Decide on output format for interoperability between parsers

[alanrice]

Currently outputting an array of objects. Maybe JSON would be better?

[alanrice]

{
  "sequences":[
    {
      "name":"SEQUENCE_1",
      "seq":"MTEITAAMVKELRESTGAGMMDCKNALSETNGDFDKAVQLLREKGLGKAAKKADRLAAEGLVSVKVSDDFTIAAMRPSYLSYEDLDMTFVENEYKALVAELEKENEERRRLKDPNKPEHKIPQFASRKQLSDAILKEAEEKIKEELKAQGKPEKIWDNIIPGKMNSFIADNSQLDSKLTLMGQFYVMDDKKTVEQVIAEKEKEFGGKIKIVEFICFEVGEGLEKKTEDFAAEVAAQL"
    },
    {
      "name":"SEQUENCE_2",
      "seq":"SATVSEINSETDFVAKNDQFIALTKDTTAHIQSNSLQSVEELHSSTINGVKFEEYLKSQIATIGENLVVRRFATLKAGANGVVNGYIHTNGRVGVVIAAACDSAEVASKSRDLLRQICMH"
    }
  ]
}

If unique headers can be assumed then:

{
  "SEQUENCE_1":{
    "seq":"MTEITAAMVKELRESTGAGMMDCKNALSETNGDFDKAVQLLREKGLGKAAKKADRLAAEGLVSVKVSDDFTIAAMRPSYLSYEDLDMTFVENEYKALVAELEKENEERRRLKDPNKPEHKIPQFASRKQLSDAILKEAEEKIKEELKAQGKPEKIWDNIIPGKMNSFIADNSQLDSKLTLMGQFYVMDDKKTVEQVIAEKEKEFGGKIKIVEFICFEVGEGLEKKTEDFAAEVAAQL",
    "seqType": "protein",
  },
  "SEQUENCE_2":{
    "seq":"SATVSEINSETDFVAKNDQFIALTKDTTAHIQSNSLQSVEELHSSTINGVKFEEYLKSQIATIGENLVVRRFATLKAGANGVVNGYIHTNGRVGVVIAAACDSAEVASKSRDLLRQICMH",
    "seqType": "protein",
  }
}

[alanrice]

This is kind of two issues in one: deciding on output format (should be json for interoperability) and also how to deal with non-unique headers.
Non-unique FASTA headers (discussed online in places previously) I think should be dealt with by checking that the sequence is not duplicated also. If it is duplicated then that redundant entry can be ignored, if it is a novel sequence then throw an error.