Skip to content
Projects
Groups
Snippets
Help
Loading...
Help
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
K
kicad-source-mirror
Project
Project
Details
Activity
Releases
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Commits
Open sidebar
Elphel
kicad-source-mirror
Commits
a6a1af9d
Commit
a6a1af9d
authored
Sep 29, 2013
by
Maciej Sumiński
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
Restored previous version of math_utils.h. WARNING: Mac OS build is probably broken now.
parent
d895a90b
Changes
3
Hide whitespace changes
Inline
Side-by-side
Showing
3 changed files
with
56 additions
and
83 deletions
+56
-83
CMakeLists.txt
common/CMakeLists.txt
+1
-1
math_util.cpp
common/math/math_util.cpp
+0
-80
math_util.h
include/math/math_util.h
+55
-2
No files found.
common/CMakeLists.txt
View file @
a6a1af9d
...
@@ -156,7 +156,7 @@ set(COMMON_SRCS
...
@@ -156,7 +156,7 @@ set(COMMON_SRCS
view/view_item.cpp
view/view_item.cpp
view/view_group.cpp
view/view_group.cpp
math/math_util.cpp
#
math/math_util.cpp
system/fcontext.s
system/fcontext.s
tool/tool_base.cpp
tool/tool_base.cpp
...
...
common/math/math_util.cpp
deleted
100644 → 0
View file @
d895a90b
/*
* This program source code file is part of KICAD, a free EDA CAD application.
*
* Copyright (c) 2005 Michael Niedermayer <michaelni@gmx.at>
* Copyright (C) CERN
* @author Tomasz Wlostowski <tomasz.wlostowski@cern.ch>
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, you may find one here:
* http://www.gnu.org/licenses/old-licenses/gpl-2.0.html
* or you may search the http://www.gnu.org website for the version 2 license,
* or you may write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
*/
#include <math/math_util.h>
// explicit specializations for integer types, taking care of overflow.
template
<>
int
rescale
(
int
numerator
,
int
value
,
int
denominator
)
{
return
(
int
)
(
(
int64_t
)
numerator
*
(
int64_t
)
value
/
(
int64_t
)
denominator
);
}
template
<>
int64_t
rescale
(
int64_t
numerator
,
int64_t
value
,
int64_t
denominator
)
{
uint64_t
r
=
0
;
int64_t
sign
=
(
(
numerator
<
0
)
?
-
1
:
1
)
*
(
denominator
<
0
?
-
1
:
1
)
*
(
value
<
0
?
-
1
:
1
);
uint64_t
a
=
abs
(
numerator
);
uint64_t
b
=
abs
(
value
);
uint64_t
c
=
abs
(
denominator
);
r
=
c
/
2
;
if
(
b
<=
INT_MAX
&&
c
<=
INT_MAX
)
{
if
(
a
<=
INT_MAX
)
return
sign
*
(
(
a
*
b
+
r
)
/
c
);
else
return
sign
*
(
a
/
c
*
b
+
(
a
%
c
*
b
+
r
)
/
c
);
}
else
{
uint64_t
a0
=
a
&
0xFFFFFFFF
;
uint64_t
a1
=
a
>>
32
;
uint64_t
b0
=
b
&
0xFFFFFFFF
;
uint64_t
b1
=
b
>>
32
;
uint64_t
t1
=
a0
*
b1
+
a1
*
b0
;
uint64_t
t1a
=
t1
<<
32
;
int
i
;
a0
=
a0
*
b0
+
t1a
;
a1
=
a1
*
b1
+
(
t1
>>
32
)
+
(
a0
<
t1a
);
a0
+=
r
;
a1
+=
a0
<
r
;
for
(
i
=
63
;
i
>=
0
;
i
--
)
{
a1
+=
a1
+
(
(
a0
>>
i
)
&
1
);
t1
+=
t1
;
if
(
c
<=
a1
)
{
a1
-=
c
;
t1
++
;
}
}
return
t1
*
sign
;
}
};
include/math/math_util.h
View file @
a6a1af9d
...
@@ -29,7 +29,6 @@
...
@@ -29,7 +29,6 @@
#include <cmath>
#include <cmath>
#include <cstdlib>
#include <cstdlib>
#include <stdint.h>
#include <stdint.h>
#include <climits>
/**
/**
* Function rescale()
* Function rescale()
...
@@ -37,9 +36,63 @@
...
@@ -37,9 +36,63 @@
* Scales a number (value) by rational (numerator/denominator). Numerator must be <= denominator.
* Scales a number (value) by rational (numerator/denominator). Numerator must be <= denominator.
*/
*/
template
<
typename
T
>
T
rescale
(
T
numerator
,
T
value
,
T
denominator
)
template
<
typename
T
>
static
T
rescale
(
T
numerator
,
T
value
,
T
denominator
)
{
{
return
numerator
*
value
/
denominator
;
return
numerator
*
value
/
denominator
;
}
}
// explicit specializations for integer types, taking care of overflow.
template
<>
int
rescale
(
int
numerator
,
int
value
,
int
denominator
)
{
return
(
int
)
(
(
int64_t
)
numerator
*
(
int64_t
)
value
/
(
int64_t
)
denominator
);
}
template
<>
int64_t
rescale
(
int64_t
numerator
,
int64_t
value
,
int64_t
denominator
)
{
int64_t
r
=
0
;
int64_t
sign
=
(
(
numerator
<
0
)
?
-
1
:
1
)
*
(
denominator
<
0
?
-
1
:
1
)
*
(
value
<
0
?
-
1
:
1
);
int64_t
a
=
std
::
abs
(
numerator
);
int64_t
b
=
std
::
abs
(
value
);
int64_t
c
=
std
::
abs
(
denominator
);
r
=
c
/
2
;
if
(
b
<=
INT_MAX
&&
c
<=
INT_MAX
)
{
if
(
a
<=
INT_MAX
)
return
sign
*
(
(
a
*
b
+
r
)
/
c
);
else
return
sign
*
(
a
/
c
*
b
+
(
a
%
c
*
b
+
r
)
/
c
);
}
else
{
uint64_t
a0
=
a
&
0xFFFFFFFF
;
uint64_t
a1
=
a
>>
32
;
uint64_t
b0
=
b
&
0xFFFFFFFF
;
uint64_t
b1
=
b
>>
32
;
uint64_t
t1
=
a0
*
b1
+
a1
*
b0
;
uint64_t
t1a
=
t1
<<
32
;
int
i
;
a0
=
a0
*
b0
+
t1a
;
a1
=
a1
*
b1
+
(
t1
>>
32
)
+
(
a0
<
t1a
);
a0
+=
r
;
a1
+=
((
uint64_t
)
a0
)
<
r
;
for
(
i
=
63
;
i
>=
0
;
i
--
)
{
a1
+=
a1
+
(
(
a0
>>
i
)
&
1
);
t1
+=
t1
;
if
(
(
uint64_t
)
c
<=
a1
)
{
a1
-=
c
;
t1
++
;
}
}
return
t1
*
sign
;
}
};
#endif // __MATH_UTIL_H
#endif // __MATH_UTIL_H
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment