[julia-users] Professors working on Julia in UCSD or UCLA

Hi Guys, I've received admits for M.S. in Computer Science from both UCLA and UCSD. Are there any professors working on Julia/packages in Julia in either of these colleges?

Re: [julia-users] GUI toolkits for Julia: Gtk vs Tk vs others

On 28 March 2016 at 19:58, Tim Holy wrote: > > Specifically, Gtk.jl has a more sophisticated internal design than Tk.jl. > (Meaning, I was referring to the julia code.) Among other things, it makes > use > of introspection, see https://wiki.gnome.org/Projects/GObjectIntrospection, > to > auto-gene

Re: [julia-users] GUI toolkits for Julia: Gtk vs Tk vs others

On Monday, March 28, 2016 07:38:12 PM Daniel Carrera wrote: > I wish I knew more about GUI toolkits. Somehow I've never managed to find > the time. In which way does Gtk have a more sophisticated internal design? Specifically, Gtk.jl has a more sophisticated internal design than Tk.jl. (Meaning,

[julia-users] Re: API deployments on JuliaBox

Hi Eric, Yes, it should be possible to send `get` variables to a Julia API. The API needs to accept them as named keyword arguments. You can see some examples of such APIs at https://github.com/JuliaWeb/JuliaWebAPI.jl Best, Tanmay On Thursday, March 24, 2016 at 12:04:41 AM UTC+5:30, Eric Eisam

Re: [julia-users] GUI toolkits for Julia: Gtk vs Tk vs others

Hello, I wish I knew more about GUI toolkits. Somehow I've never managed to find the time. In which way does Gtk have a more sophisticated internal design? My understanding is that Gtk doesn't work so well on Mac and Windows. The screenshots on the Gtk website certainly make it look out of place

[julia-users] Re: Error: Kernel dead in Jupyter

Temporary workaround you can try: using WinRPM unshift!(WinRPM.sources, "http://download.opensuse.org/repositories/home:/kelman:/branches:/windows:/mingw:/win64/openSUSE_13.2";) WinRPM.update() WinRPM.install("zeromq") On Monday, March 28, 2016 at 3:40:10 PM UTC-7, Tony Kelman wrote: > > Looks

[julia-users] Re: Error: Kernel dead in Jupyter

Looks like an issue with the source of binaries that we're using on Windows for many precompiled libraries. Something changed about 4 days ago, probably https://build.opensuse.org/package/rdiff/windows:mingw:win64/mingw64-runtime?linkrev=base&rev=39, causing builds since then to segfault - ht

[julia-users] Re: Error: Kernel dead in Jupyter

Could you provide more information about what the notebook is doing? You can usually bisect what causes the segfault by running the notebook one line at a time. Random guess: are you reloading code anywhere? On Monday, March 28, 2016 at 2:28:48 PM UTC-4, Zahirul ALAM wrote: > > Hi, > > I install

Re: [julia-users] Interest in BandedMatrices.jl Package?

I realized BLAS.gbmv works with both orderings using the ’T’ option, so its probably worth supporting both formats. Anyone know if there is a BandedMatrix*BandedMatrix routine in BLAS/LAPack? I have it implemented in Julia, but it would probably be a lot faster using an inbuilt version. > On

Re: [julia-users] Announcing JuDE: autocomplete and jump to definition support for Atom

Mike can step in, but I think it is indeed the plan to have a Julia IDE bundle, causing the exact packages required on the julia and atom side to be invisible implementation details from the user's POV. On Monday, March 28, 2016 at 1:27:01 PM UTC-4, David Anthoff wrote: > > I think the user expe

[julia-users] git protocol packages

Is there a way to still support git protocol (as opposed to https) packages with the new libgit2 based package system? I have a fair number of private packages on a local server, and it sure would be nice to be able to fetch those with SSH key authentication.

[julia-users] Error: Kernel dead in Jupyter

Hi, I installed Julia in a windows 10 machine and I get the following error when I try to run IJulia notebook. I could not find any related post. using Julia version 4.5. I got very pissed at Windows and went ahead did a clean reinstall but the problem will not go away. Surely it is a bug somew

[julia-users] Re: [ECCN] What packages are included in the basic distribution of Julia

Seems Julia is in the clear, even with git. I only looked at the SHA1/git side now, and assuming, git only has SHA1: http://www.apache.org/dev/crypto.html "DO DIGEST ALGORITHMS SUCH AS MD5 AND SHA1 REQUIRE NOTIFICATION? No. One-way algorithms

[julia-users] Re: [ECCN] What packages are included in the basic distribution of Julia

On Friday, March 18, 2016 at 4:39:38 PM UTC, Tony Kelman wrote: > > No Julia packages are distributed by default, only the Julia code included > in Base. > > libgit2 is only used on julia master, command line git is included and > used on julia 0.4 and earlier versions. > I found: http://sta

Re: [julia-users] ANN: Quantumoptics.jl - a quantum optics toolbox for julia

Thank you very much! It's always comforting to get nice feedback after a tedious work like writing documentation - especially if it helps to fix a mistake. On Saturday, March 26, 2016 at 9:46:45 AM UTC+1, Eric Forgy wrote: > > Hi Sebastian, > > I agree with the above comments. The documentation

RE: [julia-users] Announcing JuDE: autocomplete and jump to definition support for Atom

I think the user experience for almost all people would be much easier if they install Atom, then add one package for julia support and everything works. Having multiple Atom packages (julia-language, ink, julia-client, latex-completion and Jude right now) makes things confusing for most people,

Re: [julia-users] Re: Package build/clone error.

On Mon, Mar 28, 2016 at 1:21 PM, Tony Kelman wrote: > If you're not sure whether something is a bug, it's best to post to the > mailing list first. If someone who knows the subject well responds and says > "yes this looks like a bug, please open an issue" then it's fine to do so, > but better to g

[julia-users] Re: Package build/clone error.

If you're not sure whether something is a bug, it's best to post to the mailing list first. If someone who knows the subject well responds and says "yes this looks like a bug, please open an issue" then it's fine to do so, but better to get a second opinion from someone else on where the error m

Re: [julia-users] Interest in BandedMatrices.jl Package?

On Sunday, March 27, 2016 06:43:19 PM Sheehan Olver wrote: > I've split out the BandedMatrix data structure in BandedMatrices.jl: > > https://github.com/ApproxFun/BandedMatrices.jl > > Right now the entries are in the "wrong" order for LAPack: the rows are > stored (instead of the columns) .

Re: [julia-users] Almost at 500 packages!

Thanks Cedric and Tim. Tom (@tbreloff on github) had suggested status tags[0] for tagging packages on various criteria, or stages of development. He wrote a script which I merged, so it would be nice if various package authors submitted data for their package status. If this works, then perhaps we

[julia-users] Benders decomposition in HuMP

I have been developing a Benders decomposition model to solve the fixed-cost transportation problem. I want to introduce feasibility and optimality cuts. However, I don't know how to get the sum of infeasibilities after the phase I of the simplex method in case of an infeasible problem. I am us

[julia-users] Package build/clone error.

My specific issue: I would like to add the "CHull" package ( https://github.com/davidavdav/CHull.jl) but cannot install it from github. I first tried just Pkg.add("CHull") but this did not work; I suspect this is because CHull is not in the list of files in metadata.jl (although please correc

Re: [julia-users] hdf5 not installing on linux cluster node

Le lundi 28 mars 2016 à 10:16 -0400, Erik Schnetter a écrit : > Thanks! The GSL code should really be an example for BinDeps. Yes, I figured I would add it to the docs, as Cairo is quite complex for a first contact with BinDeps: https://github.com/JuliaLang/BinDeps.jl/pull/208 Regards > For the

Re: [julia-users] hdf5 not installing on linux cluster node

See . -erik On Mon, Mar 28, 2016 at 10:16 AM, Erik Schnetter wrote: > Thanks! The GSL code should really be an example for BinDeps. > > For the record, these two lines install GSL from source: > ```Julia > # build from source > provides(Sources, URI

Re: [julia-users] hdf5 not installing on linux cluster node

Thanks! The GSL code should really be an example for BinDeps. For the record, these two lines install GSL from source: ```Julia # build from source provides(Sources, URI("http://ftp.gnu.org/gnu/gsl/gsl-1.16.tar.gz";), libgsl) provides(BuildProcess, Autotools(libtarget = "libgsl.la"), libgsl) ```

Re: [julia-users] hdf5 not installing on linux cluster node

At the moment, you need to install HDF5 outside of Julia, e.g. using "apt-get" or "yum". Julia will then pick up this HDF5 installation. Package managers usually split packages into many small pieces, so you might have to install three or five packages. Look for packages named "hdf5" or "libhdf5",

Re: [julia-users] Almost at 500 packages!

I wonder if anyone has ever tried to build a "package2vec" or even "function2vec", along the lines of "word2vec". There's so much usable information in code and comments, and it would be cool to provide that info in an Atom tooltip, say. On Monday, March 28, 2016 at 9:40:59 AM UTC-4, Tim Holy w

Re: [julia-users] hdf5 not installing on linux cluster node

Can anyone provide more basic instructions? I am still not sure how to proceed.

Re: [julia-users] Almost at 500 packages!

I sometimes find https://github.com/svaksha/Julia.jl can be useful. It's a hard problem to keep all this organized, though, so more searchable tools might be a good idea. I'm not aware of any efforts to provide this, so it seems like an area ripe for an impactful contribution! --Tim On Monday,

Re: [julia-users] GUI toolkits for Julia: Gtk vs Tk vs others

Gtk.jl has a more sophisticated internal design and is sometimes 5x or so faster for certain time-consuming rendering operations. For new code, I would certainly recommend it over Tk.jl, and I am slowly migrating my older visualization packages to it. Best, --Tim On Sunday, March 27, 2016 05:0

Re: [julia-users] hdf5 not installing on linux cluster node

Le lundi 28 mars 2016 à 08:41 -0400, Erik Schnetter a écrit : > The HDF5 package does not know how to install HDF5 on Linux; it > requires HDF5 to be installed on the system. (This is different e.g. > on Darwin, where it can install its own HDF5 library.) The solution is > to update the HDF5 packag

Re: [julia-users] Almost at 500 packages!

Jean-François, have you looked at Julia.jl? No tags, but it's nicely categorized. On Sunday, March 27, 2016 at 10:11:16 PM UTC-4, Jean-François Baffier wrote: > > Maybe I missed a a more recent thread but I would like how things are > going for the packages organization. It is currently hard to

Re: [julia-users] hdf5 not installing on linux cluster node

The HDF5 package does not know how to install HDF5 on Linux; it requires HDF5 to be installed on the system. (This is different e.g. on Darwin, where it can install its own HDF5 library.) The solution is to update the HDF5 package to use BinDeps to install HDF5 from source. Installing HDF5 is just

[julia-users] Re: GUI toolkits for Julia: Gtk vs Tk vs others

A better indication of what to go for may be looking at the testable packages which depend on Gtk, Tk & etc. A lot of good work is put into Gtk, but the hours we spend on resolving issues (on Windows and OsX) are not so interesting. I think Gtk needs a base of users who at least reports issues

Re: [julia-users] GUI toolkits for Julia: Gtk vs Tk vs others

Le dimanche 27 mars 2016 à 17:09 -0700, Daniel Carrera a écrit : > Hello, > > When it comes to GUI toolkits in Julia, Gtk seems to be the main > choice, followed by Tk. At least in terms of development effort: > > Gtk.jl -- 444 commits, 23 contributors > Tk.jl -- 235 commits, 28 contributors > Py

Re: [julia-users] Almost at 500 packages!

Skimming pkg.julialang.org is pleasantly impossible with 904 registered packages. This smells exponential growth. Is there a tool to simply download the text files in all of the packages? Last time I checked, I wasn't able to do that with github commands. I think it would be impossible to fore