Any ideas or implementations floating about for encoding the current date including the milliseconds into the shortest possible string length?
e.g I want 31/10/2011 10:41:45 in the shortest string possible (ideally 5 characters) - obviously decodable.
If it is impossible to get down to 5 characters, then the year is optional.
edit: it doesn't actually need to be decodable. It just needs to be a unique string.
An time_t is 31 bits. Add 10 bits for up to 1000 milliseconds: That's 41 bits. You want 5 characters: That's 8 bits for the 1st 4 characters + 9 bits for the last one.
Using Chinese ideograms, you should easily be able to find a range of 256 consecutive chars for each of the 1st 4 chars and a range of 512 for the last one.
Needless to say your encoded date will look... chinese! But it should do the trick ;-)
BTW, you don't have to stick to Chinese. You might even want to choose a different Unicode 256 chars range for each character. Of course, you'll want to find sequences of 256/512 printable chars.
Now let's say we skip the year. We're down to 86400 x 366 seconds per year = 31622400 seconds. Including millisecs : 31622400000. That's 35 bits. Great: We're down at 7 bits per character. Easy! :-)
you can use the Ticks:
var ticks = System.DateTime.Now.Ticks;
this is a 64bit number. You get the Time back by calling:
var timeBack = new System.DateTime(ticks);
of course this are 8 bytes but I don't think you can get this more compact (easily).
No can do: The total ms in an year (365 days) is 31,536,000,000 (=365*24*60*60*1000). You need 34.87628063 bits of information to store that value (log2 31,536,000,000). You probably meant "printable characters" BUT you would need 7 bits/character to store 35 bits in 5 characters. As an example base64 is 6 bits/character of information, so 6 characters. Ascii85 would be a little better, but still you would need around 5.5 characters, so 6 characters.
Clearly if you meant 5 BYTES, everything changes. You can store 34.84 years (in ms) in that space.
And if you meant 5 C# PRINTABLE AND UNPRINTABLE CHARACTERS (each C# character is 16 bits), then it's even better. 10 bytes! DateTime in C# is only 8 bytes and it uses ticks (they are a VERY VERY VERY small part of a second)!
BUT if you meant 5 C# PRINTABLE CHARACTERS characters, then use Serge's response. It's very good and show us that the world is a big place (and show us that why good questions are so much important: they let us see the world in new ways).
You can use ASCII characters to represent the numbers and drop the formatting, for example:
31/10/2011 10:41:45
*/*/** *:*:*
*******
That's 7, you can drop 2 if you don't want to include the full year. Obviously the * are actual characters relating to a number, A could be 1 etc, or even use the proper ASCII codes.
Related
When processing a file from a telecom company, I came across this in the specifications :
When reading in that data, how can I convert that format to something usable in c# ? I have no idea what MMMMMMMMSS format is !!
The only logical explanation I can think of is the following:
Since this is a call duration representation, let's say that a call duration was 10:10:5. I assume they want to represent this in minutes and seconds only. Hence considering the given format, it would be represented like this: 61005 which is 610 minutes and 5 seconds, then the 5 remaining bytes can be filled with trailing zeros, or with space characters (since you mentioned that's what they used to represent a value).
Hope that helps.
I would expect each of these to be zero-padded. Regardless, Split the last two characters off to derive seconds and cents, respectively. The first 8 characters represent minutes and dollars. A one minute (exactly) call would be 7 zeros followed by a 1 followed by two zeros. A ten minute and ten second call would be 6 zeros followed by 1010.
I need to encode current datetime into some unique string to store it in database.
I found this article how to generate a unique token which expires after 24 hours? but for me generated token is to long (34 symbols)
Is there some other similar way to encode shorter string?
Perfect size <= 10 symbols.
I'd ask why? What are you trying to do? If you want to timestamp something like a log entry then just use the datetime value - every database type I know has a built in date/time type.
If you're trying to generate a unique Id for use as something like a primary key then this would be a bad idea - I've yet to find a good case for using a date based unique id.
It would be much better to have an auto-incrementing integer or even GUID value. If you wanted you could then add a timestamp column to the database
You can use a tick (DateTime.Ticks for instance) but don't store the tick as simple string, encode the bits. If you use a long tick (64bit) you should consider ASCII85 encoding of the bytes so it wont exceed 10 symbols.
var tickBytes = BitConverter.GetBytes(DateTime.UtcNow.Ticks);
string encodedTicks = new Ascii85().Encode(tickBytes);
If you chose a 32bit tick, base 64 should be fine.
For a readable tick with precision to second (less precise than the previous solution)
long origin = new DateTime(2014, 7, 24).Ticks / TimeSpan.TicksPerSecond;
long customTicks = (DateTime.UtcNow.Ticks / TimeSpan.TicksPerSecond) - origin;
string readableTicks = customTicks.ToString(CultureInfo.InvariantCulture);
That will stay on 10 chars or less for ~300 years.
Okay, if you want it from "about now" to some point in the future, and you want seconds granularity, and you want ASCII symbols, let's assume base64.
With 8 characters of base64, we can encode 6 bytes of data. That will give us 248 different values, which allows about 9 million years-worth of seconds. Given that range, we might as well use the DateTime.Ticks property and divide by ticks-per-second, not worrying about the epoch. Full code coming later if you want it, but as a list of steps:
Take DateTime.UtcNow.Ticks
Divide by TimeSpan.TicksPerSecond
Convert the result into a byte[], e.g. with BitConverter.GetBytes(long)
Encode the least-significant 6 bits (I'm hopeless with endianness - either the first or last 6 bytes of the byte[] as base64 using Convert.ToBase64String
I have a website where you can buy stuff, and we want to format the orderID that goes to our portal in certain way. I am using the string.format method to format it like this:
Portal.OrderID = string.Format( "{0}{1:0000000}-{2:000}",
"Z",
this.Order.OrderID,
"000");
So we want it to look like this basically Z0545698-001. My question is, if I am using string.format will it blow up if this.Order.OrderID is greater than 7 characters?
If so, how can I keep the same formatting (i.e. Z 1234567 - 000) but have the first set of numbers (the 1-7) be a minimum of 7 (with any numbers less than 7 in length have leading 0's). And then have anything greater than 7 in length just extend the formatting so I could get an order number like Z12345678-001?
how can I keep the same formatting (i.e. Z 1234567 - 000) but have the first set of numbers (the 1-7) be a minimum of 7 (with any numbers less than 7 in length have leading 0's). And then have anything greater than 7 in length just extend the formatting so I could get an order number like Z12345678-001?
Use exactly the code that you have, because that's what it does.
I want to create pin codes and serial numbers for scratch papers , I have already generated unique 10 digit numbers , now I want to turn that 10 digit number to a 16 digit number (with check digit in the end) . The thing is that the function that does this should be reversible so by seeing the 16 digit number I can check whether it is valid or not .(if it is not generated by me it should not be valid) .
this is how I have generated the 10 digit unique random codes :
Guid PinGuid;
byte[] Arr;
UInt32 PINnum = 0;
while (PINnum.ToString().Length != 10)
{
PinGuid = Guid.NewGuid();
Arr = PinGuid.ToByteArray();
PINnum = BitConverter.ToUInt32(Arr, 0);
}
return PINnum.ToString();
I would be grateful if you can give me a hint on how to do it .
First off, I would avoid GUID since some prefixes are reserved for special applications. Which means that these areas of the GUID may not be allocated uniformly on creation, so you may not get exactly 10 digits of randomness like you plan.
Also since your loop waits for the GUID to become the right size you could do it more efficiently.
10 digits = 10**10
Log_2(10) = approx 3322/1000
So you need approx 33 bits for 10 digit number. Since you want your number to be exactly 10 digits, you can either pad numbers less than 10^10 with leading zeroes, or you can generate only numbers between 10^9 and 10^10 - 1.
If you take the latter case you need 9*10^9 numbers in your space -- giving you all numbers from 1 followed by nine zeroes up to 9 followed by 9 9s.
Then you would like to convert this space of numbers into a larger space, to expand it by a factor of 5 and include one more digit as a check digit.
Pick a check digit function as anything you like. You could simply sum (mod 10) the original 10 digits, or choose something more complicated.
Presumably you do not want people to be able to generate valid instances. So if you are really serious about your security, you should modify any suggestions you get from the net before deploying them.
I would do something along the lines of :
Generate a uniform 10digit number with no leading zeroes by
randomTenDigits = 10**9 + rand(9*10**9)
Using an encryption scheme (like AES 256 or even RSA or El-Gamal since their slower speed will no be so important since input length is small ) encrypt this 10 digit number using a secret key only you and others you trust are aware of. Perhaps you can concatenate the 10 digit number 10 times, and then concatenate that result with some other secret that you choose, and then finally encrypt this expanded secret of which the 10 digit number is a part.
Take some choice 5 digits (around 17 bits) of the resulting ciphertext, and append these to your 10 digit number.
Generate 1 digit of check digit by whatever method you desire.
As you will note the real security of this scheme is not from a check digit, it is from the secret key you can use to authenticate the 16 digit number. The test you will use to authenticate it is: does the given 10 digit number when concatenated with other secrets I have, encrypt, using a secret key only I know, to the given 5 digit number presented with it.
Since the difficulty for an attacker of forging one of your numbers depends on the difficulty of
discovering your secret keys and other info
discovering which method of encryption you use
discovering which part of the resulting cipher text you emit for the 5 digit secret, or
simply brute forcing the 5 digits to discover the correct pairing, and since 5 digits is not a big space to search, I would suggest instead generating larger numbers. 10 or 16 digits is not really a huge space to search. So instead of digits I would use upper and lower case letters plus digits plus space and full stop to give you 64 letters in your alphabet. Then if you used 16 you get around 96 bits of security.
However if numbers are non-negotiable and the size of 10 digits for your base space is also non-negotiable, doing it this way is probably the most secure. You may be able to set up your system to deter people from brute forcing it, though you should consider what if someone acquires a piece of your hardware through a vendor. I believe it is easier to design security in rather than design in a mechanism for detecting people trying to brute force query your system.
However if serious dough is on the line ( like millions ) the security you employ should really be first class. Equivalent to the kind of security you would employ to protect a pin number to a million dollar bank account. The more secure you are the longer you can carry on your biz with credibility and trust.
So along these lines I would suggest increasing the size of your secrets to make it infeasible for someone to simply try all combinations and forge a valid one, and in particular thinking about how to design your system to make it difficult to break for people with lots of skills and motivation (money). You really can't be too careful.
I would keep it simple. Put PINnum.ToString() into a buffer. Place a filler digit at 5 intervals. The first four could be random garbage and the last could be a check digit, or you could make each filler a check digit for its section. Here is an example.
buf = PINnum.ToString();
int chkdgit = function to create your checkdigit
Random rnd = new Random();
int i = rnd.Next(1001,9999);
fillbuf = i.toString();
return buf[0] + buf[1] + fillbuf[0] + buf[2] .... chkdgit.toString();
its a rather simple approach, but if your security needs aren't at level 1, it might suffice
In this code I am debugging, I have this code snipit:
ddlExpYear.SelectedItem.Value.Substring(2).PadLeft(2, '0');
What does this return? I really can't run this too much as it is part of a live credit card application. The DropDownList as you could imagine from the name contains the 4-digit year.
UPDATE: Thanks everyone. I don't do a lot of .NET development so setting up a quick test isn't as quick for me.
It takes the last two digits of the year and pads the left side with zeroes to a maximum of 2 characters. Looks like a "just in case" for expiration years ending in 08, 07, etc., making sure that the leading zero is present.
This prints "98" to the console.
class Program
{
static void Main(string[] args)
{
Console.Write("1998".Substring(2).PadLeft(2, '0'));
Console.Read();
}
}
Of course you can run this. You just can't run it in the application you're debugging. To find out what it's doing, and not just what it looks like it's doing, make a new web application, put in a DropDownList, put a few static years in it, and then put in the code you've mentioned and see what it does. Then you'll know for certain.
something stupid. It's getting the value of the selected item and taking the everything after the first two characters. If that is only one character, then it adds a '0' to the beginning of it, and if it is zero characters, the it returns '00'. The reason I say this is stupid is because if you need the value to be two characters long, why not just set it like that to begin with when you are creating the drop down list?
It looks like it's grabbing the substring from the 3rd character (if 0 based) to the end, then if the substring has a length less than 2 it's making the length equal to 2 by adding 0 to the left side.
PadLeft ensures that you receive at least two characters from the input, padding the input (on the left side) with the appropriate character. So input, in this case, might be 12. You get "12" back. Or input might be 9, in which case, you get "09" back.
This is an example of complex chaining (see "Is there any benefit in Chaining" post) gone awry, and making code appear overly complex.
The substring returns the value with the first two characters skipped, the padleft pads the result with leading zeros:
string s = "2014";
MessageBox.Show(s.Substring(2).PadLeft(2, 'x')); //14
string s2 = "14";
MessageBox.Show(s2.Substring(2).PadLeft(2, 'x')); //xx
My guess is the code is trying to convert the year to a 2 digit value.
The PadLeft only does something if the user enters a year that is either 2 or 3 digits long.
With a 1-digit year, you get an exception (Subsring errs).
With a 2-digit year (07, 08, etc), it will return 00. I would say this is an error.
With a 3-digit year (207, 208), which the author may have assumed to be typos, it would return the last digit padded with a zero -- 207 -> 07; 208 -> 08.
As long as the user must choose a year and isn't allowed to enter a year, the PadLeft is unnecessary -- the Substring(2) does exactly what you need given a 4-digit year.
This code seems to be trying to grab a 2 digit year from a four digit year (ddlexpyear is the hint)
It takes strings and returns strings, so I will eschew the string delimiters:
1998 -> 98
2000 -> 00
2001 -> 01
2012 -> 12
Problem is that it doesn't do a good job. In these cases, the padding doesn't actually help. Removing the pad code does not affect the cases it gets correct.
So the code works (with or without the pad) for 4 digit years, what does it do for strings of other lengths?
null: exception
0: exception
1: exception
2: always returns "00". e.g. the year 49 (when the Jews were expulsed from rome) becomes "00". This is bad.
3: saves the last digit, and puts a "0" in front of it. Correct in 10% of cases (when the second digit is actually a zero, like 304, or 908), but quite wrong in the remainder (like 915, 423, and 110)
5: just saves the 3rd and 4th digits, which is also wrong, "10549" should probably be "49" but is instead "54".
as you can expect the problem continues in higher digits.
OK so it's taking the value from the drop down, ABCD
Then it takes the substring from position 2, CD
And then it err, left pads it with 2 zeros if it needs too, CD
Or, if you've just ended X, then it would substring to X and pad to OX
It's taking the last two digits of the year, then pad to the left with a "0".
So 2010 would be 10, 2009 would be 09.
Not sure why the developer didn't just set the value on the dropdown to the last two digits, or why you would need to left pad it (unless you were dealing with years 0-9 AD).