I have a problem with junk in one QSO record when using the date filter on importing ADIF.
I use a small live file for operating(on my laptop), then import this into a fully merged log on my desktop machine. On the laptop, I changed the export spec to use 'Oldest entries on top', and now when I import and filter out everything but the last couple of days, the first imported QSO gets the leftovers from the skipped entries in (at least) IOTA, State, QSL VIA, etc.
Looks like these fields aren't cleared when the other QSOs are skipped.
Oops, forgot to mention -
CQRLOG Ver. 2.5.2(001)
openSUSE 15.3
HI!
Can you reproduce this with couple of small adif files imported to new empty log?
In that case please make a description how to operate and send those small adif files to me so I can try to make it happen also here.
Perhaps there is a bug laying somewhere.
When I tried to import two small adif files made with "oldest on top" I did not manage to find any errors.
--
Saku
OH1KH
Attached is my current Live.ADI (masquerading as Live.txt). To reproduce the problem:
1) Rename Live.txt to Live.ADI
2) Start CQRLOG, create and open a new log file.
3) Ctrl+O to get QSO list, ctrl+I to import
4) Navigate to Live.ADI, click Open
5) In the selection box, click on Filter, then in the calendars, Choose 2021-11-01 to 2021-11-02, then Import and Close.
6) In the QSO list, select the first (oldest) one - YV5JLO.
7) On my system, it shows an IOTA entry of AF-004(obviously wrong), and QSL VIA EB7DX(also wrong). Both of these data appear elsewhere in the import file.
File:
Yes !
Found clear bug.
If qso was rejected by date the typevariable collecting data from adif record was not cleared before new adif qso record read.
I have made a pull request to official source.
If you can compile by yourself you can test the fix with source from https://github.com/OH1KH/cqrlog/tree/adif_import_err
When pull request is accepted you find it from official source from https://github.com/ok2cqr/cqrlog
--
Saku
OH1KH
HI !
Very good explanation! Thanks!
I was able to reproduce the problem. I will look at the source and do some debugs to see where it goes wrong.
Thanks!
--
Saku
OH1KH