An additional bit of information from /tmp/ofx.log: AQB5: Sending: ------------------------------------- OFXHEADER:100 DATA:OFXSGML VERSION:102 SECURITY:NONE ENCODING:USASCII CHARSET:1252 COMPRESSION:NONE OLDFILEUID:NONE NEWFILEUID:20200318081757.000
<OFX>... git: Sending: ------------------------------------- <?xml version="1.0" encoding="utf-8" standalone="no"?> <?OFX OFXHEADER="200" VERSION="220" SECURITY="NONE" OLDFILEUID="NONE"?> <OFX>... > On Mar 17, 2020, at 10:40 AM, John Ralls <jra...@ceridwen.us> wrote: > > > >> On Mar 17, 2020, at 9:45 AM, John Ralls <jra...@ceridwen.us> wrote: >> >> >> >>> On Mar 15, 2020, at 1:21 PM, John Ralls <jra...@ceridwen.us> wrote: >>> >>>> >>>> On Mar 15, 2020, at 12:47 PM, Martin Preuss <mar...@aqbanking.de> wrote: >>>> >>>> Hi, >>>> >>>> Am 15.03.20 um 20:17 schrieb John Ralls: >>>>> I just made a MacOS build with the latest AQBanking so I could test it >>>>> with my bank (USAA). No go, and a bit of troubleshooting reveals that >>>>> it's because AQBanking now uses the XML-based OFX V2. AFAIK that's not >>>>> supported by any US bank. >>>> [...] >>>> >>>> OFX2 is the only version I could get access to a test server for, so >>>> thats what I implemented so far. >>>> >>>> However, since the OFX spec state that banks should also accept XML like >>>> OFX requests (i.e. OFX files with closing XML elements) the code might >>>> be adjustable for this... >>>> >>>> I will try to add a mode in which the plugin sends the headers >>>> previously used for OFX1 and still send requests with closing XML >>>> elements, at least according to the specs I read that should work... >>>> >>>> Anyway, as I wrote before I can't test it myself. There is only person - >>>> so far - I heard of from the US who tested the latest code with a real >>>> server and I really appreciate his help, but he seems to be using the >>>> prebuilt Windows version which makes the debug process... well... >>>> difficult. >>>> >>>> It would be extremely helpfull to have someone who could compile and >>>> test a version from GIT, but given the different timezones the possible >>>> OFX users and I live in that might be time consuming for both sides... >>> >>> Martin, >>> >>> As I said here and in my bug report, AQB 5's OFX backend works well and has >>> done for many years. If you just update it for the new AQB6 and Gwen5 API >>> it has a good chance of working. >>> >>> I can build from git and test. So far there's no need for debugging, simply >>> comparing the OFX from 5.8 and 6.1 showed me the problem: OFX DirectConnect >>> servers using V1 can't parse the XML you send in the request and 400, see >>> the transcripts in the bug report. >>> >>> The current OFX spec doesn't matter if the bank is using software written >>> for OFX V1 as nearly all in the US do. >> >> Martin, >> >> Frank noticed yesterday that you'd pushed a large OFX V1 change to AQBanking >> master, so I pulled it and tested it. Unfortunately its a total fail because >> the V1 code formats the OFX in XML. >> >> V1 servers cannot parse XML and return a 400 error. > > Martin, > > Sorry, that's not quite right. It's not XML, but it does still 400. The > headers match those sent by AQB5, but the AQB5 request block is > > <OFX><SIGNONMSGSRQV1><SONRQ><DTCLIENT>20200315112517<USERID>XXXXXXXXXX > <USERPASS>XXXXXXXXX > <LANGUAGE>ENG<FI><ORG>USAA<FID>24591</FI><APPID>QWIN<APPVER>2300</SONRQ></SIGNONMSGSRQV1><CREDITCARDMSGSRQV1><CCSTMTTRNRQ><TRNUID>20200315112517<CLTCOOKIE>1<CCSTMTRQ><CCACCTFROM><ACCTID>XXXXXXXXXXXXXXXX</CCACCTFROM><INCTRAN><DTSTART>20200314000000<DTEND>20200315112517<INCLUDE>Y</INCTRAN></CCSTMTRQ></CCSTMTTRNRQ></CREDITCARDMSGSRQV1></OFX> > > unrolled and indented: > <OFX> > <SIGNONMSGSRQV1> > <SONRQ> > <DTCLIENT>20200315112517 > <USERID>XXXXXXXXXX > <USERPASS>XXXXXXXXX > <LANGUAGE>ENG > <FI> > <ORG>USAA > <FID>24591 > </FI> > <APPID>QWIN > <APPVER>2300 > </SONRQ> > </SIGNONMSGSRQV1> > <CREDITCARDMSGSRQV1> > <CCSTMTTRNRQ> > <TRNUID>20200315112517 > <CLTCOOKIE>1 > <CCSTMTRQ> > <CCACCTFROM> > <ACCTID>XXXXXXXXXXXXXXXX > </CCACCTFROM> > <INCTRAN> > <DTSTART>20200314000000 > <DTEND>20200315112517 > <INCLUDE>Y > </INCTRAN> > </CCSTMTRQ> > </CCSTMTTRNRQ> > </CREDITCARDMSGSRQV1> > </OFX> > > > and git V1 sends > <OFX> > <SIGNONMSGSRQV1> > <SONRQ> > <DTCLIENT>20200317095100</DTCLIENT> > <USERID>XXXXXXXX</USERID> > <USERPASS>XXXXXXXX</USERPASS> > <LANG>ENG</LANG> > <FI> > <ORG>USAA</ORG> > <FID>24591</FID> > </FI> > <APPID>QWIN</APPID> > <APPVER>2300</APPVER> > </SONRQ> > </SIGNONMSGSRQV1> > <CREDITCARDMSGSRQV1> > <CCSTMTTRNRQ> > <TRNUID>20200317095103</TRNUID> > <CCSTMTRQ> > <CCACCTFROM> > <BANKID>24591</BANKID> > <ACCTID>XXXXXXXXXXXX</ACCTID> > </CCACCTFROM> > <INCTRAN> > <DTSTART>20200314000000</DTSTART> > <DTEND>20200317000000</DTEND> > <INCLUDE>Y</INCLUDE> > </INCTRAN> > </CCSTMTRQ> > </CCSTMTTRNRQ> > </CREDITCARDMSGSRQV1> > </OFX> > > Aside from the extra whitespace, which shouldn't matter, the git code has > closing tags on every element while the AQB5 code has closing tags only on > grouping elements. > > USERID is specified in the OFX 1.6 DTD as > <!ELEMENT USERID - o %IDTYPE> > and according to > https://en.wikipedia.org/wiki/Standard_Generalized_Markup_Language#OMITTAG > the "- o" in the middle means that it should have an opening tag and no > closing tag. > > Regards, > John Ralls > > > > > > _______________________________________________ > gnucash-user mailing list > gnucash-user@gnucash.org > To update your subscription preferences or to unsubscribe: > https://lists.gnucash.org/mailman/listinfo/gnucash-user > If you are using Nabble or Gmane, please see > https://wiki.gnucash.org/wiki/Mailing_Lists for more information. > ----- > Please remember to CC this list on all your replies. > You can do this by using Reply-To-List or Reply-All. _______________________________________________ gnucash-user mailing list gnucash-user@gnucash.org To update your subscription preferences or to unsubscribe: https://lists.gnucash.org/mailman/listinfo/gnucash-user If you are using Nabble or Gmane, please see https://wiki.gnucash.org/wiki/Mailing_Lists for more information. ----- Please remember to CC this list on all your replies. You can do this by using Reply-To-List or Reply-All.
