http://blog.gmane.org/gmane.comp.science.graph.graph-tool.bugs hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/695 <pre>#104: No setup.py makes installation with pip impossible new defect --------------------------------------------------------------------- Reporter: jeffrey.finkelstein&lt; at &gt;gmail.com Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- Without a `setup.py` file, `graph-tool` cannot be installed using `pip`. This is a pain if one wants to distribute a Python package which depends on it. -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/104&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. _______________________________________________ graph-tool-tickets mailing list graph-tool-tickets&lt; at &gt;skewed.de http://lists.skewed.de/mailman/listinfo/graph-tool-tickets </pre> graph-tool 2012-05-23T20:18:14 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/686 <pre>#103: Compilation on Archlinux fails new defect --------------------------------------------------------------------- Reporter: muellner&lt; at &gt;math.stanfod.edu Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- Compilation on Archlinux fails. I have Python 2 and Python 3 installed on my system. To make the configure script work, I set the environment variable PYTHON_VERSION=2.7. I attach config.log and the output of make, in the hope that the source of the error can be determined by that. -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/103&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. _______________________________________________ graph-tool-tickets mailing list graph-tool-tickets&lt; at &gt;skewed.de http://lists.skewed.de/mailman/listinfo/graph-tool-tickets </pre> graph-tool 2012-05-12T17:15:11 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/681 <pre>#102: ImportError: No module named graph_tool.all new defect --------------------------------------------------------------------- Reporter: subusurendran&lt; at &gt;gmail.com Owner: tiago peixoto Priority: major Milestone: 3.0 --------------------------------------------------------------------- Hai, I am new in using graph-tool module. After the installation i tried to execute the python statement "import graph_tool.all as gt" to check the installation. But I am getting an error message"ImportError: No module named graph_tool.all". Environment: UBUNTU 11.04 Python 2.7.1 [GCC 4.5.2] Can anyone help me to rectify the problem -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/102&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. _______________________________________________ graph-tool-tickets mailing list graph-tool-tickets&lt; at &gt;skewed.de http://lists.skewed.de/mailman/listinfo/graph-tool-tickets </pre> graph-tool 2012-05-09T13:50:24 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/677 <pre>#101: Git Build fails on Mac OS X Lion new defect --------------------------------------------------------------------- Reporter: oscarvarto&lt; at &gt;gmail.com Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- My ./configure output is here: http://pastebin.com/ZsKdTBrP I installed boost with macports: `sudo port install boost +python27` (boost 1.49) Don't know how to fix this... -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/101&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. _______________________________________________ graph-tool-tickets mailing list graph-tool-tickets&lt; at &gt;skewed.de http://lists.skewed.de/mailman/listinfo/graph-tool-tickets </pre> graph-tool 2012-05-08T16:49:50 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/668 <pre>#100: graph-tool cannot find gvc library when it's installed in a non-standard location new defect --------------------------------------------------------------------- Reporter: spivakov&lt; at &gt;ebi.ac.uk Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- The problem occurred when another, older version of gvc was found in a standard location, while the version I was looking for was installed into a custom folder, specified in LD_LIBRARY_PATH. My solution to this, which unfortunately isn't elegant, was to hardcode the library name into draw/ init .py instead of using ctypes.util.find_library. The problem with find_library is that it doesn't consult LD_LIBRARY_PATH, making the search algorithm rather inflexible. libgv = ctypes.CDLL("libgvc.so.6.0.0") # libname = ctypes.util.find_library("gvc") # if libname is None: # raise OSError() # libgv = ctypes.CDLL(libname) Hopefully, something more robust can be done </pre> graph-tool 2012-04-09T21:45:27 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/660 <pre>#99: graph-tool does not compile with boost 1.49.0 new defect --------------------------------------------------------------------- Reporter: Mikhail Spivakov Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- CXX graph_betweenness.lo ... boost/property_map/property_map.hpp:354: error: no match for 'operator[]' in '(const boost::unchecked_vector_property_map&lt;long double, boost::adj_list_edge_property_map&lt;boost::bidirectional_tag, long unsigned int, long unsigned int&amp;, long unsigned int, boost::property&lt;boost::edge_index_t, long unsigned int, boost::no_property&gt;, boost::edge_index_t&gt; &gt;&amp;)((const boost::unchecked_vector_property_map&lt;long double, boost::adj_list_edge_property_map&lt;boost::bidirectional_tag, long unsigned int, long unsigned int&amp;, long unsigned int, boost::property&lt;boost::edge_index_t, long unsigned int, boost::no_property&gt;, boost::edge_index_t&gt; &gt;*)(&amp; pa))[k]' ... /homes/spivakov/n/boost_1_49_0/boost/p</pre> graph-tool 2012-04-05T18:25:26 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/655 <pre>#98: PropertyMaps 'reuse' values assigned to deleted vertices/edges new defect --------------------------------------------------------------------- Reporter: Wojciech Waśko &lt;wojciechwasko&lt; at &gt;gmail.com&gt; Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- While I was unit-testing my code, I came across some strange behaviour, which I traced back to `graph-tool`. If I have some property maps and delete vertices/edges from the corresponding graph and then add some other vertex/edge, the property map 'reuses' the previously assigned values, corresponding to already-deleted vertices/edges. It is best illustrated by the following: {{{#!python&gt;&gt;&gt; import graph_tool as gt Traceback (most recent call last): File "&lt;stdin&gt;", line 1, in &lt;module&gt; File "/usr/lib/python2.7/site-packages/graph_tool/__init__.py", line 403, in __getitem__ return self.__map[self.__key_trans(k)] ValueError: invalid edge descriptor 'eprop' Traceback (mos</pre> graph-tool 2012-03-28T10:06:30 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/651 <pre>#97: Deleting edges is broken new defect --------------------------------------------------------------------- Reporter: Wojciech Waśko &lt;wojciechwasko&lt; at &gt;gmail.com&gt; Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- [3a528888d] broke removing edges; the problem is at line 1043 of file `./graph_tool/__init__.py` Is: {{{ return libcore.remove_edge(self, edge) }}} Should be: {{{ return libcore.remove_edge(self.__graph, edge) }}} The [3a528888d] version would produce the following error: {{{#!python &lt;Edge object with source '0' and target '0' at 0x16d8dd0&gt; Traceback (most recent call last): File "&lt;stdin&gt;", line 1, in &lt;module&gt; File "/usr/lib/python2.7/site-packages/graph_tool/__init__.py", line 1043, in remove_edge return libcore.remove_edge(self, edge) Boost.Python.ArgumentError: Python argument types in graph_tool.libgraph_tool_core.remove_edge(Graph, Edge) did not match C++ signature: </pre> graph-tool 2012-03-21T12:01:08 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/641 <pre>#96: Compilation Fail On Arch Linux new defect --------------------------------------------------------------------- Reporter: lee.mcculler&lt; at &gt;gmail.com Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- I'm trying to compile graph_tool on arch using GCC 4.6.3 and boost 1.49.0 against the current git master and I get the following compilation error. I don't know how to parse all of the verbose template output so I'm just putting all of it (attached) -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/96&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. _______________________________________________ graph-tool-tickets mailing list graph-tool-tickets&lt; at &gt;skewed.de http://lists.skewed.de/mailman/listinfo/graph-tool-tickets </pre> graph-tool 2012-03-20T18:45:15 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/635 <pre>#95: Edges are reported not to be instances of graph-tool's Edge class new defect --------------------------------------------------------------------- Reporter: Wojciech Waśko &lt;wojciechwasko&lt; at &gt;gmail.com&gt; Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- Today, I came across the following bug: edges returned from `graph_tool.Graph.add_edge()` or `graph_tool.Graph.edge()` fail test for `isintance` against graph-tool's `Edge`. See the snippet below: {{{#!python &lt;class 'graph_tool.libgraph_tool_core.Vertex'&gt; &lt;class 'graph_tool.libgraph_tool_core.Vertex'&gt; &lt;class 'graph_tool.libgraph_tool_core.Edge'&gt; &lt;class 'graph_tool.libgraph_tool_core.Edge'&gt; True True False False False False }}} -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/95&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. _______________________________________________ graph-tool-tickets mailing list </pre> graph-tool 2012-03-09T13:33:10 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/634 <pre>#94: Access underlying graph from Vertex or Edge new enhancement --------------------------------------------------------------------- Reporter: Wojciech Waśko &lt;wojciechwasko&lt; at &gt;gmail.com&gt; Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- It would be nice if one could access the underlying graph from `Vertex`/`Edge` context. The primary motivation is that in order to set a property of a vertex/edge, you need to do it through `Graph.vertex_properties`/`Graph.edge_properties`, thus needing the `Graph` object itself. I don't know the implementation details of `graph-tool`, but I can see that internally, there is some kind of access to the underlying graph from `Vertex`/`Edge` context: {{{#!python True True False False }}} -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/94&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. ______________________________________</pre> graph-tool 2012-03-09T09:35:49 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/632 <pre>#93: Problem Installing graph tool on Macosx new defect --------------------------------------------------------------------- Reporter: halehashki&lt; at &gt;gmail.com Owner: tiago peixoto Priority: major Milestone: (None) --------------------------------------------------------------------- Dear Sir/Mam, I did download MacPort .dmg file and installed it.looks work fine.but when I want to use it to install graph tool gives me this error: port install py-graph-tool Error: Insufficient privileges to write to MacPorts install prefix. I will be thankful if you help me solve the problem. Thanks, Haleh -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/93&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. _______________________________________________ graph-tool-tickets mailing list graph-tool-tickets&lt; at &gt;skewed.de http://lists.skewed.de/mailman/listinfo/graph-tool-tickets </pre> graph-tool 2012-03-07T17:03:42 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/630 <pre>#92: Incorrect modularity evaluation new defect --------------------------------------------------------------------- Reporter: gamma2&lt; at &gt;users.sourceforge.net Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- The current implementation of the modularity mixes edge weights and unweighted degrees. Given the modularity definition: Q = 1/(2m) * sum_ij [ A(i,j) - k(i)*k(j)/(2m) ] d(c_i,c_j) In the code, while A(i,j) uses the weights, k(i)*k(j) is calculated using just the degree of the vertices, instead of summing the weights of the edges. The paper where Newman defined the weighted modularity is http://arxiv.org/pdf/condmat/0407503.pdf , page 7, formula 9. I know you refer to the 2006's paper, where only vertex degree is used, however the notation is the same. You can easily check it by creating a graph with 2 vertices, connected by an edge with weight 0.1, and putting them in the same community. Using your modularity implementati</pre> graph-tool 2012-02-28T10:30:47 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/628 <pre>#91: Installation on PythonXY new enhancement --------------------------------------------------------------------- Reporter: jhinebau&lt; at &gt;mie.utoronto.ca Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- Is it possible to install this tool kit into a PythonXY framework? A collaborative project is starting based in Python. Some of our team members cannot work from Linux machines due to security restrictions at their work-places. Are there instructions available for getting the Graph-Tool libraries installed on a Windows XP machine? That machine will preferably use PythonXY. -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/91&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. _______________________________________________ graph-tool-tickets mailing list graph-tool-tickets&lt; at &gt;skewed.de http://lists.skewed.de/mailman/listinfo/graph-tool-tickets </pre> graph-tool 2012-02-24T23:54:38 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/622 <pre>#89: Drawing vertices labels does not work new defect --------------------------------------------------------------------- Reporter: wojciechwasko&lt; at &gt;gmail.com Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- I've been trying to add labels to nodes when drawing a graph. The documentation is not clear on that, however I guess it should be done through the `vpropos` attribute of the `graph_tool.draw.graph_draw` function. This does not seem to work. Below, I attach the commands I used to create the plot, and the plot itself. {{{ ##### [12:54:17] # [dare2be&lt; at &gt;PC10CIS:~] ##### $ python Python 2.7.2 (default, Oct 27 2011, 01:40:22) [GCC 4.6.1 20111003 (Red Hat 4.6.1-10)] on linux2 Type "help", "copyright", "credits" or "license" for more information. &lt;PropertyMap object with key type 'Vertex' and value type 'vector&lt;double&gt;', for Graph 0x1e969d0, at 0x1e96a90&gt; }}} -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/89&gt; graph-</pre> graph-tool 2012-02-21T13:49:28 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/620 <pre>#88: Tarball is missing stuff new defect --------------------------------------------------------------------- Reporter: dmitrijs ledkovs Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- Tarball release is missing following things: * src/graph/topology/graph_reciprocity.cc * doc/ first one doesn't seem to be used in the Makefile.am since beginning of time.... Maybe it's junk? Maybe it should be built? Docs are missing as well. probably `$ make dist` is playing up. -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/88&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. _______________________________________________ graph-tool-tickets mailing list graph-tool-tickets&lt; at &gt;skewed.de http://lists.skewed.de/mailman/listinfo/graph-tool-tickets </pre> graph-tool 2012-02-13T20:39:54 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/615 <pre>#87: Compilation problem with Fedora 15 / gcc 4.6 new defect --------------------------------------------------------------------- Reporter: andreas.putz&lt; at &gt;afcc-auto.com Owner: tiago peixoto Priority: normal Milestone: 2.2 --------------------------------------------------------------------- I am trying to install graph-tool on my Fedora 15 notebook. The reported error showed up on a google search (http://answerpot.com/showthread.php?3073566-graphtool+compile+error+on+F15), but has not been brought to a conclusion. There is also an older thread in a related issue (http://groups.google.com/group/boost-list/browse_thread/thread/eca7c42f287f0f00) I am compiling the code on a Fedora 14 machine at the moment, and that seems to work (The compilation is not done yet though). The error described below manifests in the stable release as well as the trunk. = System = {{{ [root&lt; at &gt;hex graph-tool-trunk]# python -V Python 2.7.1 [root&lt; at &gt;hex graph-tool-trunk]# gcc -v Using built-in specs. COLLECT_GCC=/usr/bin/gcc</pre> graph-tool 2012-02-03T19:32:51 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/611 <pre>#86: "Error: agstrfree lost'' while using graph_tool.draw.draw_graph in multithreaded app new defect --------------------------------------------------------------------- Reporter: ujujinsan Owner: tiago peixoto Priority: minor Milestone: (None) --------------------------------------------------------------------- First of all I am not sure whether it's a defect or anything, but please read on: I am using graph_tool extensively in my project, and I recently added creating graphs in threads(thousands of graphs). During graph creation graph_tool.draw.draw_graph is called in order to acquire node location which is used later on. After adding threading I get: 'Error: agstrfree lost' For example: Error: agstrfree lost \N Error: agstrfree lost 0.3 Error: agstrfree lost #a40000 Error: agstrfree lost 0.105 Error: agstrfree lost 0.3 Error: agstrfree lost 0.105 Error: agstrfree lost 1.0 Error: agstrfree lost 8.0 Error: agstrfree lost 0.105 Error: agstrfree lost 0.3 Error: agstrfree lost \N Erro</pre> graph-tool 2012-01-24T18:06:02 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/608 <pre>#85: Compile problem on Ubuntu new defect --------------------------------------------------------------------- Reporter: gamma2&lt; at &gt;users.sourceforge.net Owner: tiago peixoto Priority: normal Milestone: (None) --------------------------------------------------------------------- I just compiled the source on Ubuntu, but I had to add {{{ #include &lt;boost/unordered_map.hpp&gt; }}} in source/graph/gml.hh because the make process raised the following error: {{{ In file included from graph_io.cc:38:0: gml.hh:202:5: error: ‘unordered_map’ does not name a type ...more type definitions and errors relative to the previous line... }}} Please check if the problem is relative to graph_tool or boost includes, otherwise I should check my configuration. -- Ticket URL: &lt;URL:http://graph-tool.skewed.de/ticket/85&gt; graph-tool &lt;URL:http://graph-tool.skewed.de/&gt; An efficient python module for graph analysis and manipulation. _______________________________________________ graph-tool-tickets mailing list graph-too</pre> graph-tool 2012-01-09T21:34:14 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/607 <pre>#85: paleo diet - Become Thinner And Healthier With The Paleo Diet ----------------------------+--------------------------- Reporter: Everett Melton | Owner: tiago peixoto Type: defect | Status: new Priority: normal | Milestone: Component: graph-tool | Version: 2.2 Keywords: | ----------------------------+--------------------------- Do you think you're tired of the ''' [http://www.aboutpaleodiet.org paleo food list]''' you're on? So deprived from sacrificing the produce you want to enjoy? Everyone has our recommended weight that you should maintain to look good and feel good. It makes us feel light and balanced, allowing us to transfer without having restrictions acknowledge that we can easily since we're also in good shape. '''diet''' has evolved from the moments, most of them claim they are helpful but once experimented with by weight watchers, they may be just going for incorrect anticipation. There are plenty of diets that indi</pre> graph-tool 2012-01-07T01:34:21 http://comments.gmane.org/gmane.comp.science.graph.graph-tool.bugs/606 <pre>#85: Test this ------------------------+--------------------------- Reporter: locke411&lt; at &gt;… | Owner: tiago peixoto Type: defect | Status: new Priority: normal | Milestone: Component: graph-tool | Version: 2.2 Keywords: | ------------------------+--------------------------- Hi </pre> graph-tool 2012-01-04T17:34:21 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.comp.science.graph.graph-tool.bugs