HI Again,
I guess I need a pointer from you. I have created by first HDF5 file, and
it contains a packet table (yeah).
When I tried to attach attributes, I got an error
HDF5-DIAG: Error detected in HDF5 (1.9.131) thread 0:
#000: H5A.c line 242 in H5Acreate2(): not a location
major: Invalid arguments to routine
minor: Inappropriate type
#001: H5Gloc.c line 253 in H5G_loc(): invalid object ID
major: Invalid arguments to routine
minor: Bad value
>From my searching on the Web and your kind warning below (I fell into the
hole anyway), I see
what you mean. However, I am not exactly sure of your solution. Do you
mean while I created
the packet table as
hid_t ptTable = H5PTcreate_fl (fileID, "PCS Weather Table", memType,
chunkSize, compress);
that I need to close the table, and then open the table as a dataset as
below?
hid_t ptAsDs = H5Dopen (ptTable, "PCS Weather Table", H5P_DEFAULT);
Use of ptTable does not work in this open statement. I tried using the
file identifier in the
H5Dopen in conjunction with the packet table name, and this seems to work
(I have other
errors). Is this the accepted way to do this?
Thanks,
Michele
On Tue, Oct 23, 2012 at 7:41 AM, Mitchell, Scott - IS <
scott.mitch...@exelisinc.com> wrote:
> Michele,
>
>
>
> I’m dealing with a similar timestamped data streams and use Packet Tables
> almost exclusively.
>
>
>
> My solution for the unit labels problem was to attach an attribute to the
> Packet Table. The Unit attribute is an array of strings, one per column.
>
>
>
> Another important trick, for HDF5.net at least, you need to open the
> Packet Table as a (H5D) dataset to attach/read an attribute.
>
>
>
> I’m not sure what you mean by (4) Packet Tables not having the concept of
> a column?
>
>
>
>
>
> Scott
>
>
>
> *From:* Hdf-forum [mailto:hdf-forum-boun...@hdfgroup.org] *On Behalf Of
> *Michele
> De La Pena
> *Sent:* Monday, October 22, 2012 5:54 PM
> *To:* hdf-forum@hdfgroup.org
> *Subject:* [Hdf-forum] Newbie questions regarding tables
>
>
>
> HI Folks,
>
> I am an HDF5 newbie, so please be patient. I have a few basic
> questions after reading the tutorials and such. First, I will say that
> I have telemetry data (timestamp with state information) which I need
> to stream to an HDF file. The data nicely fits the model of a table.
>
> My Sys Admin has installed HDF5 version 1.9.131.
>
> My first question is regard to your table APIs. I see that an HDF
> file can consist of the fundamental constructs of group and/or dataset.
> However, it seems the table and packetTable are also fundamental in
> that I do not have to package either of these tables inside of a group to
> make a valid HDF file. (1) Is this the way these are intended to be used?
>
> My problem with the regular table is that during its creation, it does
> not hand back an identifier so I cannot attach attributes directly to
> the table. I would like the columns in the table to have "units" and
> a "description" as meta-data for the table. I can put the regular
> table and an attributes into a group (where the attributes
> are attached to the group) which seems not too bad overall.
> (2) Is this what one would ordinarily do for this issue?
> (3) Maybe I made a mistake, but I cannot h5dump or h5stat the file I
> have generated with a regular table. Is this expected?
>
> (4) As for the packet table, I can attach attributes, but the packet table
> does not really have the concept of columns as far as I can tell. So
> is it assumed that meta-data will be handled in the attributes dataset
> which can attach to the packet table?
>
> Is it that I am just to bound by my conventional view of the world, and
> HDF is more general. Can I make the packetTable actually anything
> I want it to be via attributes?
>
> Thanks,
> Michele
>
> --
> Michele D. De La Pena Scientific Software Engineer
> Large Binocular Telescope Obs. email: mdelap...@lbto.org
> 933 N. Cherry Avenue phone: 520-626-1402
> Tucson, AZ 85721-0065 cell: 520-419-8825
>
> ------------------------------
>
> Email addresses of ITT Exelis employees have changed from itt.com to
> exelisinc.com. Please update your favorites and contact information to
> reflect these changes.
>
> This e-mail and any files transmitted with it may be proprietary and are
> intended solely for the use of the individual or entity to whom they are
> addressed. If you have received this e-mail in error please notify the
> sender. Please note that any views or opinions presented in this e-mail are
> solely those of the author and do not necessarily represent those of Exelis
> Inc. The recipient should check this e-mail and any attachments for the
> presence of viruses. Exelis Inc. accepts no liability for any damage caused
> by any virus transmitted by this e-mail.
>
> _______________________________________________
> Hdf-forum is for HDF software users discussion.
> Hdf-forum@hdfgroup.org
> http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org
>
>
--
Michele D. De La Pena Scientific Software Engineer
Large Binocular Telescope Obs. email: mdelap...@lbto.org
933 N. Cherry Avenue phone: 520-626-1402
Tucson, AZ 85721-0065 cell: 520-419-8825
_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@hdfgroup.org
http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org
