Thanks for the quick reply! I might be even better off just installing Linux, I've tried it on a virtual instance and it was more or less just an apt-get install on Ubuntu which is quite a bit easier :) I was wondering: the specific graphs I'm looking for are induced isomorphic subgraphs. G can be any undirected graph, and subgraph S is a chorless cycle (i.e. a 'ring' with n vertices). I specifically want to find all (unique) vertex induced subgraphs of S in G, or in other words all chordless subgraphs of length n in G. With vm, em = gt.subgraph_isomorphism(S,G) I can find subgraphs, but they are not neccesarily vertex induced: the found isomorphic subgraph (lets call them F) can have more edges than S I can see two ways to detect if the found subgraph F is isomorphic to S 1. (for this specific case) check if the number of edges in S == number of edges in F - as they are cyclic this works 2. check with gt.isomorphism(S,F) Only: how do I get F? I use vmask, emask = gt.mark_subgraph(G,S, vm[i], em[i]) in a loop as in http://projects.skewed.de/graph-tool/doc/topology.html#graph_tool.topology.s..., but after masking G.set_vertex_filter(vmask) still gt.isomorphism(S,G) == False How do I generate a new graph from the output of gt.subgraph_isomorphism(S,G) which has all vertices and edges of F but no more (i.e. how do I define F as a graph which will possibly give gt.isomorphism(F,S) == True when this is the case). Or should I be able to check the number of edges in the filtered version of G as per my method 1? How? Thanks for your assistance! Jelle Veraa On 3 January 2013 11:42, Tiago Peixoto [via Main discussion list for the graph-tool project] <ml-node+s982480n4024890h64@n3.nabble.com> wrote:
Hi there,
It seems the configure script cannot find some libraries:
c:/users/jelle.veraa/dropbox/ke-works-jelleveraa/random/mingw/bin/../lib/gcc/mingw32/4.7.2/../../../../mingw32/bin/ld.exe:
cannot find -lbz2 c:/users/jelle.veraa/dropbox/ke-works-jelleveraa/random/mingw/bin/../lib/gcc/mingw32/4.7.2/../../../../mingw32/bin/ld.exe:
cannot find -lexpat
and it screws up finding extra link flags for python:
configure:17030: checking python extra linking flags configure:17037: result: None configure:17044: checking consistency of all components of python development environment configure:17070: gcc -o conftest.exe -g -O2 -Ic:\Python27\include conftest.c -lm -Lc:\Python27\Lib\config -lpython27 None >&5 gcc.exe: error: None: No such file or directory
Make sure all the necessary libraries are installed in the correct path. For the last bug, the check is done in m4/ax_python_devel.m4:
if test -z "$PYTHON_EXTRA_LDFLAGS"; then PYTHON_EXTRA_LDFLAGS=`$PYTHON -c "import distutils.sysconfig; \ conf = distutils.sysconfig.get_config_var; \ print (conf('LINKFORSHARED'))"` fi
For some reason, it is printing 'None', instead of an empty string. This could be changed to something like,
if test -z "$PYTHON_EXTRA_LDFLAGS"; then PYTHON_EXTRA_LDFLAGS=`$PYTHON -c "import distutils.sysconfig; \ conf = distutils.sysconfig.get_config_var; \ x = conf('LINKFORSHARED'); \ x = x if x is not None else ''; \ print (x)"` fi
But this is unlikely to fix the problem, since I think it is very strange that this value points to None in the first place.
It would be nice to have graph-tool working on windows, but it is a very weird platform, and I do not have access to a windows machine to understand all its idiosyncrasies.
Cheers, Tiago
On 01/02/2013 02:27 PM, Jelle Veraa wrote:
Dear all,
Happy new year!
Starting of with what I'd actually like to do (building the program is of course a means to an end :) - I'd like to use the library for its subgraph isomorphism algorithm specifically - should you be able to provide me a binary of the library or know of another fast implementation for subgraph isomorphism that might help me as well.
I'm trying to install the library on a 64 bit Windows 7 machine running 32 bit python 2.7. Installing the dependencies (from binaries) works, but running ./config from MinGW/MSys gives me an error. it exits with "Could not link test program to Python. Maybe the main Python library has been installed in some non-standard library path. If so, pass it to configure, via the LDFLAGS environment variable.", but the solution seems less straightforward - I've already tried all kinds of LDFLAGS versions pointing towards my Python folder (at C:\Python27, or /c/Python27 in MinGW).
here is the log: http://bpaste.net/show/82ASZHcSnWmGsBIglFEo/
From line 2055/configure:16996 things start going wrong (a bit is posted below). There is a reference to Lib\config, but the config directory does not exist. Furthermore the output 'None' at configure:17037 is injected into the command line for gcc.exe which also gives an error.
I've been googling for an answer for some hours now bu no luck. I see that you need python-dev on linux for instance, but I've got Windows so 'apt-get python-dev' won't work :)
Can anyone help me?
thanks, Jelle
configure:16996: result: -Lc:\Python27\Lib\config -lpython27 configure:17003: checking for Python site-packages path configure:17009: result: c:\Python27\Lib\site-packages configure:17016: checking python extra libraries configure:17023: result: configure:17030: checking python extra linking flags configure:17037: result: None configure:17044: checking consistency of all components of python development environment configure:17070: gcc -o conftest.exe -g -O2 -Ic:\Python27\include conftest.c -lm -Lc:\Python27\Lib\config -lpython27 None >&5 gcc.exe: error: None: No such file or directory
_______________________________________________ graph-tool mailing list [hidden email] <http://user/SendEmail.jtp?type=node&node=4024890&i=0> http://lists.skewed.de/mailman/listinfo/graph-tool
-- Tiago de Paula Peixoto <[hidden email]<http://user/SendEmail.jtp?type=node&node=4024890&i=1>>
_______________________________________________ graph-tool mailing list [hidden email] <http://user/SendEmail.jtp?type=node&node=4024890&i=2> http://lists.skewed.de/mailman/listinfo/graph-tool
*signature.asc* (565 bytes) Download Attachment<http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com/attachment/4024890/0/signature.asc> -- Tiago de Paula Peixoto <tiago@skewed.de>
------------------------------ If you reply to this email, your message will be added to the discussion below:
http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com/... To unsubscribe from Error installing graph-tool on windows, click here<http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=4024889&code=amVsbGUudmVyYWFAa2Utd29ya3MuY29tfDQwMjQ4ODl8MTk0ODgyODA5> . NAML<http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
-- View this message in context: http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com/... Sent from the Main discussion list for the graph-tool project mailing list archive at Nabble.com.