Please keep "us" in the loop... I'm sure I'm going to want to duplicate these efforts for my station... Thanks!
Paul On Tuesday, February 23, 2021 at 4:06:39 AM UTC-6 gjr80 wrote: > Hi, > > You more or less have it. CSV imports require a header row so that > wee_import can identify the source data fields and map them to WeeWX > field as per the field map. If the data you intend to import has no header > row then you can add one, the field names are not important, the important > part is making sure you use the correct field name in the field map. > > There is no --test, I think you mean --dry-run instead. Best thing is to > work through the CSV import step-by-step instructions > <http://weewx.com/docs/utilities.htm#Step-by-step_instructions> in the > Utilities Guide, but I would also revise all of the CSV related sections as > well. > > In terms of checking the import results, that really comes down to you and > how much confirmation you require. In order of increasing complexity you > might consider the following: > > 1. did the import executed without error? > 2. did the reported number of records processed and imported match your > expectations? > 3. were there any errors in the log? > 4. were there any unexplained failures to save a record to database in the > log? (a record may not be saved to database because a record for that > timestamp already exists) > 5. do the NOAA month and year reports for the period imported have > sensible/expected values? > 6. did the number of records saved to database match your expectation? > (the only way to confirm the number of records saved to database is to > count the success 'record added' messga ein the log, unfortunately > wee_import is unable to determine how many records are actually saved to > database, it can only tell how many records it processed and attempted to > save to database) > 7. does each field of the imported data in the database match the > corresponding field/record in the source? (this one is for hardcore users, > there are no tools for doing this it is a manual process) > > I would do 1-5 but 6 and 7, well that depends on how hard core you are :) > > Gary > > > On Tuesday, 23 February 2021 at 16:37:29 UTC+10 [email protected] > wrote: > >> sorry, my touchpad has caused early posting! >> To continue, >> 5. Edit the import file to include the field names. >> 6. run wee-import --test only >> 7. evaluate results, make corrections >> 8. run final import >> 9. restart weewx >> >> any way to verify success? >> >> >> On Monday, February 22, 2021 at 10:27:44 PM UTC-8 David Prellwitz wrote: >> >>> I'm trying to import (wee_import) two weeks of data from my tempest >>> repository held at Weatherflow. The API page example of my data shows no >>> header in the csv file. How should i handle this import? should i edit the >>> file by adding the first reacord and inserting the proper field names? >>> i assume the normal approach is to: >>> 1. create the import config file. >>> 2. extract the WeatherFlow data use their API. >>> 3. stop weewx, >>> 4. copy current db as backup, >>> 5. edit >>> 6. >>> >>> -- 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 [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/f24eae16-ed16-49d1-9156-f85c6ef8b3a2n%40googlegroups.com.
