How to correctly parse an HL7 message using python HL7Apy?

Question

Here is my code:

from hl7apy.parser import parse_message

hl7 = open("hl7.txt", "r").read()
msg = parse_message(hl7)
print(msg.children)

result:

[<Segment MSH>]

It shows only the first segment, seems simple but I don't know what i'm doing wrong.

I've tried from a text file, passing the message directly or even with another HL7 message, but always got same results.

Here is the message:

MSH|^~\&|SendingAPP|TEST|||20080617143943||ORU^R01|1|P|2.3.1||||||UNICODE
PID|1||7393670^^^^MR||Joan^JIang||19900804000000|Female
PV1|1||nk^^001
OBR|1||20071207011|00001^Automated Count^99MRC||20080508140600|20080508150616|||John||||20080508150000||||||||||HM||||||||TEST

Here is my message in notepad++ where all characters are shown:

Make sure all the HL7 separator characters are correct in message. Please refer to [this](https://stackoverflow.com/a/59153332/5779732) answer for more details. — Amit Joshi, Jan 03 '20 at 06:20

badger0053 · Answer 1 · 2020-01-03T00:15:07.100

I think your issue is the the MLLP constant for HL7apy is a Carriage return \r. If you replace the new line characters \n the groups will parse out fine

from hl7apy.parser import parse_message
from hl7apy.core import Group, Segment

hl7 = """
MSH|^~\&|SendingAPP|TEST|||20080617143943||ORU^R01|1|P|2.3.1||||||UNICODE
PID|1||7393670^^^^MR||Joan^JIang||19900804000000|Female PV1|1||nk^^001
OBR|1||20071207011|00001^Automated
Count^99MRC||20080508140600|20080508150616|||John||||20080508150000||||||||||HM||||||||TEST
"""

msg = parse_message(hl7.replace('\n', '\r'), find_groups=True, validation_level=2)
print(msg.children)
print(msg.children[1].children)

for segment in msg.children:
    if isinstance(segment, Segment):
        for attribute in segment.children:
            print(attribute, attribute.value)
    if isinstance(segment, Group):
        for group in segment.children:
            for group_segment in group.children:
                for attribute in group_segment.children:
                    print(attribute, attribute.value)

Thanks for your answer, the thing is: this is a straight to the point sample code to my probleme. Actualy, the hl7 messages are recieved over tcp/ip and since they are not parsed perfectly it become difficult to adapt my code for all the use cases, it's not just about accessing the informations — infantry, Jan 03 '20 at 23:50

sqlab · Answer 2 · 2020-01-09T06:01:54.110

Additionaly to the answer above you can use an exception handler

from hl7apy import parser
from hl7apy.core import Group, Segment
from hl7apy.exceptions import UnsupportedVersion

and use it like this

 try:
     msg = parser.parse_message(hl7)
 except UnsupportedVersion:
     msg = parser.parse_message(hl7.replace("n", "r"))

If your message is correct you can acces the contents of fields like this

 print(msg.ORU_R01_PATIENT_RESULT.ORU_R01_PATIENT.PID.PID_7.value)
 print(msg.ORU_R01_PATIENT_RESULT.ORU_R01_PATIENT.ORU_R01_VISIT.PV1.PV1_1.value)
 print(msg.ORU_R01_PATIENT_RESULT.ORU_R01_ORDER_OBSERVATION.OBR.OBR_3.value)

So an enhanced answer showing all groups and subgroups - but still not perfect - can be e.g.

from hl7apy import parser
from hl7apy.core import Group, Segment
from hl7apy.exceptions import UnsupportedVersion

hl7 = open("hl7.txt", "r").read()

try:
    msg = parser.parse_message(hl7.replace('\n', '\r'), find_groups=True, validation_level=2)
except UnsupportedVersion:
    msg = parser.parse_message(hl7.replace('\n', '\r'), find_groups=True, validation_level=2)

indent = "    "
indent_seg = "    "
indent_fld = "        "

def subgroup (group, indent):
    indent = indent + "    "
    print (indent , group)
    for group_segment in group.children:
        if isinstance(group_segment, Group):
            subgroup (group_segment)
        else: 
            print(indent_seg, indent ,group_segment)
            for attribute in group_segment.children:
                print(indent_fld, indent ,attribute, attribute.value)


def showmsg (msg):
    print(msg.children[1])
    for segment in msg.children:
        if isinstance(segment, Segment):
            print (indent ,segment)
            for attribute in segment.children:
                print(indent_fld, indent, attribute, attribute.value)
        if isinstance(segment, Group):
            for group in segment.children:
                print (indent,group)
                for group_segment in group.children:
                    if isinstance (group_segment, Group): 
                        subgroup (group_segment, indent)
                    else:
                        print(indent_seg, indent ,group_segment)
                        for attribute in group_segment.children:
                            print(indent_fld, indent, attribute, attribute.value)

showmsg (msg)

can give a result like this

<Group ORU_R01_PATIENT_RESULT>
    <Segment MSH>
             <Field MSH_1 (FIELD_SEPARATOR) of type ST> |
             <Field MSH_2 (ENCODING_CHARACTERS) of type ST> ^~\&
             <Field MSH_3 (SENDING_APPLICATION) of type HD> SendingAPP
             <Field MSH_4 (SENDING_FACILITY) of type HD> TEST
             <Field MSH_7 (DATE_TIME_OF_MESSAGE) of type TS> 20080617143943
             <Field MSH_9 (MESSAGE_TYPE) of type MSG> ORU^R01
             <Field MSH_10 (MESSAGE_CONTROL_ID) of type ST> 1
             <Field MSH_11 (PROCESSING_ID) of type PT> P
             <Field MSH_12 (VERSION_ID) of type VID> 2.3.1
             <Field MSH_18 (CHARACTER_SET) of type ID> UNICODE
    <Group ORU_R01_PATIENT>
         <Segment PID>
             <Field PID_1 (SET_ID_PID) of type SI> 1
             <Field PID_3 (PATIENT_IDENTIFIER_LIST) of type CX> 7393670^^^^MR
             <Field PID_5 (PATIENT_NAME) of type XPN> Joan^JIang
             <Field PID_7 (DATE_TIME_OF_BIRTH) of type TS> 19900804000000
             <Field PID_8 (SEX) of type IS> Female
        <Group ORU_R01_VISIT>
             <Segment PV1>
                 <Field PV1_1 (SET_ID_PV1) of type SI> 1
                 <Field PV1_3 (ASSIGNED_PATIENT_LOCATION) of type PL> nk^^001
    <Group ORU_R01_ORDER_OBSERVATION>
         <Segment OBR>
             <Field OBR_1 (SET_ID_OBR) of type SI> 1
             <Field OBR_3 (FILLER_ORDER_NUMBER) of type EI> 20071207011
             <Field OBR_4 (UNIVERSAL_SERVICE_ID) of type CE> 00001^Automated Count^99MRC
             <Field OBR_6 (REQUESTED_DATE_TIME) of type TS> 20080508140600
             <Field OBR_7 (OBSERVATION_DATE_TIME) of type TS> 20080508150616
             <Field OBR_10 (COLLECTOR_IDENTIFIER) of type XCN> John
             <Field OBR_14 (SPECIMEN_RECEIVED_DATE_TIME) of type TS> 20080508150000
             <Field OBR_24 (DIAGNOSTIC_SERV_SECT_ID) of type ID> HM
             <Field OBR_32 (PRINCIPAL_RESULT_INTERPRETER) of type NDL> TEST1
    <Group ORU_R01_ORDER_OBSERVATION>
         <Segment OBR>
             <Field OBR_1 (SET_ID_OBR) of type SI> 2
             <Field OBR_3 (FILLER_ORDER_NUMBER) of type EI> 20071207011
             <Field OBR_4 (UNIVERSAL_SERVICE_ID) of type CE> 00001^Automated Count^99MRC
             <Field OBR_6 (REQUESTED_DATE_TIME) of type TS> 20080508140600
             <Field OBR_7 (OBSERVATION_DATE_TIME) of type TS> 20080508150616
             <Field OBR_10 (COLLECTOR_IDENTIFIER) of type XCN> John
             <Field OBR_14 (SPECIMEN_RECEIVED_DATE_TIME) of type TS> 20080508150000
             <Field OBR_24 (DIAGNOSTIC_SERV_SECT_ID) of type ID> HM
             <Field OBR_32 (PRINCIPAL_RESULT_INTERPRETER) of type NDL> TEST2
        <Group ORU_R01_OBSERVATION>
             <Segment OBX>
                 <Field OBX_1 (SET_ID_OBX) of type SI> 1
                 <Field OBX_2 (VALUE_TYPE) of type ID> 2

infantry · Accepted Answer · 2020-01-31T09:35:03.623

2

by disabling find_groups like this msg = parser.parse_message(hl7, find_groups=False) My messages are now parsed correctly.

Thx to @sqlab and @badger0053 for help, i was able to make a progression.

edited Jan 31 '20 at 09:35

answered Jan 26 '20 at 00:37

infantry

336
1
5
15

How to correctly parse an HL7 message using python HL7Apy?

3 Answers3