Thank you Gary.
This might be the error from my side. I just copied the whole bunch of logs 
into my import directory.
I will make subdir for every year, then I am going to test again.
Maybe going backward from 2020 to 2005 might be a good idea to find out my 
problems.

Version of my WeeWX is 4.4.0, running on a fresh Raspian image, installed 
on Friday.

Let me test tonight and I will report you back.

Peter

gjr80 schrieb am Montag, 1. Februar 2021 um 04:38:54 UTC+1:

> Peter,
>
> I ran wee_import under WeeWX v4.4.0/python3 using the import config file 
> and log files posted and the import processed without error:
>
> gary@buster31:~$ /home/weewx/bin/wee_import 
> --import-config=/var/tmp/wd-import-test.conf --from=2020-12-01T00:00 
> --to=2020-12-31T23:59 --verbose
> Using WeeWX configuration file /home/weewx/weewx.conf
>
> Starting wee_import...
> Weather Display monthly log files in the '/tmp/wdisplay' directory will be 
> imported
> The following options will be used:
>      config=/home/weewx/weewx.conf, 
> import-config=/var/tmp/wd-import-test.conf
>
>      from=2020-12-01T00:00, to=2020-12-31T23:59
>      dry-run=None, calc_missing=True, ignore_invalid_data=True
>      monthly logs are in METRIC units
>      tranche=250, interval=1
>      UV=True, radiation=True ignore extreme temperature and humidity=True
> Using database binding 'wx_binding', which is bound to database 'weewx.sdb'
> Destination table 'archive' unit system is '0x01' (US).
> Missing derived observations will be calculated.
> Observations timestamped after 2020-12-01 00:00:00 AEST (1606744800) and 
> up to and
> including 2020-12-31 23:59:00 AEST (1609423140) will be imported.
>
> Starting import ...
> Records covering multiple periods have been identified for import.
> Obtaining raw import data for period 1 ...
> Raw import data read successfully for period 1.
> Mapping raw import data for period 1 ...
> Mapped 44639 records.
> Raw import data mapped successfully for period 1.
> Saving mapped data to archive for period 1 ...
> 44639 records identified for import.
> Proceeding will save all imported records in the WeeWX archive.
> Are you sure you want to proceed (y/n)? y
> Unique records processed: 44639; Last timestamp: 2020-12-31 23:59:00 AEST 
> (1609423140)
> Mapped data saved to archive successfully for period 1.
> Calculating missing derived observations ...
> Processing record: 44639; Last record: 2021-01-01 00:00:00 AEST 
> (1609423200)
> Recalculating daily summaries...
> Records processed: 44000; time: 2020-12-31 13:20:00 AEST (1609384800)
> Finished recalculating daily summaries
> Finished calculating missing derived observations
> Finished import
> 44639 records were processed and 44639 unique records imported in 249.23 
> seconds.
> Those records with a timestamp already in the archive will not have been
> imported. Confirm successful import in the WeeWX log file.
>
> So I am confident there is nothing wrong with the import config file and 
> log files posted. Granted it was a setup.py install but should not really 
> make a difference to wee_import. 
>
> You said you had 15 years of WD data, when you did the import were there 
> other older log files in /tmp/wdisplay? The reason I ask is that whilst 
> you can limit the time frame over which wee_import imports data, all that 
> does is apply a filter to the final processing/saving of records. 
> wee_import will still go through and read all log files it finds starting 
> at the oldest. So if you had 15 years of WD data in a directory and tried 
> to import the most recent month all 15 years of log files will be read and 
> processed in a limited manner. The initial processing of the raw import 
> data involves assembling the individual date and time components from each 
> row in the log. That appears to be where your import failed. When I wrote 
> the WD component of wee_import I found a number of data inconsistencies 
> in the WD log files I was using so I guess it is possible you may have 
> similar inconsistencies. Have a look through the oldest WD log file set in 
> /tmp/wdisplay and see if you can spot an obvious error/inconsistency in 
> the date and time data. If you can't spot anything could you post the 
> respective log files here. Alternatively, if you let me know what version 
> of WeeWX you are running I can provide you with an instrumented version of 
> wee_import to log details of the offending record/log.
>
> Gary
> On Monday, 1 February 2021 at 00:59:35 UTC+10 peter.su...@gmail.com wrote:
>
>> Hi Gary,
>>
>> thanks for your quick answer.
>> Please check the attached file. All I edited was the path to data and 
>> filenames. I limited it to only one type of file, with my last change.
>>
>> No, problem when it takes time.
>> Thanks for checking out
>>
>> Peter
>> gjr80 schrieb am Sonntag, 31. Januar 2021 um 13:58:21 UTC+1:
>>
>>> Hi,
>>>
>>> Nothing leaps out at me having a quick look at the output and log files 
>>> you posted. I'll endeavour to have a look at this tomorrow (bit late here). 
>>> Could I ask you to post the import config file you are using, 
>>> /etc/weewx/wd-import-test.conf. There should be nothing sensitive in 
>>> the file.
>>>
>>> Gary
>>>
>>> On Sunday, 31 January 2021 at 22:28:32 UTC+10 peter.su...@gmail.com 
>>> wrote:
>>>
>>>> Hello,
>>>> after using my Vantage pro 2 with Weather Display for about 15 years, I 
>>>> think now its time for a change. :)
>>>> I am making the first tests for transferring data, before switch the 
>>>> live system.
>>>>
>>>> My problem, I cannot import the log data from wdisplay.
>>>> I'm getting this error message:
>>>>
>>>> pi@WEEWX-DEMO:/etc/weewx $ wee_import 
>>>> --import-config=wd-import-test.conf --from=2020-12-01T00:00 
>>>> --to=2020-12-31T23:59 --verbose
>>>> Using WeeWX configuration file /etc/weewx/weewx.conf
>>>> Starting wee_import...
>>>> Weather Display monthly log files in the '/tmp/wdisplay' directory will 
>>>> be imported
>>>> The following options will be used:
>>>>      config=/etc/weewx/weewx.conf, 
>>>> import-config=/etc/weewx/wd-import-test.conf
>>>>      from=2020-12-01T00:00, to=2020-12-31T23:59
>>>>      dry-run=None, calc_missing=True, ignore_invalid_data=True
>>>>      monthly logs are in METRIC units
>>>>      tranche=250, interval=1
>>>>      UV=True, radiation=True ignore extreme temperature and 
>>>> humidity=True
>>>> Using database binding 'wx_binding', which is bound to database 
>>>> 'weewx.sdb'
>>>> Destination table 'archive' unit system is '0x01' (US).
>>>> Missing derived observations will be calculated.
>>>> Observations timestamped after 2020-12-01 00:00:00 CET (1606777200) and 
>>>> up to and
>>>> including 2020-12-31 23:59:00 CET (1609455540) will be imported.
>>>> Starting import ...
>>>> Records covering multiple periods have been identified for import.
>>>> Period 1 ...
>>>> Obtaining raw import data for period 1 ...
>>>> Traceback (most recent call last):
>>>>   File "/usr/share/weewx/wee_import", line 903, in <module>
>>>>     main()
>>>>   File "/usr/share/weewx/wee_import", line 833, in main
>>>>     source_obj.run()
>>>>   File "/usr/share/weewx/weeimport/weeimport.py", line 369, in run
>>>>     _raw_data = self.getRawData(period)
>>>>   File "/usr/share/weewx/weeimport/wdimport.py", line 598, in getRawData
>>>>     _ts = "%s" % int(time.mktime(_tt))
>>>> OverflowError: mktime argument out of range
>>>>
>>>>
>>>> Any ideas? File content is looking ok.
>>>> It is happening with all time tests across this 15 years.
>>>> Here a month sample data.
>>>>
>>>> Thanks for any ideas...
>>>> Peter
>>>>
>>>>

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to weewx-user+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/weewx-user/c810aafa-0c39-483c-8e5c-10af1fec8eb4n%40googlegroups.com.

Reply via email to