FILLBAD is the jCommon module of jLab.

 FILLBAD  Linearly interpolate over bad data points. 
 
    DATA=FILLBAD(DATA,FLAG), where DATA is a column vector and "bad" points
    within DATA are equal to FLAG, fills the bad points with a linear 
    interpolation between the closest adjacent good points. 
 
    FILLBAD(DATA) uses a value of NAN for FLAG.
 
    If DATA is a matrix, interpolation is performed along each column. 
 
    FILLBAD(DATA,FLAG,MAXN) or FILLBAD(DATA,INDEX,MAXN) interpolates over
    gaps having a maximum length MAXN, which defaults to INF. 
 
    For complex-valued data, the interpolation is performed on both the 
    real and imaginary parts separately.  For a FLAG of NAN, this means 
    the data must contain NAN+SQRT(-1)*NAN.
 
    FILLBAD(...,STR) uses the interpolation method specified in STR.
 
    DATAC=FILLBAD(DATAC,...) also works when the input DATAC is a cell
    array of numerical arrays, DATAC{1}=DATA1, DATAC{2}=DATA2, etc. The 
    output is then a cell array of the identical size.
 
    'fillbad --t' runs a test.
 
    Usage: data=fillbad(data,flag);
           data=fillbad(data,flag,maxnan);   
    _________________________________________________________________
    This is part of JLAB --- type 'help jlab' for more information 
    (C) 2000--2020 J.M. Lilly --- type 'help jlab_license' for details