The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
rgexf is back on CRAN! The package was archived on CRAN on 2026-05-18 because outstanding check issues were not corrected in time. This release fixes those issues and returns the package to CRAN.
plot.gexf() now renders graphs as an interactive
htmlwidget powered by sigma.js v3
and graphology. Node
positions, colours, and sizes are read from the viz:*
attributes in the GEXF document.
New sigmajs() function creates a sigma.js htmlwidget
from a gexf object or a path to a .gexf file.
Shiny helpers sigmajsOutput() and
renderSigmajs() are also available.
The legacy gexf-js file-server renderer is preserved as
plot_gexfjs(). The gexfjs() htmlwidget (inline
iframe approach) is likewise kept for backward compatibility.
gexfjs() now accepts a gexf object in
addition to a file path, matching sigmajs().
gexfjs() has an explicit sizing policy: the widget
now defaults to the full available width ("100%") and a
height of "400px" in knitr documents, the browser, and
Shiny alike. Both gexfjs() and sigmajs() honor
user-supplied width/height, and their
documented defaults were clarified.
The gexfjs() widget rendered an empty main canvas:
the bundled config.js (which sets zoomLevel,
showEdges, and other gexf-js defaults) was never loaded
into the iframe, making every screen coordinate
NaN.
Icons and images in the gexfjs() widget (zoom, lens,
edge buttons, search icon, Gephi logo) did not display: relative
url(...) references in the CSS cannot resolve inside a
srcdoc iframe, so images are now inlined as base64 data
URIs.
The gexfjs() widget’s node-attribute panel was blank
and search did not work: the generated DOM was missing the
#leftcontent element, the search form id had a typo
(reacherche), and the autocomplete list was clipped by the
title bar.
We now have a hex sticker!
Users can now cite properly rgexf with JOSS (see
citation(package="rgexf")).
Added a few extra breaks across the examples (suggested by @corneliusfritz).
Improved documentation regarding spells and dynamic graphs (suggested by @jonjoncardoso).
New head() function allows a glimpse of the
n first nodes/edges.
Passing colors with four values (alpha) no longer fails during checks. (reported by @IsabelFE).
The summary function prints nodes’ attributes as expected.
Hex colors now work (#41 reported by @milnus).
gexf.to.igraph correctly processes edge attributes
(#38 reported by @balachia).
Time range is now correctly computed (#19).
Non-integer ids were incorrectly processed when reading GEXF files.
Function write.gexf has a new argument for
specifying the GEXF version for now it only changes the header.
plot.gexf method now uses gexf-js
instead of sigma.js.
Now igraph.to.gexf passes arguments to
gexf. Before, it was only passing position.
This way users have more flexibility specifying attributes.
gexf’s nodesVizAtt now has defaults for
color, size, and position; this is a requirement of
gexf-js.
gexf ’s nodesVizAtt color and size now
support passing a default for all the nodes. Also, color can be
specified as a character scalar (name of the) color, or as an integer
(number in colors()).
gexf has a new argument,
rescale.node.size. When set to TRUE, the
size vector in nodesVizAtt is rescaled such
that when calling the plot method the largest node spans roughly 5% of
the plot.
read.gexf now parses viz attributes (color,
position, and size).
igraph.to.gexf and vice versa now pass viz
attributes and layout.
gexf has been introduced as an eventual
replacement of write.gexf since it makes more sense.
write.gexf should only be for writing the gexf file on the
disk (effective starting vers 0.17.0).Modernization of the project (roxygen, new CRAN standards, etc.)
Updated emails.
Remove broken links.
Some data management functions were fully ported to R.
New option -digits- in several functions allows controlling for printing precision (reported in issue 16, thanks Nico!).
Function -igraph.to.gexf- now imports attributes.
Option -defaultedgetype- in the -write.gexf- function now works for static graphs (reported in issue 17, thanks Avitus!).
-datetime- time format in -write.gexf- now works for other formats different from numeric datetime types (reported as issue 15, thanks Thomas Ullmann!).
Correcting errors in -plot.gexf- method, now visual attributes are imported correctly.
Included class checks in -gexf- class functions (thanks to Samuel Finegold).
write.gexf does not fails when dynamics different
from double are passed (thank you, Samuel
Finegold!).
Passing weights from igraph to gexf has now been fix (thank you Roberto Brunelli!).
Fixing encoding issues thanks to Yong Cha’s suggestion (thank you!).
New S3 method plot.gexf, implemented by Joshua B
Kunst, shows the graph in the web browser by means of Sigma.js
(!).
New functions igraph.to.gexf and
gexf.to.igraph converts objects between igraph
and gexf classes. Colors and attributes are
preserved.
Improving general documentation.
Just starting to add new options to
add.gexf.node/edge, more precisely, passing
attributes.
sprintf error when using other formats rather than
double (issue 10).
in .addNodesEdges, add support to case “!attributes
&& vizattributes” (issue 9).
New functions add.node.spell and
add.edge.spell now allow to work with nodes and edges time
spells.
New function check.dpl.edges, written in C, analyzes
links and reports duplicates (marking them) and number of times the same
link is repeated considering if the graph is directed or not.
New function switch.edges, also written in C, allows
to order links representations (source and target) in order to set the
smallest id as source and the highest as target.
Function write.gexf now has an improved error
handler. Now parses objects before even opening the XML graph.
Edges support id assignment and labeling.
Small bug detected in viz attributes fixed.
Edges thickness viz att XML representation
(viz:thickness) was replaced by viz:size (as
it should be).
“Library” replaced by “Package” everywhere
(ups!)
.addNodesEdges rewritten now works faster in most of
CPUs (some of them with very high speedups) (Thanks to Duncan Temple
Lang, RXML author)
Several code routines have been extracted from “bigger functions” and written as functions themselves.
New functions new.gexf.graph,
add.gexf.node, add.gexf.edge,
rm.gexf.node and rm.gexf.edge allow to build
and manipulate gexf objects from scratch.
New function read.gexf allows to import gexf files
as gexf class objects.
gexf function now it is called
write.gexf.
Edges now allow weighting.
Viz Attributes (color, shape, size, etc.) can be included in both, nodes and edges.
Real-life datasets have been included.
New function edge.list builds a dataframe of nodes
from an edge list.
New methods for gexf objects:
print.gexf and summary.gexf.
gexf class objects are now a standard.
Function gexf is now named
write.gexf.
Faster net build.
More demos + improve ones.
Cleaner code.
Fixing big issue at attvalues XML tag: it was replaced from “att” to “attvalue”.
Fixing problem with XML value printing: Leading spaces where removed from XML values at ids, source, target, etc.
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.