[MOAB-dev] h5m: GLOBAL_ID

[Grindeanu, Iulian R.]

the start_id attribute refers to the unique file id (or  we call it sometimes unique entity id)  for each entity in an h5m format
There is no relation with the GLOBAL_ID tag.
file id is a moab convention
GLOBAL_ID corresponds slightly to global id for a cubit file
(so when you convert a cubit file, entities usually have global ids)

for moab, global ids are tags for entities, and usually global ids for vertices start from 1
global ids for hexas start from 1 too. They are just attributes/tags associated to the data, they can be set, reset, without modifying the mesh. (at your own risk:)

 file ids are unique identificators strictly related to moab hdf5 format
When sets are specified in h5m format, they are specified in terms of file ids (as they are unique, there is no confusion)
Also, connectivities are also specified using file ids (and not global ids for vertices, for example)

http://ftp.mcs.anl.gov/pub/fathom/moab-docs/h5mmain.html


________________________________
From: moab-dev-bounces at mcs.anl.gov [moab-dev-bounces at mcs.anl.gov] on behalf of Nico Schlömer [nico.schloemer at gmail.com]
Sent: Monday, November 09, 2015 4:34 AM
To: moab-dev at mcs.anl.gov
Subject: [MOAB-dev] h5m: GLOBAL_ID

Hi everyone,

I understand that in h5m, every entity is associated with a GLOBAL_ID. This ID is specified in the GLOBAL_ID array of the tags groups associated with, e.g., nodes or elements.

Now, I see that that in the example [1], the GLOBAL_ID array for nodes is
```
1 2 3 ... 409
```
and for elements it's
```
0 ... 758
```
To resolve the conflicts, the connectivity dataset of elements as the attribute `start_id = 410`, so I can just add that value to the above IDs and have them globally unique.

What's confusing me is that coordinates dataset of nodes also has an attribute `start_id` and it's value is `1`! So, no adding to GLOBAL_ID array in this case, for some reason?

Cheers,
Nico


[1] http://chunk.io/f/d24c8c0a3ebc413bbd7c40209767a21c
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/moab-dev/attachments/20151109/45c699e6/attachment.html>