From randre at gmail.com  Mon Feb 24 21:02:53 2020
From: randre at gmail.com (=?UTF-8?B?Um9nZXIgQW5kcsOp?=)
Date: Mon, 24 Feb 2020 12:02:53 -0800
Subject: [Rgeos-devel] How best to work with row.names output from
	gUnaryUnion?
Message-ID: <CAOL+ROF_Ta9OdDOzBZadOGVv82M5DqfV=jqLoY6PfrGZuBDn3A@mail.gmail.com>

Hello,

I'm slowly learning the implementation details of rgeos after being a
longtime user of GDAL and PostGIS.  I was doing some work that necessitated
doing some spatial unions which are  aggregated by a feature attribute and
was surprised to see that the original dissolve field (or `group by` field)
wasn't returned in a data frame, but rather as row IDs.  This was
inconvenient for me, so I wrote a short helper function to give me back my
desired output.

dissolvePolygons <- function (
    input_SPDF, # SPDF object to dissolve
    diss_field # name of column to dissolve by, as string
    ) {
    diss_command <- sprintf("rgeos::gUnaryUnion(input_SPDF,
input_SPDF$%s)", diss_field)
    data_command <- sprintf("data.frame(%s = row.names(out_polys))",
diss_field)
    out_polys <- eval(parse(text = diss_command))
    data <- eval(parse(text = data_command))
    output_SPDF <- SpatialPolygonsDataFrame(out_polys,data)

    return(output_SPDF)
    }

# Usage
zones <- dissolvePolygons(sample_tracts, "zone")

I'm curious what a preferred method is to work with the original row
output?  I was finding it problematic to tie it back to other data and was
hoping to find some examples of how to do so.

Thanks,

Roger
--
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/rgeos-devel/attachments/20200224/644cf961/attachment.html>

From Roger.Bivand at nhh.no  Mon Feb 24 21:31:38 2020
From: Roger.Bivand at nhh.no (Roger Bivand)
Date: Mon, 24 Feb 2020 20:31:38 +0000
Subject: [Rgeos-devel] How best to work with row.names output
	from	gUnaryUnion?
In-Reply-To: <CAOL+ROF_Ta9OdDOzBZadOGVv82M5DqfV=jqLoY6PfrGZuBDn3A@mail.gmail.com>
References: <CAOL+ROF_Ta9OdDOzBZadOGVv82M5DqfV=jqLoY6PfrGZuBDn3A@mail.gmail.com>
Message-ID: <cfe3c10165804409a4b231a588a32c64@nhh.no>

The list is not used. Either email  the package maintainer, post on R-sig-geo, or - much better - use the sf package for all new projects involving GEOS in R - sf does do what you appear to want. 

--
Roger Bivand
Norwegian School of Economics
Helleveien 30, 5045 Bergen, Norway
Roger.Bivand at nhh.no


________________________________________
Fra: Rgeos-devel <rgeos-devel-bounces at lists.r-forge.r-project.org> på vegne av Roger André <randre at gmail.com>
Sendt: mandag 24. februar 2020 21.02
Til: rgeos-devel at lists.r-forge.r-project.org
Emne: [Rgeos-devel] How best to work with row.names output from gUnaryUnion?

Hello,

I'm slowly learning the implementation details of rgeos after being a longtime user of GDAL and PostGIS.  I was doing some work that necessitated doing some spatial unions which are  aggregated by a feature attribute and was surprised to see that the original dissolve field (or `group by` field) wasn't returned in a data frame, but rather as row IDs.  This was inconvenient for me, so I wrote a short helper function to give me back my desired output.

dissolvePolygons <- function (
    input_SPDF, # SPDF object to dissolve
    diss_field # name of column to dissolve by, as string
    ) {
    diss_command <- sprintf("rgeos::gUnaryUnion(input_SPDF, input_SPDF$%s)", diss_field)
    data_command <- sprintf("data.frame(%s = row.names(out_polys))", diss_field)
    out_polys <- eval(parse(text = diss_command))
    data <- eval(parse(text = data_command))
    output_SPDF <- SpatialPolygonsDataFrame(out_polys,data)

    return(output_SPDF)
    }

# Usage
zones <- dissolvePolygons(sample_tracts, "zone")

I'm curious what a preferred method is to work with the original row output?  I was finding it problematic to tie it back to other data and was hoping to find some examples of how to do so.

Thanks,

Roger
--