Andreas Beger
Andreas Beger
It's a lot faster for analytical queries than SQLite, although updates/new inserts might take longer.
Updating the DB takes a lot of time right now. SQLite doesn't have bulk inserts, but it seems that using a transaction for bulk inserts might be faster? - Use...
There are some events that have wrong cameo codes, '13y': ```r query_icews("select * from events where cameo_code=='13y' limit 5; ") event_id event_date source_name source_sectors 1 25333471 20170104 Head of Government...
There are several issues in the "Events.2017.20201006.tab" file. I'm gonna use this issue to document them in one place. - [ ] 1. the dates are in a non-standard ""...
Some of the text field values include outer double-quotes in their value, e.g.: ``` > query_icews("select * from events where event_id = 25326166;") event_id event_date source_name source_sectors 1 25326166 20170101...
The 2017 events file is missing a "CAMEO column", but it does have "Event Text". 1. Fix for future reads: there already is a file-specific non-standard treatment for dates in...
Prepare for release: * [x] Check that description is informative * [x] Check licensing of included files * [x] `devtools::build_readme()` * [x] `usethis::use_cran_comments()` * [x] `devtools::check(remote = TRUE, manual =...
The **icews** package will not properly work with the CRAN version of **dataverse** (0.2.0 from 2017 as I'm writing this). This was brought up here in issues #51 and #58,...
1. Get the dataverse URLs: `get_doi()` does this in principle, but it's not the most intuitive name. Maybe something like `dataverse_url(repo)`? 2. Listing what's in the repos, basically some wrapper...
On Windows, the test database set up for testthat sometimes does not get deleted when calling `unlink()`. I have no idea why. This is specifically when using `setup_mock_environment()`, which creates...