code you've ever seen.
They're good at what they do, but that doesn't mean you have to like how it looks.
If it's not "Find a zip code" or "find a phone number" then I don't use them and subject myself to write a subroutine for, in this example, counting words.
bowen on
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
If it's not "Find a zip code" or "find a phone number" then I don't use them and subject myself to write a subroutine for, in this example, counting words.
I find them great when used for matching. For example I wouldn't want to write my own file extension matching code. The key to writing regexp is to document the intent and use cases of what passes. I also split the regexp into smaller pieces that I document.
Because they look like someone took a dump on your monitor. The longer or more complex the regex string, the worse it gets.
Long regex expressions can't be parsed by a human without spending an inordinate amount of time trying to figure out what it's supposed to do. It's like the worst perl code you've ever seen.
I don't find this to be a problem with Python. I use the re.VERBOSE flag so I can split the regex across multiple lines and have comments there to explain what each line does, and raw strings negate the need to use double backslashes for things like "\s" which helps quite a bit with readability.
Barrakketh on
Rollers are red, chargers are blue....omae wa mou shindeiru
Thanks for the help. After looking at was given, I couldn't get it to work at all so I just went ahead and used a stringTokenizer.
I have a new question that needs fresh eyes. I'm trying to create a hashtable without being allowed to use the java hash table already implemented at all. So I'm trying to do chaining on my array.
The first problem is, I believe the list and listiterator I'm using is only creating one list and adding elements only to that one list. I need it to be for every slot on the array so it functions like a hash table w/ chaining. Do I change my array from holding objects, to holding lists instead?
The second problem is, since my variable class is abstract and the actual nodes placed onto the array at the moment are classes that extend variable (like IntVar or FloatVar), I'm not able to use my getValue() methods for those specific classes. What am I doing wrong? Do I need a specific get method in my abstract class first?
My teacher gave me this advice:
Only when you actually need the value (in setting/getting it in ASSIGN, PRINT etc.) you have to check what the dataType is, and cast you Variable object into either IntVariable or FloatVariable. And there you have two similar pieces of code, one for handling integers and one for floats. But it's good the have the part that deals with the value separate from the parts that don't - in this way you don't duplicate some code. It's not good to put checking e.g. whether variable exists into setValue, because then you have to put it in two places (in setValue for float, and in setValue for int, like you have it in your code below). Better to have a separate function in Variable class that checks this, and you call it prior to calling setValue.
But for my abstract class, if I put setValue(so so), and my other classes need to overwrite this method, how do I determine which value it is. Same with getValue? Do I have an if statement in my abstract class that says:
If(dataType.equals("INT"))
do this
else
do this other thing
My code so far:
import java.util.*;
public class Calculator
{
private static final int DEFAULT_SIZE = 128;
private static Variable[] table;
private static LinkedListIterator itr;
private static LinkedList list;
private static String command;
private static Scanner scan;
public static void main(String[] args)
{
scan = new Scanner(System.in);
StringTokenizer st;
table = new Variable[DEFAULT_SIZE];
for(int i = 0; i < DEFAULT_SIZE; i++)
{
table[i] = null;
}//end table setup
while(true)
{
command = scan.next();
if(command.equalsIgnoreCase("QUIT"))
{
break;
}
else
{
String temp = scan.next();
st = new StringTokenizer(temp, ("((v))"));
Integer tempKey = new Integer(st.nextToken());
int key = tempKey.intValue();
/*---------------------------------------------------
* BEGIN LISTING COMMANDS
--------------------------------------------------*/
if(command.equalsIgnoreCase("INT"))
{
int hash = (key % DEFAULT_SIZE);
put(command, key, hash);
}//end if command == INT
else if(command.equalsIgnoreCase("FLOAT"))
{
int hash = (key % DEFAULT_SIZE);
put(command, key, hash);
}//end command FLOAT
else if(command.equalsIgnoreCase("ASSIGN"))
{
int hash = (key % DEFAULT_SIZE);
assign(key, hash);
}//end command Assign
else if(command.equalsIgnoreCase("ADDASSIGN"))
{
}
else if(command.equalsIgnoreCase("PRINT"))
{
int hash = (key % DEFAULT_SIZE);
print(key, hash);
}
else if(command.equalsIgnoreCase("INT2FLOAT"))
{
}
else if(command.equalsIgnoreCase("FLOAT2INT"))
{
}
else if(command.equalsIgnoreCase("QUIT"))
{
break;
}
}//end else loop for all commands except quit
}//end while loop all commands
}//end Main
/*---------------------------------------------------------
*Command: put
*Purpose: create variable at the hash point on
*the table and make it either an IntVar or FloatVar object
*Input: Command(type), key, and hash point
*Output: Error if variable has already been declared
--------------------------------------------------------*/
public static void put(String dataType, int key, int hash)
{
while(true)
{
if(table[hash] != null && table[hash].getName() == key)
{
System.out.println("ERROR VARIABLE ALREADY DECLARED");
break;
}
else if(table[hash] != null && table[hash].getName() != key)
{
if(itr.current.next == null)
table[hash] = new FloatVar(dataType, key, itr.retrieve());
else
{
list = new LinkedList();
list.insert(dataType, key, itr);
}
}
else
{
table[hash] = new FloatVar(dataType, key, null);
break;
}
}//end while
}//end method put
/*---------------------------------------------------------
*Command: assign
*Purpose: Assign a value to the object on the table
*Input: key, hash
*Output:
* * Error if variable has already been declared
* * Error if value mismatches type of object
* * Error if variable has not been assigned
--------------------------------------------------------*/
public static void assign(int key, int hash)
{
//search for key
while(table[hash] != null)
{
if(key == table[hash].getName())
{
if(table[hash].dataType.equalsIgnoreCase("INT"))
{
try
{
int value = scan.nextInt();
table[hash].setValue(scan.nextInt());
table[hash].setAssigned(true);
break;
}
catch(InputMismatchException e)
{
System.out.println("ERROR TYPE MISMATCH");
break;
}
}//end if
else if(table[hash].dataType.equalsIgnoreCase("FLOAT"))
{
try
{
double value = scan.nextDouble();
table[hash].setValue(value);
table[hash].setAssigned(true);
break;
}
catch(InputMismatchException e)
{
System.out.println("ERROR TYPE MISMATCH");
break;
}
}//end else
}//end if
else
{
System.out.println("ERROR VARIABLE NOT DECLARED");
break;
}
}//end while
}//end method assign
/*---------------------------------------------------------
*Command: print
*Purpose: print the object at the hash point
*Input: key and hash
*Output:
* * Error if variable has not been declared
* * Error if variable declared but has no value
--------------------------------------------------------*/
public static void print(int key, int hash)
{
while(true)
{
if(table[hash] != null && table[hash].wasAssigned)
{
System.out.println("PASS: "+table[hash].getName()+" "+table[hash].dataType);
break;
}
else if(table[hash] != null && table[hash].wasAssigned == false)
{
System.out.println("ERROR DECLARED BUT HAS NO VALUE");
break;
}
else
{
System.out.println("ERROR VARIABLE NOT DECLARED");
break;
}
}
}//end method print
static abstract class Variable
{
int name;
String dataType;
boolean wasAssigned;
Variable next;
Variable(String dataType, int name, Variable next)
{
this.name = name;
this.dataType = dataType;
this.wasAssigned = false;
this.next = next;
}
private int getName()
{
return this.name;
}
abstract void setValue(int value);
abstract void setValue(double value);
abstract void setDataType(String dataType);
abstract void setAssigned(boolean assign);
}
private static class IntVar extends Variable
{
int value;
IntVar(String dataType, int name, Variable next)
{
super(dataType, name, null);
}
void setValue(int value)
{
if(this.dataType == "FLOAT")
{
System.out.println("ERROR TYPE MISMATCH");
}
else
{
this.value = value;
this.dataType = "INT";
}
}
private int getValue()
{
return this.value;
}
void setValue(double value)
{
System.out.println("ERROR TYPE MISMATCH");
}
void setDataType(String dataType)
{
this.dataType = dataType;
}
void setAssigned(boolean assign)
{
this.wasAssigned = assign;
}
}
private static class FloatVar extends Variable
{
double value;
FloatVar(String dataType, int name, Variable next)
{
super(dataType, name, null);
}
void setValue(int value)
{
System.out.println("ERROR TYPE MISMATCH");
}
private double getValue()
{
return this.value;
}
void setValue(double value)
{
if(this.dataType == "INT")
{
System.out.println("ERROR TYPE MISMATCH");
}
else
{
this.value = value;
this.dataType = "FLOAT";
}
}
void setDataType(String dataType)
{
this.dataType = dataType;
}
void setAssigned(boolean assign)
{
this.wasAssigned = assign;
}
}
private static class LinkedList
{
Variable head;
private boolean isEmpty()
{
return head.next == null;
}
private void makeEmpty()
{
head.next = null;
}
private void insert(String type, int key, LinkedListIterator p)
{
if( p != null && p.current != null)
{
if(type.equalsIgnoreCase("INT"))
p.current.next = new IntVar(type, key, p.current.next);
else if(type.equalsIgnoreCase("FLOAT"))
p.current.next = new FloatVar(type, key, p.current.next);
}
}
}//end class LinkedList
private static class LinkedListIterator
{
Variable current;
LinkedListIterator(Variable theNode)
{
current = theNode;
}
private boolean isValid()
{
return current != null;
}
private Variable retrieve()
{
if(isValid())
{
return current;
}
else
return null;
}
private void advance()
{
if(isValid())
{
current = current.next;
}
}
}
}//end class Calculator
I love regexps for shell scripting and similar quick jobs, but if I find myself hardcoding them into more permanent programs I usually reevaluate my design... depends on the situation of course.
Monkey Ball WarriorA collection of mediocre hatsSeattle, WARegistered Userregular
edited November 2010
Most of my regexs in java look like this:
/* This reads Some Number, and maybe a dash, then some letters,
* then some letters, then maybe a number, then one dot, then
* either "txt" or "java"
*/
private static final String FILENAME_REGEX = "(@%*Y*&^*D^&[email protected]*&F*
$*(*GD*F&[email protected]&#G(HF*GB#@*OGF*&%[email protected]#(*FDOOOOMPIE*$!!"
You have to translate it into english in the comment, at the very least so if you messed up you can know what it was you were trying to do in the first place.
Monkey Ball Warrior on
"I resent the entire notion of a body as an ante and then raise you a generalized dissatisfaction with physicality itself" -- Tycho
0
KakodaimonosCode fondlerHelping the 1% get richerRegistered Userregular
edited November 2010
I always like the RFC 2822 Regex. That's a prime example of a WTF regex. But if I'm doing a lot of parsing and file spelunking, I usually would do it in a Regex. Though lately, I've been messing around with Clojure and Chukwa/Hadoop as a possibility for large data set work.
So my friend said there might be a job in his area on base, and this job would require the person to be good at Remedy. I've never heard of this language... Is there somewhere I can look to read about it?
So my friend said there might be a job in his area on base, and this job would require the person to be good at Remedy. I've never heard of this language... Is there somewhere I can look to read about it?
The only thing that I know of is actually a ticketing system, although there is probably some customization involved.
So my friend said there might be a job in his area on base, and this job would require the person to be good at Remedy. I've never heard of this language... Is there somewhere I can look to read about it?
I use it, creating tickets for broken computers, to be passed around to the people who actually fix them. It's pretty awful, things disappear all the time, and it creates a lot of extra work. On the plus side, if you can get it working decently then all your colleagues will love you.
So my friend said there might be a job in his area on base, and this job would require the person to be good at Remedy. I've never heard of this language... Is there somewhere I can look to read about it?
The only thing that I know of is actually a ticketing system, although there is probably some customization involved.
This is almost certainly what it is. Remedy is pretty highly customizable so that it can be tailored to the organization using it. I did a tiny bit of work with it years ago building a searchable knowledgebase in it for a helpdesk I was consulting for. I don't really remember anything other than that I did it at this point and that I'm certain I barely scratched the surface.
I think there's a lot of customizing what fields are displayed, what they're named, probably database back end design, etc.
Of course, depending on the job "know remedy" means "can enter trouble tickets into a ticket tracking system, update them, and close them". I've seen this listed for many helpdesk jobs and I have no idea why, because it's like 10 minutes of on the job training that you're going to get whether you've done it before or not and any moron can do it.
as a side note, I wonder if I can make a single post in this thread without editing it.
So I'm taking a community college intro to programming with C++ class right now. I'm starting to get the appeal of coding. As obtuse as the language can be sometimes and I sometimes feel like its going over my head, I AM slowly learning it. I just spent the last 4 hours working on one problem for the homework, and by god it fucking works. Shit yeah. Nothing like pure unadulterated satisfaction to end my night before going to bed.
I always like the RFC 2822 Regex. That's a prime example of a WTF regex. But if I'm doing a lot of parsing and file spelunking, I usually would do it in a Regex. Though lately, I've been messing around with Clojure and Chukwa/Hadoop as a possibility for large data set work.
So I'm taking a community college intro to programming with C++ class right now. I'm starting to get the appeal of coding. As obtuse as the language can be sometimes and I sometimes feel like its going over my head, I AM slowly learning it. I just spent the last 4 hours working on one problem for the homework, and by god it fucking works. Shit yeah. Nothing like pure unadulterated satisfaction to end my night before going to bed.
It is the best feeling. Especially at 4 AM.
bowen on
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
I always like the RFC 2822 Regex. That's a prime example of a WTF regex. But if I'm doing a lot of parsing and file spelunking, I usually would do it in a Regex. Though lately, I've been messing around with Clojure and Chukwa/Hadoop as a possibility for large data set work.
I always like the RFC 2822 Regex. That's a prime example of a WTF regex. But if I'm doing a lot of parsing and file spelunking, I usually would do it in a Regex. Though lately, I've been messing around with Clojure and Chukwa/Hadoop as a possibility for large data set work.
Spoilered for OMG:
...I think I see a face.
Really? I see a sailboat.
It's a schooner.
Tofystedeth on
0
Monkey Ball WarriorA collection of mediocre hatsSeattle, WARegistered Userregular
edited November 2010
Suprisingly hard to find this out via google, perhaps I'm asking it the wrong question.
If I do something like this in Java
boolean var = true;
var |= someMethod();
Will that even bother to evaluate someMethod()? I believe this concept is called lazy evaluation? I know this works on || and &&...
Monkey Ball Warrior on
"I resent the entire notion of a body as an ante and then raise you a generalized dissatisfaction with physicality itself" -- Tycho
I understand the difference between the two and I understand the arguments, but I gotta say... the ease of use and scalability of Flash really overshadows the complexity and unstable nature of Ajax methods, (mainly due to browser support issues) and it's almost forcing me to use Flash for simple dynamic information and that upsets me because the major drawback to flash is the ability to "edit on the fly" when changes need to be made. I WANT to use HTML output.
This cross-browser shit is really really starting to piss me off as newer better technologies are taking fooooooreeevvveerr to be implimented, and then of course are broken in their various forms.
I'm not using framework specific setups. In my quest to educate myself I am doing a lot from scratch, and the sheer volume of development time and code to do simple things just boggles my mind.
I have tinkered with jQuery and Prototype though.
For standard in-line editing and simple field updating it's pretty "easy" but when you start getting into full interactivity, dragging and dropping with dynamic cell distribution and display it becomes a whole ballgame of "does it look like shit in IE." (I fucking hate IE)
It's just frustrating because i'm trying to force myself to use the "popular working" methods, and after 7 years of working with Flash I am dealing with growing pains. I have a lot of love and respect for people who do this for a living, lol.
I'm not using framework specific setups. In my quest to educate myself I am doing a lot from scratch, and the sheer volume of development time and code to do simple things just boggles my mind.
I have tinkered with jQuery and Prototype though.
For standard in-line editing and simple field updating it's pretty "easy" but when you start getting into full interactivity, dragging and dropping with dynamic cell distribution and display it becomes a whole ballgame of "does it look like shit in IE." (I fucking hate IE)
It's just frustrating because i'm trying to force myself to use the "popular working" methods, and after 7 years of working with Flash I am dealing with growing pains. I have a lot of love and respect for people who do this for a living, lol.
Use the jQuery ui library. They've solved 99.9% of those cross-browser DOM issues you're needlessly beating your head against.
Standards-compliant animation is easy mode, with the right tools.
I'm not using framework specific setups. In my quest to educate myself I am doing a lot from scratch, and the sheer volume of development time and code to do simple things just boggles my mind.
I have tinkered with jQuery and Prototype though.
For standard in-line editing and simple field updating it's pretty "easy" but when you start getting into full interactivity, dragging and dropping with dynamic cell distribution and display it becomes a whole ballgame of "does it look like shit in IE." (I fucking hate IE)
It's just frustrating because i'm trying to force myself to use the "popular working" methods, and after 7 years of working with Flash I am dealing with growing pains. I have a lot of love and respect for people who do this for a living, lol.
The "popular working" methods are using jQuery, prototype, moo tools, etc.
I think starting with doing it by hand as you've done is the way to go, but now that you understand how it works and that it's a total pain in the balls? Grab a framework and make it enjoyable again.
Posts
my $dbh; # GOTO is gone. You're also not continually reconnecting to the DB even when already #connected anymore, which will make your DB server admins happy # Are you sure you really want to be doing this, though? That's still pretty nasty if it can't connect. while(!$dbh) { #eval is like a try{} in other languages... kind of. # it also suppresses warnings and stuff going to stderr # you can get the return code if you need it. eval{$dbh = DBI->connect($dsn,$user,$pass);} } # Move your prepare outside of the loop. Inside the loop is defeating the point # of using prepare() in the first place. # also, don't use the ; at the end of your sql statement. It doesn't actually hurt anything # but is still generally considered wrong to do my $sth = $dbh->prepare ('select id, UNIX_TIMESTAMP(last_active) from machine_state'); for (;;) { $sth->execute; while(my @row = $sth->fetchrow_array()) { $current_time = time; if (($row[1] != 0)) { if ($current_time > ($row[1]+45)) {turn_off($row[0]);} } } sleep(5); }Read about eval because there are some other things you may need to know about it.
prepare() should be called outside of loops. If you prepare outside of the loop and then just execute() inside of the loop you save the db server some work setting up the execution plan repeatedly and some network traffic and so on.
You should also really try to use the strict pragma and be sure to declare your variables with my (in most cases... on rare occasions you'll use our). This makes Perl behave like a proper programming language with sane variable scoping so that crazy things don't start happening as your script inevitably gets longer and more complex.
You probably don't want DBI->connect() being called repeatedly if it can't connect. If it can't connect, there's likely a problem and hammering the server and blowing cpu cycles on the client like that isn't going to help anything and can make matters worse.
I would put that while() loop with the DBI->connect() with a sleep() or wait() that steps up. Something like after first failure to connet wait 5 seconds, then 30 seconds, then 1 minute, then 5 minutes, then 10 minutes, then 30 minutes, then 1 hour and hold at retrying every hour until the script is manually restarted.
I need to do a word count on some text. Thats it.
The thing is, I can't figure out how to deal with stuff like double spaces, and what exactly defines a word.
(In java)
If anybody could point me in the right direction I would be very appreciative.
I don't do them enough in java to quickly whip up an example, but basically for a very naive system you could just count the number of matches to "<beginning of line><any characters except whitespace>" or "<space><any characters except whitespace>" . There's a special char, \s, which will match space or tab whitespace so it's nice and easy to just match "whitespace" in general, and \S matches non-whitespace, assuming the "group of characters between spaces" definition of word. If it's the other, regexp has a way to count characters.
I suspect Java has a way to just count the number of matches using the Matcher class (as opposed to String.matches()), but I haven't looked at the docs for it.
In a worst case scenario it probably has a way to just dump the matches into an array, but depending on just how large this text is, you may not want to do that. In an even worse scenario, you could even do that with String.split() and split on whitespace and dump to an array. I would start with the regular expressions and seeing if the Matcher class has a method to just return the number of matches, though, because that's almost certainly the better way.
When dealing with removing words from a sentence you do want to use a regular expression. Look at \b and \B which detect word and non word boundaries.
God, I always forget about \b and \B. Note that \b doesn't count ' (apostrophe) or - (hyphen) as a word character and so Jaziek's would be 2 words without a little bit more trickery using groups between []
I'd still look into using the actual regex stuff unless this is like intro to java programming class and this week's chapter is on the String class since using that means dumping it into an array whereas the Matcher class has a groupCount() method (I just checked) which returns the number of matches.
\b and \B will do what he wants aside from the caveat I mentioned above. They don't match on the computer definition of word, they are like \w and \W except in java they support unicode and \w and \W don't.
Of course my solution of matching on non-space characters could also give problematic results if things like & are not supposed to count as a word.
I'll have input that has the command then the variable to be used.
Ex: INT v298
I need the scanner to separate the char from the int.
Do I just have my scanner scan the line, pass it to a string tokenizer which separates the two strings (1 the command, 2 the variable), then have the tokenizer also separate the char from the int, or do I just use a delimiter to ignore the v? If I do that, how do I set it up?
Would it be like so:
Scanner scan = new Scanner(System.in).useDelimiter("v");
In particular, I would recommend against using Java regexes to match across line boundaries unless you are familiar with Java's regex engine. For example, Java does not normalize line breaks into \n like C does, and it has a few other oddities.
Because they look like someone took a dump on your monitor. The longer or more complex the regex string, the worse it gets.
Long regex expressions can't be parsed by a human without spending an inordinate amount of time trying to figure out what it's supposed to do. It's like the worst perl code you've ever seen.
They're good at what they do, but that doesn't mean you have to like how it looks.
I find them great when used for matching. For example I wouldn't want to write my own file extension matching code. The key to writing regexp is to document the intent and use cases of what passes. I also split the regexp into smaller pieces that I document.
I have a new question that needs fresh eyes. I'm trying to create a hashtable without being allowed to use the java hash table already implemented at all. So I'm trying to do chaining on my array.
The first problem is, I believe the list and listiterator I'm using is only creating one list and adding elements only to that one list. I need it to be for every slot on the array so it functions like a hash table w/ chaining. Do I change my array from holding objects, to holding lists instead?
The second problem is, since my variable class is abstract and the actual nodes placed onto the array at the moment are classes that extend variable (like IntVar or FloatVar), I'm not able to use my getValue() methods for those specific classes. What am I doing wrong? Do I need a specific get method in my abstract class first?
My teacher gave me this advice:
But for my abstract class, if I put setValue(so so), and my other classes need to overwrite this method, how do I determine which value it is. Same with getValue? Do I have an if statement in my abstract class that says:
If(dataType.equals("INT"))
do this
else
do this other thing
My code so far:
import java.util.*; public class Calculator { private static final int DEFAULT_SIZE = 128; private static Variable[] table; private static LinkedListIterator itr; private static LinkedList list; private static String command; private static Scanner scan; public static void main(String[] args) { scan = new Scanner(System.in); StringTokenizer st; table = new Variable[DEFAULT_SIZE]; for(int i = 0; i < DEFAULT_SIZE; i++) { table[i] = null; }//end table setup while(true) { command = scan.next(); if(command.equalsIgnoreCase("QUIT")) { break; } else { String temp = scan.next(); st = new StringTokenizer(temp, ("((v))")); Integer tempKey = new Integer(st.nextToken()); int key = tempKey.intValue(); /*--------------------------------------------------- * BEGIN LISTING COMMANDS --------------------------------------------------*/ if(command.equalsIgnoreCase("INT")) { int hash = (key % DEFAULT_SIZE); put(command, key, hash); }//end if command == INT else if(command.equalsIgnoreCase("FLOAT")) { int hash = (key % DEFAULT_SIZE); put(command, key, hash); }//end command FLOAT else if(command.equalsIgnoreCase("ASSIGN")) { int hash = (key % DEFAULT_SIZE); assign(key, hash); }//end command Assign else if(command.equalsIgnoreCase("ADDASSIGN")) { } else if(command.equalsIgnoreCase("PRINT")) { int hash = (key % DEFAULT_SIZE); print(key, hash); } else if(command.equalsIgnoreCase("INT2FLOAT")) { } else if(command.equalsIgnoreCase("FLOAT2INT")) { } else if(command.equalsIgnoreCase("QUIT")) { break; } }//end else loop for all commands except quit }//end while loop all commands }//end Main /*--------------------------------------------------------- *Command: put *Purpose: create variable at the hash point on *the table and make it either an IntVar or FloatVar object *Input: Command(type), key, and hash point *Output: Error if variable has already been declared --------------------------------------------------------*/ public static void put(String dataType, int key, int hash) { while(true) { if(table[hash] != null && table[hash].getName() == key) { System.out.println("ERROR VARIABLE ALREADY DECLARED"); break; } else if(table[hash] != null && table[hash].getName() != key) { if(itr.current.next == null) table[hash] = new FloatVar(dataType, key, itr.retrieve()); else { list = new LinkedList(); list.insert(dataType, key, itr); } } else { table[hash] = new FloatVar(dataType, key, null); break; } }//end while }//end method put /*--------------------------------------------------------- *Command: assign *Purpose: Assign a value to the object on the table *Input: key, hash *Output: * * Error if variable has already been declared * * Error if value mismatches type of object * * Error if variable has not been assigned --------------------------------------------------------*/ public static void assign(int key, int hash) { //search for key while(table[hash] != null) { if(key == table[hash].getName()) { if(table[hash].dataType.equalsIgnoreCase("INT")) { try { int value = scan.nextInt(); table[hash].setValue(scan.nextInt()); table[hash].setAssigned(true); break; } catch(InputMismatchException e) { System.out.println("ERROR TYPE MISMATCH"); break; } }//end if else if(table[hash].dataType.equalsIgnoreCase("FLOAT")) { try { double value = scan.nextDouble(); table[hash].setValue(value); table[hash].setAssigned(true); break; } catch(InputMismatchException e) { System.out.println("ERROR TYPE MISMATCH"); break; } }//end else }//end if else { System.out.println("ERROR VARIABLE NOT DECLARED"); break; } }//end while }//end method assign /*--------------------------------------------------------- *Command: print *Purpose: print the object at the hash point *Input: key and hash *Output: * * Error if variable has not been declared * * Error if variable declared but has no value --------------------------------------------------------*/ public static void print(int key, int hash) { while(true) { if(table[hash] != null && table[hash].wasAssigned) { System.out.println("PASS: "+table[hash].getName()+" "+table[hash].dataType); break; } else if(table[hash] != null && table[hash].wasAssigned == false) { System.out.println("ERROR DECLARED BUT HAS NO VALUE"); break; } else { System.out.println("ERROR VARIABLE NOT DECLARED"); break; } } }//end method print static abstract class Variable { int name; String dataType; boolean wasAssigned; Variable next; Variable(String dataType, int name, Variable next) { this.name = name; this.dataType = dataType; this.wasAssigned = false; this.next = next; } private int getName() { return this.name; } abstract void setValue(int value); abstract void setValue(double value); abstract void setDataType(String dataType); abstract void setAssigned(boolean assign); } private static class IntVar extends Variable { int value; IntVar(String dataType, int name, Variable next) { super(dataType, name, null); } void setValue(int value) { if(this.dataType == "FLOAT") { System.out.println("ERROR TYPE MISMATCH"); } else { this.value = value; this.dataType = "INT"; } } private int getValue() { return this.value; } void setValue(double value) { System.out.println("ERROR TYPE MISMATCH"); } void setDataType(String dataType) { this.dataType = dataType; } void setAssigned(boolean assign) { this.wasAssigned = assign; } } private static class FloatVar extends Variable { double value; FloatVar(String dataType, int name, Variable next) { super(dataType, name, null); } void setValue(int value) { System.out.println("ERROR TYPE MISMATCH"); } private double getValue() { return this.value; } void setValue(double value) { if(this.dataType == "INT") { System.out.println("ERROR TYPE MISMATCH"); } else { this.value = value; this.dataType = "FLOAT"; } } void setDataType(String dataType) { this.dataType = dataType; } void setAssigned(boolean assign) { this.wasAssigned = assign; } } private static class LinkedList { Variable head; private boolean isEmpty() { return head.next == null; } private void makeEmpty() { head.next = null; } private void insert(String type, int key, LinkedListIterator p) { if( p != null && p.current != null) { if(type.equalsIgnoreCase("INT")) p.current.next = new IntVar(type, key, p.current.next); else if(type.equalsIgnoreCase("FLOAT")) p.current.next = new FloatVar(type, key, p.current.next); } } }//end class LinkedList private static class LinkedListIterator { Variable current; LinkedListIterator(Variable theNode) { current = theNode; } private boolean isValid() { return current != null; } private Variable retrieve() { if(isValid()) { return current; } else return null; } private void advance() { if(isValid()) { current = current.next; } } } }//end class CalculatorYou have to translate it into english in the comment, at the very least so if you messed up you can know what it was you were trying to do in the first place.
Spoilered for OMG:
The only thing that I know of is actually a ticketing system, although there is probably some customization involved.
See here.
I use it, creating tickets for broken computers, to be passed around to the people who actually fix them. It's pretty awful, things disappear all the time, and it creates a lot of extra work. On the plus side, if you can get it working decently then all your colleagues will love you.
I think there's a lot of customizing what fields are displayed, what they're named, probably database back end design, etc.
Of course, depending on the job "know remedy" means "can enter trouble tickets into a ticket tracking system, update them, and close them". I've seen this listed for many helpdesk jobs and I have no idea why, because it's like 10 minutes of on the job training that you're going to get whether you've done it before or not and any moron can do it.
as a side note, I wonder if I can make a single post in this thread without editing it.
...I think I see a face.
Switch: SW-3515-0057-3813 FF XIV: Q'vehn Tia
It is the best feeling. Especially at 4 AM.
Really? I see a sailboat.
It's a schooner.
If I do something like this in Java
Will that even bother to evaluate someMethod()? I believe this concept is called lazy evaluation? I know this works on || and &&...
Damn, but ||= doesn't exist. Guess I'll have to type out exactly what I mean.
I understand the difference between the two and I understand the arguments, but I gotta say... the ease of use and scalability of Flash really overshadows the complexity and unstable nature of Ajax methods, (mainly due to browser support issues) and it's almost forcing me to use Flash for simple dynamic information and that upsets me because the major drawback to flash is the ability to "edit on the fly" when changes need to be made. I WANT to use HTML output.
This cross-browser shit is really really starting to piss me off as newer better technologies are taking fooooooreeevvveerr to be implimented, and then of course are broken in their various forms.
</endrant>
I have tinkered with jQuery and Prototype though.
For standard in-line editing and simple field updating it's pretty "easy" but when you start getting into full interactivity, dragging and dropping with dynamic cell distribution and display it becomes a whole ballgame of "does it look like shit in IE." (I fucking hate IE)
It's just frustrating because i'm trying to force myself to use the "popular working" methods, and after 7 years of working with Flash I am dealing with growing pains. I have a lot of love and respect for people who do this for a living, lol.
Standards-compliant animation is easy mode, with the right tools.
I think starting with doing it by hand as you've done is the way to go, but now that you understand how it works and that it's a total pain in the balls? Grab a framework and make it enjoyable again.