Reading C code written for R packages. Where to find definitions for RANDIN, UNIF, EPS, etc?

Question

I am interested in precisely understanding C code for R packages. But I have come across code such as RANDIN, RANDOUT, UNIF, EPS etc (are they macros?) for which I don't know where to find the implementations/definitions.

Where can I find the code (and explanations if any) for these capitalised 'expressions'?

An example is the VR_onlineSOM function inside the 'class' R package. This function has the following code written in C:

void
VR_onlineSOM(double *data, double *codes, double *nhbrdist,
         double *alpha, double *radii,
         Sint *pn, Sint *pp, Sint *pncodes, Sint *rlen)
{
    int n = *pn, p = *pp, ncodes = *pncodes;
    int i, j, k, nearest = 0 /* -Wall */, nind;
    double dm, dist, tmp;
    unsigned int cd; /* avoid spurious warning from gcc pre-4.3.0 */

    RANDIN;
    for (k = 0; k < *rlen; k++) {
    /* pick a random data point */
    i = (int)(n * UNIF);
    /* find the nearest code 'near' */
    nind = 0; dm = DOUBLE_XMAX;
    for (cd = 0; cd < ncodes; cd++) {
        dist = 0.0;
        for (j = 0; j < p; j++) {
        tmp = data[i + j*n] - codes[cd + j*ncodes];
        dist += tmp * tmp;
        }
        if (dist <= dm * (1 + EPS)) {
        if (dist < dm * (1 - EPS)) {
            nind = 0;
            nearest = cd;
        } else {
            if(++nind * UNIF < 1.0) nearest = cd;
        }
        dm = dist;
        }
        /* update all codes within radii[k] of 'nearest' */
        for (cd = 0; cd < ncodes; cd++) {
        if(nhbrdist[cd + ncodes*nearest] > radii[k]) continue;
        for(j = 0; j < p; j++)
            codes[cd + j*ncodes] += alpha[k] *
            (data[i + j*n] - codes[cd + j*ncodes]);
        }
    }
    }
    RANDOUT;
}

Answer 1

Yes they are some macros and constantes defined in the begining of the class.c.

#define EPS 1e-4            /* relative test of equality of distances */
#define RANDIN  GetRNGstate()
#define RANDOUT PutRNGstate()
#define UNIF unif_rand()