Google Git
Sign in
gem5 / public / gem5-resources / 8ccd059d5dcaaeffe15cd93c17f1e76e92907662 / . / src / parsec / disk-image / parsec / parsec-benchmark / pkgs / apps / vips / src / doc / src / vipsmanual / vipsmanualse3.html
blob: a4abe8bc038e7ca971cdceb0155f0cc82d7ba656 [file] [log] [blame]
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<html >
<head><title>The VImage class</title>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<meta name="generator" content="TeX4ht (http://www.cse.ohio-state.edu/~gurari/TeX4ht/)">
<meta name="originator" content="TeX4ht (http://www.cse.ohio-state.edu/~gurari/TeX4ht/)">
<!-- 3,html -->
<meta name="src" content="vipsmanual.tex">
<meta name="date" content="2010-06-09 21:39:00">
<link rel="stylesheet" type="text/css" href="vipsmanual.css">
</head><body
>
<!--l. 1--><div class="crosslinks"><p class="noindent">[<a
href="vipsmanualse4.html" >next</a>] [<a
href="vipsmanualse2.html" >prev</a>] [<a
href="vipsmanualse2.html#tailvipsmanualse2.html" >prev-tail</a>] [<a
href="#tailvipsmanualse3.html">tail</a>] [<a
href="vipsmanualch1.html#vipsmanualse3.html" >up</a>] </p></div>
<h3 class="sectionHead"><span class="titlemark">1.3 </span> <a
id="x8-110001.3"></a>The <span
class="pcrr7t-">VImage </span>class</h3>
<!--l. 3--><p class="noindent" >The <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> class is a layer over the VIPS <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">IMAGE</span></span></span>
type. It automates almost all of the image creation and
destruction issues that complicate the C API, it automates
error handling, and it provides a convenient system for
composing operations.
<!--l. 8--><p class="noindent" >
<h4 class="subsectionHead"><span class="titlemark">1.3.1 </span> <a
id="x8-120001.3.1"></a>Constructors</h4>
<!--l. 10--><p class="noindent" >There are two principal constructors for <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span>:
<div class="verbatim" id="verbatim-5">
VImage::VImage(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;name,
&#x00A0;<br />&#x00A0;&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;mode&#x00A0;=&#x00A0;"r"&#x00A0;);
&#x00A0;<br />VImage::VImage();
</div>
<!--l. 16--><p class="nopar" >
<!--l. 18--><p class="indent" > The first form creates a new <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span>, linking it to the
named file. <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">mode</span></span></span> sets the mode for the file: it can take the
following values:
<dl class="description"><dt class="description">
<span
class="pcrb7t-">"r"</span> </dt><dd
class="description">The named image file is opened read-only. This is
the default mode.
</dd><dt class="description">
<span
class="pcrb7t-">"w"</span> </dt><dd
class="description">A <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> is created which, when written to,
will write pixels to disc in the specified file. Any
existing file of this name is deleted.
</dd><dt class="description">
<span
class="pcrb7t-">"t"</span> </dt><dd
class="description">As the <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">"w"</span></span></span> mode, but pixels written to the
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> will be saved in a temporary memory
buffer.
</dd><dt class="description">
<span
class="pcrb7t-">"p"</span> </dt><dd
class="description">This creates a special &#8216;partial&#8217; image. Partial
images represent intermediate results, and are used
to join VIPS operations together, see&#x00A0;<span
class="cmsy-10">§</span><a
href="#x8-160001.3.5">1.3.5<!--tex4ht:ref: sec:compute --></a>.
</dd><dt class="description">
<span
class="pcrb7t-">"rw"</span> </dt><dd
class="description">As the <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">"r"</span></span></span> mode, but the image is mapped
into your address space read-write. This mode is
useful for paintbox-style applications which need
to directly modify an image. See <span
class="cmsy-10">§</span><a
href="vipsmanualse18.html#x26-1140004.2.8">4.2.8<!--tex4ht:ref: sec:inplace --></a>.
</dd></dl>
<!--l. 46--><p class="indent" > The second form of constructor is shorthand for:
<div class="verbatim" id="verbatim-6">
VImage(&#x00A0;"VImage:1",&#x00A0;"p"&#x00A0;)
</div>
<!--l. 50--><p class="nopar" >
<!--l. 52--><p class="noindent" >It is used for representing intermediate results of
computations.
<!--l. 55--><p class="indent" > Two further constructors are handy for wrapping
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> around existing images.
<div class="verbatim" id="verbatim-7">
VImage(&#x00A0;void&#x00A0;&#x22C6;buffer,
&#x00A0;<br />&#x00A0;&#x00A0;int&#x00A0;width,&#x00A0;int&#x00A0;height,&#x00A0;int&#x00A0;bands,
&#x00A0;<br />&#x00A0;&#x00A0;TBandFmt&#x00A0;format&#x00A0;);
&#x00A0;<br />VImage(&#x00A0;void&#x00A0;&#x22C6;image&#x00A0;);
</div>
<!--l. 63--><p class="nopar" >
<!--l. 65--><p class="noindent" >The first constructor makes a <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> from an area
of memory (perhaps from another image processing
system), and the second makes a <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> from an
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">IMAGE</span></span></span>.
<!--l. 70--><p class="indent" > In both these two cases, the VIPS C++ API does not
assume responsibility for the resources: it&#8217;s up to you to
make sure the buffer is freed.
<!--l. 73--><p class="indent" > The Python interface adds the usual <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">frombuffer</span></span></span> and
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">fromstring</span></span></span> methods.
<div class="verbatim" id="verbatim-8">
VImage.fromstring&#x00A0;(string,
&#x00A0;<br />&#x00A0;&#x00A0;width,&#x00A0;height,&#x00A0;bands,&#x00A0;format)&#x00A0;-&#x003E;
&#x00A0;<br />&#x00A0;&#x00A0;VImage
</div>
<!--l. 80--><p class="nopar" >
<div class="verbatim" id="verbatim-9">
VImage.frombuffer&#x00A0;(buffer,
&#x00A0;<br />&#x00A0;&#x00A0;width,&#x00A0;height,&#x00A0;bands,&#x00A0;format)&#x00A0;-&#x003E;
&#x00A0;<br />&#x00A0;&#x00A0;VImage
</div>
<!--l. 86--><p class="nopar" >
<!--l. 88--><p class="noindent" >Use <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">fromstring</span></span></span> to avoid worries about object lifetime,
but you&#8217;ll see a lot of copies and high memory use. Use
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">frombuffer</span></span></span> for speed, but you have to manage object
lifetime yourself.
<!--l. 93--><p class="indent" > They are useful for moving images into VIPS from other
image processing libraries. There&#8217;s also a utility function,
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">vips_from_PIL_mode</span></span></span>, which turns a PIL mode into a
VIPS band, format, type triple.
<div class="verbatim" id="verbatim-10">
VImage.vips_from_PIL_mode&#x00A0;(mode)&#x00A0;-&#x003E;
&#x00A0;<br />&#x00A0;&#x00A0;(bands,&#x00A0;format,&#x00A0;type)
</div>
<!--l. 100--><p class="nopar" >
<!--l. 102--><p class="indent" > See also <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">tobuffer</span></span></span> and <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">tostring</span></span></span> below.
<!--l. 104--><p class="noindent" >
<h4 class="subsectionHead"><span class="titlemark">1.3.2 </span> <a
id="x8-130001.3.2"></a>File conversion</h4>
<!--l. 106--><p class="noindent" >VIPS can read and write a number of different file formats.
Information about file format conversion is taken from the
filename. For example:
<div class="verbatim" id="verbatim-11">
VImage&#x00A0;jim(&#x00A0;"fred.jpg"&#x00A0;);
</div>
<!--l. 111--><p class="nopar" >
<!--l. 113--><p class="noindent" >This will decompress the file <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">fred.jpg</span></span></span> to a memory
buffer, wrap a VIPS image around the buffer and build a
reference to it called <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">jim</span></span></span>.
<!--l. 117--><p class="indent" > Options are passed to the file format converters embedded
in the filename. For example:
<div class="verbatim" id="verbatim-12">
VImage&#x00A0;out(&#x00A0;"jen.tif:deflate",&#x00A0;"w"&#x00A0;);
</div>
<!--l. 122--><p class="nopar" >
<!--l. 124--><p class="noindent" >Writing to the descriptor <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">out</span></span></span> will cause a TIFF image to be
written to disc with deflate compression.
<!--l. 128--><p class="indent" > See the manual page for <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">im_open(3)</span></span></span> for details of all
the file formats and conversions available. See the man page
for <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VipsFormat(3)</span></span></span> for a lower-level API which lets you
control more of the detail of reading and writing data and is
more suitable for large files.
<!--l. 133--><p class="noindent" >
<h4 class="subsectionHead"><span class="titlemark">1.3.3 </span> <a
id="x8-140001.3.3"></a>Projection functions</h4>
<!--l. 135--><p class="noindent" >A set of member functions of <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> provide access to the
fields in the header:
<div class="verbatim" id="verbatim-13">
int&#x00A0;Xsize();
&#x00A0;<br />int&#x00A0;Ysize();
&#x00A0;<br />int&#x00A0;Bands();
&#x00A0;<br />TBandFmt&#x00A0;BandFmt();
&#x00A0;<br />TCoding&#x00A0;Coding();
&#x00A0;<br />TType&#x00A0;Type();
&#x00A0;<br />float&#x00A0;Xres();
&#x00A0;<br />float&#x00A0;Yres();
&#x00A0;<br />int&#x00A0;Length();
&#x00A0;<br />TCompression&#x00A0;Compression();
&#x00A0;<br />short&#x00A0;Level();
&#x00A0;<br />int&#x00A0;Xoffset();
&#x00A0;<br />int&#x00A0;Yoffset();
</div>
<!--l. 152--><p class="nopar" >
<!--l. 154--><p class="noindent" >Where <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">TBandFmt</span></span></span>, <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">TCoding</span></span></span>, <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">TType</span></span></span> and <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">TCompression</span></span></span>
are <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">enum</span></span></span>s for the types in the VIPS file header. See
section&#x00A0;<span
class="cmsy-10">§</span><a
href="vipsmanualse2.html#x7-80001.2.1">1.2.1<!--tex4ht:ref: sec:header --></a> for an explanation of all of these fields.
<!--l. 159--><p class="indent" > Two functions give access to the filename and history
fields maintained by the VIPS IO system.
<div class="verbatim" id="verbatim-14">
char&#x00A0;&#x22C6;filename();
&#x00A0;<br />char&#x00A0;&#x22C6;Hist();
</div>
<!--l. 165--><p class="nopar" >
<!--l. 167--><p class="indent" > You can get and set extra metadata fields with
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">meta_get()</span></span></span> and <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">meta_set()</span></span></span>. They read and write
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">GValue</span></span></span> objects, see <span
class="cmsy-10">§</span><a
href="vipsmanualse8.html#x14-430002.2.6">2.2.6<!--tex4ht:ref: sec:meta --></a>.
<div class="verbatim" id="verbatim-15">
void&#x00A0;meta_set(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field,&#x00A0;GValue&#x00A0;&#x22C6;value&#x00A0;);
&#x00A0;<br />void&#x00A0;meta_get(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field,&#x00A0;GValue&#x00A0;&#x22C6;value_copy&#x00A0;);
&#x00A0;<br />GType&#x00A0;meta_get_type(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field&#x00A0;);
</div>
<!--l. 175--><p class="nopar" >
<!--l. 177--><p class="indent" > A set of convenience functions build on these two to
provide accessors for common types.
<div class="verbatim" id="verbatim-16">
int&#x00A0;meta_get_int(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field&#x00A0;)
&#x00A0;<br />double&#x00A0;meta_get_double(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field&#x00A0;)
&#x00A0;<br />const&#x00A0;char&#x00A0;&#x22C6;meta_get_string(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field&#x00A0;)
&#x00A0;<br />void&#x00A0;&#x22C6;meta_get_area(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field&#x00A0;)
&#x00A0;<br />void&#x00A0;&#x22C6;meta_get_blob(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field,&#x00A0;size_t&#x00A0;&#x22C6;length&#x00A0;)
&#x00A0;<br />
&#x00A0;<br />void&#x00A0;meta_set(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field,&#x00A0;int&#x00A0;value&#x00A0;)
&#x00A0;<br />void&#x00A0;meta_set(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field,&#x00A0;double&#x00A0;value&#x00A0;)
&#x00A0;<br />void&#x00A0;meta_set(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field,&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;value&#x00A0;)
&#x00A0;<br />void&#x00A0;meta_set(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field,
&#x00A0;<br />VCallback&#x00A0;free_fn,&#x00A0;void&#x00A0;&#x22C6;value&#x00A0;)
&#x00A0;<br />void&#x00A0;meta_set(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;field,
&#x00A0;<br />VCallback&#x00A0;free_fn,&#x00A0;void&#x00A0;&#x22C6;value,&#x00A0;size_t&#x00A0;length&#x00A0;)
</div>
<!--l. 194--><p class="nopar" >
<!--l. 196--><p class="indent" > The <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">image()</span></span></span> member function provides access to the
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">IMAGE</span></span></span> descriptor underlying the C++ API. See the <span
class="cmsy-10">§</span><a
href="vipsmanualse7.html#x13-360002.1">2.1<!--tex4ht:ref: sec:appl --></a> for
details.
<div class="verbatim" id="verbatim-17">
void&#x00A0;&#x22C6;image();
</div>
<!--l. 201--><p class="nopar" >
<!--l. 203--><p class="indent" > The <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">data()</span></span></span> member function returns a pointer to an
array of pixel data for the image.
<div class="verbatim" id="verbatim-18">
void&#x00A0;&#x22C6;data()&#x00A0;const;
</div>
<!--l. 208--><p class="nopar" >
<!--l. 210--><p class="noindent" >This can be very slow and use huge amounts of RAM.
<!--l. 213--><p class="indent" > The Python interface adds <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">tobuffer</span></span></span> and <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">tostring</span></span></span>.
These operations call <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">data()</span></span></span> to generate the image pixels
and then either copy it and return the copy as a string, or
wrap the pixels up as a Python buffer object.
<!--l. 218--><p class="indent" > Use <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">tostring</span></span></span> to avoid worries about object lifetime,
but you&#8217;ll see a lot of copies and high memory use. Use
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">tobuffer</span></span></span> for speed, but you have to manage object
lifetime yourself.
<!--l. 222--><p class="indent" > They are useful for moving images from VIPS into other
image processing libraries. There&#8217;s also a utility function,
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">PIL_mode_from_vips</span></span></span>, which makes a PIL mode from
a VIPS image.
<div class="verbatim" id="verbatim-19">
VImage.PIL_mode_from_vips&#x00A0;(vips-image)&#x00A0;-&#x003E;
&#x00A0;<br />&#x00A0;&#x00A0;mode
</div>
<!--l. 229--><p class="nopar" >
<!--l. 231--><p class="indent" > See also <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">frombuffer</span></span></span> and <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">fromstring</span></span></span> above.
<!--l. 233--><p class="noindent" >
<h4 class="subsectionHead"><span class="titlemark">1.3.4 </span> <a
id="x8-150001.3.4"></a>Assignment</h4>
<!--l. 235--><p class="noindent" ><span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> defines copy and assignment, with reference-counted,
pointer-style semantics. For example, if you write:
<div class="verbatim" id="verbatim-20">
VImage&#x00A0;fred(&#x00A0;"fred.v"&#x00A0;);
&#x00A0;<br />VImage&#x00A0;jim(&#x00A0;"jim.v"&#x00A0;);
&#x00A0;<br />
&#x00A0;<br />fred&#x00A0;=&#x00A0;jim;
</div>
<!--l. 243--><p class="nopar" >
<!--l. 245--><p class="indent" > This will automatically close the file <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">fred.v</span></span></span>, and make
the variable <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">fred</span></span></span> point to the image <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">jim.v</span></span></span> instead. Both
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">jim</span></span></span> and <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">fred</span></span></span> now point to the same underlying image
object.
<!--l. 249--><p class="indent" > Internally, a <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> object is just a pointer to a
reference-counting block, which in turn holds a pointer to
the underlying VIPS <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">IMAGE</span></span></span> type. You can therefore
efficiently pass <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> objects to functions by value, and
return <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> objects as function results.
<!--l. 254--><p class="noindent" >
<h4 class="subsectionHead"><span class="titlemark">1.3.5 </span> <a
id="x8-160001.3.5"></a>Computing with <span
class="pcrr7t-">VImage</span>s</h4>
<!--l. 257--><p class="noindent" >All VIPS image processing operations are member
functions of the <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> class. For example:
<div class="verbatim" id="verbatim-21">
VImage&#x00A0;fred(&#x00A0;"fred.v"&#x00A0;);
&#x00A0;<br />VImage&#x00A0;jim(&#x00A0;"jim.v"&#x00A0;);
&#x00A0;<br />
&#x00A0;<br />VImage&#x00A0;result&#x00A0;=&#x00A0;fred.cos()&#x00A0;+&#x00A0;jim;
</div>
<!--l. 265--><p class="nopar" >
<!--l. 267--><p class="indent" > Will apply <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">im_costra()</span></span></span> to <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">fred.v</span></span></span>, making an
image where each pixel is the cosine of the corresponding
pixel in <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">fred.v</span></span></span>; then add that image to <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">jim.v</span></span></span>. Finally,
the result will be held in <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">result</span></span></span>.
<!--l. 271--><p class="indent" > VIPS is a demand-driven image processing system:
when it computes expressions like this, no actual pixels
are calculated (although you can use the projection
functions on images &#8212; <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">result.BandFmt()</span></span></span> for
example). When you finally write the result to a file (or
use some operation that needs pixel values, such as
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">min()</span></span></span>, find minimum value), VIPS evaluates all of the
operations you have called to that point in parallel. If you
have more than one CPU in your machine, the load
is spread over the available processors. This means
that there is no limit to the size of the images you can
process.
<!--l. 280--><p class="indent" > <span
class="cmsy-10">§</span><a
href="vipsmanualse18.html#x26-1060004.2">4.2<!--tex4ht:ref: sec:packages --></a> lists all of the VIPS packages. These general rules
apply:
<ul class="itemize1">
<li class="itemize">VIPS operation names become C++ member
function names by dropping the <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">im_</span></span></span> prefix, and
if present, the <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">tra</span></span></span> postfix, the <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">const</span></span></span> postfix and
the <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">_vec</span></span></span> postfix. For example, the VIPS operation
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">im_extract()</span></span></span> becomes <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">extract()</span></span></span>, and
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">im_costra()</span></span></span> becomes <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">cos()</span></span></span>.
</li>
<li class="itemize">The <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> object to which you apply the
member function is the first input image, the
member function returns the first output. If there is
no image input, the member is declared <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">static</span></span></span>.
<!--l. 297--><p class="noindent" >For example, <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">im_project(3)</span></span></span> returns two
images. You can call it from Python like this:
<div class="verbatim" id="verbatim-22">
hout&#x00A0;=&#x00A0;VImage.VImage&#x00A0;()
&#x00A0;<br />vout&#x00A0;=&#x00A0;im.project&#x00A0;(hout)
</div>
<!--l. 303--><p class="nopar" >
<!--l. 305--><p class="noindent" >In other words, <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">.project()</span></span></span> writes the second result
to the <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> you pass as an argument.
</li>
<li class="itemize"><span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">INTMASK</span></span></span> and <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">DOUBLEMASK</span></span></span> types become <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VMask</span></span></span>
objects, <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">im_col_display</span></span></span> types become <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VDisplay</span></span></span>
objects.
</li>
<li class="itemize">Several C API functions can map to the same
C++ API member. For example, <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">im_andimage</span></span></span>,
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">im_andimage_vec</span></span></span> and <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">im_andimageconst</span></span></span> all
map to the member <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">andimage</span></span></span>. The API relies on
overloading to discriminate between these functions.
</li></ul>
<!--l. 321--><p class="indent" > This part of the C++ API is generated automatically
from the VIPS function database, so it should all be
up-to-date.
<!--l. 324--><p class="indent" > There are a set of arithmetic operators defined for your
convenience. You can generally write any arithmetic
expression and include <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> in there.
<div class="verbatim" id="verbatim-23">
VImage&#x00A0;fred(&#x00A0;"fred.v"&#x00A0;);
&#x00A0;<br />VImage&#x00A0;jim(&#x00A0;"jim.v"&#x00A0;);
&#x00A0;<br />
&#x00A0;<br />Vimage&#x00A0;v&#x00A0;=&#x00A0;int((fred&#x00A0;+&#x00A0;jim)&#x00A0;/&#x00A0;2);
</div>
<!--l. 332--><p class="nopar" >
<!--l. 334--><p class="noindent" >
<h4 class="subsectionHead"><span class="titlemark">1.3.6 </span> <a
id="x8-170001.3.6"></a>Writing results</h4>
<!--l. 336--><p class="noindent" >Once you have computed some result, you can write it to a
file with the member <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">write()</span></span></span>. It takes the following
forms:
<div class="verbatim" id="verbatim-24">
VImage&#x00A0;write(&#x00A0;const&#x00A0;char&#x00A0;&#x22C6;name&#x00A0;);
&#x00A0;<br />VImage&#x00A0;write(&#x00A0;VImage&#x00A0;out&#x00A0;);
&#x00A0;<br />VImage&#x00A0;write();
</div>
<!--l. 343--><p class="nopar" >
<!--l. 345--><p class="indent" > The first form simply writes the image to the named file.
The second form writes the image to the specified <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span>
object, for example:
<div class="verbatim" id="verbatim-25">
VImage&#x00A0;fred(&#x00A0;"fred.v"&#x00A0;);
&#x00A0;<br />VImage&#x00A0;jim(&#x00A0;"jim&#x00A0;buffer",&#x00A0;"t"&#x00A0;);
&#x00A0;<br />
&#x00A0;<br />Vimage&#x00A0;v&#x00A0;=&#x00A0;(fred&#x00A0;+&#x00A0;42).write(&#x00A0;jim&#x00A0;);
</div>
<!--l. 353--><p class="nopar" >
<!--l. 355--><p class="noindent" >This creates a temporary memory buffer called <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">jim</span></span></span>, and
fills it with the result of adding 42 to every pixel in
<span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">fred.v</span></span></span>.
<!--l. 359--><p class="indent" > The final form of <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">write()</span></span></span> writes the image to a
memory buffer, and returns that.
<!--l. 362--><p class="noindent" >
<h4 class="subsectionHead"><span class="titlemark">1.3.7 </span> <a
id="x8-180001.3.7"></a>Type conversions</h4>
<!--l. 364--><p class="noindent" >Two type conversions are defined: you can cast <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VImage</span></span></span> to
a <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VDMask</span></span></span> and to a <span class="obeylines-h"><span class="verb"><span
class="pcrr7t-">VIMask</span></span></span>.
<div class="verbatim" id="verbatim-26">
operator&#x00A0;VDMask();
&#x00A0;<br />operator&#x00A0;VIMask();
</div>
<!--l. 370--><p class="nopar" >
<!--l. 372--><p class="indent" > These operations are slow and need a lot of memory!
Emergencies only.
<!--l. 1--><div class="crosslinks"><p class="noindent">[<a
href="vipsmanualse4.html" >next</a>] [<a
href="vipsmanualse2.html" >prev</a>] [<a
href="vipsmanualse2.html#tailvipsmanualse2.html" >prev-tail</a>] [<a
href="vipsmanualse3.html" >front</a>] [<a
href="vipsmanualch1.html#vipsmanualse3.html" >up</a>] </p></div>
<!--l. 1--><p class="indent" > <a
id="tailvipsmanualse3.html"></a>
</body></html>