SATF/SASF Grammar support

[goetzrrGit]

Closes #1496
I added support for SATF and SASF formats, which includes:

• 85% of the grammar rules in SATF are now supported
• I've also added support for 60% of the grammar rules in SASF, using references from Psyche SASF and NISAR

I created a converter which will eventually be a npm plugin that can take SeqN data and generate most of the SATF/SASF content. This is because each mission has slight variation so they can use this to generate 90% of the SATF/SASF and fill in the mission specific parts in their adaptation.

There's also a way to convert a SATF/SASF file into SeqN format. Not everything is carried over as SeqN doesn't support all the features of SATF/SASF

TESTING
You can find a bunch of SATF/SASF here: https://jpl365prod-my.sharepoint.com/:f:/g/personal/shaheer_khan_jpl_nasa_gov/EplqvV9nI9ZJo6JT1KVzoVwBKZHzFNbVGXWi2qUaN6k1yg?e=zIbLEa

This is hard to test as it requires hacking the Sequence Editor to use the top editor as a SATF/SASF editor and the bottom editor as a SeqN. Here is the modification I did.

  // Reconfigure sequence editor.
          editorSequenceView.dispatch({
            effects: [
              compartmentSeqLanguage.reconfigure(
                setupLanguageSupport(
                  $sequenceAdaptation.autoComplete(
                    parsedChannelDictionary,
                    parsedCommandDictionary,
                    nonNullParsedParameterDictionaries,
                    librarySequences,
                  ),
                ),
              ),
              compartmentSeqLanguage.reconfigure(setupSatfLanguageSupport()), <---- Add this line
              compartmentSeqLinter.reconfigure(
                inputLinter(
                  parsedChannelDictionary,
                  parsedCommandDictionary,
                  nonNullParsedParameterDictionaries,
                  librarySequences,
                ),
              ),
              compartmentSeqTooltip.reconfigure(
                sequenceTooltip(parsedChannelDictionary, parsedCommandDictionary, nonNullParsedParameterDictionaries),
              ),
              ...($sequenceAdaptation.autoIndent
                ? [compartmentSeqAutocomplete.reconfigure(indentService.of($sequenceAdaptation.autoIndent()))]
                : []),
            ],
          });

    if ($sequenceAdaptation?.modifyOutput !== undefined && output !== undefined) {
      const modifiedOutput = $sequenceAdaptation.modifyOutput(output, parameterDictionaries, channelDictionary);
      if (modifiedOutput === null) {
        output = 'modifyOutput returned null. Verify your adaptation is correct';
      } else if (modifiedOutput === undefined) {
        output = 'modifyOutput returned undefined. Verify your adaptation is correct';
      } else if (typeof modifiedOutput === 'object') {
        output = JSON.stringify(modifiedOutput);
      } else {
        output = `${modifiedOutput}`;
      }
    }
    
    const seqN = await satfToSequence(sequence);  <--------- Add this line
    output = seqN.header + '\n' + seqN.sequences.join('\n'); <----- Add this line

    editorOutputView.dispatch({ changes: { from: 0, insert: output, to: editorOutputView.state.doc.length } });

    if (output !== undefined) {
      dispatch('sequence', { input: sequence, output });
    }

[joswig]

Syntax highlighting looks good in my testing.

Were you able to test for parse errors in all of the linked files?

[joswig]

Are we good proceeding with this TODO?

[goetzrrGit]

This is related to NASA-AMMOS/seq-json-schema#32

Request blocks will have a special time tag in SeqN that we can use. This is just a reminder.

[joswig]

Should the literals representing token and rule names use the satfConstants?

[goetzrrGit]

We should and I updated and used them from the satfConstants

[cohansen]

Is there a list somewhere of remaining work? You mention that this is about 60% of the current grammar rules.

[goetzrrGit]

Is there a list somewhere of remaining work? You mention that this is about 60% of the current grammar rules.

The remaining work is to verify that both satf/sasf sis are 100% implemented. The rules I left out are things that PSYCHE, NISAR are not using here is a rough list of unsupported steps:

for,
while,
pause,
spawn,
call,
wait_change,
Delay_by,
Delay_until,
Test_and_set,
command_wait,
noop,
re_command,
reservation,
if/else conditional

lgtm, but will defer to @joswig for the final approval

[cohansen]

LGTM, I did some manual testing and I think @joswig and @duranb had good feedback.

[joswig]

lgtm