Running Measure Test via OpenStudio CLI Returns Error
I was modifying a copy of a reporting measure. To save test time, I am using OpenStudio CLI. My modified OS measure was from the OpenStudio Result IP measure and I was testing how to create my own sections using the example template within the same measure. I did this by modifying a copy of os_lib_reporting.rb
. See my modified file below:
# Custom OS reporting
require 'json'
module OsLib_Reporting_Custom
#setup - get model, sql, and setup web assets path
def self.setup(runner)
results = {}
#get the last model
model = runner.lastOpenStudioModel
if model.empty?
runner.registerError('Cannot find last model.')
return false
end
model = model.get
#get the last idf
workspace = runner.lastEnergyPlusWorkspace
if workspace.empty?
runner.registerError('Cannot find last idf file.')
return false
end
workspace = workspace.get
#get the last sql file
sqlFile = runner.lastEnergyPlusSqlFile
if sqlFile.empty?
runner.registerError('Cannot find last sql file.')
return false
end
sqlFile = sqlFile.get
model.setSqlFile(sqlFile)
#populate hash to pass to measure
results[:model] = model
#results[:workspace] = workspace
results[:sqlFile] = sqlFile
results[:web_asset_path] = OpenStudio.getSharedResourcesPath / OpenStudio::Path.new('web_assets')
return results
end
#create template section
def self.template_section(model, sqlFile, runner, name_only = false)
#array to hold tables
template_tables = []
#gather data for section
@template_section = {}
@template_section[:title] = 'Template Section'
@template_section[:tables] = template_tables
#stop here if only name is requested - this is used to populate display name for arguments
if name_only == true
return @template_section
end
#create table
template_table_01 = {}
template_table_01[:title] = 'Sample Table 1'
template_table_01[:header] = ['Bldg', 'Energy']
template_table_01[:units] = ['', 'kWh']
template_table_01[:data] = []
#add rows to table
template_table_01[:data] << ['Bldg 1', 1000]
template_table_01[:data] << ['Bldg 2', 9870]
template_table_01[:data] << ['Bldg 3', 431]
#create chart
template_table_01[:chart_type] = 'simple_pie' #search for value options
template_table_01[:chart] = []
template_table_01[:data].each do |row|
template_table_01[:chart] << JSON.generate(label: row[0], value: row[1])
end
#add table to array of tables
template_tables << template_table_01
#use helper method that generates additional table for section
#template_tables << OsLib_Reporting_Custom.template_table(model, sqlFile, runner)
return @template_section
end
end
I added this file into my measure.rb
as shown below:
require 'erb'
require 'json'
require "#{File.dirname(__FILE__)}/resources/os_lib_reporting"
require "#{File.dirname(__FILE__)}/resources/os_lib_reporting_custom"
require "#{File.dirname(__FILE__)}/resources/os_lib_schedules"
require "#{File.dirname(__FILE__)}/resources/os_lib_helper_methods"
require_relative 'resources/Siz.Model'
# start the measure
class ReportFormatV1 < OpenStudio::Measure::ReportingMeasure
# human readable name
def name
# Measure name should be the title case of the class name.
return "Report Format v1"
end
Besides renaming all instances of OsLib_Reporting
to OsLib_Reporting_Custom
, I haven't edited measure.rb
yet as I was excited to see my simple section template.
So what I did next was to proceed to my test file. Once again, I modified a copy of the existing test file (now named report_format_v_1_test.rb
). Within it, I simply updated the expected number of arguments; e.g.,
def test_number_of_arguments_and_argument_names
# create an instance of the measure
measure = ReportFormatV1.new
# get arguments and test that they are what we are expecting
arguments = measure.arguments()
assert_equal(1, arguments.size) #originally this was 0
end
Finally, I ...
The measure you based things on is a modified version of OpenStudio Results measures and I don’t know the state of the measure or tests, but it seems like it ran for you. It is unclear if the error you have. Is based on a measure only with the few lines of code above or is based on the other measure so it is hard to diagnose the error. I would suggest you make a fresh measure that comes with its own tests as a starting point. You can do this in the OpenStuido Application or Parametric Analysis Tool (PAT). We want add make new measure to CLI but have not yet.
I also wanted to point you to an example reporting measure that may be a useful starting point. It pulls in some model data but also some external data that can be replaced with what you want to report. https://github.com/NREL/openstudio-co...