<div class="gmail_extra">Hi Steve,<br>It happens :-)<br>Thanks for the clarification. Actually I was a bit confused also. You may call me Razequl or my nick name Zibon.<br>I think your test cases will be very helpful for testing the new implementation. <br>
<br>Razequl<br><br><br><br><div class="gmail_quote">On Mon, Apr 23, 2012 at 10:16 AM, Stephen Woodbridge <span dir="ltr"><<a href="mailto:woodbri@swoodbridge.com" target="_blank">woodbri@swoodbridge.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Hi Razequl,<br>
<br>
I can be slow when I'm distracted by other things - I did not connect Razequl as being the same person as Khondoker that you proposal was submitted under until now when I was reviewing your proposal again. So please ignore my first bullet below and accept my apology.<br>
<br>
I have developed some test cases for the TRSP algorithm and we might be able to use them as tests for the bidirectional algorithm. I also have extensive datasets covering large areas that we can potentially use for performance testing.<br>
<br>
Thank you for your introduction.<br>
<br>
Best regards,<br>
-Steve<div class="HOEnZb"><div class="h5"><br>
<br>
On 4/22/2012 2:01 PM, Stephen Woodbridge wrote:<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Hi Razequl,<br>
<br>
I have a few of comments in no particular order:<br>
<br>
* We have a GSoC student that has proposed a project to do a<br>
Bi-directional routing algorithm and we should know by tomorrow if that<br>
was accepted or not. That said, while if it is accepted then having two<br>
is a little redundant unless one supports turn restrictions like TRSP<br>
and one does not.<br>
<br>
* Boost as the advantage that is a generic library that is well tested<br>
and most of the development is about how to use the library to do what<br>
you want. On the other hand it is more restrictive if you want to do<br>
something similar to Boost but not supported in the existing code. You<br>
might be able to reuse some of the TRSP code which would be good if you<br>
can do that.<br>
<br>
* Testing - I think you need to consider making some simple unit test<br>
examples that allow you to validate your code. This would be a<br>
collection of tests that you can apply to a relatively simple network<br>
that is easy to debug and trace the execution. Once this is done and<br>
working it is reasonable to throw a larger network like osm2pgrouting to<br>
look at performance tuning.<br>
<br>
* One of the keys to making quality releases is our ability to automate<br>
sanity testing of the code before we release it. While we do not have<br>
any automated testing yet, it is very high on my list of things that<br>
need to get done, so I will be pushing everyone to support this.<br>
<br>
As Jay said, if you have any specific questions that we can help with,<br>
please ask on the list.<br>
<br>
Best regards,<br>
-Steve<br>
<br>
On 4/22/2012 1:00 PM, Razequl Islam wrote:<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Hi Jay,<br>
Thanks for your reply and suggestion. I have already gone through the<br>
source of basic dijkstra, A* and the new variant trsp (turn restricted<br>
shortest path) as suggested by Steve. I found that dijkstra and A* has<br>
been implemented using the boost library while trsp has its own<br>
structure and implementation. It also uses STL for priority queue. I am<br>
thinking of implementing like trsp using my own data structure.<br>
I am still playing with these. I need some sample database, I am trying<br>
to install osm2pgrouting but struggling.<br>
Thanks for your help once again.<br>
<br>
Razequl<br>
<br>
On Sat, Apr 21, 2012 at 12:34 AM, Jay Mahadeokar<br>
<<a href="mailto:jai.mahadeokar@gmail.com" target="_blank">jai.mahadeokar@gmail.com</a> <mailto:<a href="mailto:jai.mahadeokar@gmail.com" target="_blank">jai.mahadeokar@gmail.<u></u>com</a>>> wrote:<br>
<br>
Hi Razequl,<br>
<br>
Welcome to pgRouting. Good implementation of Bi-directional<br>
searching algorithm will be a very useful addition to pgRouting. I<br>
would suggest you to:<br>
<br>
- See the pgRouting source.<br>
- Study typical pgRouting algorithm implementation link<br>
shortest_path, play around with it, compile and debug etc. You will<br>
need to use cmake and other tools for the same.<br>
- Think about design the data structures and modules that will be<br>
needed for Bi-directional algorithm implementation and discuss the<br>
same in the list.<br>
<br>
Steve and Daniel will surely provide useful insights and guidance<br>
for you. :-)<br>
Let us know if you have any queries.<br>
<br>
Regards,<br>
<br>
On Fri, Apr 20, 2012 at 10:12 PM, Razequl Islam<br>
<<a href="mailto:ziboncsedu@gmail.com" target="_blank">ziboncsedu@gmail.com</a> <mailto:<a href="mailto:ziboncsedu@gmail.com" target="_blank">ziboncsedu@gmail.com</a>>> wrote:<br>
<br>
Hi,<br>
I am Razequl Islam, student of Dhaka university, Bangladesh. I<br>
am new to the group. I am interested in implementing bi<br>
directional searching algorithms in pgrouting, Your feedback and<br>
suggestions are welcome.<br>
<br>
Best Regards,<br>
Razequl<br>
<br>
<br>
______________________________<u></u>_________________<br>
pgrouting-dev mailing list<br>
<a href="mailto:pgrouting-dev@lists.osgeo.org" target="_blank">pgrouting-dev@lists.osgeo.org</a> <mailto:<a href="mailto:pgrouting-dev@lists.osgeo.org" target="_blank">pgrouting-dev@lists.<u></u>osgeo.org</a>><br>
<a href="http://lists.osgeo.org/mailman/listinfo/pgrouting-dev" target="_blank">http://lists.osgeo.org/<u></u>mailman/listinfo/pgrouting-dev</a><br>
<br>
<br>
<br>
<br>
--<br>
-Jay Mahadeokar<br>
Mtech, II Year,<br>
IIT Kanpur,<br>
India.<br>
<br>
______________________________<u></u>_________________<br>
pgrouting-dev mailing list<br>
<a href="mailto:pgrouting-dev@lists.osgeo.org" target="_blank">pgrouting-dev@lists.osgeo.org</a> <mailto:<a href="mailto:pgrouting-dev@lists.osgeo.org" target="_blank">pgrouting-dev@lists.<u></u>osgeo.org</a>><br>
<a href="http://lists.osgeo.org/mailman/listinfo/pgrouting-dev" target="_blank">http://lists.osgeo.org/<u></u>mailman/listinfo/pgrouting-dev</a><br>
<br>
<br>
<br>
<br>
______________________________<u></u>_________________<br>
pgrouting-dev mailing list<br>
<a href="mailto:pgrouting-dev@lists.osgeo.org" target="_blank">pgrouting-dev@lists.osgeo.org</a><br>
<a href="http://lists.osgeo.org/mailman/listinfo/pgrouting-dev" target="_blank">http://lists.osgeo.org/<u></u>mailman/listinfo/pgrouting-dev</a><br>
</blockquote>
<br>
______________________________<u></u>_________________<br>
pgrouting-dev mailing list<br>
<a href="mailto:pgrouting-dev@lists.osgeo.org" target="_blank">pgrouting-dev@lists.osgeo.org</a><br>
<a href="http://lists.osgeo.org/mailman/listinfo/pgrouting-dev" target="_blank">http://lists.osgeo.org/<u></u>mailman/listinfo/pgrouting-dev</a><br>
</blockquote>
<br>
______________________________<u></u>_________________<br>
pgrouting-dev mailing list<br>
<a href="mailto:pgrouting-dev@lists.osgeo.org" target="_blank">pgrouting-dev@lists.osgeo.org</a><br>
<a href="http://lists.osgeo.org/mailman/listinfo/pgrouting-dev" target="_blank">http://lists.osgeo.org/<u></u>mailman/listinfo/pgrouting-dev</a><br>
</div></div></blockquote></div><br></div>