[Sandy McQuay]

Thanks @Jim Gilbert!

That worked . . . after I “deleted” the t_financial site (renamed the directory and removed the site in Server Administrator).  I couldn’t find any real documentation for deleting a site . . .

The permissions on the folder now are:

drwxrwxr-x. 22 hci staff 4096 Oct 18 15:15 t_financial

Underlying files and directories appear to be correct as well.

Fingers crossed!

 

[Sandy McQuay]

Happy retirement, Rob!  Best wishes!

[Sandy McQuay]

This is great information.  We’re just starting to plan our upgrade from 6.2.2 on RHEL 7.9 to 2022.09.03 on RHEL x.x (not sure yet, whatever is certified).  We also have Sectra PACS.

We’ll keep this in mind as we move forward.

[Sandy McQuay]

Jim – The file is one big long record (i.e. no cr/lf).  I’m attaching a hex dump of the file.  In testing tool, I tried New Line Terminated and EOF with same results and also selected “process all records” on both NL/EOF.

Charlie – I’m intrigued by your suggestion.  I may see what I can come up with.  I have more experience with HL7, so it might be easier to figure out the repeat.

Thanks for the suggestions.  I’ll try to remember to post the solution.

Sandy

Attachments:

You must be logged in to view attached files.

[Sandy McQuay]

Hi Jeff,

My xlate is configured the same way.  Be sure to check “Store original message in metadata for downstream processing” on the route from the source interface, then on the database-outbound protocol configuration, check “Use cached pre-xlate message as whole message.”

I’ll attach some screen shots.

Attachments:

You must be logged in to view attached files.

[Sandy McQuay]

I figured out my problem.  I was so focused on the protocol set up, that I failed to look at the other tabs, including the fact that Outbound Only was checked on Inbound tab.

I made a copy of the aforementioned outbound-database for the database-inbound thread.  Guess I only looked at the protocol and routing information.

 

[Sandy McQuay]

I tried both selecting and not selecting the Table Schema in the config.  Still doesn’t read the db.

Thanks!

[Sandy McQuay]

😳  😳  😳

So, for the uuencode::uudecode command, it comes down to transposing the CR/LF mapping:

Code:

set DecryptWF [string map {\X0A\\X0D\ “n”} $DecryptWF]

instead of

Code:

set DecryptWF [string map {\X0D\\X0A\ “n”} $DecryptWF]

I guess when I combined the separate lines into one, I used the X0A line and tacked on the X0D.

This creates a readable TIF file:

Code:

package require uuencode
set WaveForm [lindex [split [hl7_get_field $data ZPD 3] ^] 2]

# Reinsert HL7 delimiter characters according to MUSE specs
set DecryptWF [string map {\F\ |} $WaveForm]
set DecryptWF [string map {\S\ ^} $DecryptWF]
set DecryptWF [string map {\T\ &} $DecryptWF]
set DecryptWF [string map {\R\ ~} $DecryptWF]
set DecryptWF [string map {\E\ \} $DecryptWF]
set DecryptWF [string map {\X0D\\X0A\ “n”} $DecryptWF]

set decodeWF [lindex [uuencode::uudecode $DecryptWF] 0 2]
msgset $mh $decodeWF
return “{CONTINUE $mh}”

However, even with mapping the CR/LF properly, I still can’t exec uudecode from the engine.  I’ll continue working with Infor on this.

I will probably move forward with using the uuencode package for this interface.

Many thanks to all who offered suggestions, especially David Barr and Charlie Bursell.  These guys went above and beyond!  This group is an awesome resource!

[Sandy McQuay]

Holy crap Charlie!  I just took your “sandy.tcl” proc as is and ran it in the engine without making any changes.  It created a readable TIF!

Now I have to pick it apart and see what I’m doing wrong in my code.

[Sandy McQuay]

Also, binary is not an option in 6.2 TPS tester.  I tried them all . . . no dice!

[Sandy McQuay]

It’s currently working on 5.7.  I’m trying to get it to work in 6.2

[Sandy McQuay]

Charlie,

I appreciate your tenacity in trying to find a resolution for this.  Unfortunately, the TCL uuencode::uudecode command returns nothing for me:

Code:

Please Wait ……
Command Issued: hcitpstest -r run -x ASCII -f nl -c sms_ib_data -e “hcitpstestshowbydisp ” /cloverleaf/cis6.2/integrator/clinical/data/MUSE_EKG_20180413141455.NL “tps_decodeMUSE”
Command output:

Wave form after uuencode::uudecode command:

Message should be here —><—

CONTINUE: ''

Granted, I  didn’t take your entire script, but I did take your suggestion on the CR/LF substitution and changed the decoding command to:

Code:

set msg [uuencode::uudecode $data]

Sorry my code is so messy.  It was originally written in the early 2000’s and has worked with subsequent upgrades, so it wasn’t a priority to streamline.  We don’t seem to have any performance issues in 5.7.  Also, I don’t have a formal programming education, so I typically just find what works and go with it.  🙂

[Sandy McQuay]

All,

I’m still not able to get any of these suggestions to work within the engine.  Infor support offered the exact same suggestion that Tao posted (is Tao with Infor?), but still no luck.

I’m beginning to think there may be something weird in my environment if everyone else is able to get these suggestions to work.

I’ve asked Infor for a WebEx so they can see it in my environment.

I’ll post once we find a resolution . . . or let you know I had to take the uudecode outside of the engine.

Thanks for all the work and research, especially David and Charlie!

[Sandy McQuay]

Unfortunately, these suggestions still don’t work for me.

Here’s my shell script (bash not csh):

Code:

#!/usr/bin/bash -f

uudecode -o /dev/stdout $1

And my tcl script:

Code:

       run {
           # ‘run’ mode always has a MSGID; fetch and process it
           keylget args MSGID mh

           set path “/usr/integrator/prod/muse/tmp”
           set zpd3 [lindex [split [lindex [split [lsearch -inline -regexp
              [split [msgget $mh] r] ^ZPD] |] 3] ^] 2]
           echo “ZPD-3:n$zpd3”

           # Substitutions
           set data [string map {\F\ |} $zpd3]
           set data [string map {\S\ ^} $data]
           set data [string map {\T\ &} $data]
           set data [string map {\R\ ~} $data]
           regsub -all {\X0A\} $data “n” data
           regsub -all {\X0D\} $data “r” data
           set data [string map {\E\ \} $data]

           set fh [open $path/EKG_DECRYPTED.tmp w]
           puts -nonewline $fh $data
           close $fh

           echo current encoding is [encoding system]   Tried with and without these lines
           encoding system identity                     <—/

           if {[catch {exec decode.sh $path/EKG_DECRYPTED.tmp} msg]} {
               echo nERROR executing uudecode $msg
               echo To Error DBn
               #msgset $mh USERDATA "Error executing uudecode: $msg"  <—— This line gives me an error "expected integer but got "Error executing uudecode: uudecode: /usr/integrato"
               return "{ERROR $mh}"
           }

           msgset $mh $msg
           return "{CONTINUE $mh}"
           #lappend dispList "CONTINUE $mh"      
       }

And here’s the error I get in the route tester and engine (still works in tps tester):

Code:

ERROR executing uudecode uudecode: /usr/integrator/prod/muse/tmp/EKG_DECRYPTED.tmp: No `begin’ line

Here’s the version of uudecode I have (in case that makes a difference):

Code:

[hci@cloverleaf-prd scripts]$ uudecode -v
uudecode (GNU sharutils) 4.13.3

I appreciate all the help and suggestions.  I will open a ticket with support and report this as a bug.  Until it’s resolved, I’ll take the uudecode processing outside the engine.

[Sandy McQuay]

Oops!  Let’s try again.

[Author]

Replies