allow the use of "all" to define catchment and nexus subsets

README.md

@@ -47,18 +47,19 @@ If the software is configurable, describe it in detail, either here or in other
 
 To run the *ngen* engine, the following command line positional arguments are supported:
 - _catchment_data_path_ -- path to catchment data geojson input file.
-- _catchment subset ids_ -- list of comma separated ids (NO SPACES!!!) to subset the catchment data, i.e. 'cat-0,cat-1'
+- _catchment subset ids_ -- list of comma separated ids (NO SPACES!!!) to subset the catchment data, i.e. 'cat-0,cat-1', an empty string or "all" will use all catchments in the hydrofabric
 - _nexus_data_path_ -- path to nexus data geojson input file
-- _nexus subset ids_ -- list of comma separated ids (NO SPACES!!!) to subset the nexus data, i.e. 'nex-0,nex-1'
+- _nexus subset ids_ -- list of comma separated ids (NO SPACES!!!) to subset the nexus data, i.e. 'nex-0,nex-1', an empty string or "all" will use all nexus points
 - _realization_config_path_ -- path to json configuration file for realization/formulations associated with the hydrofabric inputs
 - _partition_config_path_ -- path to the partition json config file, when using the driver with [distributed processing](doc/DISTRIBUTED_PROCESSING.md).
 - `--subdivided-hydrofabric` -- an explicit, optional flag, when using the driver with [distributed processing](doc/DISTRIBUTED_PROCESSING.md), to indicate to the driver processes that they should operate on process-specific subdivided hydrofabric files.
 
 An example of a complete invocation to run a subset of a hydrofabric.  If the realization configuration doesn't contain catchment definitions for the subset keys provided, the default `global` configuration is used.  Alternatively, if the realization configuration contains definitions that are not in the subset (or hydrofabric) keys, then a warning is produced and the formulation isn't created.
 `./cmake-build-debug/ngen ./data/catchment_data.geojson "cat-27,cat-52" ./data/nexus_data.geojson "nex-26,nex-34" ./data/example_realization_config.json`
 
-To simulate every catchment in the input hydrofabric, leave the subset lists empty, i.e.:
+To simulate every catchment in the input hydrofabric, leave the subset lists empty, or use "all" i.e.:
 `ngen ./data/catchment_data.geojson "" ./data/nexus_data.geojson "" ./data/refactored_example_realization_config.json`
+`ngen ./data/catchment_data.geojson "all" ./data/nexus_data.geojson "all" ./data/refactored_example_realization_config.json`
 
 Examples specific to running with with distributed processing can be found [here](doc/DISTRIBUTED_PROCESSING.md#examples).
 

src/NGen.cpp

@@ -78,9 +78,9 @@ int main(int argc, char *argv[]) {
     //Use "positional args"
     //arg 0 is program name
     //arg 1 is catchment_data file path
-    //arg 2 is catchment subset ids, comma seperated string of ids (no spaces!), "" for all
+    //arg 2 is catchment subset ids, comma seperated string of ids (no spaces!), "all" for all
     //arg 3 is nexus_data file path
-    //arg 4 is nexus subset ids, comma seperated string of ids (no spaces!), "" for all
+    //arg 4 is nexus subset ids, comma seperated string of ids (no spaces!), "all" for all
     //arg 5 is realization config path
     //arg 7 is the partition file path
     //arg 8 is an optional flag that driver should, if not already preprocessed this way, subdivided the hydrofabric
@@ -176,7 +176,11 @@ int main(int argc, char *argv[]) {
 
         //split the subset strings into vectors
         boost::split(catchment_subset_ids, argv[2], [](char c){return c == ','; } );
+        if( catchment_subset_ids.size() == 1 && catchment_subset_ids[0] == "all")
+          catchment_subset_ids.pop_back();
         boost::split(nexus_subset_ids, argv[4], [](char c){return c == ','; } );
+        if( nexus_subset_ids.size() == 1 && nexus_subset_ids[0] == "all")
+          nexus_subset_ids.pop_back();
         //If a single id or no id is passed, the subset vector will have size 1 and be the id or the ""
         //if we get an empy string, pop it from the subset list.
         if(nexus_subset_ids.size() == 1 && nexus_subset_ids[0] == "") nexus_subset_ids.pop_back();