Working with Cross-Sections#
Cross-sections get some special treatment in pytuflow as the contents of the GIS file or cross-section database (e.g. Flood Modeller) are loaded into a database. This is done as querying the cross-section data programmatically is often very useful, especially if pytuflow is being used in a software that uses a UI (like a QGIS plugin), or script that generates figures that require cross-section data.
Viewing TUFLOW Cross-Sections#
The below example uses EG14_001.tcf from the TUFLOW Example Model Dataset
to demonstrate how to view the TUFLOW cross-section data in pytuflow. Note, despite being referenced in a GIS
layer, GDAL python bindings are not required to load TUFLOW cross-sections.
The first step is to load the TUFLOW model via the TCF class:
>>> from pytuflow import TCF
>>> tcf = TCF('path/to/EG14_001.tcf')
Next, we need to find the cross-section input. The cross-section input we want to view is in the GIS file
gis\1d_xs_EG14_001_L.shp. We can find this input using the TCF.find_input() method:
>>> xs_inp = tcf.find_input('1d_xs_EG14_001_L')[0]
>>> print(xs_inp)
<GisInput> Read GIS Table Links == gis\1d_xs_EG14_001_L.shp
The database can be accessed via the GisInput.cf attribute. The cf attribute
is a list of all the control files or databases that are loaded from the input. Another example of this is the
Geometry Control File input, which would have a list of all the geometry control files loaded from the input.
>>> xs_db = xs_inp.cf[0]
>>> print(xs_db)
<CrossSectionDatabase> 1d_xs_EG14_001_L.shp
In this case, only one database instance will be loaded. Multiple instances of the database can be loaded if
the file path uses a variable (e.g. 1d_xs_<<~s1~>>_001_L), which could be resolved into mulitiple file paths.
The cross-section database is lazy loaded (i.e. only loaded when it is first accessed), so the initial call to
this database will take longer than subsequent calls. Like other database class in pytuflow, the cross-section
database is stored in a DataFrame, which can be accessed via the
CrossSectionDatabase.df attribute:
>>> print(xs_db.df)
Source Type Flags Column_1 Column_2 Column_3 Column_4 Column_5 Column_6
ID
..\csv\1d_xs_M14_C99.csv ..\csv\1d_xs_M14_C99.csv XZ
..\csv\1d_xs_M14_C100.csv ..\csv\1d_xs_M14_C100.csv XZ
..\csv\1d_xs_M14_C101.csv ..\csv\1d_xs_M14_C101.csv XZ
..\csv\1d_xs_M14_C102.csv ..\csv\1d_xs_M14_C102.csv XZ
..\csv\1d_xs_M14_C103.csv ..\csv\1d_xs_M14_C103.csv XZ
The cross-section ID in this case is the file path to the CSV file, however, TUFLOW allows multiple cross-sections to be stored in a single CSV file. TUFLOW uses columns names to determine where the data is in the CSV file for this situation. Similarly, pytuflow will use the file path + the column name as the ID. This ensures a unique ID for each cross-section.
Again, similar to other pytuflow database classes, the CrossSectionDatabase.value()
method can be used to get the cross-section data for a given ID.
>>> xs_db.value(r'..\csv\1d_xs_M14_C99.csv')
X Z
0 0.00000 38.4567
1 1.16450 38.2227
2 6.74383 37.4142
3 6.74534 37.4140
4 7.58031 36.8805
5 7.58061 36.8803
6 8.83271 35.9894
7 8.83344 35.9889
8 10.99330 36.0249
9 10.99470 36.0249
10 11.40040 36.0340
... ... ...
26 42.39770 37.6635
27 43.05550 37.6766
28 44.40290 37.7324
Viewing Flood Modeller Cross-Sections#
Flood Modeller cross-sections are also supported in pytuflow. If you are using the the command: XS Database ==, then
you can use a similar approach to the above example to obtain the cross-section database. However, if you are
linking Flood Modeller with TUFLOW, then the .dat file won’t be present in the TUFLOW model. You can
load the Flood Modeller .dat file manually to access the cross-section data (see example below).
There are alternative methods to load the Flood Modeller cross-section data, for instance, using the Flood Modeller API. However, if you have a script that needs to handle both TUFLOW cross-sections and Flood Modeller cross-sections, then you may find using pytuflow preferable as it provides a consistent interface for accessing the cross-section data.
The TUFLOW Example Model Dataset does not contain a Flood Modeller cross-section or linked example. However, TUFLOW
does have a Flood Modeller tutorial model
that can be used to demonstrate this functionality. The below example uses FMT_M01_001.dat from the completed
tutorial model.
>>> from pytuflow import CrossSectionDatabase
>>> xs_db = CrossSectionDatabase('path/to/FMT_M01_001.dat')
The above step loads the cross-section database from the Flood Modeller .dat file into a CrossSectionDatabase.
The database instance is the same as the TUFLOW cross-section database, however the DataFrame will have different columns, but can
be accessed in the same way:
>>> print(xs_db.df)
Name Type
ID
RIVER_SECTION_FC01.40 FC01.40 RIVER
RIVER_SECTION_FC01.39 FC01.39 RIVER
RIVER_SECTION_FC01.38 FC01.38 RIVER
RIVER_SECTION_FC01.37 FC01.37 RIVER
RIVER_SECTION_FC01.36 FC01.36 RIVER
RIVER_SECTION_FC01.35 FC01.35 RIVER
... ... ...
RIVER_SECTION_FC02.02 FC02.02 RIVER
RIVER_SECTION_FC02.01 FC02.01 RIVER
RIVER_SECTION_FC02.01d FC02.01d RIVER
The cross-section data can be accessed via the CrossSectionDatabase.value() method.
The key for the cross-section data can either by the ID or the Name value of the cross-section. Names within Flood Modeller are
not always unique for the whole model, however are unique within the context of river sections, so it is safe to use it
this way:
>>> xs_db.value('RIVER_SECTION_FC01.40')
x y n rel_path_len chan_marker easting northing deactivation_marker sp_marker path_marker
0 0.000 56.244 0.09 1.0 NaN 0.0 0.0 NaN NaN
1 3.223 55.707 0.09 0.0 NaN 0.0 0.0 NaN NaN
2 3.621 55.656 0.09 0.0 NaN 0.0 0.0 NaN NaN
3 3.628 55.656 0.09 0.0 NaN 0.0 0.0 NaN NaN
4 4.714 55.555 0.09 0.0 NaN 0.0 0.0 NaN NaN
5 7.523 55.284 0.09 0.0 NaN 0.0 0.0 NaN NaN
6 8.544 55.219 0.09 0.0 NaN 0.0 0.0 NaN NaN
7 11.506 54.266 0.09 0.0 NaN 0.0 0.0 NaN NaN
... ... ... ... ... ... ... ...
50 43.466 52.803 0.09 0.0 NaN 0.0 0.0 NaN NaN
51 43.476 52.803 0.09 0.0 NaN 0.0 0.0 NaN NaN
52 44.660 52.820 0.09 0.0 NaN 0.0 0.0 NaN NaN