|
Hello, Rakesh Patil V. wrote: Hi...I'm keeping this on the list so everybody will benefit. The "tag" name in the XML becomes the ID when LAS starts and the XML is converted to the new form. E.g. <datasets> <xxxx name = "my data set" ... becomes <datasets> <dataset ID="xxxx" name="my data set" ... We don't use the file name for this because the ID of the data set must be unique for all data sets (as must the tag name by the way) installed in the LAS. If two data sets (/data/ds1/temperature.nc and /data/ds2/temperature.nc) are added with file names temperature.nc then the ID's assigned by the old addXML.pl code (temperature_nc) would not be unique and the XML would not work properly. You can supply a name for the data set when you invoke addXML.sh like this: addXML.sh -n /data/datasets/sla_try.nc -d:"Name in UI for users to select" or if the name you want to use for the name attribute is in a global attribute in the file use this: addXML.sh -n /data/datasets/sla_try.nc -g attribute_name
To me this seems irrelevant. The ID is how LAS refers to the data set internally, but doesn't play a role in the user interface. addXML.sh does not provide you any options for choosing the ID, but at the end of the day if you want a data set to have a particular ID you can always edit the file addXML.sh produces and set the ID to what ever you want. It will be your job to make sure they are globally unique for your LAS a job addXML.sh does for you automatically. No I mean to say this is a new version of the tool that does this particular job. addXML.pl is old unsupported code. Please don't use it. We cannot promise it will produce compatible XML. It will be removed from the distribution in a future release. Can't do it with our tools. You can only edit them if that's what you want. Roland
|