Hi Jeremy, While I only have experience with PyTables, I found that it has a very nice to use, Pythonic API. And it is for lovers [1].
Hope this helps, Matt [1] http://www.youtube.com/watch?v=Nzx0HAd3FiI On Thu, Dec 5, 2013 at 11:02 AM, Jeremy Sanders <[email protected]> wrote: > Dear all > > I'm currently reworking the import code (started yesterday, actually). I'm > trying to reorganise it so that all the code is defined in one place (i.e. > the Operation class, the LinkedFile class, the command line command > [necessary for saved files] and the code for the import dialog). This should > make it much easier to write or experiment with a non-plugin import. Plugin > imports are generally ok, except you don't have much user interface > flexibility. Maybe the new code should work in a week or two (see my > import_improvement branch, which doesn't work yet). > > If I were to write some code, I'd like to know exactly what you'd like to > get out of the HDF file (as Steve says). It would be very helpful to have > some example files, and some idea what you'd like to get out of the file. > I've never used the format, so it's hard to know what common operations > should be and how to streamline the UI. > > Also, if anyone has experience reading HDF5 in Python, which is the better > python module to use? > > Thanks > > Jeremy > > > > On 05/12/13 16:09, Steve Richardson wrote: >> >> Hi, >> >> I've found that making custom data import plugins for Veusz is pretty >> easy. My method consists of two parts: >> >> 1) Write python code that reads your data format. Get this working as a >> standalone python script. Doing this first, and separate from any Veusz >> coding, lets you find and fix bugs easily. >> 2) Write the Veusz plugin for your data format, following the example at >> http://barmag.net/veusz-wiki/ImportPlugins as a guide. This code will >> call the functions/classes you wrote in step 1. >> >> I've done this for a few custom data formats that I work with. >> >> There are several issues that you will probably encounter with HDF5. It >> is a very flexible file format, which is good for data storage, but will >> make it harder for you to write code to read your files. There are >> several python modules that can help; I'd take a look at >> http://www.h5py.org and http://www.pytables.org since they both use HDF5. >> >> Regardless of what approach you take, there will probably end up being >> some limitations with what you can do with HDF5. Without getting into >> the guts of Veusz (to define new data formats), you will probably not be >> able to do things with HDF5 that would be useful for large datasets. >> E.g., from the h5py site, one of the features is being able to "slice >> multi-terabyte datasets stored on disk, as if they were real NumPy >> arrays." I'd love to be corrected, but I don't think you will be able to >> do that with a Veusz plugin. >> >> As I get some time, I'll probably take a shot at making a simple HDF5 >> plugin, since it would be useful for me, too. If I come up with >> anything, I'll be sure to share it. >> >> Steve >> >> >> On Wed, Dec 4, 2013 at 4:06 PM, J Lister <[email protected] >> <mailto:[email protected]>> wrote: >> >> I'm really needing some HDF5 read in support in this. I know many >> aerospace engineers that would enjoy using this package, but cannot >> due >> their data being in HDF5. I would be happy to work on a plug-in or >> something, but I need more direction. >> >> >> _______________________________________________ >> Veusz-discuss mailing list >> [email protected] <mailto:[email protected]> >> https://mail.gna.org/listinfo/veusz-discuss >> >> >> >> >> >> _______________________________________________ >> Veusz-discuss mailing list >> [email protected] >> https://mail.gna.org/listinfo/veusz-discuss >> > > > _______________________________________________ > Veusz-discuss mailing list > [email protected] > https://mail.gna.org/listinfo/veusz-discuss _______________________________________________ Veusz-discuss mailing list [email protected] https://mail.gna.org/listinfo/veusz-discuss
