Error: reached end of input before finding ISA segment identifier about stupidedi HOT 6 CLOSED

Hi Josh,

I'm assuming you have something like parser, result = parser.read(...), where parser is a StateMachine instance.

The second return value from this call will always be Result.failure(...) because reaching the end of the file is an error to the tokenizer -- it's unknown to the tokenizer if the caller expected this or not. However, it's labeled a non-fatal error: you can test for more serious errors with result.fatal?, which might happen if the file has invalid delimiters, etc.

The first return value, parser is the new StateMachine which points at the last segment of the parse tree constructed by reading the input. You might check parser.empty? to make sure that the contents of the file were read into the parse tree -- the tokenizer skips past everything until it finds a valid ISA segment.

from stupidedi.

Here's what I've got...

input = File.open("public/ups/UPSS_06_13_2013_1020110058.DAT", encoding: "ISO-8859-1")
=> #<File:public/ups/UPSS_06_13_2013_1020110058.DAT>

parser, result = parser.read(Stupidedi::Reader.build(input))
=> [#<Stupidedi::Builder::StateMachine:0x3fdf82ffb0a8 ...>, Either.failure("reached end of input before finding ISA segment identifier")]

parser.empty?
=> false

parser.segment
=> Either.success(#<Stupidedi::Zipper::EditedCursor:0x007fbf05ff6358 @node=IEA, @path=root/4/1054, @parent=#<Stupidedi::Zipper::EditedCursor:0x007fbf28739b98 @node=InvalidEnvelopeVal(), @path=root/4, @parent=#<Stupidedi::Zipper::RootCursor:0x007fbf3714b5b0 @node=TransmissionVal(), @path=root>>>)

parser.segment.map(&:node)
=> Either.success(IEA)

parser.segment.map(&:parent).map(&:node)
=> Either.success(InvalidEnvelopeVal())

parser.element(1)
=> Either.failure("invalid segment")

parser.first.map(&:first?)
=> Either.success(true)

parser.next
=> Either.failure("cannot move to next after last segment")

Something just feels...off. I've run through the documentation and tried various methods to access the data and just can't seem to get it.

Maybe I'm doing something wrong?

from stupidedi.

Looks like you just need to rewind the parser to the start of the parse tree -- the #read method leaves it positioned on the last segment that was parsed (IEA).

Take a look at bin/edi-pp for an example, but parser = parser.first will give you Either.success(#<Stupidedi::Builder::StateMachine:...>), where the parser is positioned at the start. You'll need to use #map, #tap, and #flatmap to manipulate the StateMachine inside the Either wrapper.

Secondly, your document is QM214, which I recently merged into the contrib namespace. You'll need to use config = Stupidedi::Config.contrib to parse this file -- otherwise you'll see mostly InvalidSegmentVal values in the parse tree. Be sure to pull the latest commits, I fixed a few issues just now.

Since these were contributed by other authors and I don't have the official grammar for this transaction, you probably need to make changes to the grammar. For instance, the grammar in Stupidedi for QM214 doesn't allow MAN, N1, or CD3 segments but your document has these. If you need the information contained in these segments, you'll need to find some documentation like this or this from your EDI trading partner. If you don't need those segments, you can probably leave things as-is -- the parser will add InvalidSegmentVal nodes in their place, which generally are ignored when moving around the parse tree.

-- Kyle

from stupidedi.

Gotcha! Okay, getting there. One more question...how do I select the second of a duplicate element?

For example, there's this segment:

ST|214|000056390~
B10|0|0|UPSN~
L11|QVD|TN|O4~
L11|1Z6VX489YN03534815|2I~
N1|SH|NA|25|6VX489~
N1|BT||25|6VX489|01~
LX|1~
AT7|X8|AJ|||20130613|003551|LT~
AT7|||AB|BG|20130613~
L11|2U-UPS WILL ATTEMPT DELIVERY|REC|KAGE WILL NOT BE TRANSFERRED T~
MAN|CP|E1|PITTSBURGH DOWNTOWN|CP|PA|US~
CD3|||R4~
N1|UP|NA~
SE|14|000056390~

Then I do this:

a.flatmap{|m| m.find(:GS) }.flatmap{|m| m.find(:ST) }.tap do |m|
   el(m.find(:L11), 3){|e| puts "Transaction Reference Number: #{e}" }
   el(m.find(:L11)){|e| puts "Tracking Number: #{e}" }
end

That first L11 works correctly (and pulls O4). But I want to get the next L11 to retrieve 1Z6VX489YN03534815

(Really appreciate your help so far!)

from stupidedi.

It seems like the second element of L11 specifies which "type" of L11 it is, so I would probably do something like this, since TN or 2I probably mean something specific and you'll want to act on them differently:

el(m.find(:L11, nil, "TN"), 3){|e| ... }
el(m.find(:L11, nil, "2I"), 3){|e| ... }

You'll need to check specs for QM 214 to be certain, but usually that's how these documents work. If you can have many L11*...*TN segments or many L11*..*2I segments in a row, or if the second element is not a qualifier, you should do something like this:

# Loop through each L11 sequentially
m.iterate(:L11) do |l11|
  el(l11, 3){|el| ... }
end

from stupidedi.

I've also added a method #sequence which isn't documented yet. It's just shorthand for chaining multiple #find calls. Here's an example:

a.flatmap{|m| m.sequence(:GS, :ST) }.tap do |m|
   ...
end

from stupidedi.