search

pkmcfarland / TSTools

Python tools for GPS position time series.
3 stars 1 forks source link

new timeSeries.TimeSeries.readPosFile() function #16

Open pkmcfarland opened 4 years ago

pkmcfarland commented 4 years ago

The timeSeries.TimeSeries class holds information associated with GNSS position time series as well as functions to read that information in from common time series file types and functions to do some manipulations of the time series. There already exists a function to read in data from a UNR-style .txyz2 time series file. It would be great to have a similar function to read in time series from the PBO-style .pos files.

An example file for station AB01 from the UNAVCO archive can be found at this page

pkmcfarland commented 4 years ago

Just realized that, as written, the timeSeries.TimeSeries class only stores positions in XYZ, dXdYdZ, or ENU during a given instantiation of a TimeSeries object. The .pos file format gives positions in XYZ and ENU in the same file. So either the user would have to pick what kind of coordinates they want or we would have to re-write the TimeSeries class to be able to hold both types of coordinates at once.

At this point I'm leaning toward the latter option...

bstressler commented 4 years ago

Created a new branch and added to the repo. Preliminary class and reading functions are written- currently using pandas to parse and store data. Still need to add:

pkmcfarland commented 4 years ago

Sounds good. Thanks Bryan. I just re-read this thread and remembered that I was going to update the TimeSeries class so that it could hold more than one type of coordinate time series (i.e. XYZ, dXdYdZ, and ENU all at the same time). But I probably won't be able to get to that until after my defense to be honest. Here's what I'm thinking to get around that. Since you're writing a PBO pos file class, maybe the function that converts the file class to a timeSeries.TimeSeries object can take an argument where the user specifies what type of coordinate they want.

What do you think?

bstressler commented 4 years ago

That’s exactly what I was thinking. It’ll be easy to swap the data from the pandas data frame to the time series object. I’ll check out the existing code and do my best to mimic it.

On Mon, Dec 30, 2019 at 5:22 PM Phillip K McFarland < notifications@github.com> wrote:

Sounds good. Thanks Bryan. I just re-read this thread and remembered that I was going to update the TimeSeries class so that it could hold more than one type of coordinate time series (i.e. XYZ, dXdYdZ, and ENU all at the same time). But I probably won't be able to get to that until after my defense to be honest. Here's what I'm thinking to get around that. Since you're writing a PBO pos file class, maybe the function that converts the file class to a timeSeries.TimeSeries object can take an argument where the user specifies what type of coordinate they want.

What do you think?

— You are receiving this because you were assigned. Reply to this email directly, view it on GitHub https://github.com/twistmcflip/TSTools/issues/16?email_source=notifications&email_token=ANAI5TMWK73IRXPN4HH47TLQ3JYBDA5CNFSM4JKI3JEKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEH3KU5I#issuecomment-569813621, or unsubscribe https://github.com/notifications/unsubscribe-auth/ANAI5TJ6CENVFGGIG6FKOVLQ3JYBDANCNFSM4JKI3JEA .

pkmcfarland commented 4 years ago

Awesome. Thanks!