Apologise for a vague title.
What is the best way to pass a finite number of properties(e.g. read, write, execute, ...). I think I could do this like these permissions - count the sum of some numbers assigned to properties - 1,2,4,8,16... Is it the best easy to do this and what algorithm should I use to get the summands - for example if I have 17 - how could I calculate that this is 16 and 1?
Thanks.
You could use Bitwise Operators. This way you can compare the actual bits of an integer, Here's how it works:
Bit Pattern
8 4 2 1
1 0 0 0 = 8
0 1 1 1 = 7
0 1 1 0 = 6
0 1 0 1 = 5
0 1 0 0 = 4
0 0 1 1 = 3
0 0 1 0 = 2
0 0 0 1 = 1
0 0 0 0 = 0
So,
6 & 4 = 4
Means show me the bits that are present in both 6 and 4.
Compare '0 1 1 0' and '0 1 0 0' and you will see that the second bit '4' is the only bit populated in both, so return 4.
Or
7 & 3 = 3
Means show me the bits that are present in both 7 and 3.
Compare '0 1 1 1' and '0 0 1 1' and you will see that the third (2) and fourth (1) bits are populated in both, so return 3 (2 + 1).
Related
I've array MxN of zeros (0). I need to fill it with 20 truths (1) by random postions.
An example
1 1 1 0 1 0 1
0 0 0 0 0 0 0
0 1 0 1 0 1 1
0 0 0 0 1 1 1
0 0 1 0 0 0 0
0 0 1 0 1 1 0
1 1 0 1 1 0 0
I fill it by non-determine algorithm:
$amountUnits = 20;
while($amountUnits > 0) {
$i = rand(0, $M-1);
$j = rand(0, $N-1);
if(!$grid[$i][$j]) {
$grid[$i][$j] = 1;
$amountUnits--;
}
}
But it's solved in random time. How can I fill it throught seed or something else? And where I can read about this problem?
Please excuse my bad English.
Thank you!
Fill a 1-dimensional array x times with 1 and the rest with 0. shuffle () shuffles the array.
array_chunk divides the array so that a matrix (a 2-dimensional array) is created.
$rows = 7;
$cols = 7;
$fillTrue = 20;
$arr = array_fill(0,$fillTrue,1)+array_fill($fillTrue, $rows*$cols-$fillTrue,0);
shuffle($arr);
$arr = array_chunk($arr,$cols);
Updated
So it turns out that PHP has a shuffle function, and #jspit's answer shows you how to use it. His answer is the php way to implement my answer.
Original Answer
Fill the array with 0's, and then add all of the 1's in the first rows and columns. So if you have an array that's 4 rows and 5 columns, and you want 7 1's, it would look like this:
1 1 1 1 1
1 1 0 0 0
0 0 0 0 0
0 0 0 0 0
Now, do a Fischer-Yates shuffle to distribute the 1's around randomly.
Normally, you do that shuffle on a 1-dimensional array. But you can map a 1-dimensional array index to two dimensions with some simple arithmetic, as shown below in pseudo-code. (Sorry, but I'm not really fluent in php.)
// shuffle a 2-dimensional array
for (i = numRows * numCols; i > 1; --i)
j = random(0, i) // picks a number from 0 to i-1
// get row and column for i
irow = (i-1) / numCols
icol = (i-1) % numCols
// get row and column for j
jrow = j / numCols
jcol = j % numCols
// and swap the items
temp = array[irow, icol]
array[irow, icol] = array[jrow, jcol]
array[jrow, jcol] = temp
It seems like you could do this (since it's really just a chopped up string of random digits?):
function truths(int $truths, int $many): array {
$facts = array_fill(0, $truths, true);
$statements = array_fill(0, $many, false);
foreach ($statements as $assertion => $statement) {
$statements[$assertion] = $facts[$assertion] ?? false;
}
shuffle($statements);
return $statements;
}
Then you give a function the number of truths and grid dimensions, and it will generate the output:
function truth_grid(int $truths, int $rows, int $columns) {
return chunk_split(implode(' ', array_map('intval',
truths($truths, $rows * $columns)
)), $columns * 2);
}
Given:
$set = [[4, 3, 3], [10, 4, 5], [12, 3, 9], [5, 2, 10],];
foreach($set as [$truths, $rows, $columns]) {
print_r(
sprintf('%d of %d x %d', $truths, $rows, $columns).PHP_EOL.PHP_EOL.
truth_grid($truths, $rows, $columns).PHP_EOL
);
}
The readout of which is:
4 of 3 x 3
1 0 0
1 0 1
1 0 0
10 of 4 x 5
1 0 0 0 1
0 1 0 0 1
1 1 0 0 0
1 1 1 0 1
12 of 3 x 9
0 0 0 0 1 1 0 0 0
0 0 0 1 1 1 1 0 1
1 1 1 0 0 1 0 0 1
5 of 2 x 10
0 0 0 0 0 0 0 0 1 1
0 0 1 0 0 0 1 0 0 1
And if you're looking for chunked arrays of the same shape:
function truth_chunks(int $truths, int $rows, int $columns) {
return array_chunk(truths($truths, $rows * $columns), $columns);
}
id price tax item discount
1 500 15 100 0
2 0 15 200 1
3 100 15 500 0
4 0 15 300 0
5 1000 15 88 0
6 0 15 190 1
7 0 15 120 0
Select All if both price and discount or not equal to zero.
I don't want to select if the value of price and discount both are zero in mysql
Not to select only if price and isDiscount both are zero... (only one case)
Price IsDiscount Select
1 1 Yes
!0 0 yes
0 1 Yes
0 0 no
I might be missing something here, ...but isn't it as simple as
SELECT
id,tax,...
FROM
tablename
WHERE
price!=0
OR discount!=0
?
Here is your simple answer.
the query is as follows :
select * from orders WHERE price+discount > 0
select * from from_your_table WHERE price+discount > 0
SELECT
id,tax,...
FROM
tablename
WHERE
(price >0) or (discount>0) or (discount>0 and price>0)
I am building a word unscrambler (php/mysql) that takes user input of between 2 and 8 letters and returns words of between 2 and 8 letters that can be made from those letters, not necessarily using all of the letters, but definitely not including more letters than supplied.
The user will enter something like MSIKE or MSIKEI (two i's), or any combination of letters or multiple occurrences of a letter.
The query below will find all occurrences of words that contain M, S, I, K, or E.
However, the query below also returns words that have multiple occurrences of letters not requested. For example, the word meek would be returned, even though it has two e's and the user didn't enter two e's, or the word kiss, even though the user didn't enter s twice.
SELECT word
FROM words
WHERE word REGEXP '[msike]'
AND has_a=0
AND has_b=0
AND has_c=0
AND has_d=0
(we skip e) or we could add has_e=1
AND has_f=0
...and so on...skipping letters m, s, i, k, and e
AND has_w=0
AND has_x=0
AND has_y=0
AND has_z=0
Note the columns has_a, has_b, etc are either 1 if the letter occurs in the word or 0 if not.
I am open to any changes to the table schema.
This site: http://grecni.com/texttwist.php is a good example of what I am trying to emulate.
Question is how to modify the query to not return words with multiple occurrences of a letter, unless the user specifically entered a letter multiple times. Grouping by word length would be an added bonus.
Thanks so much.
EDIT: I altered the db per the suggestion of #awei, The has_{letter} is now count_{letter} and stores the total number of occurrences of the respective letter in the respective word. This could be useful when a user enters a letter multiple times. example: user enters MSIKES (two s).
Additionally, I have abandoned the REGEXP approach as shown in the original SQL statement. Working on doing most of the work on the PHP side, but many hurdles still in the way.
EDIT: Included first 10 rows from table
id word alpha otcwl ospd csw sowpods dictionary enable vowels consonants start_with end_with end_with_ing end_with_ly end_with_xy count_a count_b count_c count_d count_e count_f count_g count_h count_i count_j count_k count_l count_m count_n count_o count_p count_q count_r count_s count_t count_u count_v count_w count_x count_y count_z q_no_u letter_count scrabble_points wwf_points status date_added
1 aa aa 1 0 0 1 1 1 aa a a 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 2 1 2015-11-12 05:39:45
2 aah aah 1 0 0 1 0 1 aa h a h 0 0 0 2 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 6 5 1 2015-11-12 05:39:45
3 aahed aadeh 1 0 0 1 0 1 aae hd a d 0 0 0 2 0 0 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 9 8 1 2015-11-12 05:39:45
4 aahing aaghin 1 0 0 1 0 1 aai hng a g 1 0 0 2 0 0 0 0 0 1 1 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 6 10 11 1 2015-11-12 05:39:45
5 aahs aahs 1 0 0 1 0 1 aa hs a s 0 0 0 2 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 4 7 6 1 2015-11-12 05:39:45
6 aal aal 1 0 0 1 0 1 aa l a l 0 0 0 2 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 3 4 1 2015-11-12 05:39:45
7 aalii aaiil 1 0 0 1 1 1 aaii l a i 0 0 0 2 0 0 0 0 0 0 0 2 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 6 1 2015-11-12 05:39:45
8 aaliis aaiils 1 0 0 1 0 1 aaii ls a s 0 0 0 2 0 0 0 0 0 0 0 2 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 6 6 7 1 2015-11-12 05:39:45
9 aals aals 1 0 0 1 0 1 aa ls a s 0 0 0 2 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 4 4 5 1 2015-11-12 05:39:45
10 aardvark aaadkrrv 1 0 0 1 1 1 aaa rdvrk a k 0 0 0 3 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 2 0 0 0 1 0 0 0 0 0 8 16 17 1 2015-11-12 05:39:45
Think you've already done the hard work with your revised schema. All you need to do now is modify the query to look for <= the number of counts of each letter as specified by the user.
E.g. if the user entered "ALIAS":
SELECT word
FROM words
WHERE count_a <= 2
AND count_b <= 0
AND count_c <= 0
AND count_d <= 0
AND count_e <= 0
AND count_f <= 0
AND count_g <= 0
AND count_h <= 0
AND count_i <= 1
AND count_j <= 0
AND count_k <= 0
AND count_l <= 1
AND count_m <= 0
AND count_n <= 0
AND count_o <= 0
AND count_p <= 0
AND count_q <= 0
AND count_r <= 0
AND count_s <= 1
AND count_t <= 0
AND count_u <= 0
AND count_v <= 0
AND count_w <= 0
AND count_x <= 0
AND count_y <= 0
AND count_z <= 0
ORDER BY CHAR_LENGTH(word), word;
Note: As requested, this is ordering by word length, then alphabetically. Have used <= even for <= 0 just to make it easier to modify by hand for other letters.
This returns "aa", "aal" and "aals" (but not "aalii" or "aaliis" since they both have two "i"s).
See SQL Fiddle Demo.
Since you have two different requirements, I suggest implementing both two different solutions.
Where you don't care about dup letters, build a SET datatype with the 26 letters. Populate the bits according what the word has. This ignores duplicate letters. This also facilitates looking for words with a subset of the letters: (the_set & ~the_letters) = 0.
Where you do care about dups, sort the letters in the word and store that as the key. "msike" becomes "eikms".
Build a table that contains 3 columns:
eikms -- non unique index on this
msike -- the real word - probably good to have this as the PRIMARY KEY
SET('m','s','i',','k','e') -- for the other situation.
msikei and meek would be entered as
eikms
msikei
SET('m','s','i',','k','e') -- (or, if more convenient: SET('m','i','s','i',','k','e')
ekm
meek
SET('e','k','m')
REGEXP is not practical for your task.
Edit 1
I think you also need a column that indicates whether there are any doubled letters in the word. That way, you can distinguish that kiss is allowed for msikes but for for msike.
Edit 2
A SET or an INT UNSIGNED can hold 1 bit for each of the 26 letters -- 0 for not present, 1 for present.
msikes and msike would both go into the set with exactly 5 bits turned on. The value to INSERT would be 'm,s,i,k,e,s' for msikes. Since the rest needs to involve Boolean arithmetic, maybe it would be better to use INT UNSIGNED. So...
a is 1 (1 << 0)
b is 2 (1 << 1)
c is 4 (1 << 2)
d is 8 (1 << 3)
...
z is (1 << 25)
To INSERT you use the | operator. bad becomes
(1 << 1) | (1 << 0) | (1 << 3)
Note how the bits are laid out, with 'a' at the bottom:
SELECT BIN((1 << 1) | (1 << 0) | (1 << 3)); ==> 1011
Similarly 'ad' is 1001. So, does 'ad' match 'bad'? The answer comes from
SELECT b'1001' & ~b'1011' = 0; ==> 1 (meaning 'true')
That means that all the letters in 'ad' (1001) are found in 'bad' (1011). Let's introduce "bed", which is 11010.
SELECT b'11010' & ~b'1011' = 0; ==> FALSE because of 'e' (10000)
But 'dad' (1001) will work fine:
SELECT b'1001' & ~b'1011' = 0; ==> TRUE
So, now comes the "dup" flag. Since 'dad' has dup letters, but 'bad' did not, your rules say that it is not a match. But it took the "dup" to finish the decision.
If you have not had a course in Boolean arithmetic, well, I have just presented the first couple of chapters. If I covered it too fast, find a math book on such and jump in. "It's not rocket science."
So, back to what code is needed to decide whether my_word has a subset of letters and whether it is allowed to have duplicate letters:
SELECT $my_mask & ~tbl.mask = 0, dup FROM tbl;
Then do the suitable AND / OR between to finish the logic.
With the limited Regex support on MySQL, best I can do is a PHP script for generating the query, presuming it only includes English letters. It seems making an expression to exclude invalid words is easier than one that includes them.
<?php
$inputword = str_split('msikes');
$counter = array();
for ($l = 'a'; $l < 'z'; $l++) {
$counter[$l] = 0;
}
foreach ($inputword as $l) {
$counter[$l]++;
}
$nots = '';
foreach ($counter as $l => $c) {
if (!$c) {
$nots .= $l;
unset($counter[$l]);
}
}
$conditions = array();
if(!empty($nots)) {
// exclude words that have letters not given
$conditions[] = "[" . $nots . "]'";
}
foreach ($counter as $l => $c) {
$letters = array();
for ($i = 0; $i <= $c; $i++) {
$letters[] = $l;
}
// exclude words that have the current letter more times than given
$conditions[] = implode('.*', $letters);
}
$sql = "SELECT word FROM words WHERE word NOT RLIKE '" . implode('|', $conditions) . "'";
echo $sql;
Something like this might work for you:
// Input Word
$WORD = strtolower('msikes');
// Alpha Array
$Alpha = range('a', 'z');
// Turn it into letters.
$Splited = str_split($WORD);
$Letters = array();
// Count occurrence of each letter, use letter as key to make it unique
foreach( $Splited as $Letter ) {
$Letters[$Letter] = array_key_exists($Letter, $Letters) ? $Letters[$Letter] + 1 : 1;
}
// Build a list of letters that shouldn't be present in the word
$ShouldNotExists = array_filter($Alpha, function ($Letter) use ($Letters) {
return ! array_key_exists($Letter, $Letters);
});
#### Building SQL Statement
// Letters to skip
$SkipLetters = array();
foreach( $ShouldNotExists as $SkipLetter ) {
$SkipLetters[] = "`has_{$SkipLetter}` = 0";
}
// count condition (for multiple occurrences)
$CountLetters = array();
foreach( $Letters as $K => $V ) {
$CountLetters[] = "`count_{$K}` <= {$V}";
}
$SQL = 'SELECT `word` FROM `words` WHERE '.PHP_EOL;
$SQL .= '('.implode(' AND ', $SkipLetters).')'.PHP_EOL;
$SQL .= ' AND ('.implode(' AND ', $CountLetters).')'.PHP_EOL;
$SQL .= ' ORDER BY LENGTH(`word`), `word`'.PHP_EOL;
echo $SQL;
I have a fairly simple (at least I thought simple) task I'm trying to achieve but for the life of me I can't think of the best way of doing it in php.
Note I'm trying to avoid using nested multi-dimensional arrays.
What I want to do is a simple 2 axis reference table which I can query/look up against.
0 | 1 | 2 | 3 | 4 | 5
0 5 0 0 0 0 0
-
1 0 2 0 1 0 0
-
2 0 0 0 7 0 1
-
3 1 2 4 10 1 0
-
4 0 0 0 0 0 0
-
5 0 3 4 0 1 0
Then I want to query it so lookup_table(3,3) would return 10. I also need to update the values in the table so like add_lookup_table(3,3,1) would change location 3,3 to 11 or some variation of.
So far I know I can achieve it with multidim arrays however want to avoid this because its just going to be a pain in the ass getting and setting each time.
I thought about a list() but php's lists aren't really lists so that goes out the window.
What I'm after is suggestions/recommendations for a better way of doing this.
you can just use a single array like so
$array = array(
5, 0, 0, 0, 0, 0,
0, 2, 0, 1, 0 , 0
..etc...
);
then each key is just $array[$row * 6 + $column]
so $array[3*6 + 3]
or define your array as something like
$array["3.3"] = 10;
both of the above are BAD ways to do it when a multidimensional array exists and is well suited for the task.
you list the reason you dont want to use it as "getting and setting each time" - what ever the solution, you will have to get and set to populate it.
The real solution here seems to be the one you wish to avoid
you MAY want to check
http://php.net/manual/en/spl.datastructures.php
That has extra structures
I'm trying to Calculate Income Points for MLM network Users (i.e).,
I'm having five types of kits such as Rs.0/-, Rs.1000/-, Rs.2000/-, Rs.4000/-, Rs.10000/- .
Once the kit is purchased a unique pin and user id will generate for each individual kit like
If 2 number of 2000/- kit purchased means userid and pin will be
2000100 539fdda37c435 and
2000101 5395b0d8b66d1 .....
Then Based on new user sponsor id(Sponsor id will be old user), registration will be proceed with Group left and right.
Now what's my problem is While calculating income point of a user how can i take their children user from top to bottom. I'd strucked in MYSQL Query.
For Example.,
now how can i get each node's value for a parent node(1) to root node(12) from this dynamic tree in php ??
I struck in this loop concept.,
Thanks IN Advance.,
database architecture
id par lev lef rgt status wallet userstatus
1 1 1 2 3 1 0 0
2 2 2 4 5 1 0 0
6 3 2 6 7 1 0 0
8 4 3 8 9 1 0 0
9 5 3 10 11 1 0 0
10 6 3 12 13 1 0 0
11 7 3 0 0 1 0 0