pydsstools is an experimental Cython based Python library to manipulate HEC-DSS database file. It supports regular/irregular time-series, paired data series and spatial grid records. It is compatible with 64-bit Python on Windows 10 and Ubuntu like linux distributions. For the later, zlib, math, quadmath, and gfortran libraries must be installed. dssvue python library provides graphical user interface for HEC-DSS. There is also a Rust binding hecdss for HEC-DSS.
HEC-DSS is designed to be optimal for storing and retrieving large sets, or series, of data. HEC-DSS is not a relational database, but a database that is designed to retrieve and store large amounts of data quickly that are not necessarily interlinked to other sets of data, like relational databases are. Additionally, HEC-DSS provides a flexible set of utility programs and is easy to add to a user's application program. These are the features that distinguish HEC-DSS from most commercial relational database programs and make it optimal for scientific applications.
HEC-DSS uses a block of sequential data as the basic unit of storage. Each block contains a series of values of a single variable over a time span appropriate for most applications. The basic concept underlying HEC-DSS is the organization of data into records of continuous, applications-related elements as opposed to individually addressable data items. This approach is more efficient for scientific applications than a relational database system because it avoids the processing and storage overhead required to assemble an equivalent record from a relational database.
Data is stored in blocks, or records, within a file and each record is identified by a unique name called a "pathname." Each time data is stored or retrieved from the file, its pathname is used to access its data. Information about the record (e.g., units) is stored in a "header array." This includes the name of the program writing the data, the number of times the data has been written to, and the last written date and time. HEC-DSS documents stored data completely via information contained in the pathname and stored in the header so no additional information is required to identify it. One data set is not directly related to another so there is no need to update other areas of the database when a new data set is stored. The self-documenting nature of the database allows information to be recognized and understood months or years after it was stored.
Because of the self-documenting nature of the pathname and the conventions adopted, there is no need for a data dictionary or data definition file as required with other database systems. In fact, there are no database creation tasks or any database setup. Both HEC-DSS utility programs and applications that use HEC-DSS will generate and configure HEC-DSS database files automatically. There is also no pre-allocation of space; the software automatically expands the file size as needed.
HEC-DSS references data sets, or records, by their pathnames. A pathname may consist of up to 391 characters and is, by convention, separated into six parts, which may be up to 64 characters each. Each part is delimited by a slashe "/", and is labeled "A" through "F", as follows: /A/B/C/D/E/F/.
A list of the pathnames in a DSS file is called a "catalog." In version 6, the catalog was a separate file; in version 7, the catalog is constructed directly from pathnames in the file.
Multi-user access mode is handled automatically by HEC-DSS. The user does not need to do anything to turn it on. Multi-user access allows multiple users, multiple processes, to read and write to the same HEC-DSS file at the same time. This is true for a network drive as well as a local drive. You can have a shared network HEC-DSS file that has several processes reading and writing to it at the same time. The only drawback is that file access may be slower, depending on the operating system.
- USACE, Hydrologic Engineering Center (July, 2019). HEC Data Storage System Guide (Draft).
Sample dss file available in examples folder.
Write regular time-series data to example.dss
Notes: The interval must be [any] integer greater than 0 for regular time-series. Actual time-series interval implied from E-Part of pathname The values attribute can be list, array or numpy array
from datetime import datetime
from pydsstools.heclib.dss import HecDss
from pydsstools.core import TimeSeriesContainer,UNDEFINED
dss_file = "example.dss"
pathname = "/REGULAR/TIMESERIES/FLOW//1HOUR/Ex1/"
tsc = TimeSeriesContainer()
tsc.pathname = pathname
tsc.startDateTime = "15JUL2019 19:00:00"
tsc.numberValues = 7
tsc.units = "cfs"
tsc.type = "INST"
tsc.interval = 1
tsc.values = [100,UNDEFINED,500,5000,10000,24.1,25]
fid = HecDss.Open(dss_file)
fid.deletePathname(tsc.pathname)
fid.put_ts(tsc)
ts = fid.read_ts(pathname)
fid.close()
Read and plot regular time-series
from pydsstools.heclib.dss import HecDss
import matplotlib.pyplot as plt
import numpy as np
dss_file = "example.dss"
pathname = "/REGULAR/TIMESERIES/FLOW//1HOUR/Ex1/"
startDate = "15JUL2019 19:00:00"
endDate = "15AUG2019 19:00:00"
fid = HecDss.Open(dss_file)
ts = fid.read_ts(pathname,window=(startDate,endDate),trim_missing=True)
times = np.array(ts.pytimes)
values = ts.values
plt.plot(times[~ts.nodata],values[~ts.nodata],"o")
plt.show()
fid.close()
Write irregular time-series data
Notes: The interval must be [any] integer <= 0 for irregular time-series. DParts: IR-MONTH, IR-YEAR, IR-DECADE, IR-CENTURY
from datetime import datetime
from pydsstools.heclib.dss import HecDss
from pydsstools.core import TimeSeriesContainer, UNDEFINED
dss_file = "example.dss"
pathname = "/IRREGULAR/TIMESERIES/FLOW//IR-DECADE/Ex3/"
tsc = TimeSeriesContainer()
tsc.numberValues = 5
tsc.pathname = pathname
tsc.units ="cfs"
tsc.type = "INST"
tsc.interval = -1
tsc.values = [100,UNDEFINED,500,5000,10000]
tsc.times = [datetime(1900,1,12),datetime(1950,6,2,12),
datetime(1999,12,31,23,0,0),datetime(2009,1,20),
datetime(2019,7,15,5,0)]
with HecDss.Open(dss_file) as fid:
status = fid.put_ts(tsc)
Read irregular time-series data
from pydsstools.heclib.dss import HecDss
dss_file = "example.dss"
pathname = "/IRREGULAR/TIMESERIES/FLOW//IR-DECADE/Ex3/"
with HecDss.Open(dss_file) as fid:
ts = fid.read_ts(pathname,regular=False,window_flag=0)
print(ts.pytimes)
print(ts.values)
print(ts.nodata)
print(ts.empty)
Write paired data series
import numpy as np
from pydsstools.heclib.dss import HecDss
from pydsstools.core import PairedDataContainer
dss_file = "example.dss"
pathname ="/PAIRED/DATA/FREQ-FLOW///Ex5/"
pdc = PairedDataContainer()
pdc.pathname = pathname
pdc.curve_no = 2
pdc.independent_axis = list(range(1,10))
pdc.data_no = 9
pdc.curves = np.array([[5,50,500,5000,50000,10,100,1000,10000],
[11,11,11,11,11,11,11,11,11]],dtype=np.float32)
pdc.labels_list = ['Column 1','Elevens']
pdc.independent_units = 'Number'
pdc.dependent_units = 'Feet'
fid = HecDss.Open(dss_file)
fid.put_pd(pdc)
fid.close()
Read paired data-series
Notes: Row and column/curve indices start at 1 (not zero)
from pydsstools.heclib.dss import HecDss
dss_file = "example.dss"
pathname ="/PAIRED/DATA/FREQ-FLOW///Ex5/"
#labels_list = ['Column 1','Elevens']
with HecDss.Open(dss_file) as fid:
read_all = fid.read_pd(pathname)
row1,row2 = (2,4)
col1,col2 = (1,2)
read_partial = fid.read_pd(pathname,window=(row1,row2,col1,col2))
Pre-allocate paired data-series
from pydsstools.heclib.dss import HecDss
dss_file = "example.dss"
pathname ="/PAIRED/PREALLOCATED DATA/FREQ-FLOW///Ex7/"
with HecDss.Open(dss_file) as fid:
rows = 10
curves = 15
fid.preallocate_pd((rows,curves),pathname=pathname)
Write individual curve data in pre-allocated paired data-series
from pydsstools.heclib.dss import HecDss
dss_file = "example.dss"
pathname ="/PAIRED/PREALLOCATED DATA/FREQ-FLOW///Ex7/"
with HecDss.Open(dss_file) as fid:
curve_index = 5
curve_label = 'Column 5'
curve_data = [10,20,30,40,50,60,70,80,90,100]
fid.put_pd(curve_data,curve_index,pathname=pathname,labels_list=[curve_label])
curve_index = 2
curve_label = 'Column 2'
curve_data = [41,56,60]
row1,row2 = (5,7)
fid.put_pd(curve_data,curve_index,window = (row1,row2),
pathname=pathname,labels_list=[curve_label])
Read Spatial Grid
from pydsstools.heclib.dss.HecDss import Open
dss_file = "example.dss"
pathname = "/GRID/RECORD/DATA/01jan2001:1200/01jan2001:1300/Ex9/"
with Open(dss_file) as fid:
dataset = fid.read_grid(pathname)
grid_array = dataset.read()
profile = dataset.profile
# if rasterio library is installed
# raster attribute is available for dataset object
# save grid as geotiff with epsg 2868 for coordinate reference system
try:
dataset.raster.save_tiff(r'grid_dataset.tif', crs=2868)
except:
pass
else:
print('grid data saved as grid_dataset.tif')
Write Spatial Grid record
import numpy as np
import numpy.ma as ma
from affine import Affine
from pydsstools.heclib.dss.HecDss import Open
from pydsstools.heclib.utils import gridInfo
dss_file = "example.dss"
pathname_out1 = "/GRID/RECORD/DATA/01jan2019:1200/01jan2019:1300/Ex10a/"
pathname_out2 = "/GRID/RECORD/DATA/01jan2019:1200/01jan2019:1300/Ex10b/"
with Open(dss_file) as fid:
# Type 1: data is numpy array
# np.nan is considered as nodata
data = np.reshape(np.array(range(100),dtype=np.float32),(10,10))
data[0] = np.nan # assign nodata to first row
grid_info = gridInfo()
cellsize = 100 # feet
xmin,ymax = (1000,5000) # grid top-left corner coordinates
affine_transform = Affine(cellsize,0,xmin,0,-cellsize,ymax)
grid_info.update([('grid_type','specified'),
('grid_crs','unknown'),
('grid_transform',affine_transform),
('data_type','per-aver'),
('data_units','mm'),
('opt_time_stamped',False)])
fid.put_grid(pathname_out1,data,grid_info)
# Type 2: data is numpy masked array, where masked values are considered nodata
data = np.reshape(np.array(range(100),dtype=np.float32),(10,10))
data = ma.masked_where((data >= 10) & (data <30),data) # mask second and third rows
fid.put_grid(pathname_out2,data,grid_info)
Read DSS-6 Spatial Grid record Copy DSS-6 Grid to DSS-7 file
from pydsstools.heclib.dss.HecDss import Open
from pydsstools.heclib.utils import dss_logging
dss_logging.config(level='Diagnostic')
dss6_file = "example_dss6.dss"
dss7_file = "example.dss"
pathname_in = "/SHG/MAXTEMP/DAILY/08FEB1982:0000/08FEB1982:2400/PRISM/"
pathname_out = "/SHG/MAXTEMP/DAILY/08FEB1982:0000/08FEB1982:2400/Ex11/"
with Open(dss6_file) as fid:
dataset = fid.read_grid(pathname_in)
data = dataset.read()
profile = dataset.profile
with Open(dss6_file) as fidin, Open(dss7_file) as fidout:
dataset = fidin.read_grid(pathname_in)
fidout.put_grid(pathname_out,dataset,compute_range = True) # recomputing range limit table
Read pathname catalog
from pydsstools.heclib.dss.HecDss import Open
dss_file = "example.dss"
pathname_pattern ="/PAIRED/*/*/*/*/*/"
with Open(dss_file) as fid:
path_list = fid.getPathnameList(pathname_pattern,sort=1)
print('list = %r' % path_list)
Copy dss record
from pydsstools.heclib.dss.HecDss import Open
dss_file = "example.dss"
pathname_in ="/PAIRED/DATA/FREQ-FLOW///Ex5/"
pathname_out ="/PAIRED/DATA/FREQ-FLOW///Ex12/"
with Open(dss_file) as fid:
fid.copy(pathname_in,pathname_out)
Delete dss record
from pydsstools.heclib.dss.HecDss import Open
dss_file = "example.dss"
pathname ="/PAIRED/DATA/FREQ-FLOW///Ex12/"
with Open(dss_file) as fid:
fid.deletePathname(pathname)
Spatial Analysis on grid
# Notes
# Experimental geospatial methods for grid
# Not 100% sure about gridinfo that is computed for the cropped grid esp. for SHG and HRAP
# Will apreciate user feedbacks on this
# This example code was tested using the following libraries
# gdal 3.2.2
# matplotlib 3.4.4
# rasterio 1.2.1
# Potential rasterio issue with CRS
# https://github.com/mapbox/rasterio/blob/master/docs/faq.rst#why-cant-rasterio-find-projdb
# Unset PROJ_LIB environmental variable (i.e., SET PROJ_LIB= )
from pydsstools.heclib.dss.HecDss import Open
from pydsstools.heclib.utils import BoundingBox
dss_file = "example.dss"
pathname = r"/SHG/LCOLORADO/PRECIP/02JAN2020:1500/02JAN2020:1600/Ex15/"
pathname_out = r"/SHG/LCOLORADO/PRECIP/02JAN2020:1500/02JAN2020:1600/Ex15 OUT/"
fid = Open(dss_file)
ds0 = fid.read_grid(pathname)
if not getattr(ds0,'raster',None) is None:
ds0.raster.plot(mask_zeros = True, title = 'Original Spatial Grid')
bbox = BoundingBox(-50000,6*10**5,50000,7*10**5)
ds1 = ds0.raster.mask(bbox,crop = False)
ds1.raster.plot(mask_zeros = True, title = 'Clipped Spatial Grid')
ds2 = ds1.raster.mask(bbox,crop = True)
ds2.raster.plot(mask_zeros = True, title = 'Cropped Spatial Grid')
fid.put_grid(pathname_out,ds2)
The cython extension and grid version 6 libraries are compiled while installing pydsstools from source files. This requires proper configuration of build tools and python environment in the command prompt. The included cmd_intel_environ.bat can be used in Windows 10 provided that Intel Parallel Studio XE 2019 and Microsoft Visual Studio 2019 Community Edition or equivalents are available.
python setup.py install
or
pip install https://github.com/gyanz/pydsstools/zipball/master
Download suitable wheel file from dist directory and run th following command:
pip install wheel_file
Windows 10 wheel files
Python 3.6: pydsstools-X.x-cp36-cp36m-win_amd64.whl
Python 3.7: pydsstools-X.x-cp37-cp37m-win_amd64.whl
Python 3.8: pydsstools-X.x-cp38-cp38m-win_amd64.whl
Python 3.9: pydsstools-X.x-cp39-cp39m-win_amd64.whl
Ubuntu 20.04 LTS wheel files
Python 3.8: pydsstools-X.x-cp38-cp38-linux_x86_64.whl
All contributions, bug reports, bug fixes, documentation improvements, enhancements and ideas are welcome. Feel free to ask questions on my email.
This program is a free software: you can modify and/or redistribute it under MIT license.