tellervo-python
Python client to the Tellervo dendrochronology suite.
Note that this software is under development. It may eat your pet hamster.
Installation
Releases are in the Python Package Index. You can install the package with pip install tellervo-python. This package requires a copy of Python 3.3 or more recent, so on some linux distros you might need to use pip3.
Bleeding-edge development can be found on Github, if you like being in the danger zone.
Example
Here is a quick example that queries a Tellervo server with a search request. The server replies with TRiDaS XML. We gather the fields we want from the XML reply and stick the information into a list of dictionaries.
import tellervo as tel
USERNAME = 'username'
SERVER_URL = "https://tellervo.example.edu/ourserver/"
# Password is entered when we setup the connection below.
# Build a search query.
# Here we're looking for a comprehensive sample form any sample with a
# sampling date is after 2015-01-01.
search_query = tel.build_searchrequest(return_object = 'sample',
search_name = 'samplingdate',
search_operator = '>',
search_value = '2015-01-01',
results_format = 'comprehensive')
# Open connection to Tellervo server, throw our query at it, close connection.
# Note that using the `with as` statement will close the connection to the
# server, even if something goes wrong.
with tel.Connection(SERVER_URL, USERNAME, 'MyPassword!') as con:
reply = con.execute(search_query)
# Parse TRiDaS XML from server.
select_data = []
for site in reply.body.content.iterchildren():
sample_dict = {'site_name': site.title.text,
'tree_name': site.element.title.text,
'tree_taxon': site.element.taxon.get('normal'),
'sample_name': site.element.sample.title.text}
select_data.append(sample_dict)For more on parsing the XML replies, see lxml.
If you want to refine things further, we can easily take our select_data list and turn it into a pandas DataFrame. We then sort the data.
import pandas as pd
d = pd.DataFrame(select_data, dtype = 'str')
d = d.sort_values(by = ['site_name', 'tree_name', 'sample_name'])
