[Gnumed-devel] important bug: invoice PDFs can get assigned to the wrong patient, was: [Gnumed-bugs] <bug>: on saving printing document in overview

[Karsten Hilbert]

The first response to this bug is the attached SQL which
finds wrongly assigned bills. If bill_patient and
doc_patient differ from each other that bill should be
examined closely.

Deleting the faulty PDF from within the doc_patient document
tree and regenerating the PDF from within the bill_patient
will fix the situation for that bill.

Karsten

On Wed, Sep 05, 2012 at 04:04:46PM +0200, Karsten Hilbert wrote:
> Date: Wed, 5 Sep 2012 16:04:46 +0200
> From: Karsten Hilbert <address@hidden>
> To: address@hidden
> Cc: Stephan Reus <address@hidden>
> Subject: Re: [Gnumed-bugs] <bug>: on saving printing document in overview
>  scr
> User-Agent: Mutt/1.5.21 (2010-09-15)
> 
> Hello Marc,
> 
> thanks for your report. I have seen a very similar report
> by another user (which I have CCed) before.
> 
> Can we try to retrace the exact steps ?
> 
> > user comment  : on saving printing document in overview screen
> 
> Do you mean that you were in the overview screen, then you
> printed a document, GNUmed asked you which episode to save
> it under, and then the bug came up ?
> 
> Let's see what the log gives us. First:
> 
> > 2012-09-05 08:33:04  ERROR     gm.datetime 
> > (/usr/share/gnumed/Gnumed/pycommon/gmDateTime.py::pydt_strftime() #262): 
> > Python cannot strftime() this <datetime>
> > Traceback (most recent call last):
> >   File "/usr/share/gnumed/Gnumed/pycommon/gmDateTime.py", line 260, in 
> > pydt_strftime
> >     return dt.strftime(format).decode(encoding, 'replace')
> > ValueError: year=193 is before 1900; the datetime strftime() methods 
> > require year >= 1900
> > 2012-09-05 08:33:26  CRITICAL  gm.db 
> > (/usr/share/gnumed/Gnumed/pycommon/gmBusinessDBObject.py::refetch_payload() 
> > #504): [cIdentity:795]: cannot reload, payload changed
> 
> GNUmed seems to think that something is odd (likely the date
> of birth) with Patient 795. That was most probably a typo
> some time ago when entering the year.
> 
> Please try to correct this and see if you can evoke the bug
> again - it may have been followup error (which would still
> be a bug but that might give us hints).
> 
> Next, given this:
> 
> > 2012-09-05 08:33:45  ERROR     gm.docs 
> > (/usr/share/gnumed/Gnumed/business/gmDocuments.py::update_data_from_file() 
> > #308): [/tmp/gnumed/gm-wefjPq/gm-L-Template-QRjPiF-instance.tex] is not a 
> > readable file
> > 2012-09-05 08:33:45  ERROR     gm.docs 
> > (/usr/share/gnumed/Gnumed/business/gmDocuments.py::add_part() #513): cannot 
> > import binary data from 
> > [/tmp/gnumed/gm-wefjPq/gm-L-Template-QRjPiF-instance.tex] into document part
> > 2012-09-05 08:33:45  ERROR     gm.docs 
> > (/usr/share/gnumed/Gnumed/business/gmDocuments.py::add_parts_from_files() 
> > #533): cannot instantiate document part object
> 
> and this:
> 
> > client version: 1.2.1
> 
> and this:
> 
>       1.2.2 CHANGELOG
> 
>       FIX: failure to save .tex bill files
> 
> and this:
> 
>       gnumed-client-de:
>         Installiert:           1.2.3-1
>         Installationskandidat: 1.1.17-1
>         Paket-Pinning: 1.1.17-1
>         Versionstabelle:
>        *** 1.2.3-1 601
>                10 ftp://ftp.de.debian.org/debian/ experimental/main i386 
> Packages
>               100 /var/lib/dpkg/status
>            1.1.17-1 601
>               990 http://ftp.de.debian.org/debian/ wheezy/main i386 Packages
>               990 ftp://ftp.de.debian.org/debian/ testing/main i386 Packages
>                50 ftp://ftp.de.debian.org/debian/ unstable/main i386 Packages
>            0.7.10-3 601
>               500 http://ftp.de.debian.org/debian/ squeeze/main i386 Packages
> 
> I would like to ask you to upgrade to 1.2.3
> 
>       apt-get install -t experimental gnumed-client-de
> 
> and try again whether you can reproduce the bug.
> 
> I don't, however, think we've gotten to the root of the
> problem just yet. Next we find this:
> 
> > 2012-09-05 11:10:13  DEBUG     gm.gui 
> > (/usr/share/gnumed/Gnumed/wxpython/gmExceptionHandlingWidgets.py::handle_uncaught_exception_wx()
> >  #186): unhandled exception caught:
> > Traceback (most recent call last):
> >   File "/usr/lib/python2.7/dist-packages/wx-2.8-gtk2-unicode/wx/_core.py", 
> > line 14665, in <lambda>
> >     lambda event: event.callable(*event.args, **event.kw) )
> >   File "/usr/share/gnumed/Gnumed/wxpython/gmEMRBrowser.py", line 266, in 
> > __update_text_for_selected_node
> >     with_co_encountlet_hints = True
> >   File "/usr/share/gnumed/Gnumed/business/gmEMRStructItems.py", line 1793, 
> > in format
> >     raise ValueError(msg)
> > ValueError: <patient>.ID = 216 but encounter 2839 belongs to patient 744
> 
> Which tells us that GNUmed seems greatly confused at being
> told to reconcile two different patients. So in the vain of
> better being safe than sorry it throws up its hands in
> despair. This now begs the question why.
> 
>       *************************************************
>       Did you - by any chance - attempt to print a bill
>       NOT belonging to the active patient ?
>       *************************************************
> 
> Sheeesh, trying that it turns out GNUmed will do major BS
> and create/link the PDF for/to the *ACTIVE* patient rather
> than the one that's linked to the bill %-//
> 
> This still does not quite explain the exception you saw (at
> least I cannot evoke one) but this needs fixing immediately.
> 
> I'm looking into the issue right now including developing an
> SQL query pulling up the faulty invoices (which is gladly
> possible by cross-checking
> 
> bill->fk_receiver_identity/bill_item->patient/document->patient
> 
> You assured us that you were printing while the (new,
> patient) overview was active:
> 
> > user comment  : on saving printing document in overview screen
> 
> However, the above code belongs to the EMR tree browser:
> 
> >   File "/usr/share/gnumed/Gnumed/wxpython/gmEMRBrowser.py", line 266, in 
> > __update_text_for_selected_node
> 
> which should, however, only be run when the EMR tree plugin
> is active.
> 
> Can you therefore describe the exact steps needed to
> reproduce including what state the GUI was in at each step ?
> 
> Karsten
> -- 
> GPG key ID E4071346 @ gpg-keyserver.de
> E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346
> 

-- 
GPG key ID E4071346 @ gpg-keyserver.de
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346

screenshot_001.png
Description: PNG image