Skip to content

Commit 5e8f5b9

Browse files
goualard-fgoualard-f
committed
Initial commit
0 parents  commit 5e8f5b9

27 files changed

+1839
-0
lines changed

‎.gitignore

+6
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,6 @@
1+
*.jl.cov
2+
*.jl.*.cov
3+
*.jl.mem
4+
/docs/build/
5+
/docs/Manifest.toml
6+
Manifest.toml

‎License.md

+165
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,165 @@
1+
GNU LESSER GENERAL PUBLIC LICENSE
2+
Version 3, 29 June 2007
3+
4+
Copyright (C) 2007 Free Software Foundation, Inc. <http://fsf.org/>
5+
Everyone is permitted to copy and distribute verbatim copies
6+
of this license document, but changing it is not allowed.
7+
8+
9+
This version of the GNU Lesser General Public License incorporates
10+
the terms and conditions of version 3 of the GNU General Public
11+
License, supplemented by the additional permissions listed below.
12+
13+
0. Additional Definitions.
14+
15+
As used herein, "this License" refers to version 3 of the GNU Lesser
16+
General Public License, and the "GNU GPL" refers to version 3 of the GNU
17+
General Public License.
18+
19+
"The Library" refers to a covered work governed by this License,
20+
other than an Application or a Combined Work as defined below.
21+
22+
An "Application" is any work that makes use of an interface provided
23+
by the Library, but which is not otherwise based on the Library.
24+
Defining a subclass of a class defined by the Library is deemed a mode
25+
of using an interface provided by the Library.
26+
27+
A "Combined Work" is a work produced by combining or linking an
28+
Application with the Library. The particular version of the Library
29+
with which the Combined Work was made is also called the "Linked
30+
Version".
31+
32+
The "Minimal Corresponding Source" for a Combined Work means the
33+
Corresponding Source for the Combined Work, excluding any source code
34+
for portions of the Combined Work that, considered in isolation, are
35+
based on the Application, and not on the Linked Version.
36+
37+
The "Corresponding Application Code" for a Combined Work means the
38+
object code and/or source code for the Application, including any data
39+
and utility programs needed for reproducing the Combined Work from the
40+
Application, but excluding the System Libraries of the Combined Work.
41+
42+
1. Exception to Section 3 of the GNU GPL.
43+
44+
You may convey a covered work under sections 3 and 4 of this License
45+
without being bound by section 3 of the GNU GPL.
46+
47+
2. Conveying Modified Versions.
48+
49+
If you modify a copy of the Library, and, in your modifications, a
50+
facility refers to a function or data to be supplied by an Application
51+
that uses the facility (other than as an argument passed when the
52+
facility is invoked), then you may convey a copy of the modified
53+
version:
54+
55+
a) under this License, provided that you make a good faith effort to
56+
ensure that, in the event an Application does not supply the
57+
function or data, the facility still operates, and performs
58+
whatever part of its purpose remains meaningful, or
59+
60+
b) under the GNU GPL, with none of the additional permissions of
61+
this License applicable to that copy.
62+
63+
3. Object Code Incorporating Material from Library Header Files.
64+
65+
The object code form of an Application may incorporate material from
66+
a header file that is part of the Library. You may convey such object
67+
code under terms of your choice, provided that, if the incorporated
68+
material is not limited to numerical parameters, data structure
69+
layouts and accessors, or small macros, inline functions and templates
70+
(ten or fewer lines in length), you do both of the following:
71+
72+
a) Give prominent notice with each copy of the object code that the
73+
Library is used in it and that the Library and its use are
74+
covered by this License.
75+
76+
b) Accompany the object code with a copy of the GNU GPL and this license
77+
document.
78+
79+
4. Combined Works.
80+
81+
You may convey a Combined Work under terms of your choice that,
82+
taken together, effectively do not restrict modification of the
83+
portions of the Library contained in the Combined Work and reverse
84+
engineering for debugging such modifications, if you also do each of
85+
the following:
86+
87+
a) Give prominent notice with each copy of the Combined Work that
88+
the Library is used in it and that the Library and its use are
89+
covered by this License.
90+
91+
b) Accompany the Combined Work with a copy of the GNU GPL and this license
92+
document.
93+
94+
c) For a Combined Work that displays copyright notices during
95+
execution, include the copyright notice for the Library among
96+
these notices, as well as a reference directing the user to the
97+
copies of the GNU GPL and this license document.
98+
99+
d) Do one of the following:
100+
101+
0) Convey the Minimal Corresponding Source under the terms of this
102+
License, and the Corresponding Application Code in a form
103+
suitable for, and under terms that permit, the user to
104+
recombine or relink the Application with a modified version of
105+
the Linked Version to produce a modified Combined Work, in the
106+
manner specified by section 6 of the GNU GPL for conveying
107+
Corresponding Source.
108+
109+
1) Use a suitable shared library mechanism for linking with the
110+
Library. A suitable mechanism is one that (a) uses at run time
111+
a copy of the Library already present on the user's computer
112+
system, and (b) will operate properly with a modified version
113+
of the Library that is interface-compatible with the Linked
114+
Version.
115+
116+
e) Provide Installation Information, but only if you would otherwise
117+
be required to provide such information under section 6 of the
118+
GNU GPL, and only to the extent that such information is
119+
necessary to install and execute a modified version of the
120+
Combined Work produced by recombining or relinking the
121+
Application with a modified version of the Linked Version. (If
122+
you use option 4d0, the Installation Information must accompany
123+
the Minimal Corresponding Source and Corresponding Application
124+
Code. If you use option 4d1, you must provide the Installation
125+
Information in the manner specified by section 6 of the GNU GPL
126+
for conveying Corresponding Source.)
127+
128+
5. Combined Libraries.
129+
130+
You may place library facilities that are a work based on the
131+
Library side by side in a single library together with other library
132+
facilities that are not Applications and are not covered by this
133+
License, and convey such a combined library under terms of your
134+
choice, if you do both of the following:
135+
136+
a) Accompany the combined library with a copy of the same work based
137+
on the Library, uncombined with any other library facilities,
138+
conveyed under the terms of this License.
139+
140+
b) Give prominent notice with the combined library that part of it
141+
is a work based on the Library, and explaining where to find the
142+
accompanying uncombined form of the same work.
143+
144+
6. Revised Versions of the GNU Lesser General Public License.
145+
146+
The Free Software Foundation may publish revised and/or new versions
147+
of the GNU Lesser General Public License from time to time. Such new
148+
versions will be similar in spirit to the present version, but may
149+
differ in detail to address new problems or concerns.
150+
151+
Each version is given a distinguishing version number. If the
152+
Library as you received it specifies that a certain numbered version
153+
of the GNU Lesser General Public License "or any later version"
154+
applies to it, you have the option of following the terms and
155+
conditions either of that published version or of any later version
156+
published by the Free Software Foundation. If the Library as you
157+
received it does not specify a version number of the GNU Lesser
158+
General Public License, you may choose any version of the GNU Lesser
159+
General Public License ever published by the Free Software Foundation.
160+
161+
If the Library as you received it specifies that a proxy can decide
162+
whether future versions of the GNU Lesser General Public License shall
163+
apply, that proxy's public statement of acceptance of any version is
164+
permanent authorization for you to choose that version for the
165+
Library.

‎Project.toml

+11
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,11 @@
1+
name = "MicroFloatingPoints"
2+
uuid = "1b3dacd6-29d6-4be1-b0be-55e841209a5b"
3+
authors = ["Frédéric Goualard <[email protected]>"]
4+
version = "0.1.0"
5+
6+
[deps]
7+
Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4"
8+
PyPlot = "d330b81b-6aea-500a-939a-2ce795aea3ee"
9+
Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
10+
Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
11+
TestSetExtensions = "98d24dd4-01ad-11ea-1b02-c9a08f80db04"

‎README.md

+7
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,7 @@
1+
# MicroFloatingPoints
2+
3+
*The MicroFloatingPoint package offers a parameterized type to manipulate [IEEE 754 standard](https://en.wikipedia.org/wiki/IEEE_754)-compliant floating-point numbers with a very small size (from 5 bits up to 32 bits).
4+
5+
## Getting started
6+
7+
TODO

‎docs/make.jl

+8
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,8 @@
1+
push!(LOAD_PATH,"../src")
2+
using Documenter, MicroFloatingPoint
3+
4+
makedocs(
5+
sitename="The MicroFloatingPoint Documentation",
6+
authors = "Frédéric Goualard"
7+
# modules = [MicroFloatingPoint],
8+
)

‎docs/src/index.md

+43
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,43 @@
1+
# MicroFloatingPoint
2+
3+
`MicroFloatingPoint` is a Julia library to manipulate small [IEEE-754](https://en.wikipedia.org/wiki/IEEE_754)-compliant floating-point numbers that are smaller or equal to the `Float32` format mandated by the standard.
4+
5+
The library may serve to exemplify the behavior of IEEE 754 floating-point numbers in a systematic way through the use of very small formats.
6+
7+
## A guided tour
8+
9+
```@meta
10+
DocTestSetup = quote
11+
using MicroFloatingPoint
12+
end
13+
```
14+
15+
```jldoctest
16+
julia> a = Floatmu{8,23}(2.7)
17+
2.700000047683716
18+
```
19+
20+
```@meta
21+
CurrentModule = MicroFloatingPoint
22+
```
23+
24+
```@docs
25+
Floatmu{szE,szf}
26+
```
27+
28+
```@docs
29+
convert(::Type{Float64}, x::Floatmu{szE,szf} where {szE, szf})
30+
convert(::Type{Float32}, x::Floatmu{szE,szf} where {szE, szf})
31+
```
32+
33+
```@docs
34+
parse(::Type{Floatmu{szE,szf}}, str::AbstractString) where {szE, szf}
35+
```
36+
37+
```@docs
38+
signbit(x::Floatmu{szE,szf}) where {szE, szf}
39+
```
40+
41+
# Developer Documentation
42+
43+
Internals.

‎src/MFPPlot.jl

+57
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,57 @@
1+
# MFPPlot --
2+
#
3+
# Copyright 2019--2021 University of Nantes, France.
4+
#
5+
# This file is part of the MicroFloatingPoint library.
6+
#
7+
# The MicroFloatingPoint library is free software; you can redistribute it and/or
8+
# modify it under the terms of the GNU Lesser General Public License as published
9+
# by the Free Software Foundation; either version 3 of the License, or (at your
10+
# option) any later version.
11+
#
12+
# The MicroFloatingPoint library is distributed in the hope that it will be useful,
13+
# but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
14+
# or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
15+
# for more details.
16+
#
17+
# You should have received copies of the GNU General Public License and the
18+
# GNU Lesser General Public License along with the MicroFloatingPoint Library.
19+
# If not, see https://www.gnu.org/licenses/.
20+
21+
module MFPPlot
22+
23+
using PyPlot
24+
using MicroFloatingPoint
25+
26+
export real_line
27+
28+
"""
29+
real_line(sze, szf, start::Float64, stop::Float64)
30+
31+
Draw the real line between `start` and `stop` and display all floating-point
32+
numbers with `sze` bits exponent and `szf` bits fractional part.
33+
"""
34+
function real_line(start = -floatmax(Floatmu{szE,szf}),
35+
stop = floatmax(Floatmu{szE,szf})) where {szE,szf}
36+
plt.axis("off")
37+
fig=plot([start,stop],[0,0],"k-",color="black")
38+
for v in FloatmuIterator(start,stop)
39+
if issubnormal(v)
40+
plot([v,v],[-.4,.4],"k-",color="purple")
41+
if v != 0
42+
text(v,-.5,string(v),fontsize=8,ha="center",va="top",
43+
rotation=90,color="purple")
44+
else
45+
text(0,-.5,L"\pm 0",fontsize=8,ha="center",va="top",
46+
rotation=90,color="purple")
47+
end
48+
else
49+
plot([v,v],[-.5,.5],"k-",color="blue")
50+
text(v,-.6,string(v),fontsize=8,ha="center",va="top",
51+
rotation=90,color="blue")
52+
end
53+
end
54+
return fig
55+
end
56+
57+
end # Module

‎src/MFPRandom.jl

+45
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,45 @@
1+
# MFPRandom --
2+
#
3+
# Copyright 2019--2021 University of Nantes, France.
4+
#
5+
# This file is part of the MicroFloatingPoint library.
6+
#
7+
# The MicroFloatingPoint library is free software; you can redistribute it and/or
8+
# modify it under the terms of the GNU Lesser General Public License as published
9+
# by the Free Software Foundation; either version 3 of the License, or (at your
10+
# option) any later version.
11+
#
12+
# The MicroFloatingPoint library is distributed in the hope that it will be useful,
13+
# but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
14+
# or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
15+
# for more details.
16+
#
17+
# You should have received copies of the GNU General Public License and the
18+
# GNU Lesser General Public License along with the MicroFloatingPoint Library.
19+
# If not, see https://www.gnu.org/licenses/.
20+
21+
module MFPRandom
22+
23+
import Random.rand
24+
using MicroFloatingPoint
25+
26+
"""
27+
irandint(rng,n)
28+
29+
Draw a n bits integer at **random**.
30+
31+
First, compute a 64 bits integer, then discard the lowest `(64-n)` bits, which
32+
are usually the less *random* ones.
33+
"""
34+
function irandint(n)
35+
return rand(UInt64) >> (64-n)
36+
end
37+
38+
39+
function rand(::Type{Floatmu{szE,szf}}) where {szE,szf}
40+
f = irandint(szf)
41+
v = (UInt64(MicroFloatingPoint.bias(Floatmu{szE,szf})) << szf) | f
42+
return Floatmu{szE,szf}(v,nothing) - Floatmu{szE,szf}(1.0)
43+
end
44+
45+
end # Module

0 commit comments

Comments
 (0)