geotext

0.4.0 verified Fri May 01 auth: no python maintenance

Geotext extracts country and city mentions from text using a simple word-matching approach (not ML). Current version 0.4.0, last updated 2018, project effectively in maintenance mode. Does not support Python 3.6+ out of the box.

pip install geotext

Common errors

error ModuleNotFoundError: No module named 'geotext' ↓

cause Wrong import: import geotext instead of from geotext import GeoText

fix

Use: from geotext import GeoText

error ImportError: No module named enum ↓

cause Python 2.7 environment missing enum backport. geotext uses enum.

fix

Install enum34: pip install enum34, or switch to Python 3.4+.

Warnings

deprecated Project unmaintained since 2018, no Python 3.6+ compatibility fixes. Use Python 2.7 or backport yourself. ↓

fix Consider alternatives: geonamescache, countryinfo, or spacy geo components.

gotcha GeoText does lowercase matching by default, so 'PARIS' or 'Paris' work, but mixed case may cause misses. ↓

fix Normalize input text to lowercase before passing to GeoText if needed.

Install

pip install git+https://github.com/elyase/geotext.git

Imports

GeoText
wrong
```
import geotext
```
correct
```
from geotext import GeoText
```
Module does not expose GeoText at top-level. Must use from-import.

Quickstart

Extract city and country mentions from text.

from geotext import GeoText

places = GeoText("I live in Paris and work in Berlin")
print(places.cities)  # ['Paris', 'Berlin']
print(places.countries)  # []