pyEXP TblData evaluation seems garbled

[michael-petersen]

When I run the following code:

coefsread = pyEXP.coefs.Coefs.factory('coef1.txt')
C = coefsread.getAllCoefs()
print(C[1,:])

coefsread = pyEXP.coefs.Coefs.factory('coef1.txt')
C = coefsread.getAllCoefs()
print(C[1,:])

I get

---- Coefs::factory: error opening as HDF5, trying EXP native and ascii table
---- Coefs::factory: read ascii and created TableData
[6.43503391e-08 5.98943807e-08 6.66737480e-08 6.90625039e-08
 7.15178637e-08 7.40412731e-08 7.66340220e-08 7.92974728e-08
 8.20330039e-08 8.48420093e-08 0.00000000e+00]
---- Coefs::factory: error opening as HDF5, trying EXP native and ascii table
---- Coefs::factory: read ascii and created TableData
[6.20909780e-08 6.43503391e-08 6.66737480e-08 6.90625039e-08
 7.15178637e-08 7.40412731e-08 7.66340220e-08 7.92974728e-08
 8.20330039e-08 8.48420093e-08 0.00000000e+00]

i.e. different results for the same file (when they should be the same). Given that some of the values are the same, I suspect an addressing/fencepost/memory issue somewhere. Some additional details:

• If the factory call is not made again, this behaviour is not seen (i.e. repeated calls to getAllCoefs() is not a problem)
• The first column (i.e. C[0,:]) is never corrupted

Is this reproducible for others?

In this example, coef1.txt is

0.0 5.989438071816292e-08 1.7521875701449447e-07 3.9311880010652795e-07 7.53369789956869e-07 1.2751284622875935e-06 
1.0 6.209097800900433e-08 1.8154483036637094e-07 4.070318022256943e-07 7.793656989784116e-07 1.3177604283519646e-06 
2.0 6.435033914548556e-08 1.8804753707487184e-07 4.2132141142163527e-07 8.060373687770234e-07 1.3614357663063152e-06 
3.0 6.6673747957078e-08 1.9473013728275417e-07 4.359940131048028e-07 8.333949731692655e-07 1.4061784948108052e-06 
4.0 6.906250388877457e-08 2.015961249825961e-07 4.5105661980662236e-07 8.614500869788951e-07 1.452001928641754e-06 
5.0 7.151786373387439e-08 2.0864903203324786e-07 4.6651630858432454e-07 8.90214365388944e-07 1.498922133315071e-06 
6.0 7.404127307599703e-08 2.1589242819040464e-07 4.823802208392562e-07 9.196995431536603e-07 1.5469552318864821e-06 
7.0 7.663402198200968e-08 2.2332992113495078e-07 4.986555621305018e-07 9.499174338041825e-07 1.5961174030538262e-06 
8.0 7.929747283965667e-08 2.3096515649909962e-07 5.153496019838852e-07 9.808799288481356e-07 1.6464248792582182e-06 
9.0 8.203300389017387e-08 2.3880181789034833e-07 5.32469673696419e-07 1.0125989969633122e-06 1.6978939447843994e-06 
10.0 8.484200927363003e-08 2.468436269132721e-07 5.500231741362771e-07 1.0450866831856246e-06 1.7505409338606263e-06 

[The9Cat]

This really helps. Not sure what's up yet, but this should be enough for me to find it.

[michael-petersen]

Sorted in #58 .