[Mail list Bots EDI software] X12 translation for a newbie..HELP!!

[Jack]

Hi, folks! I am a babe-in-the-woods newbie to Bots and Python, and I'm
having some real difficulty in processing a X12 to CSV translation for
a 271 transaction. I get the message:

InMessageParseError: line:1 pos:1; record:"ISA" not in grammar; looked
in grammar until mandatory record: "[['ST']]"

when processing the file, which has one ISA and IEA segment at each
end. I have the X12.py file present in the grammar folder, and I'm
using the 271004010.py grammer in the translation, but Bots is not
recognizing the ISA segment. What am I doing wrong? Do I need to
import the x12.py code into the 271 message grammar, or is there
something else I need to do for Bots to recognize the segment? Any
help would be greatly appreciated!

[henk-jan ebbers]

Hi Jack,


what probably is the problem is that you SHOULD have messagetype x12 in
the route.
probably you used something like 271004010 for messagetyep.

(please take a good look at the plugins. these spell it all out. just
follow the examples.)


kind regards,
henk-jan

[Jack]

Thanks for your help, Henk-Jan. I resolved the ISA problem, but I have
a further question(maybe I'm just being dense about this, but..). I
have an x12 message to read, that has multiple nested segments in a
hierarchy, and a parent segment can have more than 1 entry of a child
segment under it(i.e. more than one 'NM1' segment under a 'HL'
segment). I need to send specific fields to a CSV file from this data.
In order to find a field in a group of records two levels down in a
segment, do I need to perform multiple 'getloop' statements to go
through each level of records, or can I go directly to the field with
one 'get' statement? I'm just having a lot of difficulty in
understanding how the 'get' and 'getloop' statements should function
against a hierarchal structure like the x12/271 message. Any help
would be greatly appreciated!

[henk-jan ebbers]

the arguments used in get and getloop are called mpath.
a mpath identifies the place in the message; use segments and code
values to identify
get({'BOTSID': 'XX', XX01:'NO'}{'BOTSID': 'YY','YY01':None}
(get content of YY01 field in YY segment under a XX segment where XX01
has value 'NO'

(so: get() can fetch an item as 'deep' as you want, using segment ID's
but also other codes to filter)

I use getloop() only for things like lines in an invoice: I want to
process all lines (not just fetch one line).

in getloop() you can also be used like:
getloop({'BOTSID': 'ZZ', ZZ01:'NO'}{'BOTSID': 'YY','YY01':'15'}
returns YY segments (where YY01 has value 15 nested under ZZ with ZZ01
is 'NO'


hope this helps,
henk-jan

[henk-jan ebbers]

Jack,

thinking about your question some more info:

the 271 message with HL structures in health care is quite complicated.
(yes, health care.....)

You indicate you are starting with EDI: weel, you are starting with a
one of the more complicated messages.
IMO: this can be confusing, as you get a lot of new information. The
structure of this message takes some time to 'settle down in your mind'.
At least that is my experience.
so be aware that you have a tough start (most people start with a 'easy'
message like orders/850).

Time ago I took a look at the HL structure (I know this type of
structure from some edifact messages).
Bots can handle this very well.

Most important issue here is to set up a good inhouse-format. And a good
format would be: as convenient for your application as can be: try to
keep the CSV import as simple as possible.
When setting up the mapping: think of it as filling the CSV file,
fetching from the 271
(the opposite would be: walk over the 271, try to generate CSV records;
IMHO a confusing way).

[Jack]

Thank you for your insight and further info, Henk-Jan. I ought to
write up a manual page for get(), getloop(), put(), putloop() based on
the info you've given me already! Your info does help A LOT with
understanding how Bots works in accessing information.

My problem was in thinking about the tree as something where I had to
read at a specific level to get to the next level of data, instead of
providing a path to the specific piece of info I need in one
statement. I agree with you that the 271 document is a hard place to
start, but it's the document I have to read for the project on which
I'm working. This info should simplify what I'm trying to do
immensely. Again, thank you for your efforts and consideration!

Jack