How to include Date Diff Adiff queries?

[neogeomat]

how to query attic date using the date, diif , adiff parameters as stated here?
https://wiki.openstreetmap.org/wiki/Overpass_API/Overpass_QL#date

[mpadge]

Hey, I didn't realise that those QL features had been implemented - thanks so much for the info. At the moment, this is not possible, but i'll try to implement it soon.

[neogeomat]

I wrote this much, couldn't get the datetime_string inside quotes as required in the query.

add_date <- function (opq,datetime=Sys.time())
{
  datetime_string <- as.character(t,format = "%Y-%m-%dT%H:%M:%SZ")
  opq$prefix <- paste('[date:',datetime_string,']',opq$prefix,sep = "")
  print(opq$prefix)
}
add_date(op)

[mpadge]

You'd be most welcome to submit a PR if you'd like. You can easily insert quotes by escaping: \". Until you or I implement this, a workaround would be to call opq() separately, then modify the string in there by adding "date:..", then pass it to add_osm_feature() %>% osmdata_sc/f/p().

[TomBor]

Would be great to have this functionality!

[mpadge]

Relevant news from r-devel NEWS:

R now provides a syntax for specifying raw character constants similar to the one used in C++: r"(...)" with ... any character sequence, except that it must not contain the closing sequence )". This makes it easier to write strings that contain backslashes or both single and double quotes. For more details see ?Quotes.

[Robinlovelace]

Impressive!

[mpadge]

That commit implements opq() parameters datetime (for dates) and datetime2 (for differences in data between 2 dates). Re-opened to document in vignette.

[mpadge]

Also have to add the datetime values to the output metadata

[jmaspons]

Adiff queries still missing. It requires to parse a different xml response format containing the new and old status of the objects. I would reopen the issue

[mpadge]

@jmaspons but there's no way of representing the information from an "Adiff" query in any of the data structures here. This package remains in essence a spatial package, and that Adiff information is effectively unstructured, and generally not spatial. Or even worse: You can never know in advance whether an Adiff will reflect actual spatial differences, or other, non-spatial differences. One of the biggest challenges of this package is rectifying the non-spatial and spatial aspects; for an Adiff that task becomes far more difficult.

This is not to say it would be impossible, just way more trouble than what it would be worth. I presume - and this is just a presumption - that even Diff queries are relatively little-used within this package, and that when anybody wants time comparisons, then a standard Diff will almost always suffice. The use-cases for ADiff are very specialised, and accordingly of equivalently little general relevance. Processing ADiff results would be too much work for too little general boost in package functionality here, I'd say.

[jmaspons]

@mpadge OSM is a database with more than spatial data. I'm working in implementing queries with out tags; which return simpler structures which can easily fit in a data.frame.

Currently, adiff results works with osmdata_xml. We can leave the user to play with the xml and take advantage of all the nice infrastructure to build queries and get results from overpass. Later, perhaps we can offer adiff results with osmdata_data.frame (see #285) and out tags; queries. And who knows, perhaps in the future somebody can solve the problem with adiff queries and spatial data.