#181 PipeParser should fail (or warn) on bad line endings

[James Agnew]

Per a message on the mailing list: PipeParser should give an error, or maybe even fail completely, or maybe be lenient if it detects a message that has the wrong line endings (\n instead of \r).

Currently it just puts everything in one giant MSH segment, which is about as unhelpful as it gets.

[James Agnew]

Some sample code:
String msg1 = "MSH|^~\&|SOARF|HMC|||201304110916||ADT^A02|4c73eeb1-85ee-439f-91d9-747cb20d82ed|P|2.5||2|||||||\n" +
"EVN|A02|201304110916||XFR|B0QZ_LGILLE|201304110914|939\n" +
"PID|1||4125315^^^900000^PN^^20130411~401768^^^14^MR^^20130411||Smith^Boy Gloria^^^^^L^^^20130411~Smith^Baby Gloria^^^^^L^^^20130411@20130411||201304110706|M||H|500 MARY SUE DR^^ATLANTA^GA^30721^USA^M^^White^^^20130411||^PRN^PH^^1^706^2295221||EN|S|NRL|1000038688^^^14^VCD^^20130411|||1000038533^^^14^VCD^^20130410~100039469^^^14^VN^^20130410~227922^^^900000^PN^^20021011~169813^^^14^MR^^20021011~255892163^^^900001^SS^^20130410|HIS||N|1|||||N|N||2013|||||\n" +
"PD1||||||||N||||||^L|||||||\n" +
"PV1|1|I|NCN^NCN IN53^IN5301^939|Newborn||NCN^NCN BEDS^NURS03^939|600458^Whitman^Jeffeory^H^^^^^11^L^^^PRN^^^^19560901~421805573^^^^^^^^900001^^^^HPSS~100^^^^^^^^900000^^^^DN~D41371^^^^^^^^900002^^^^UPIN~010588^^^^^^^^1788^^^^PRN~1720074198^^^^^^^^900004^^^^NPI~022667^^^^^^^^13^^^^LN~AW1082697^^^^^^^^900001^^^^DEA|||NUR|||N|Born in hospital||||IP|100039638^^^14^VN^^20130411|Mcaid|||||||||||||||||||939|||||201304110706|||||||V|\n" +
"PV2|||^Newborn (Vaginal)||||~~~~false||||||||N||||||Y||Inpatient^L^17^^^900000^XX^^ ~^^1528056066^^^900004^NPI^^ |Checked in|Newborn||||||N|||||Y||||Nursery Newborn|||||||\n" +
"ROL||UC|AT|600458^Whitman^Jeffeory^H^^^^^11 ^L^^^PRN^^^^19560901~421805573^^^^^^^^900001 ^^^^HPSS~100^^^^^^^^900000 ^^^^DN~D41371^^^^^^^^900002 ^^^^UPIN~010588^^^^^^^^1788 ^^^^PRN~1720074198^^^^^^^^900004 ^^^^NPI~022667^^^^^^^^13 ^^^^LN~AW1082697^^^^^^^^900001 ^^^^DEA|201304110706||||Phys|||^WPN^PH^^0^706^8762130^0~^ORN^FX^^0^706^8762168\n" +
"OBX|1|NM|9272-6^Apgar 1^LN||0||||||||||||\n" +
"OBX|2|NM|9274-2^Apgar 5^LN||0||||||||||||\n" +
"PDA||||||N|||N\n" +
"ZPV|||||||||||||||||||||||||||||||||||||||||||||||||\"\"||||||||||||^^mostCurrEncTypeCd^mostCurrEncTypeCd^IP||||||||100039638^^No SCD^14^ZAVN^^20130411||201304110706|\n";

     HapiContext context = new DefaultHapiContext();
     Parser p = context.getGenericParser();
     Message hl7msg;

     try {

         hl7msg = p.parse(msg1);
     } catch (EncodingNotSupportedException e) {
         e.printStackTrace();
         return;
     } catch (HL7Exception e) {
         e.printStackTrace();
         return;
     }

    //ORU_R01 adtMsg = (ORU_R01)hl7msg;
    ADT_A02 adtMsg = (ADT_A02)hl7msg;
    PID pid  = adtMsg.getPID();
    XPN [] names = adtMsg.getPID().getPatientName();

    System.out.println(adtMsg.printStructure());

    ca.uhn.hl7v2.model.v25.segment.MSH msh = adtMsg.getMSH();
    System.out.println("patientName length: " + pid.getPatientName().length);

    String msgTrigger = msh.getMessageType().getTriggerEvent().getValue();
    System.out.println("messageTrigger: " + msgTrigger);

[Christian Ohr]

Maybe add a ParserConfiguration item that forces a String.replace(\n,\r) before parsing starts?