[mapserver-users] New method for Python mapscript

[Sean Gillies]

Greetings Python mapscript users,

Recently, I've found the use for getting mapserver map images as
Python string objects.  Rather than fill in the empty Python
getImageToVar() method, I decided to write a new interface for
a method I'm calling getImageString().  Mostly I disagreed with
the notion of getting to a variable through its name as a string.

I'm including the code that you might paste into your mapscript.i
file within the imageObj definition.  You will need to modify your
setup.py script so that -DUSE_GD_PNG and -DUSE_GD_JPEG as passed as
arguments to swig.  I've only tested this with gd-1.8.4 but will be
moving on to gd-2 next week.

So, now we can do stuff like this:

    import MapScript

    mapob = MapScript.mapObj(...)
    imgob = mapObj.draw()
    data = imgob.getImageString(...)
    png = file("map.png", "wb")
    png.write(data)
    png.close()

This is trivial.  It's more useful for creating Zope image objects
from a Zope Python script like this:

    import StringIO
    import MapScript

    mapob = MapScript.mapObj(...)
    imgob = mapObj.draw()
    data = StringIO.StringIO(imgob.getImageString(...))
    self.manage_addImage('map.png', data, "", "", 'image/png')

Hope that you find this of use and happy to hear any other ideas.

cheers,
Sean
-------------- next part --------------
// Method getImageString renders the imageObj into image data and returns
// it as a string.  Inspired by and used like the saveImage() method.  Python
// only at this time.  Questions and comments to Sean Gillies <sgillies at i3.com>
#ifdef SWIGPYTHON
    PyObject *getImageString(
        int imagetype,
        int transparent,
        int interlace,
        int quality)
    {
        unsigned char *imgbytes;
        int size;
        PyObject *imgstring;

        if (interlace) gdImageInterlace(self->bytes, 1);
        if (transparent) gdImageColorTransparent(self->bytes, 0);

        switch(imagetype) {
            case(MS_GIF):
                msSetError(MS_MISCERR, "GIF output is not available.", "saveImageString()");
                return(MS_FAILURE);
                break;
            case(MS_PNG):
#ifdef USE_GD_PNG
                imgbytes = gdImagePngPtr(self->bytes, &size);
#else
                msSetError(MS_MISCERR, "PNG output is not available.", "saveImageString()");
                return(MS_FAILURE);
#endif
                break;
            case(MS_JPEG):
#ifdef USE_GD_JPEG
                imgbytes = gdImageJpegPtr(self->bytes, &size, quality);
#else
                msSetError(MS_MISCERR, "JPEG output is not available.", "saveImageString()");
                return(MS_FAILURE);
#endif
                break;
            case(MS_WBMP):
#ifdef USE_GD_WBMP
                imgbytes = gdImageWBMPPtr(image->bytes, &size, 1);
#else
                msSetError(MS_MISCERR, "WBMP output is not available.", "saveImageString()");
                return(MS_FAILURE);
#endif
                break;
            default:
                msSetError(MS_MISCERR, "Unknown output image type.", "saveImageString()");
                return(MS_FAILURE);
        }

        // Create a Python string from the (char *) imgbytes.
        // The gdImage*Ptr functions return a size for just this purpose.
        imgstring = PyString_FromStringAndSize(imgbytes, size);

        // The gd docs recommend gdFree()
        gdFree(imgbytes);

        return imgstring;
    }
#endif