| Metamath
Proof Explorer Theorem List (p. 138 of 499) | < Previous Next > | |
| Bad symbols? Try the
GIF version. |
||
|
Mirrors > Metamath Home Page > MPE Home Page > Theorem List Contents > Recent Proofs This page: Page List |
||
| Color key: | (1-30893) |
(30894-32416) |
(32417-49836) |
| Type | Label | Description |
|---|---|---|
| Statement | ||
| Theorem | flltp1 13701 | A basic property of the floor (greatest integer) function. (Contributed by NM, 24-Feb-2005.) |
| ⊢ (𝐴 ∈ ℝ → 𝐴 < ((⌊‘𝐴) + 1)) | ||
| Theorem | fllep1 13702 | A basic property of the floor (greatest integer) function. (Contributed by Mario Carneiro, 21-May-2016.) |
| ⊢ (𝐴 ∈ ℝ → 𝐴 ≤ ((⌊‘𝐴) + 1)) | ||
| Theorem | fraclt1 13703 | The fractional part of a real number is less than one. (Contributed by NM, 15-Jul-2008.) |
| ⊢ (𝐴 ∈ ℝ → (𝐴 − (⌊‘𝐴)) < 1) | ||
| Theorem | fracle1 13704 | The fractional part of a real number is less than or equal to one. (Contributed by Mario Carneiro, 21-May-2016.) |
| ⊢ (𝐴 ∈ ℝ → (𝐴 − (⌊‘𝐴)) ≤ 1) | ||
| Theorem | fracge0 13705 | The fractional part of a real number is nonnegative. (Contributed by NM, 17-Jul-2008.) |
| ⊢ (𝐴 ∈ ℝ → 0 ≤ (𝐴 − (⌊‘𝐴))) | ||
| Theorem | flge 13706 | The floor function value is the greatest integer less than or equal to its argument. (Contributed by NM, 15-Nov-2004.) (Proof shortened by Fan Zheng, 14-Jul-2016.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℤ) → (𝐵 ≤ 𝐴 ↔ 𝐵 ≤ (⌊‘𝐴))) | ||
| Theorem | fllt 13707 | The floor function value is less than the next integer. (Contributed by NM, 24-Feb-2005.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℤ) → (𝐴 < 𝐵 ↔ (⌊‘𝐴) < 𝐵)) | ||
| Theorem | flflp1 13708 | Move floor function between strict and non-strict inequality. (Contributed by Brendan Leahy, 25-Oct-2017.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ) → ((⌊‘𝐴) ≤ 𝐵 ↔ 𝐴 < ((⌊‘𝐵) + 1))) | ||
| Theorem | flid 13709 | An integer is its own floor. (Contributed by NM, 15-Nov-2004.) |
| ⊢ (𝐴 ∈ ℤ → (⌊‘𝐴) = 𝐴) | ||
| Theorem | flidm 13710 | The floor function is idempotent. (Contributed by NM, 17-Aug-2008.) |
| ⊢ (𝐴 ∈ ℝ → (⌊‘(⌊‘𝐴)) = (⌊‘𝐴)) | ||
| Theorem | flidz 13711 | A real number equals its floor iff it is an integer. (Contributed by NM, 11-Nov-2008.) |
| ⊢ (𝐴 ∈ ℝ → ((⌊‘𝐴) = 𝐴 ↔ 𝐴 ∈ ℤ)) | ||
| Theorem | flltnz 13712 | The floor of a non-integer real is less than it. (Contributed by Glauco Siliprandi, 11-Dec-2019.) |
| ⊢ ((𝐴 ∈ ℝ ∧ ¬ 𝐴 ∈ ℤ) → (⌊‘𝐴) < 𝐴) | ||
| Theorem | flwordi 13713 | Ordering relation for the floor function. (Contributed by NM, 31-Dec-2005.) (Proof shortened by Fan Zheng, 14-Jul-2016.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ ∧ 𝐴 ≤ 𝐵) → (⌊‘𝐴) ≤ (⌊‘𝐵)) | ||
| Theorem | flword2 13714 | Ordering relation for the floor function. (Contributed by Mario Carneiro, 7-Jun-2016.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ ∧ 𝐴 ≤ 𝐵) → (⌊‘𝐵) ∈ (ℤ≥‘(⌊‘𝐴))) | ||
| Theorem | flval2 13715* | An alternate way to define the floor function. (Contributed by NM, 16-Nov-2004.) |
| ⊢ (𝐴 ∈ ℝ → (⌊‘𝐴) = (℩𝑥 ∈ ℤ (𝑥 ≤ 𝐴 ∧ ∀𝑦 ∈ ℤ (𝑦 ≤ 𝐴 → 𝑦 ≤ 𝑥)))) | ||
| Theorem | flval3 13716* | An alternate way to define the floor function, as the supremum of all integers less than or equal to its argument. (Contributed by NM, 15-Nov-2004.) (Proof shortened by Mario Carneiro, 6-Sep-2014.) |
| ⊢ (𝐴 ∈ ℝ → (⌊‘𝐴) = sup({𝑥 ∈ ℤ ∣ 𝑥 ≤ 𝐴}, ℝ, < )) | ||
| Theorem | flbi 13717 | A condition equivalent to floor. (Contributed by NM, 11-Mar-2005.) (Revised by Mario Carneiro, 2-Nov-2013.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℤ) → ((⌊‘𝐴) = 𝐵 ↔ (𝐵 ≤ 𝐴 ∧ 𝐴 < (𝐵 + 1)))) | ||
| Theorem | flbi2 13718 | A condition equivalent to floor. (Contributed by NM, 15-Aug-2008.) |
| ⊢ ((𝑁 ∈ ℤ ∧ 𝐹 ∈ ℝ) → ((⌊‘(𝑁 + 𝐹)) = 𝑁 ↔ (0 ≤ 𝐹 ∧ 𝐹 < 1))) | ||
| Theorem | adddivflid 13719 | The floor of a sum of an integer and a fraction is equal to the integer iff the denominator of the fraction is less than the numerator. (Contributed by AV, 14-Jul-2021.) |
| ⊢ ((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℕ0 ∧ 𝐶 ∈ ℕ) → (𝐵 < 𝐶 ↔ (⌊‘(𝐴 + (𝐵 / 𝐶))) = 𝐴)) | ||
| Theorem | ico01fl0 13720 | The floor of a real number in [0, 1) is 0. Remark: may shorten the proof of modid 13797 or a version of it where the antecedent is membership in an interval. (Contributed by BJ, 29-Jun-2019.) |
| ⊢ (𝐴 ∈ (0[,)1) → (⌊‘𝐴) = 0) | ||
| Theorem | flge0nn0 13721 | The floor of a number greater than or equal to 0 is a nonnegative integer. (Contributed by NM, 26-Apr-2005.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 0 ≤ 𝐴) → (⌊‘𝐴) ∈ ℕ0) | ||
| Theorem | flge1nn 13722 | The floor of a number greater than or equal to 1 is a positive integer. (Contributed by NM, 26-Apr-2005.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 1 ≤ 𝐴) → (⌊‘𝐴) ∈ ℕ) | ||
| Theorem | fldivnn0 13723 | The floor function of a division of a nonnegative integer by a positive integer is a nonnegative integer. (Contributed by Alexander van der Vekens, 14-Apr-2018.) |
| ⊢ ((𝐾 ∈ ℕ0 ∧ 𝐿 ∈ ℕ) → (⌊‘(𝐾 / 𝐿)) ∈ ℕ0) | ||
| Theorem | refldivcl 13724 | The floor function of a division of a real number by a positive real number is a real number. (Contributed by Alexander van der Vekens, 14-Apr-2018.) |
| ⊢ ((𝐾 ∈ ℝ ∧ 𝐿 ∈ ℝ+) → (⌊‘(𝐾 / 𝐿)) ∈ ℝ) | ||
| Theorem | divfl0 13725 | The floor of a fraction is 0 iff the denominator is less than the numerator. (Contributed by AV, 8-Jul-2021.) |
| ⊢ ((𝐴 ∈ ℕ0 ∧ 𝐵 ∈ ℕ) → (𝐴 < 𝐵 ↔ (⌊‘(𝐴 / 𝐵)) = 0)) | ||
| Theorem | fladdz 13726 | An integer can be moved in and out of the floor of a sum. (Contributed by NM, 27-Apr-2005.) (Proof shortened by Fan Zheng, 16-Jun-2016.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝑁 ∈ ℤ) → (⌊‘(𝐴 + 𝑁)) = ((⌊‘𝐴) + 𝑁)) | ||
| Theorem | flzadd 13727 | An integer can be moved in and out of the floor of a sum. (Contributed by NM, 2-Jan-2009.) |
| ⊢ ((𝑁 ∈ ℤ ∧ 𝐴 ∈ ℝ) → (⌊‘(𝑁 + 𝐴)) = (𝑁 + (⌊‘𝐴))) | ||
| Theorem | flmulnn0 13728 | Move a nonnegative integer in and out of a floor. (Contributed by NM, 2-Jan-2009.) (Proof shortened by Fan Zheng, 7-Jun-2016.) |
| ⊢ ((𝑁 ∈ ℕ0 ∧ 𝐴 ∈ ℝ) → (𝑁 · (⌊‘𝐴)) ≤ (⌊‘(𝑁 · 𝐴))) | ||
| Theorem | btwnzge0 13729 | A real bounded between an integer and its successor is nonnegative iff the integer is nonnegative. Second half of Lemma 13-4.1 of [Gleason] p. 217. (For the first half see rebtwnz 12842.) (Contributed by NM, 12-Mar-2005.) |
| ⊢ (((𝐴 ∈ ℝ ∧ 𝑁 ∈ ℤ) ∧ (𝑁 ≤ 𝐴 ∧ 𝐴 < (𝑁 + 1))) → (0 ≤ 𝐴 ↔ 0 ≤ 𝑁)) | ||
| Theorem | 2tnp1ge0ge0 13730 | Two times an integer plus one is not negative iff the integer is not negative. (Contributed by AV, 19-Jun-2021.) (Proof shortened by AV, 10-Jul-2022.) |
| ⊢ (𝑁 ∈ ℤ → (0 ≤ ((2 · 𝑁) + 1) ↔ 0 ≤ 𝑁)) | ||
| Theorem | flhalf 13731 | Ordering relation for the floor of half of an integer. (Contributed by NM, 1-Jan-2006.) (Proof shortened by Mario Carneiro, 7-Jun-2016.) |
| ⊢ (𝑁 ∈ ℤ → 𝑁 ≤ (2 · (⌊‘((𝑁 + 1) / 2)))) | ||
| Theorem | fldivle 13732 | The floor function of a division of a real number by a positive real number is less than or equal to the division. (Contributed by Alexander van der Vekens, 14-Apr-2018.) |
| ⊢ ((𝐾 ∈ ℝ ∧ 𝐿 ∈ ℝ+) → (⌊‘(𝐾 / 𝐿)) ≤ (𝐾 / 𝐿)) | ||
| Theorem | fldivnn0le 13733 | The floor function of a division of a nonnegative integer by a positive integer is less than or equal to the division. (Contributed by Alexander van der Vekens, 14-Apr-2018.) |
| ⊢ ((𝐾 ∈ ℕ0 ∧ 𝐿 ∈ ℕ) → (⌊‘(𝐾 / 𝐿)) ≤ (𝐾 / 𝐿)) | ||
| Theorem | flltdivnn0lt 13734 | The floor function of a division of a nonnegative integer by a positive integer is less than the division of a greater dividend by the same positive integer. (Contributed by Alexander van der Vekens, 14-Apr-2018.) |
| ⊢ ((𝐾 ∈ ℕ0 ∧ 𝑁 ∈ ℕ0 ∧ 𝐿 ∈ ℕ) → (𝐾 < 𝑁 → (⌊‘(𝐾 / 𝐿)) < (𝑁 / 𝐿))) | ||
| Theorem | ltdifltdiv 13735 | If the dividend of a division is less than the difference between a real number and the divisor, the floor function of the division plus 1 is less than the division of the real number by the divisor. (Contributed by Alexander van der Vekens, 14-Apr-2018.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+ ∧ 𝐶 ∈ ℝ) → (𝐴 < (𝐶 − 𝐵) → ((⌊‘(𝐴 / 𝐵)) + 1) < (𝐶 / 𝐵))) | ||
| Theorem | fldiv4p1lem1div2 13736 | The floor of an integer equal to 3 or greater than 4, increased by 1, is less than or equal to the half of the integer minus 1. (Contributed by AV, 8-Jul-2021.) |
| ⊢ ((𝑁 = 3 ∨ 𝑁 ∈ (ℤ≥‘5)) → ((⌊‘(𝑁 / 4)) + 1) ≤ ((𝑁 − 1) / 2)) | ||
| Theorem | fldiv4lem1div2uz2 13737 | The floor of an integer greater than 1, divided by 4 is less than or equal to the half of the integer minus 1. (Contributed by AV, 5-Jul-2021.) (Proof shortened by AV, 9-Jul-2022.) |
| ⊢ (𝑁 ∈ (ℤ≥‘2) → (⌊‘(𝑁 / 4)) ≤ ((𝑁 − 1) / 2)) | ||
| Theorem | fldiv4lem1div2 13738 | The floor of a positive integer divided by 4 is less than or equal to the half of the integer minus 1. (Contributed by AV, 9-Jul-2021.) |
| ⊢ (𝑁 ∈ ℕ → (⌊‘(𝑁 / 4)) ≤ ((𝑁 − 1) / 2)) | ||
| Theorem | ceilval 13739 | The value of the ceiling function. (Contributed by David A. Wheeler, 19-May-2015.) |
| ⊢ (𝐴 ∈ ℝ → (⌈‘𝐴) = -(⌊‘-𝐴)) | ||
| Theorem | dfceil2 13740* | Alternative definition of the ceiling function using restricted iota. (Contributed by AV, 1-Dec-2018.) |
| ⊢ ⌈ = (𝑥 ∈ ℝ ↦ (℩𝑦 ∈ ℤ (𝑥 ≤ 𝑦 ∧ 𝑦 < (𝑥 + 1)))) | ||
| Theorem | ceilval2 13741* | The value of the ceiling function using restricted iota. (Contributed by AV, 1-Dec-2018.) |
| ⊢ (𝐴 ∈ ℝ → (⌈‘𝐴) = (℩𝑦 ∈ ℤ (𝐴 ≤ 𝑦 ∧ 𝑦 < (𝐴 + 1)))) | ||
| Theorem | ceicl 13742 | The ceiling function returns an integer (closure law). (Contributed by Jeff Hankins, 10-Jun-2007.) |
| ⊢ (𝐴 ∈ ℝ → -(⌊‘-𝐴) ∈ ℤ) | ||
| Theorem | ceilcl 13743 | Closure of the ceiling function. (Contributed by David A. Wheeler, 19-May-2015.) |
| ⊢ (𝐴 ∈ ℝ → (⌈‘𝐴) ∈ ℤ) | ||
| Theorem | ceilcld 13744 | Closure of the ceiling function. (Contributed by Glauco Siliprandi, 2-Jan-2022.) |
| ⊢ (𝜑 → 𝐴 ∈ ℝ) ⇒ ⊢ (𝜑 → (⌈‘𝐴) ∈ ℤ) | ||
| Theorem | ceige 13745 | The ceiling of a real number is greater than or equal to that number. (Contributed by Jeff Hankins, 10-Jun-2007.) |
| ⊢ (𝐴 ∈ ℝ → 𝐴 ≤ -(⌊‘-𝐴)) | ||
| Theorem | ceilge 13746 | The ceiling of a real number is greater than or equal to that number. (Contributed by AV, 30-Nov-2018.) |
| ⊢ (𝐴 ∈ ℝ → 𝐴 ≤ (⌈‘𝐴)) | ||
| Theorem | ceilged 13747 | The ceiling of a real number is greater than or equal to that number. (Contributed by Glauco Siliprandi, 2-Jan-2022.) |
| ⊢ (𝜑 → 𝐴 ∈ ℝ) ⇒ ⊢ (𝜑 → 𝐴 ≤ (⌈‘𝐴)) | ||
| Theorem | ceim1l 13748 | One less than the ceiling of a real number is strictly less than that number. (Contributed by Jeff Hankins, 10-Jun-2007.) |
| ⊢ (𝐴 ∈ ℝ → (-(⌊‘-𝐴) − 1) < 𝐴) | ||
| Theorem | ceilm1lt 13749 | One less than the ceiling of a real number is strictly less than that number. (Contributed by AV, 30-Nov-2018.) |
| ⊢ (𝐴 ∈ ℝ → ((⌈‘𝐴) − 1) < 𝐴) | ||
| Theorem | ceile 13750 | The ceiling of a real number is the smallest integer greater than or equal to it. (Contributed by Jeff Hankins, 10-Jun-2007.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℤ ∧ 𝐴 ≤ 𝐵) → -(⌊‘-𝐴) ≤ 𝐵) | ||
| Theorem | ceille 13751 | The ceiling of a real number is the smallest integer greater than or equal to it. (Contributed by AV, 30-Nov-2018.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℤ ∧ 𝐴 ≤ 𝐵) → (⌈‘𝐴) ≤ 𝐵) | ||
| Theorem | ceilid 13752 | An integer is its own ceiling. (Contributed by AV, 30-Nov-2018.) |
| ⊢ (𝐴 ∈ ℤ → (⌈‘𝐴) = 𝐴) | ||
| Theorem | ceilidz 13753 | A real number equals its ceiling iff it is an integer. (Contributed by AV, 30-Nov-2018.) |
| ⊢ (𝐴 ∈ ℝ → (𝐴 ∈ ℤ ↔ (⌈‘𝐴) = 𝐴)) | ||
| Theorem | flleceil 13754 | The floor of a real number is less than or equal to its ceiling. (Contributed by AV, 30-Nov-2018.) |
| ⊢ (𝐴 ∈ ℝ → (⌊‘𝐴) ≤ (⌈‘𝐴)) | ||
| Theorem | fleqceilz 13755 | A real number is an integer iff its floor equals its ceiling. (Contributed by AV, 30-Nov-2018.) |
| ⊢ (𝐴 ∈ ℝ → (𝐴 ∈ ℤ ↔ (⌊‘𝐴) = (⌈‘𝐴))) | ||
| Theorem | quoremz 13756 | Quotient and remainder of an integer divided by a positive integer. TODO - is this really needed for anything? Should we use mod to simplify it? Remark (AV): This is a special case of divalg 16311. (Contributed by NM, 14-Aug-2008.) |
| ⊢ 𝑄 = (⌊‘(𝐴 / 𝐵)) & ⊢ 𝑅 = (𝐴 − (𝐵 · 𝑄)) ⇒ ⊢ ((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℕ) → ((𝑄 ∈ ℤ ∧ 𝑅 ∈ ℕ0) ∧ (𝑅 < 𝐵 ∧ 𝐴 = ((𝐵 · 𝑄) + 𝑅)))) | ||
| Theorem | quoremnn0 13757 | Quotient and remainder of a nonnegative integer divided by a positive integer. (Contributed by NM, 14-Aug-2008.) |
| ⊢ 𝑄 = (⌊‘(𝐴 / 𝐵)) & ⊢ 𝑅 = (𝐴 − (𝐵 · 𝑄)) ⇒ ⊢ ((𝐴 ∈ ℕ0 ∧ 𝐵 ∈ ℕ) → ((𝑄 ∈ ℕ0 ∧ 𝑅 ∈ ℕ0) ∧ (𝑅 < 𝐵 ∧ 𝐴 = ((𝐵 · 𝑄) + 𝑅)))) | ||
| Theorem | quoremnn0ALT 13758 | Alternate proof of quoremnn0 13757 not using quoremz 13756. TODO - Keep either quoremnn0ALT 13758 (if we don't keep quoremz 13756) or quoremnn0 13757? (Contributed by NM, 14-Aug-2008.) (Proof modification is discouraged.) (New usage is discouraged.) |
| ⊢ 𝑄 = (⌊‘(𝐴 / 𝐵)) & ⊢ 𝑅 = (𝐴 − (𝐵 · 𝑄)) ⇒ ⊢ ((𝐴 ∈ ℕ0 ∧ 𝐵 ∈ ℕ) → ((𝑄 ∈ ℕ0 ∧ 𝑅 ∈ ℕ0) ∧ (𝑅 < 𝐵 ∧ 𝐴 = ((𝐵 · 𝑄) + 𝑅)))) | ||
| Theorem | intfrac2 13759 | Decompose a real into integer and fractional parts. TODO - should we replace this with intfrac 13787? (Contributed by NM, 16-Aug-2008.) |
| ⊢ 𝑍 = (⌊‘𝐴) & ⊢ 𝐹 = (𝐴 − 𝑍) ⇒ ⊢ (𝐴 ∈ ℝ → (0 ≤ 𝐹 ∧ 𝐹 < 1 ∧ 𝐴 = (𝑍 + 𝐹))) | ||
| Theorem | intfracq 13760 | Decompose a rational number, expressed as a ratio, into integer and fractional parts. The fractional part has a tighter bound than that of intfrac2 13759. (Contributed by NM, 16-Aug-2008.) |
| ⊢ 𝑍 = (⌊‘(𝑀 / 𝑁)) & ⊢ 𝐹 = ((𝑀 / 𝑁) − 𝑍) ⇒ ⊢ ((𝑀 ∈ ℤ ∧ 𝑁 ∈ ℕ) → (0 ≤ 𝐹 ∧ 𝐹 ≤ ((𝑁 − 1) / 𝑁) ∧ (𝑀 / 𝑁) = (𝑍 + 𝐹))) | ||
| Theorem | fldiv 13761 | Cancellation of the embedded floor of a real divided by an integer. (Contributed by NM, 16-Aug-2008.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝑁 ∈ ℕ) → (⌊‘((⌊‘𝐴) / 𝑁)) = (⌊‘(𝐴 / 𝑁))) | ||
| Theorem | fldiv2 13762 | Cancellation of an embedded floor of a ratio. Generalization of Equation 2.4 in [CormenLeisersonRivest] p. 33 (where 𝐴 must be an integer). (Contributed by NM, 9-Nov-2008.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝑀 ∈ ℕ ∧ 𝑁 ∈ ℕ) → (⌊‘((⌊‘(𝐴 / 𝑀)) / 𝑁)) = (⌊‘(𝐴 / (𝑀 · 𝑁)))) | ||
| Theorem | fznnfl 13763 | Finite set of sequential integers starting at 1 and ending at a real number. (Contributed by Mario Carneiro, 3-May-2016.) |
| ⊢ (𝑁 ∈ ℝ → (𝐾 ∈ (1...(⌊‘𝑁)) ↔ (𝐾 ∈ ℕ ∧ 𝐾 ≤ 𝑁))) | ||
| Theorem | uzsup 13764 | An upper set of integers is unbounded above. (Contributed by Mario Carneiro, 7-May-2016.) |
| ⊢ 𝑍 = (ℤ≥‘𝑀) ⇒ ⊢ (𝑀 ∈ ℤ → sup(𝑍, ℝ*, < ) = +∞) | ||
| Theorem | ioopnfsup 13765 | An upper set of reals is unbounded above. (Contributed by Mario Carneiro, 7-May-2016.) |
| ⊢ ((𝐴 ∈ ℝ* ∧ 𝐴 ≠ +∞) → sup((𝐴(,)+∞), ℝ*, < ) = +∞) | ||
| Theorem | icopnfsup 13766 | An upper set of reals is unbounded above. (Contributed by Mario Carneiro, 7-May-2016.) |
| ⊢ ((𝐴 ∈ ℝ* ∧ 𝐴 ≠ +∞) → sup((𝐴[,)+∞), ℝ*, < ) = +∞) | ||
| Theorem | rpsup 13767 | The positive reals are unbounded above. (Contributed by Mario Carneiro, 7-May-2016.) |
| ⊢ sup(ℝ+, ℝ*, < ) = +∞ | ||
| Theorem | resup 13768 | The real numbers are unbounded above. (Contributed by Mario Carneiro, 7-May-2016.) |
| ⊢ sup(ℝ, ℝ*, < ) = +∞ | ||
| Theorem | xrsup 13769 | The extended real numbers are unbounded above. (Contributed by Mario Carneiro, 7-May-2016.) |
| ⊢ sup(ℝ*, ℝ*, < ) = +∞ | ||
| Syntax | cmo 13770 | Extend class notation with the modulo operation. |
| class mod | ||
| Definition | df-mod 13771* | Define the modulo (remainder) operation. See modval 13772 for its value. For example, (5 mod 3) = 2 and (-7 mod 2) = 1 (ex-mod 30424). (Contributed by NM, 10-Nov-2008.) |
| ⊢ mod = (𝑥 ∈ ℝ, 𝑦 ∈ ℝ+ ↦ (𝑥 − (𝑦 · (⌊‘(𝑥 / 𝑦))))) | ||
| Theorem | modval 13772 | The value of the modulo operation. The modulo congruence notation of number theory, 𝐽≡𝐾 (modulo 𝑁), can be expressed in our notation as (𝐽 mod 𝑁) = (𝐾 mod 𝑁). Definition 1 in Knuth, The Art of Computer Programming, Vol. I (1972), p. 38. Knuth uses "mod" for the operation and "modulo" for the congruence. Unlike Knuth, we restrict the second argument to positive reals to simplify certain theorems. (This also gives us future flexibility to extend it to any one of several different conventions for a zero or negative second argument, should there be an advantage in doing so.) (Contributed by NM, 10-Nov-2008.) (Revised by Mario Carneiro, 3-Nov-2013.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → (𝐴 mod 𝐵) = (𝐴 − (𝐵 · (⌊‘(𝐴 / 𝐵))))) | ||
| Theorem | modvalr 13773 | The value of the modulo operation (multiplication in reversed order). (Contributed by Alexander van der Vekens, 14-Apr-2018.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → (𝐴 mod 𝐵) = (𝐴 − ((⌊‘(𝐴 / 𝐵)) · 𝐵))) | ||
| Theorem | modcl 13774 | Closure law for the modulo operation. (Contributed by NM, 10-Nov-2008.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → (𝐴 mod 𝐵) ∈ ℝ) | ||
| Theorem | flpmodeq 13775 | Partition of a division into its integer part and the remainder. (Contributed by Alexander van der Vekens, 14-Apr-2018.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → (((⌊‘(𝐴 / 𝐵)) · 𝐵) + (𝐴 mod 𝐵)) = 𝐴) | ||
| Theorem | modcld 13776 | Closure law for the modulo operation. (Contributed by Mario Carneiro, 28-May-2016.) |
| ⊢ (𝜑 → 𝐴 ∈ ℝ) & ⊢ (𝜑 → 𝐵 ∈ ℝ+) ⇒ ⊢ (𝜑 → (𝐴 mod 𝐵) ∈ ℝ) | ||
| Theorem | mod0 13777 | 𝐴 mod 𝐵 is zero iff 𝐴 is evenly divisible by 𝐵. (Contributed by Jeff Madsen, 2-Sep-2009.) (Proof shortened by Fan Zheng, 7-Jun-2016.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → ((𝐴 mod 𝐵) = 0 ↔ (𝐴 / 𝐵) ∈ ℤ)) | ||
| Theorem | mulmod0 13778 | The product of an integer and a positive real number is 0 modulo the positive real number. (Contributed by Alexander van der Vekens, 17-May-2018.) (Revised by AV, 5-Jul-2020.) |
| ⊢ ((𝐴 ∈ ℤ ∧ 𝑀 ∈ ℝ+) → ((𝐴 · 𝑀) mod 𝑀) = 0) | ||
| Theorem | negmod0 13779 | 𝐴 is divisible by 𝐵 iff its negative is. (Contributed by Jeff Madsen, 2-Sep-2009.) (Proof shortened by Fan Zheng, 7-Jun-2016.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → ((𝐴 mod 𝐵) = 0 ↔ (-𝐴 mod 𝐵) = 0)) | ||
| Theorem | modge0 13780 | The modulo operation is nonnegative. (Contributed by NM, 10-Nov-2008.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → 0 ≤ (𝐴 mod 𝐵)) | ||
| Theorem | modlt 13781 | The modulo operation is less than its second argument. (Contributed by NM, 10-Nov-2008.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → (𝐴 mod 𝐵) < 𝐵) | ||
| Theorem | modelico 13782 | Modular reduction produces a half-open interval. (Contributed by Stefan O'Rear, 12-Sep-2014.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → (𝐴 mod 𝐵) ∈ (0[,)𝐵)) | ||
| Theorem | moddiffl 13783 | Value of the modulo operation rewritten to give two ways of expressing the quotient when "𝐴 is divided by 𝐵 using Euclidean division." Multiplying both sides by 𝐵, this implies that 𝐴 mod 𝐵 differs from 𝐴 by an integer multiple of 𝐵. (Contributed by Jeff Madsen, 17-Jun-2010.) (Revised by Mario Carneiro, 6-Sep-2016.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → ((𝐴 − (𝐴 mod 𝐵)) / 𝐵) = (⌊‘(𝐴 / 𝐵))) | ||
| Theorem | moddifz 13784 | The modulo operation differs from 𝐴 by an integer multiple of 𝐵. (Contributed by Mario Carneiro, 15-Jul-2014.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → ((𝐴 − (𝐴 mod 𝐵)) / 𝐵) ∈ ℤ) | ||
| Theorem | modfrac 13785 | The fractional part of a number is the number modulo 1. (Contributed by NM, 11-Nov-2008.) |
| ⊢ (𝐴 ∈ ℝ → (𝐴 mod 1) = (𝐴 − (⌊‘𝐴))) | ||
| Theorem | flmod 13786 | The floor function expressed in terms of the modulo operation. (Contributed by NM, 11-Nov-2008.) |
| ⊢ (𝐴 ∈ ℝ → (⌊‘𝐴) = (𝐴 − (𝐴 mod 1))) | ||
| Theorem | intfrac 13787 | Break a number into its integer part and its fractional part. (Contributed by NM, 31-Dec-2008.) |
| ⊢ (𝐴 ∈ ℝ → 𝐴 = ((⌊‘𝐴) + (𝐴 mod 1))) | ||
| Theorem | zmod10 13788 | An integer modulo 1 is 0. (Contributed by Paul Chapman, 22-Jun-2011.) |
| ⊢ (𝑁 ∈ ℤ → (𝑁 mod 1) = 0) | ||
| Theorem | zmod1congr 13789 | Two arbitrary integers are congruent modulo 1, see example 4 in [ApostolNT] p. 107. (Contributed by AV, 21-Jul-2021.) |
| ⊢ ((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℤ) → (𝐴 mod 1) = (𝐵 mod 1)) | ||
| Theorem | modmulnn 13790 | Move a positive integer in and out of a floor in the first argument of a modulo operation. (Contributed by NM, 2-Jan-2009.) |
| ⊢ ((𝑁 ∈ ℕ ∧ 𝐴 ∈ ℝ ∧ 𝑀 ∈ ℕ) → ((𝑁 · (⌊‘𝐴)) mod (𝑁 · 𝑀)) ≤ ((⌊‘(𝑁 · 𝐴)) mod (𝑁 · 𝑀))) | ||
| Theorem | modvalp1 13791 | The value of the modulo operation (expressed with sum of denominator and nominator). (Contributed by Alexander van der Vekens, 14-Apr-2018.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → ((𝐴 + 𝐵) − (((⌊‘(𝐴 / 𝐵)) + 1) · 𝐵)) = (𝐴 mod 𝐵)) | ||
| Theorem | zmodcl 13792 | Closure law for the modulo operation restricted to integers. (Contributed by NM, 27-Nov-2008.) |
| ⊢ ((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℕ) → (𝐴 mod 𝐵) ∈ ℕ0) | ||
| Theorem | zmodcld 13793 | Closure law for the modulo operation restricted to integers. (Contributed by Mario Carneiro, 28-May-2016.) |
| ⊢ (𝜑 → 𝐴 ∈ ℤ) & ⊢ (𝜑 → 𝐵 ∈ ℕ) ⇒ ⊢ (𝜑 → (𝐴 mod 𝐵) ∈ ℕ0) | ||
| Theorem | zmodfz 13794 | An integer mod 𝐵 lies in the first 𝐵 nonnegative integers. (Contributed by Jeff Madsen, 17-Jun-2010.) |
| ⊢ ((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℕ) → (𝐴 mod 𝐵) ∈ (0...(𝐵 − 1))) | ||
| Theorem | zmodfzo 13795 | An integer mod 𝐵 lies in the first 𝐵 nonnegative integers. (Contributed by Stefan O'Rear, 6-Sep-2015.) |
| ⊢ ((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℕ) → (𝐴 mod 𝐵) ∈ (0..^𝐵)) | ||
| Theorem | zmodfzp1 13796 | An integer mod 𝐵 lies in the first 𝐵 + 1 nonnegative integers. (Contributed by AV, 27-Oct-2018.) |
| ⊢ ((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℕ) → (𝐴 mod 𝐵) ∈ (0...𝐵)) | ||
| Theorem | modid 13797 | Identity law for modulo. (Contributed by NM, 29-Dec-2008.) |
| ⊢ (((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) ∧ (0 ≤ 𝐴 ∧ 𝐴 < 𝐵)) → (𝐴 mod 𝐵) = 𝐴) | ||
| Theorem | modid0 13798 | A positive real number modulo itself is 0. (Contributed by Alexander van der Vekens, 15-May-2018.) |
| ⊢ (𝑁 ∈ ℝ+ → (𝑁 mod 𝑁) = 0) | ||
| Theorem | modid2 13799 | Identity law for modulo. (Contributed by NM, 29-Dec-2008.) |
| ⊢ ((𝐴 ∈ ℝ ∧ 𝐵 ∈ ℝ+) → ((𝐴 mod 𝐵) = 𝐴 ↔ (0 ≤ 𝐴 ∧ 𝐴 < 𝐵))) | ||
| Theorem | zmodid2 13800 | Identity law for modulo restricted to integers. (Contributed by Paul Chapman, 22-Jun-2011.) |
| ⊢ ((𝑀 ∈ ℤ ∧ 𝑁 ∈ ℕ) → ((𝑀 mod 𝑁) = 𝑀 ↔ 𝑀 ∈ (0...(𝑁 − 1)))) | ||
| < Previous Next > |
| Copyright terms: Public domain | < Previous Next > |