Python package for processing Filemaker databases following the structure of Malchukov & Haspelmath's Ditranstive constructions data base.
This package reads the Excel export from the Filemaker database and converts it into the CLDF format.
Ditrans2cldf can be installed using pip:
$ pip install ditrans2cldf
pydictionaria is built on top of cldfbench. You can set up a ditrans2cldf project using the provided cldfbench template.
$ cldfbench new --template=ditransitive_db
The conversion process involves exporting the views in the Filemaker database as
Excel files and putting them into the raw/
folder of the cldfbench. See
doc/export-tutorial.pdf
for a list of table columns
that you need to export. The exported Excel files need then to be converted to
CSV (unfortunately Filemaker's builtin CSV/TSV exports lack column headers,
otherwise we would have used those directly). After that you can run makecldf
to generate the CLDF data.
$ # convert excel to csv
$ cldfbench download cldfbench_*.py
$ # generate cldf data
$ cldfbench makecldf cldfbench_*.py
ditrans2cldf supports a configuration file in the etc/
folder of the
cldfbench, which gives more control over which Filemaker columns map to which
CLDF columns. For all supported options refer to
doc/configuration-file.md
.