Compute the square root of a positive integer using binary searchInteger square root in x86 assembly (NASM)Epilogue to a binary search to find the range of matching indexesBinary search that gets ALL matching resultsBinary search of an integer arraySend a tweet to ISP when internet speed dropsNavigating over a square spiralSwiftly counting rooms in a floor planHashTable using C

What was the intention with the Commodore 128?

Did Michelle Obama have a staff of 23; and Melania have a staff of 4?

How to use the passive form to say "This flower was watered."

What is the opposite of "hunger level"?

What should I do with the stock I own if I anticipate there will be a recession?

Gofer work in exchange for Letter of Recommendation

A Magic Diamond

Polar contour plot in Mathematica?

Meaning and structure of headline "Hair it is: A List of ..."

Unsolved Problems due to Lack of Computational Power

When does The Truman Show take place?

How to render "have ideas above his station" into German

Is it alright to say good afternoon Sirs and Madams in a panel interview?

Existence of a certain set of 0/1-sequences without the Axiom of Choice

Unconventional examples of mathematical modelling

How do I cope with haze for the photos containing sky and trees at a distance?

Heyawacky: Ace of Cups

Designing a prison for a telekinetic race

What happened after the end of the Truman Show?

Are there any OR challenges that are similar to kaggle's competitions?

Why should P.I be willing to write strong LOR even if that means losing a undergraduate from his/her lab?

What exactly happened to the 18 crew members who were reported as "missing" in "Q Who"?

A reccomended structured approach to self studying music theory for songwriting

Output with the same length always



Compute the square root of a positive integer using binary search


Integer square root in x86 assembly (NASM)Epilogue to a binary search to find the range of matching indexesBinary search that gets ALL matching resultsBinary search of an integer arraySend a tweet to ISP when internet speed dropsNavigating over a square spiralSwiftly counting rooms in a floor planHashTable using C






.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty margin-bottom:0;








3












$begingroup$


The requirement is to find the square root of a positive integer using binary search and the math property that square root of a number n is between 0 and n/2, and the required answer is "floored", meaning mySqrt(8) is to return 2.



Please comment on the efficiency, and if possible, the loop invariants in terms of correctness:



class Solution(object):
def mySqrt(self, x):
"""
:type x: int
:rtype: int

Loop invariant:

The answer is always in the range [low, high] inclusive,
except possibly:

1) low == high == mid, and mid * mid == x and
any of low, high, or mid can be returned as the answer.

2) if there is no exact answer and the floor is to be
returned, then low > high by 1. Since sq != x,
so either low or high is set.

If low gets set and gets pushed up, it is pushed up too much.
So when low > high by 1, low - 1 is the answer and it is the same
as high, because low > high by 1.

If high gets set and gets pushed down, high can be
the correct answer. When low > high, it is by 1,
and high is the correct floor value to be returned.
(since there is no perfect square root and the floor is required)

0 <= low <= answer <= high <= n//2 + 1

where answer is floor(sqrt(x)) to be found,
except if low > high and the loop will exit.

Each loop iteration always makes the range smaller.

If the range is empty at the end, low is > high by 1, and high is
the correct floored value, and low is the ceiling value, so high is returned.
"""

low = 0;
high = x//2 + 1;

while (low <= high):
mid = low + (high - low) // 2;
sq = mid * mid;
if (sq == x):
return mid;
elif (sq > x):
high = mid - 1; # sq exceeds target, so mid cannot be the answer floored, but when high is set to mid - 1, then it can be the answer
else:
low = mid + 1; # (here sq < x, and mid might be the answer floored, so when low is set to mid + 1, then low might be too big, while high is correct)

return high;









share|improve this question











$endgroup$




















    3












    $begingroup$


    The requirement is to find the square root of a positive integer using binary search and the math property that square root of a number n is between 0 and n/2, and the required answer is "floored", meaning mySqrt(8) is to return 2.



    Please comment on the efficiency, and if possible, the loop invariants in terms of correctness:



    class Solution(object):
    def mySqrt(self, x):
    """
    :type x: int
    :rtype: int

    Loop invariant:

    The answer is always in the range [low, high] inclusive,
    except possibly:

    1) low == high == mid, and mid * mid == x and
    any of low, high, or mid can be returned as the answer.

    2) if there is no exact answer and the floor is to be
    returned, then low > high by 1. Since sq != x,
    so either low or high is set.

    If low gets set and gets pushed up, it is pushed up too much.
    So when low > high by 1, low - 1 is the answer and it is the same
    as high, because low > high by 1.

    If high gets set and gets pushed down, high can be
    the correct answer. When low > high, it is by 1,
    and high is the correct floor value to be returned.
    (since there is no perfect square root and the floor is required)

    0 <= low <= answer <= high <= n//2 + 1

    where answer is floor(sqrt(x)) to be found,
    except if low > high and the loop will exit.

    Each loop iteration always makes the range smaller.

    If the range is empty at the end, low is > high by 1, and high is
    the correct floored value, and low is the ceiling value, so high is returned.
    """

    low = 0;
    high = x//2 + 1;

    while (low <= high):
    mid = low + (high - low) // 2;
    sq = mid * mid;
    if (sq == x):
    return mid;
    elif (sq > x):
    high = mid - 1; # sq exceeds target, so mid cannot be the answer floored, but when high is set to mid - 1, then it can be the answer
    else:
    low = mid + 1; # (here sq < x, and mid might be the answer floored, so when low is set to mid + 1, then low might be too big, while high is correct)

    return high;









    share|improve this question











    $endgroup$
















      3












      3








      3





      $begingroup$


      The requirement is to find the square root of a positive integer using binary search and the math property that square root of a number n is between 0 and n/2, and the required answer is "floored", meaning mySqrt(8) is to return 2.



      Please comment on the efficiency, and if possible, the loop invariants in terms of correctness:



      class Solution(object):
      def mySqrt(self, x):
      """
      :type x: int
      :rtype: int

      Loop invariant:

      The answer is always in the range [low, high] inclusive,
      except possibly:

      1) low == high == mid, and mid * mid == x and
      any of low, high, or mid can be returned as the answer.

      2) if there is no exact answer and the floor is to be
      returned, then low > high by 1. Since sq != x,
      so either low or high is set.

      If low gets set and gets pushed up, it is pushed up too much.
      So when low > high by 1, low - 1 is the answer and it is the same
      as high, because low > high by 1.

      If high gets set and gets pushed down, high can be
      the correct answer. When low > high, it is by 1,
      and high is the correct floor value to be returned.
      (since there is no perfect square root and the floor is required)

      0 <= low <= answer <= high <= n//2 + 1

      where answer is floor(sqrt(x)) to be found,
      except if low > high and the loop will exit.

      Each loop iteration always makes the range smaller.

      If the range is empty at the end, low is > high by 1, and high is
      the correct floored value, and low is the ceiling value, so high is returned.
      """

      low = 0;
      high = x//2 + 1;

      while (low <= high):
      mid = low + (high - low) // 2;
      sq = mid * mid;
      if (sq == x):
      return mid;
      elif (sq > x):
      high = mid - 1; # sq exceeds target, so mid cannot be the answer floored, but when high is set to mid - 1, then it can be the answer
      else:
      low = mid + 1; # (here sq < x, and mid might be the answer floored, so when low is set to mid + 1, then low might be too big, while high is correct)

      return high;









      share|improve this question











      $endgroup$




      The requirement is to find the square root of a positive integer using binary search and the math property that square root of a number n is between 0 and n/2, and the required answer is "floored", meaning mySqrt(8) is to return 2.



      Please comment on the efficiency, and if possible, the loop invariants in terms of correctness:



      class Solution(object):
      def mySqrt(self, x):
      """
      :type x: int
      :rtype: int

      Loop invariant:

      The answer is always in the range [low, high] inclusive,
      except possibly:

      1) low == high == mid, and mid * mid == x and
      any of low, high, or mid can be returned as the answer.

      2) if there is no exact answer and the floor is to be
      returned, then low > high by 1. Since sq != x,
      so either low or high is set.

      If low gets set and gets pushed up, it is pushed up too much.
      So when low > high by 1, low - 1 is the answer and it is the same
      as high, because low > high by 1.

      If high gets set and gets pushed down, high can be
      the correct answer. When low > high, it is by 1,
      and high is the correct floor value to be returned.
      (since there is no perfect square root and the floor is required)

      0 <= low <= answer <= high <= n//2 + 1

      where answer is floor(sqrt(x)) to be found,
      except if low > high and the loop will exit.

      Each loop iteration always makes the range smaller.

      If the range is empty at the end, low is > high by 1, and high is
      the correct floored value, and low is the ceiling value, so high is returned.
      """

      low = 0;
      high = x//2 + 1;

      while (low <= high):
      mid = low + (high - low) // 2;
      sq = mid * mid;
      if (sq == x):
      return mid;
      elif (sq > x):
      high = mid - 1; # sq exceeds target, so mid cannot be the answer floored, but when high is set to mid - 1, then it can be the answer
      else:
      low = mid + 1; # (here sq < x, and mid might be the answer floored, so when low is set to mid + 1, then low might be too big, while high is correct)

      return high;






      python algorithm mathematics binary-search






      share|improve this question















      share|improve this question













      share|improve this question




      share|improve this question








      edited 6 hours ago







      太極者無極而生

















      asked 8 hours ago









      太極者無極而生太極者無極而生

      1997 bronze badges




      1997 bronze badges























          1 Answer
          1






          active

          oldest

          votes


















          5












          $begingroup$

          • Your comments on the elif / else part are too long to be just after the statements


          • Don't use semicolons (;) in Python.


          • This is a refactored version of the code


          import math

          class Solution(object):
          def mySqrt(self, x):
          """
          :type x: int
          :rtype: int

          Returns floor(sqrt(x))
          """
          low = 0
          high = x//2 + 1

          """
          It is proved that 0 <= sqrt(x) <= x/2, so
          we run a dichotomic in [0, x/2] to find floor(sqrt(x))

          Loop analysis:
          * Initialization: low = 0 and high = x/2 + 1
          * Termination: |high-low| is reduced each iteration,
          as shown in lines high = mid - 1 and low = mid + 1.
          * Invariant: low <= floor(sqrt(x)) <= high.
          Let mid be (low + high)/2.
          - If mid^2 <= x < (mid+1)^2,
          then mid is floor(sqrt(x)) and just return it.
          - If mid^2 > x, search for values smaller than mid.
          - Otherwise, if mid^2 < x, search within higher values.
          """
          while (low <= high):
          mid = (low + high) // 2
          sq = mid * mid
          sq_next = (mid+1)*(mid+1)
          if (sq <= x < sq_next):
          return mid
          elif (sq > x):
          high = mid - 1
          else:
          low = mid + 1

          for i in range(1, 26):
          assert(math.floor(math.sqrt(i)) == Solution().mySqrt(i))





          share|improve this answer











          $endgroup$














          • $begingroup$
            I suppose if it is Python, it can be mid = (low + high) // 2 because there is infinite precision arithmetics... but it just depends whether you want it first to overflow (either 32 bit or 64 bit int) to become bignum, and then divided by 2 to make it back to a 32 bit or 64 bit int
            $endgroup$
            – 太極者無極而生
            8 hours ago







          • 1




            $begingroup$
            @太極者無極而生 Take a look to the new answer; now there is no need for returning high or low after the loop, so it is easier to reason about.
            $endgroup$
            – JnxF
            7 hours ago











          • $begingroup$
            you are using $$mid^2$$ and $$ (mid+1)^2 $$ to check for the answer and return and no need to consider how low or high gets set... interesting... it looks like it can make it simpler loop invariants
            $endgroup$
            – 太極者無極而生
            7 hours ago











          • $begingroup$
            how come you assert from 1 to 26 instead of from 1 to some larger number like... 5000 or a million
            $endgroup$
            – 太極者無極而生
            7 hours ago










          • $begingroup$
            @太極者無極而生 Code is proved to be correct mathematically, so you can check up to any number you want. Checked up to $1000000$ and it works (surely, you can try up to any number you want).
            $endgroup$
            – JnxF
            7 hours ago













          Your Answer






          StackExchange.ifUsing("editor", function ()
          StackExchange.using("externalEditor", function ()
          StackExchange.using("snippets", function ()
          StackExchange.snippets.init();
          );
          );
          , "code-snippets");

          StackExchange.ready(function()
          var channelOptions =
          tags: "".split(" "),
          id: "196"
          ;
          initTagRenderer("".split(" "), "".split(" "), channelOptions);

          StackExchange.using("externalEditor", function()
          // Have to fire editor after snippets, if snippets enabled
          if (StackExchange.settings.snippets.snippetsEnabled)
          StackExchange.using("snippets", function()
          createEditor();
          );

          else
          createEditor();

          );

          function createEditor()
          StackExchange.prepareEditor(
          heartbeatType: 'answer',
          autoActivateHeartbeat: false,
          convertImagesToLinks: false,
          noModals: true,
          showLowRepImageUploadWarning: true,
          reputationToPostImages: null,
          bindNavPrevention: true,
          postfix: "",
          imageUploader:
          brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
          contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
          allowUrls: true
          ,
          onDemand: true,
          discardSelector: ".discard-answer"
          ,immediatelyShowMarkdownHelp:true
          );



          );













          draft saved

          draft discarded


















          StackExchange.ready(
          function ()
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fcodereview.stackexchange.com%2fquestions%2f226340%2fcompute-the-square-root-of-a-positive-integer-using-binary-search%23new-answer', 'question_page');

          );

          Post as a guest















          Required, but never shown

























          1 Answer
          1






          active

          oldest

          votes








          1 Answer
          1






          active

          oldest

          votes









          active

          oldest

          votes






          active

          oldest

          votes









          5












          $begingroup$

          • Your comments on the elif / else part are too long to be just after the statements


          • Don't use semicolons (;) in Python.


          • This is a refactored version of the code


          import math

          class Solution(object):
          def mySqrt(self, x):
          """
          :type x: int
          :rtype: int

          Returns floor(sqrt(x))
          """
          low = 0
          high = x//2 + 1

          """
          It is proved that 0 <= sqrt(x) <= x/2, so
          we run a dichotomic in [0, x/2] to find floor(sqrt(x))

          Loop analysis:
          * Initialization: low = 0 and high = x/2 + 1
          * Termination: |high-low| is reduced each iteration,
          as shown in lines high = mid - 1 and low = mid + 1.
          * Invariant: low <= floor(sqrt(x)) <= high.
          Let mid be (low + high)/2.
          - If mid^2 <= x < (mid+1)^2,
          then mid is floor(sqrt(x)) and just return it.
          - If mid^2 > x, search for values smaller than mid.
          - Otherwise, if mid^2 < x, search within higher values.
          """
          while (low <= high):
          mid = (low + high) // 2
          sq = mid * mid
          sq_next = (mid+1)*(mid+1)
          if (sq <= x < sq_next):
          return mid
          elif (sq > x):
          high = mid - 1
          else:
          low = mid + 1

          for i in range(1, 26):
          assert(math.floor(math.sqrt(i)) == Solution().mySqrt(i))





          share|improve this answer











          $endgroup$














          • $begingroup$
            I suppose if it is Python, it can be mid = (low + high) // 2 because there is infinite precision arithmetics... but it just depends whether you want it first to overflow (either 32 bit or 64 bit int) to become bignum, and then divided by 2 to make it back to a 32 bit or 64 bit int
            $endgroup$
            – 太極者無極而生
            8 hours ago







          • 1




            $begingroup$
            @太極者無極而生 Take a look to the new answer; now there is no need for returning high or low after the loop, so it is easier to reason about.
            $endgroup$
            – JnxF
            7 hours ago











          • $begingroup$
            you are using $$mid^2$$ and $$ (mid+1)^2 $$ to check for the answer and return and no need to consider how low or high gets set... interesting... it looks like it can make it simpler loop invariants
            $endgroup$
            – 太極者無極而生
            7 hours ago











          • $begingroup$
            how come you assert from 1 to 26 instead of from 1 to some larger number like... 5000 or a million
            $endgroup$
            – 太極者無極而生
            7 hours ago










          • $begingroup$
            @太極者無極而生 Code is proved to be correct mathematically, so you can check up to any number you want. Checked up to $1000000$ and it works (surely, you can try up to any number you want).
            $endgroup$
            – JnxF
            7 hours ago















          5












          $begingroup$

          • Your comments on the elif / else part are too long to be just after the statements


          • Don't use semicolons (;) in Python.


          • This is a refactored version of the code


          import math

          class Solution(object):
          def mySqrt(self, x):
          """
          :type x: int
          :rtype: int

          Returns floor(sqrt(x))
          """
          low = 0
          high = x//2 + 1

          """
          It is proved that 0 <= sqrt(x) <= x/2, so
          we run a dichotomic in [0, x/2] to find floor(sqrt(x))

          Loop analysis:
          * Initialization: low = 0 and high = x/2 + 1
          * Termination: |high-low| is reduced each iteration,
          as shown in lines high = mid - 1 and low = mid + 1.
          * Invariant: low <= floor(sqrt(x)) <= high.
          Let mid be (low + high)/2.
          - If mid^2 <= x < (mid+1)^2,
          then mid is floor(sqrt(x)) and just return it.
          - If mid^2 > x, search for values smaller than mid.
          - Otherwise, if mid^2 < x, search within higher values.
          """
          while (low <= high):
          mid = (low + high) // 2
          sq = mid * mid
          sq_next = (mid+1)*(mid+1)
          if (sq <= x < sq_next):
          return mid
          elif (sq > x):
          high = mid - 1
          else:
          low = mid + 1

          for i in range(1, 26):
          assert(math.floor(math.sqrt(i)) == Solution().mySqrt(i))





          share|improve this answer











          $endgroup$














          • $begingroup$
            I suppose if it is Python, it can be mid = (low + high) // 2 because there is infinite precision arithmetics... but it just depends whether you want it first to overflow (either 32 bit or 64 bit int) to become bignum, and then divided by 2 to make it back to a 32 bit or 64 bit int
            $endgroup$
            – 太極者無極而生
            8 hours ago







          • 1




            $begingroup$
            @太極者無極而生 Take a look to the new answer; now there is no need for returning high or low after the loop, so it is easier to reason about.
            $endgroup$
            – JnxF
            7 hours ago











          • $begingroup$
            you are using $$mid^2$$ and $$ (mid+1)^2 $$ to check for the answer and return and no need to consider how low or high gets set... interesting... it looks like it can make it simpler loop invariants
            $endgroup$
            – 太極者無極而生
            7 hours ago











          • $begingroup$
            how come you assert from 1 to 26 instead of from 1 to some larger number like... 5000 or a million
            $endgroup$
            – 太極者無極而生
            7 hours ago










          • $begingroup$
            @太極者無極而生 Code is proved to be correct mathematically, so you can check up to any number you want. Checked up to $1000000$ and it works (surely, you can try up to any number you want).
            $endgroup$
            – JnxF
            7 hours ago













          5












          5








          5





          $begingroup$

          • Your comments on the elif / else part are too long to be just after the statements


          • Don't use semicolons (;) in Python.


          • This is a refactored version of the code


          import math

          class Solution(object):
          def mySqrt(self, x):
          """
          :type x: int
          :rtype: int

          Returns floor(sqrt(x))
          """
          low = 0
          high = x//2 + 1

          """
          It is proved that 0 <= sqrt(x) <= x/2, so
          we run a dichotomic in [0, x/2] to find floor(sqrt(x))

          Loop analysis:
          * Initialization: low = 0 and high = x/2 + 1
          * Termination: |high-low| is reduced each iteration,
          as shown in lines high = mid - 1 and low = mid + 1.
          * Invariant: low <= floor(sqrt(x)) <= high.
          Let mid be (low + high)/2.
          - If mid^2 <= x < (mid+1)^2,
          then mid is floor(sqrt(x)) and just return it.
          - If mid^2 > x, search for values smaller than mid.
          - Otherwise, if mid^2 < x, search within higher values.
          """
          while (low <= high):
          mid = (low + high) // 2
          sq = mid * mid
          sq_next = (mid+1)*(mid+1)
          if (sq <= x < sq_next):
          return mid
          elif (sq > x):
          high = mid - 1
          else:
          low = mid + 1

          for i in range(1, 26):
          assert(math.floor(math.sqrt(i)) == Solution().mySqrt(i))





          share|improve this answer











          $endgroup$



          • Your comments on the elif / else part are too long to be just after the statements


          • Don't use semicolons (;) in Python.


          • This is a refactored version of the code


          import math

          class Solution(object):
          def mySqrt(self, x):
          """
          :type x: int
          :rtype: int

          Returns floor(sqrt(x))
          """
          low = 0
          high = x//2 + 1

          """
          It is proved that 0 <= sqrt(x) <= x/2, so
          we run a dichotomic in [0, x/2] to find floor(sqrt(x))

          Loop analysis:
          * Initialization: low = 0 and high = x/2 + 1
          * Termination: |high-low| is reduced each iteration,
          as shown in lines high = mid - 1 and low = mid + 1.
          * Invariant: low <= floor(sqrt(x)) <= high.
          Let mid be (low + high)/2.
          - If mid^2 <= x < (mid+1)^2,
          then mid is floor(sqrt(x)) and just return it.
          - If mid^2 > x, search for values smaller than mid.
          - Otherwise, if mid^2 < x, search within higher values.
          """
          while (low <= high):
          mid = (low + high) // 2
          sq = mid * mid
          sq_next = (mid+1)*(mid+1)
          if (sq <= x < sq_next):
          return mid
          elif (sq > x):
          high = mid - 1
          else:
          low = mid + 1

          for i in range(1, 26):
          assert(math.floor(math.sqrt(i)) == Solution().mySqrt(i))






          share|improve this answer














          share|improve this answer



          share|improve this answer








          edited 7 hours ago

























          answered 8 hours ago









          JnxFJnxF

          4384 silver badges13 bronze badges




          4384 silver badges13 bronze badges














          • $begingroup$
            I suppose if it is Python, it can be mid = (low + high) // 2 because there is infinite precision arithmetics... but it just depends whether you want it first to overflow (either 32 bit or 64 bit int) to become bignum, and then divided by 2 to make it back to a 32 bit or 64 bit int
            $endgroup$
            – 太極者無極而生
            8 hours ago







          • 1




            $begingroup$
            @太極者無極而生 Take a look to the new answer; now there is no need for returning high or low after the loop, so it is easier to reason about.
            $endgroup$
            – JnxF
            7 hours ago











          • $begingroup$
            you are using $$mid^2$$ and $$ (mid+1)^2 $$ to check for the answer and return and no need to consider how low or high gets set... interesting... it looks like it can make it simpler loop invariants
            $endgroup$
            – 太極者無極而生
            7 hours ago











          • $begingroup$
            how come you assert from 1 to 26 instead of from 1 to some larger number like... 5000 or a million
            $endgroup$
            – 太極者無極而生
            7 hours ago










          • $begingroup$
            @太極者無極而生 Code is proved to be correct mathematically, so you can check up to any number you want. Checked up to $1000000$ and it works (surely, you can try up to any number you want).
            $endgroup$
            – JnxF
            7 hours ago
















          • $begingroup$
            I suppose if it is Python, it can be mid = (low + high) // 2 because there is infinite precision arithmetics... but it just depends whether you want it first to overflow (either 32 bit or 64 bit int) to become bignum, and then divided by 2 to make it back to a 32 bit or 64 bit int
            $endgroup$
            – 太極者無極而生
            8 hours ago







          • 1




            $begingroup$
            @太極者無極而生 Take a look to the new answer; now there is no need for returning high or low after the loop, so it is easier to reason about.
            $endgroup$
            – JnxF
            7 hours ago











          • $begingroup$
            you are using $$mid^2$$ and $$ (mid+1)^2 $$ to check for the answer and return and no need to consider how low or high gets set... interesting... it looks like it can make it simpler loop invariants
            $endgroup$
            – 太極者無極而生
            7 hours ago











          • $begingroup$
            how come you assert from 1 to 26 instead of from 1 to some larger number like... 5000 or a million
            $endgroup$
            – 太極者無極而生
            7 hours ago










          • $begingroup$
            @太極者無極而生 Code is proved to be correct mathematically, so you can check up to any number you want. Checked up to $1000000$ and it works (surely, you can try up to any number you want).
            $endgroup$
            – JnxF
            7 hours ago















          $begingroup$
          I suppose if it is Python, it can be mid = (low + high) // 2 because there is infinite precision arithmetics... but it just depends whether you want it first to overflow (either 32 bit or 64 bit int) to become bignum, and then divided by 2 to make it back to a 32 bit or 64 bit int
          $endgroup$
          – 太極者無極而生
          8 hours ago





          $begingroup$
          I suppose if it is Python, it can be mid = (low + high) // 2 because there is infinite precision arithmetics... but it just depends whether you want it first to overflow (either 32 bit or 64 bit int) to become bignum, and then divided by 2 to make it back to a 32 bit or 64 bit int
          $endgroup$
          – 太極者無極而生
          8 hours ago





          1




          1




          $begingroup$
          @太極者無極而生 Take a look to the new answer; now there is no need for returning high or low after the loop, so it is easier to reason about.
          $endgroup$
          – JnxF
          7 hours ago





          $begingroup$
          @太極者無極而生 Take a look to the new answer; now there is no need for returning high or low after the loop, so it is easier to reason about.
          $endgroup$
          – JnxF
          7 hours ago













          $begingroup$
          you are using $$mid^2$$ and $$ (mid+1)^2 $$ to check for the answer and return and no need to consider how low or high gets set... interesting... it looks like it can make it simpler loop invariants
          $endgroup$
          – 太極者無極而生
          7 hours ago





          $begingroup$
          you are using $$mid^2$$ and $$ (mid+1)^2 $$ to check for the answer and return and no need to consider how low or high gets set... interesting... it looks like it can make it simpler loop invariants
          $endgroup$
          – 太極者無極而生
          7 hours ago













          $begingroup$
          how come you assert from 1 to 26 instead of from 1 to some larger number like... 5000 or a million
          $endgroup$
          – 太極者無極而生
          7 hours ago




          $begingroup$
          how come you assert from 1 to 26 instead of from 1 to some larger number like... 5000 or a million
          $endgroup$
          – 太極者無極而生
          7 hours ago












          $begingroup$
          @太極者無極而生 Code is proved to be correct mathematically, so you can check up to any number you want. Checked up to $1000000$ and it works (surely, you can try up to any number you want).
          $endgroup$
          – JnxF
          7 hours ago




          $begingroup$
          @太極者無極而生 Code is proved to be correct mathematically, so you can check up to any number you want. Checked up to $1000000$ and it works (surely, you can try up to any number you want).
          $endgroup$
          – JnxF
          7 hours ago

















          draft saved

          draft discarded
















































          Thanks for contributing an answer to Code Review Stack Exchange!


          • Please be sure to answer the question. Provide details and share your research!

          But avoid


          • Asking for help, clarification, or responding to other answers.

          • Making statements based on opinion; back them up with references or personal experience.

          Use MathJax to format equations. MathJax reference.


          To learn more, see our tips on writing great answers.




          draft saved


          draft discarded














          StackExchange.ready(
          function ()
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fcodereview.stackexchange.com%2fquestions%2f226340%2fcompute-the-square-root-of-a-positive-integer-using-binary-search%23new-answer', 'question_page');

          );

          Post as a guest















          Required, but never shown





















































          Required, but never shown














          Required, but never shown












          Required, but never shown







          Required, but never shown

































          Required, but never shown














          Required, but never shown












          Required, but never shown







          Required, but never shown







          Popular posts from this blog

          19. јануар Садржај Догађаји Рођења Смрти Празници и дани сећања Види још Референце Мени за навигацијуу

          Israel Cuprins Etimologie | Istorie | Geografie | Politică | Demografie | Educație | Economie | Cultură | Note explicative | Note bibliografice | Bibliografie | Legături externe | Meniu de navigaresite web oficialfacebooktweeterGoogle+Instagramcanal YouTubeInstagramtextmodificaremodificarewww.technion.ac.ilnew.huji.ac.ilwww.weizmann.ac.ilwww1.biu.ac.ilenglish.tau.ac.ilwww.haifa.ac.ilin.bgu.ac.ilwww.openu.ac.ilwww.ariel.ac.ilCIA FactbookHarta Israelului"Negotiating Jerusalem," Palestine–Israel JournalThe Schizoid Nature of Modern Hebrew: A Slavic Language in Search of a Semitic Past„Arabic in Israel: an official language and a cultural bridge”„Latest Population Statistics for Israel”„Israel Population”„Tables”„Report for Selected Countries and Subjects”Human Development Report 2016: Human Development for Everyone„Distribution of family income - Gini index”The World FactbookJerusalem Law„Israel”„Israel”„Zionist Leaders: David Ben-Gurion 1886–1973”„The status of Jerusalem”„Analysis: Kadima's big plans”„Israel's Hard-Learned Lessons”„The Legacy of Undefined Borders, Tel Aviv Notes No. 40, 5 iunie 2002”„Israel Journal: A Land Without Borders”„Population”„Israel closes decade with population of 7.5 million”Time Series-DataBank„Selected Statistics on Jerusalem Day 2007 (Hebrew)”Golan belongs to Syria, Druze protestGlobal Survey 2006: Middle East Progress Amid Global Gains in FreedomWHO: Life expectancy in Israel among highest in the worldInternational Monetary Fund, World Economic Outlook Database, April 2011: Nominal GDP list of countries. Data for the year 2010.„Israel's accession to the OECD”Popular Opinion„On the Move”Hosea 12:5„Walking the Bible Timeline”„Palestine: History”„Return to Zion”An invention called 'the Jewish people' – Haaretz – Israel NewsoriginalJewish and Non-Jewish Population of Palestine-Israel (1517–2004)ImmigrationJewishvirtuallibrary.orgChapter One: The Heralders of Zionism„The birth of modern Israel: A scrap of paper that changed history”„League of Nations: The Mandate for Palestine, 24 iulie 1922”The Population of Palestine Prior to 1948originalBackground Paper No. 47 (ST/DPI/SER.A/47)History: Foreign DominationTwo Hundred and Seventh Plenary Meeting„Israel (Labor Zionism)”Population, by Religion and Population GroupThe Suez CrisisAdolf EichmannJustice Ministry Reply to Amnesty International Report„The Interregnum”Israel Ministry of Foreign Affairs – The Palestinian National Covenant- July 1968Research on terrorism: trends, achievements & failuresThe Routledge Atlas of the Arab–Israeli conflict: The Complete History of the Struggle and the Efforts to Resolve It"George Habash, Palestinian Terrorism Tactician, Dies at 82."„1973: Arab states attack Israeli forces”Agranat Commission„Has Israel Annexed East Jerusalem?”original„After 4 Years, Intifada Still Smolders”From the End of the Cold War to 2001originalThe Oslo Accords, 1993Israel-PLO Recognition – Exchange of Letters between PM Rabin and Chairman Arafat – Sept 9- 1993Foundation for Middle East PeaceSources of Population Growth: Total Israeli Population and Settler Population, 1991–2003original„Israel marks Rabin assassination”The Wye River Memorandumoriginal„West Bank barrier route disputed, Israeli missile kills 2”"Permanent Ceasefire to Be Based on Creation Of Buffer Zone Free of Armed Personnel Other than UN, Lebanese Forces"„Hezbollah kills 8 soldiers, kidnaps two in offensive on northern border”„Olmert confirms peace talks with Syria”„Battleground Gaza: Israeli ground forces invade the strip”„IDF begins Gaza troop withdrawal, hours after ending 3-week offensive”„THE LAND: Geography and Climate”„Area of districts, sub-districts, natural regions and lakes”„Israel - Geography”„Makhteshim Country”Israel and the Palestinian Territories„Makhtesh Ramon”„The Living Dead Sea”„Temperatures reach record high in Pakistan”„Climate Extremes In Israel”Israel in figures„Deuteronom”„JNF: 240 million trees planted since 1901”„Vegetation of Israel and Neighboring Countries”Environmental Law in Israel„Executive branch”„Israel's election process explained”„The Electoral System in Israel”„Constitution for Israel”„All 120 incoming Knesset members”„Statul ISRAEL”„The Judiciary: The Court System”„Israel's high court unique in region”„Israel and the International Criminal Court: A Legal Battlefield”„Localities and population, by population group, district, sub-district and natural region”„Israel: Districts, Major Cities, Urban Localities & Metropolitan Areas”„Israel-Egypt Relations: Background & Overview of Peace Treaty”„Solana to Haaretz: New Rules of War Needed for Age of Terror”„Israel's Announcement Regarding Settlements”„United Nations Security Council Resolution 497”„Security Council resolution 478 (1980) on the status of Jerusalem”„Arabs will ask U.N. to seek razing of Israeli wall”„Olmert: Willing to trade land for peace”„Mapping Peace between Syria and Israel”„Egypt: Israel must accept the land-for-peace formula”„Israel: Age structure from 2005 to 2015”„Global, regional, and national disability-adjusted life years (DALYs) for 306 diseases and injuries and healthy life expectancy (HALE) for 188 countries, 1990–2013: quantifying the epidemiological transition”10.1016/S0140-6736(15)61340-X„World Health Statistics 2014”„Life expectancy for Israeli men world's 4th highest”„Family Structure and Well-Being Across Israel's Diverse Population”„Fertility among Jewish and Muslim Women in Israel, by Level of Religiosity, 1979-2009”„Israel leaders in birth rate, but poverty major challenge”„Ethnic Groups”„Israel's population: Over 8.5 million”„Israel - Ethnic groups”„Jews, by country of origin and age”„Minority Communities in Israel: Background & Overview”„Israel”„Language in Israel”„Selected Data from the 2011 Social Survey on Mastery of the Hebrew Language and Usage of Languages”„Religions”„5 facts about Israeli Druze, a unique religious and ethnic group”„Israël”Israel Country Study Guide„Haredi city in Negev – blessing or curse?”„New town Harish harbors hopes of being more than another Pleasantville”„List of localities, in alphabetical order”„Muncitorii români, doriți în Israel”„Prietenia româno-israeliană la nevoie se cunoaște”„The Higher Education System in Israel”„Middle East”„Academic Ranking of World Universities 2016”„Israel”„Israel”„Jewish Nobel Prize Winners”„All Nobel Prizes in Literature”„All Nobel Peace Prizes”„All Prizes in Economic Sciences”„All Nobel Prizes in Chemistry”„List of Fields Medallists”„Sakharov Prize”„Țara care și-a sfidat "destinul" și se bate umăr la umăr cu Silicon Valley”„Apple's R&D center in Israel grew to about 800 employees”„Tim Cook: Apple's Herzliya R&D center second-largest in world”„Lecții de economie de la Israel”„Land use”Israel Investment and Business GuideA Country Study: IsraelCentral Bureau of StatisticsFlorin Diaconu, „Kadima: Flexibilitate și pragmatism, dar nici un compromis în chestiuni vitale", în Revista Institutului Diplomatic Român, anul I, numărul I, semestrul I, 2006, pp. 71-72Florin Diaconu, „Likud: Dreapta israeliană constant opusă retrocedării teritoriilor cureite prin luptă în 1967", în Revista Institutului Diplomatic Român, anul I, numărul I, semestrul I, 2006, pp. 73-74MassadaIsraelul a crescut in 50 de ani cât alte state intr-un mileniuIsrael Government PortalIsraelIsraelIsraelmmmmmXX451232cb118646298(data)4027808-634110000 0004 0372 0767n7900328503691455-bb46-37e3-91d2-cb064a35ffcc1003570400564274ge1294033523775214929302638955X146498911146498911

          Кастелфранко ди Сопра Становништво Референце Спољашње везе Мени за навигацију43°37′18″ СГШ; 11°33′32″ ИГД / 43.62156° СГШ; 11.55885° ИГД / 43.62156; 11.5588543°37′18″ СГШ; 11°33′32″ ИГД / 43.62156° СГШ; 11.55885° ИГД / 43.62156; 11.558853179688„The GeoNames geographical database”„Istituto Nazionale di Statistica”проширитиууWorldCat156923403n850174324558639-1cb14643287r(подаци)