Some context would be nice as this (very nice library) has been around for a while: was there a new release? What are we supposed to be looking for Here?
I just felt like sharing it. I was foraging around Twitter for good links and found this. I honestly didn't know it was around for a while, but it stimulated my intellectual curiosity so I shared it.
Gephi and Cytoscape (https://cytoscape.org/) are both open-source Java-based network analysis and visualization software.
Cytoscape.js is a graph theory and network visualization Javascript library. It can be compared to SigmaJS (http://sigmajs.org/) or the more generic D3 data visualization toolkit (https://d3js.org/).
Cytoscape.js is MUCH more capable than those, it can do client side network algorithms (A* etc) and traversal, Sigma & D3 are 'just' visualisation tools, where Gephi is an 'offline' tool
His ngraph has less features than cytoscape but if you need to plot large graphs it has a crazy good performance.
And his galaxies are amazing as well, https://anvaka.github.io/pm/#/?_k=9kvbgq . Precalculated positions for all NPM packages for example. Crazy stuff, really fun.
But yeah, cytoscape.js is nice and has great set of features.
Yes! I looked at a bunch graph visualization libraries in the past. Cytoscape did not impressed me. Anvaka's ngraph and VivaGraphJS seemed pretty closer to what I wanted.
Cryptoscape looks really awesome but i'm having trouble locating the source code of the demos like this one http://js.cytoscape.org/demos/images-breadthfirst-layout/ actually figured out how to open dev tools without a mouse and its simple too.
How hard is it to teach either of these libs about Clojure data structures? (Without copying them into plain js objects)
Cryptoscape - unclear how to use actual dom to style nodes, not just css classes (or if this is possible at all)
Love this library so much that I include it in my news letter[0] for code to read sections for JavaScript category
Some thing about this project:
- It has all kind of npm shortcut
- Minimal dependencies: only 2, other are for build time, not run time. It's too good that he opt to use `lodash.debounce` instead of whole `lodash`
- Pure JavaScript. TypeScript is nice but its generated code is such a beast
- Files are small(Many files are < 100 lines)
- Method are short
I can't say enough good things about this library. The library is fast, the API is great, the features are awesome, and it is actively maintained. When I was evaluating the library for my use case, I submitted a PR for a small feature on their roadmap, and the maintainer had gotten back to me, made suggestions, and merged the PR in a matter of hours.
Shout out to Max Franz for being such a dedicated, responsive, and competent maintainer for such a cool project!
19 comments
[ 3.5 ms ] story [ 24.6 ms ] threadCytoscape.js is a graph theory and network visualization Javascript library. It can be compared to SigmaJS (http://sigmajs.org/) or the more generic D3 data visualization toolkit (https://d3js.org/).
His ngraph has less features than cytoscape but if you need to plot large graphs it has a crazy good performance.
And his galaxies are amazing as well, https://anvaka.github.io/pm/#/?_k=9kvbgq . Precalculated positions for all NPM packages for example. Crazy stuff, really fun.
But yeah, cytoscape.js is nice and has great set of features.
Cryptoscape looks really awesome but i'm having trouble locating the source code of the demos like this one http://js.cytoscape.org/demos/images-breadthfirst-layout/ actually figured out how to open dev tools without a mouse and its simple too.
How hard is it to teach either of these libs about Clojure data structures? (Without copying them into plain js objects)
Cryptoscape - unclear how to use actual dom to style nodes, not just css classes (or if this is possible at all)
Some thing about this project:
- It has all kind of npm shortcut - Minimal dependencies: only 2, other are for build time, not run time. It's too good that he opt to use `lodash.debounce` instead of whole `lodash` - Pure JavaScript. TypeScript is nice but its generated code is such a beast - Files are small(Many files are < 100 lines) - Method are short
Very easy to read and follow.
---
[0]: https://betterdev.link/issues/77
Shout out to Max Franz for being such a dedicated, responsive, and competent maintainer for such a cool project!