How to read csv in a Read Conversion initialization?

[mkalin2]

Is there a good way to read a csv file in the initialize method for a read conversion?

Currently I am getting a directory not found error when I upload to COSMOS:

Error. Error processing /tmp/d20241105-334-xh1ft/TARGET_NAME/cmd_tlm/tlm.txt

ENOENT: No such file or directory @ rb_sysopen - targets/TARGET_NAME/lib/csv_files/file.csv

Initially I was assuming the local directory was the target lib directory so I was just using csv_files/file.csv as my path, but I was getting a similar error on build. Changing the path to targets/TARGET_NAME/lib/csv_files/file.csv fixed the error on build, but when I upload to COSMOS I am still getting the same error.

[ryanmelt]

When we're parsing config files, the working directory is set to the folder of the config file. I'm not sure where you are reading this CSV file, but you need to control the working directory appropriately.

Wrap your file read like this:

      relative_path_to_other_file = "whatever relative to this file"
      full_path_to_current_file = File.expand_path(File.dirname(__FILE__))
      OpenC3.set_working_dir(File.dirname(full_path_to_current_file)) do
        return File.read(relative_path_to_other_file).force_encoding("UTF-8")
      end

[mkalin2]

Sorry this doesn't seem to be working. I have tried using the relative path from tlm.txt as well as this_conversion.rb and still get the same error for both cases.

Here is the directory structure:

TARGET_NAME
├── cmd_tlm
│   ├── cmd.txt
│   └── tlm.txt
├── lib
│   ├── this_conversion.rb
│   ├── csv_files
│   │   └── file.csv
├── procedures
│   └── procedure.rb
├── public
│   └── README.txt
├── screens
│   ├── status.txt
└── target.txt

[ryanmelt]

Please provide the code that reads the file.

[mkalin2]

This function gets called in the initialize function of this_conversion.rb.

file_name = 'file.csv'

def read_csv(file_name)
      relative_path = 'csv_files/' + file_name
      curr_dir = File.expand_path(File.dirname(__FILE__))
      OpenC3.set_working_dir(File.dirname(curr_dir)) do
        return CSV.read(relative_path)
      end
end

I have also tried using File.read() rather than the csv library, but I still get the file not found error.

[mkalin2]

Here is the full stack trace:

Error: Error processing /tmp/d20241106-10-k083cs/TARGET_NAME/cmd_tlm/tlm.txt:

ENOENT : No such file or directory @ rb_sysopen - csv_files/file.csv
/usr/lib/ruby/3.2.0/csv.rb:1593:in `initialize'
/usr/lib/ruby/3.2.0/csv.rb:1593:in `open'
/usr/lib/ruby/3.2.0/csv.rb:1593:in `open'
/usr/lib/ruby/3.2.0/csv.rb:1829:in `read'
/tmp/d20241106-10-lfwb9v/gem/targets/TARGET_NAME/lib/this_conversion.rb:77:in `block in read_csv'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/top_level.rb:483:in `set_working_dir_internal'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/top_level.rb:473:in `block in set_working_dir'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/top_level.rb:472:in `synchronize'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/top_level.rb:472:in `set_working_dir'
/tmp/d20241106-10-lfwb9v/gem/targets/TARGET_NAME/lib/this_conversion.rb:76:in `read_csv'
/tmp/d20241106-10-lfwb9v/gem/targets/TARGET_NAME/lib/this_conversion.rb:27:in `initialize'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/packets/packet_config.rb:578:in `new'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/packets/packet_config.rb:578:in `process_current_item'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/packets/packet_config.rb:238:in `block in process_file'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/config/config_parser.rb:217:in `parse_loop'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/config/config_parser.rb:217:in `parse_file'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/packets/packet_config.rb:144:in `process_file'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/system/system.rb:181:in `block in add_target'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/system/system.rb:180:in `each'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/system/system.rb:180:in `add_target'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/system/system.rb:169:in `block in initialize'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/system/system.rb:169:in `each'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/system/system.rb:169:in `initialize'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/models/target_model.rb:602:in `new'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/models/target_model.rb:602:in `deploy'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/models/plugin_model.rb:251:in `block in install_phase2'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/config/config_parser.rb:217:in `parse_loop'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/config/config_parser.rb:217:in `parse_file'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/openc3/models/plugin_model.rb:243:in `install_phase2'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/../bin/openc3cli:327:in `validate_plugin'
/usr/lib/ruby/gems/3.2.0/gems/openc3-5.19.0/lib/../bin/openc3cli:850:in `<main>'

[ryanmelt]

I think my earlier code had one too many dirname calls.

Try this:

def read_csv(file_name)
      relative_path = 'csv_files/' + file_name
      full_file_path = File.expand_path(__FILE__)
      OpenC3.set_working_dir(File.dirname(full_file_path)) do
        return CSV.read(relative_path)
      end
end

[mkalin2]

That seems to have fixed it! Thanks!