Something to note:
A.
(1)Before using new codes, users are strongly recommended to read the algorithm developed by
   Yunyue Zhu and Dennis Shasha(http://cs.nyu.edu/cs/faculty/shasha/papers/burst.d/burst.html)
(2)This readme file introduces the basic structure of the new codes covering the following files

   makefile
   grb.short.h   
   grb_finder.c  
   grb_finder_publish.c
   grb_finder_swt.c

   All the above files exist in the old program except grb_finder_swt.c, which is the core 
   file of the new algorithm and used in place of grb_finder_evaldata_table.c and grb_finder_evaldata_bins.c
   It is documented in the code file.
(3)The following description will assume the reader has a sound understanding of both the old and the new algorithm.
B.
(1) In the old algorithm, grb_finder.c launches the burst search where grb_finder_evaldata_table.c is for the smaller window size 
    than 0.2 while grb_finder_evaldata_bins.c is for the rest.
    Here in the new algorithm all the window sizes are in the charge of grb_finder_swt.c.
    The grb_finder_swt.c has been fully documented for your reference.
    grb_finder.c is updated in the part where search procedures are called.
(2) The bursts reported may be a little different from those of the original algorithm. The location may has a shift of 
    several iha and idec and the burst time may change a little bit. But all of these are accptable based on the original 
    resolution requirement. 
(3) In the new detection strategy, the bursts are reported in a different manner, so grb_finder_publish.c is revised accordingly.
(4) In grb.short.h the procesure grb_finder_swt.c is declared and grb_finder_evaldata_bins.c and grb_finder_evaldata_table.c are
    commented.
(5) Finally makefile is updated accordingly.
 
C. 
Some of the current codes are specially for the current test data files. However it is very easy for users to genralize it after 
carefully read the algorithm and the code's documentation. To embed the codes seamlessly, the following points should be noted:
(1) Due to the limitation of memory, the maximal space available is no more than a upbound, so in the current codes, those data 
    set requiring of more than this memory limitation will be split so as to be able to be put into the data structure of the 
    codes.(see the code documentation in grb_finder_swt.c)

If any problem or comments,please send email to Dennis Shasha(shasha@cs.nyu.edu) or Xiaojian Zhao(xiaojian@cs.nyu.edu).
