Merge pull request #54 from openeventdata/dev

Fix gazetteer and other issues

README.md

@@ -72,6 +72,26 @@ docker run -d -p 127.0.0.1:9200:9200 -v $(pwd)/geonames_index/:/usr/share/elasti
 See the [es-geonames](https://github.com/openeventdata/es-geonames) for the code used
 to produce this index.
 
+To update the index, simply shut down the old container, re-download the index
+from s3, and restart the container with the new index.
+
+Citing
+------
+
+If you use this software in academic work, please cite as 
+
+```
+@article{halterman2017mordecai,
+  title={Mordecai: Full Text Geoparsing and Event Geocoding},
+  author={Halterman, Andrew},
+  journal={The Journal of Open Source Software},
+  volume={2},
+  number={9},
+  year={2017},
+  doi={10.21105/joss.00091}
+}
+```
+
 How does it work?
 -----------------
 
@@ -93,8 +113,8 @@ from it.
 The training data for the two models includes copyrighted text so cannot be
 shared freely, but get in touch with me if you're interested in it.
 
-API
---------
+API and Configuration
+---------------------
 
 When instantiating the `Geoparser()` module, the following options can be changed:
 
@@ -148,7 +168,8 @@ Acknowledgements
 ----------------
 
 An earlier verion of this software was donated to the Open Event Data Alliance
-by Caerus Associates.  See [Releases](https://github.com/openeventdata/mordecai/releases) or the [legacy-docker](https://github.com/openeventdata/mordecai/tree/legacy-docker) branch for the
+by Caerus Associates.  See [Releases](https://github.com/openeventdata/mordecai/releases) 
+or the [legacy-docker](https://github.com/openeventdata/mordecai/tree/legacy-docker) branch for the
 2015-2016 and the 2016-2017 production versions of Mordecai.
 
 This work was funded in part by DARPA's XDATA program, the U.S. Army Research
@@ -159,28 +180,6 @@ recommendations expressed in this material are those of the authors and do not
 necessarily reflect the views of DARPA, ARO, Minerva, NSF, or the U.S.
 government.
 
-Citing
-------
-
-Send a note if you use Mordecai! It's always interesting to hear what people
-are doing with it and whether it's doing what they want it to.
-
-If you use this software in academic work, please cite as 
-
-Andrew Halterman, (2017). Mordecai: Full Text Geoparsing and Event Geocoding. *Journal of Open Source
-Software*, 2(9), 91, doi:10.21105/joss.00091
-
-```
-@article{halterman2017mordecai,
-  title={Mordecai: Full Text Geoparsing and Event Geocoding},
-  author={Halterman, Andrew},
-  journal={The Journal of Open Source Software},
-  volume={2},
-  number={9},
-  year={2017},
-  doi={10.21105/joss.00091}
-}
-```
 
 Contributing
 ------------

mordecai/__init__.py

@@ -1,3 +1,3 @@
 from .geoparse import Geoparser
 
-__version__ = "2.0.0a6"
+__version__ = "2.0.1"

mordecai/geoparse.py

@@ -27,7 +27,7 @@
 
 class Geoparser:
     def __init__(self, es_ip="localhost", es_port="9200", verbose = False,
-                country_threshold = 0.6, n_threads = 4):
+                country_threshold = 0.6, n_threads = 4, mod_date = "2018-06-05"):
         DATA_PATH = pkg_resources.resource_filename('mordecai', 'data/')
         MODELS_PATH = pkg_resources.resource_filename('mordecai', 'models/')
         self._cts = utilities.country_list_maker()
@@ -59,6 +59,9 @@ def __init__(self, es_ip="localhost", es_port="9200", verbose = False,
                  "Mordecai needs access to the Geonames/Elasticsearch gazetteer to function.",
                  "See https://github.com/openeventdata/mordecai#installation-and-requirements",
                  "for instructions on setting up Geonames/Elasticsearch")
+        es_date = utilities.check_geonames_date(self.conn)
+        if es_date != mod_date:
+            print("You may be using an outdated Geonames index. Your index is from {0}, while the most recent is {1}. Please see https://github.com/openeventdata/mordecai/ for instructions on updating.".format(es_date, mod_date))
 
     def _feature_country_mentions(self, doc):
         """

mordecai/tests/test_mordecai.py

@@ -2,6 +2,8 @@
 import sys
 import glob
 import json
+from elasticsearch_dsl import Q
+from ..utilities import structure_results
 #from ..geoparse import Geoparser
 
 import spacy
@@ -150,3 +152,11 @@ def test_issue_45(geo):
 capital city of Digos."""
     locs = geo.geoparse(text)
     assert len(locs) > 0
+
+def test_ohio(geo):
+    # This was a problem in issue 41
+    r = Q("match", geonameid='5165418')
+    result = geo.conn.query(r).execute()
+    output = structure_results(result)
+    assert output['hits']['hits'][0]['asciiname'] == "Ohio"
+

mordecai/utilities.py

@@ -247,3 +247,9 @@ def setup_es(es_ip, es_port):
     CLIENT = Elasticsearch([{'host' : es_ip, 'port' : es_port}])
     S = Search(using=CLIENT, index="geonames")
     return S
+
+def check_geonames_date(conn):
+    r = Q("match", geonameid='4943351')
+    result = conn.query(r).execute()
+    output = structure_results(result)
+    return output['hits']['hits'][0]['modification_date']

requirements.txt

@@ -1,10 +1,10 @@
-editdistance==0.3.1
+editdistance>=0.3.1
 elasticsearch==5.4.0
 elasticsearch-dsl==5.3.0
 h5py>=2.6.0
-Keras==2.0.8
-pandas==0.19.2
-spacy==2.0.3
-https://github.com/explosion/spacy-models/releases/download/en_core_web_lg-2.0.0/en_core_web_lg-2.0.0.tar.gz
-tensorflow==1.3.0
+Keras>=2.0.8
+pandas>=0.19.2
+spacy>=2.0.3
+#https://github.com/explosion/spacy-models/releases/download/en_core_web_lg-2.0.0/en_core_web_lg-2.0.0.tar.gz
+tensorflow>=1.3.0
 numpy>=1.12

setup.py

@@ -1,7 +1,7 @@
 from setuptools import setup
 
 setup(name='mordecai',
-      version='2.0.0a6',
+      version='2.0.1',
       description='Full text geoparsing and event geocoding',
       url='https://github.com/openeventdata/mordecai/',
       author='Andy Halterman',