diff options
| author | scuri <scuri> | 2008-10-17 06:10:15 +0000 |
|---|---|---|
| committer | scuri <scuri> | 2008-10-17 06:10:15 +0000 |
| commit | 5a422aba704c375a307a902bafe658342e209906 (patch) | |
| tree | 5005011e086bb863d8fb587ad3319bbec59b2447 /html/en/storage_guide.html | |
First commit - moving from LuaForge to SourceForge
Diffstat (limited to 'html/en/storage_guide.html')
| -rw-r--r-- | html/en/storage_guide.html | 311 |
1 files changed, 311 insertions, 0 deletions
diff --git a/html/en/storage_guide.html b/html/en/storage_guide.html new file mode 100644 index 0000000..e225ae5 --- /dev/null +++ b/html/en/storage_guide.html @@ -0,0 +1,311 @@ +<!doctype HTML PUBLIC "-//IETF//DTD HTML//EN"> +<html> + +<head> +<meta http-equiv="Content-Language" content="en-us"> +<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> +<title>Storage Guide</title> +<link rel="stylesheet" type="text/css" href="../style.css"> +</head> + +<body> + +<h1>Storage Guide</h1> +<h3 align="left"><a name="read">Reading</a></h3> + + <p>When reading the file extension is not relevant to determine the file + format, but it is used to speed up the process of finding the correct format. + With few exceptions the format drivers that access multiple images can read + them in any sequence you want. </p> + <p>During the read process the original data can be converted to some options + of user data. Not all conversions are available. You can convert any data to a + bitmap version of it, and you can select any of the color mode flags<b> + IM_ALPHA</b>, <b>IM_PACKED</b> and <b>IM_TOPDOWN</b>, + regardless of the file original configuration.</p> + <p>Remember that even if all the images in the file have the same parameters + you still have to call <b>imFileReadImageInfo</b> before calling <b>imFileReadImageData</b>. </p> + <p>In the following example all the images in the file are loaded.</p> + + <pre>char format[10], compression[10]; +int error, image_count; +int width, height, color_mode, data_type; +void* data; + +imFile* ifile = imFileOpen("test.tif", &error); +if (error != IM_ERR_NONE) + // handle the error + +imFileGetInfo(ifile, format, compression, &image_count); + +for (i = 0; i < image_count, i++) +{ + error = imFileReadImageInfo(ifile, i, &width, &height, &color_mode, &data_type); + if (error != IM_ERR_NONE) + // handle the error + + // prepare data + + error = imFileReadImageData(ifile, data, 0, -1); // no bitmap convertion, use original color mode flags + if (error != IM_ERR_NONE) + // handle the error + + // store data somewhere +} + +imFileClose(ifile); </pre> + + <p>A more simple code loads only the first image in the file:</p> + + <pre>imFile* ifile = imFileOpen(file_name, &error); + +imFileReadImageInfo(ifile, 0, &width, &height, &color_mode, &data_type); + +imFileReadImageData(ifile, data, 0, -1); + +imFileClose(ifile); </pre> + + <p>If you are using the <b>imImage</b> structure it is easier:</p> + + <pre>imFile* ifile = imFileOpen(file_name, &error); + +imImage* image = imFileLoadImage(ifile, 0, &error);</pre> + <pre>// or use imFileLoadBitmap to force a bitmap conversion + +imFileClose(ifile);</pre> + + <p>Or the simplest version:</p> + + <pre>imImage* image = imFileImageLoad(file_name, 0, &error);</pre> + + +<h3 align="left"><a name="write">Writing</a></h3> + + <p>When writing there is no color space or data type conversion. Only color + mode flags can be different: <b>IM_ALPHA</b>, <b>IM_PACKED</b> and + <b>IM_TOPDOWN</b>. You just have to describe your data and the <b>imFileWriteImageData</b> will handle the color mode flag differences.</p> + <p>Of course you still have to check the error codes because, not all color + spaces and data types are supported by each format.</p> + <p>When saving a sequence of images you must provide each image in the order + that they will be in the file. For a video or animation start from frame 0 and + go on, you can not jump or change the frame order. Also when saving videos you + should not forget to save the numbers of frames per second in the attribute + "FPS", the default value is 15.</p> + <p>For all the formats it is not necessary to set the compression, each driver + will choose a default compression. But you may set it using the function <b>imFileSetInfo</b>.</p> + <p>To save several images to the same file:</p> + + <pre>int error, width, height; +void *data; + +imFile* ifile = imFileNew("test.tif", "TIFF", &error); +if (error != IM_ERR_NONE) + // handle the error + +for (i = 0; i < image_count, i++) +{ + error = imFileWriteImageInfo(ifile, width, height, IM_RGB, IM_BYTE); + if (error != IM_ERR_NONE) + // handle the error + + error = imFileWriteImageData(ifile, data); + if (error != IM_ERR_NONE) + // handle the error +} + +imFileClose(ifile); </pre> + + <p>But remember that not all file formats supports several images. To save + just one image is more simple:</p> + + <pre>imFile* ifile = imFileNew(file_name, format, &error); + +error = imFileWriteImageInfo(ifile, width, height, color_mode, data_type); + +error = imFileWriteImageData(ifile, data); + +imFileClose(ifile); </pre> + + <p>If you are using the <b>imImage</b> structure it is easier:</p> + + <pre>imFile* ifile = imFileNew(file_name, format, &error); + +error = imFileSaveImage(ifile, image); + +imFileClose(ifile);</pre> + + <p>Or the simplest version:</p> + + <pre>error = imFileImageSave(file_name, format, image);</pre> + + +<h3>Error Messages</h3> + + <p>Here is a sample error message display using IUP and IM error codes:</p> + + <pre>static void imIupErrorMessage(int error, int interactive) +{ + char* lang = IupGetLanguage(); + char *msg, *title; + if (strcmp(lang, "ENGLISH")==0) + { + title = "Error"; + switch (error) + { + case IM_ERR_OPEN: + msg = "Error Opening File."; + break; + case IM_ERR_MEM: + msg = "Insuficient memory."; + break; + case IM_ERR_ACCESS: + msg = "Error Accessing File."; + break; + case IM_ERR_DATA: + msg = "Image type not Suported."; + break; + case IM_ERR_FORMAT: + msg = "Invalid Format."; + break; + case IM_ERR_COMPRESS: + msg = "Invalid or unsupported compression."; + break; + default: + msg = "Unknown Error."; + } + } + else + { + title = "Erro"; + switch (error) + { + case IM_ERR_OPEN: + msg = "Erro Abrindo Arquivo."; + break; + case IM_ERR_MEM: + msg = "Memória Insuficiente."; + break; + case IM_ERR_ACCESS: + msg = "Erro Acessando Arquivo."; + break; + case IM_ERR_DATA: + msg = "Tipo de Imagem não Suportado."; + break; + case IM_ERR_FORMAT: + msg = "Formato Inválido."; + break; + case IM_ERR_COMPRESS: + msg = "Compressão Inválida ou não Suportada."; + break; + default: + msg = "Erro Desconhecido."; + } + } + + if (interactive) + IupMessage(title, msg); + else + printf("%s: %s", title, msg); +} +</pre> + + +<h3><a name="formats">About File Formats</a></h3> + + <p>TIFF is still the most complete format available. It could be better if + Adobe releases the revision 7, but it is on stand by. TIFF supports all the IM + image representation concepts. In fact we were partially inspired by the TIFF + specification. My suggestion is whenever possible use TIFF.</p> + <p>But TIFF may not be the ideal format for many situations. The W3C standards + include only JPEG, GIF and PNG for Web browsers. JPEG forces the image to be + RGB or Gray with a lossy compressed. GIF forces the image to be MAP with LZW + compression. PNG forces the image to be RGB, MAP, Gray or Binary, with Deflate + compression. So these characteristics are necessary to force small values for + faster downloads.</p> + <p>JPEG is to be used for photographic content, PNG should be used for the + remaining cases, but GIF is still the best to do simple animated images.</p> + <p>Except for some specific cases where a format is needed for compatibility, + the other formats are less important. TGA, PCX, RAS, SGI and BMP have almost + the same utility.</p> + <p>JP2 must be used for JPEG-2000 compression, would be nice if a new TIFF + specification includes this standard.</p> + <p>Since PNM has a textual header it is very simple to teach for students so + they can actually "see" the header. It is also a format easy to share images, + but it does not do much more than that.</p> + <p>The TIFF and the GIF format also have support for multiple images. This + does not necessarily defines an animation, pyramid nor a volume, but some + times they are used in these ways. </p> + <p>GIF became very popular to build animations for the Web, and since the LZW + patent expired Unisys realized that charging the usage isn't going to work and + so they did not renew it. LZW is fully supported at IM.</p> + <p>IM also supports video formats like AVI and WMV as external libraries. In + these cases the frames are also loaded as a sequence of individual images. + Sound is not supported.</p> + <p>TIFF, JPEG and PNG have an extensive list of attributes, most of them are + listed in the documentation, but some custom attributes may come up when + reading an image from file.</p> + +<h3><a name="filesdk">New File Formats</a></h3> + + <p>Again the easiest way is to look at the source code of an already + implemented format. The RAS, BMP, TGA and SGI formats are very simple to + follow.</p> + <p>Basically you have to implement a class that inherits from <b>imFormat</b> + and implement its virtual methods. You can use the <b>imBinFile</b> functions + for I/O or use an external SDK.</p> + <p>For more information see + <a href="doxygen/group__filesdk.html">File + Format SDK</a>.</p> + +<h3><a name="binfilemem">Memory I/O and Others</a></h3> + + <p>For the majority of the formats, with the exception of the ones that use + external SDKs, the I/O is done by the <b>imBinFile</b> module.</p> + <p>This module can be configured to access other types of media by + implementing a driver. There are some predefined drivers see + <a href="doxygen/group__binfile.html">Reference + / Utilities / Binary File Access</a>.</p> + <p>One very useful is the <b>Memory Buffer</b> where you can read and write a + file in memory. The activation is very simple, it needs to happen just before + the <b>imFileOpen/imFileNew</b> functions. But the file name must be a + pointer to an <b>imBinMemoryFileName </b>structure instead of a string. + Se the example bellow:</p> + + <pre>int old_mode = imBinFileSetCurrentModule(IM_MEMFILE); + +imBinMemoryFileName MemFileName; // This structure must exists + while the file remains open.<br> + MemFileName.buffer = NULL; // Let the library initializes the buffer, <br> + + // but it must be freed the the application, free(MemFileName.buffer) + MemFileName.size = 1024; // The initial size<br> + MemFileName.reallocate = 1.5; // The reallocation will increase 50% the + buffer.<br> + + // This is used only when writing with a variable buffer.<br> + + // Use 0 to fix the buffer size. + +int error;<br> + imFile* ifile = imFileNew((const char*)&MemFileName, "GIF", &error); + +imBinFileSetCurrentModule(old_mode); // The mode needs to be active + only for the imFileOpen/imFileNew call. + +if (error != IM_ERR_NONE) ....</pre> + + <p>Another driver interesting is the <b>Subfile</b> where you can read and + write from a file that is already open. This is very important for formats + that can have an embedded format inside. In this module the file_name + <span class="comment">is a pointer to an <b>imBinFile</b> + structure from any other module that uses the <b>imBinFile</b> functions. The + <b>imBinFileSize</b> will return the full file size, but the <b>imBinFileSeekTo</b> and + <b>imBinFileTell</b> functions will + compensate the position when the subfile was open.</span></p> + <p><span class="comment">Using </span><b>imBinFileSetCurrentModule(IM_SUBFILE)</b> just like the example above will + allow you to open a subfile using the <b>imFileOpen/imFileNew</b> + functions.</p> + + +</body> + +</html> |