<div dir="ltr"><div><p>Hi all,</p><p>I am working on the implementation of maximum flow
algorithms and their applications for pgRouting during the GSoC 2016.
This is my eigth report.</p>1) What did you get done this week?<br><br>This week I finished writing most part of the documentation, tests, and examples. PgTAP unit tests were written to test all the signatures¶meters of the max flow functions and error handling was added where it was required. I also refactored the signature of maximum cardinality matching and edge disjoint paths after discussion with the mentors. Edge-disjoint paths now also works with multiple sources/destinations.<br><br>2) What do you plan on doing next week?<br>Next week I plan to do several things:<br><ul><li>Add more PgTAP tests covering other functions and checking of the actual results.</li><li>Expand the pgr_edgedisjointpaths function in order to list all the actual paths and not just the number.</li><li>Feasibility study on implementing other <a href="https://jorlin.scripts.mit.edu/docs/papersfolder/O%28nm%29MaxFlow.pdf">possible implementations</a> of maxflow that offer even better complexities. Discussion with mentors on this topic.</li></ul><br>3) Are you blocked on anything?<br>No.<br><br><p>You can check the branch I'm working on here:</p><p><a href="https://github.com/Illedran/pgrouting/tree/gsoc-flow" target="_blank">https://github.com/Illedran/pgrouting/tree/gsoc-flow</a></p><p>Best regards,</p>Andrea Nardelli</div></div>