Add Windows script support, setup.py installation support

requirements.txt

@@ -1,2 +1,3 @@
-PyEnchant
-nltk
+nltk
+six
+PyEnchant[full]

setup.py

@@ -0,0 +1,22 @@
+from setuptools import setup, find_packages
+import sotawhat
+
+setup(
+    name='sotawhat',
+    version=str(sotawhat.__VERSION__),
+    packages=find_packages(),
+    description='arxiv-sanity query script',
+    long_description=str('I often get frustrated searching for the latest '
+                         'research results on Google and Arxiv so I wrote '
+                         'SOTAwhat, a script to query Arxiv for the latest '
+                         'abstracts and extract summaries from them. '),
+    url='https://huyenchip.com/2018/10/04/sotawhat.html',
+    license="",
+    install_requires=['six', 'nltk'],
+    extras_require={
+        'full': ['PyEnchant'],
+    },
+    entry_points={
+        'console_scripts': ['sotawhat=sotawhat.cmd_line:main'],
+    }
+)

sotawhat/__init__.py

@@ -0,0 +1 @@
+__VERSION__ = '0.0.1'

sotawhat/cmd_line.py

@@ -0,0 +1,31 @@
+import os
+import argparse
+import warnings
+from sotawhat import sotawhat
+
+
+def main():
+    parser = argparse.ArgumentParser('Query arxiv-sanity to obtain most recent sota papers.')
+
+    parser.add_argument('query', type=str,
+                        help='Keyword to query')
+
+    parser.add_argument('-c', '--count', type=int, default=5,
+                        help='Number of results to display')
+
+    parser.add_argument('-e', '--exact', action='store_true', dest='exact',
+                        help='Assume query is exact, with no spelling mistakes')
+    parser.set_defaults(exact=False)
+
+    args = parser.parse_args()
+
+    if 'nt' in os.name:
+        try:
+            import win_unicode_console
+            win_unicode_console.enable()
+        except ImportError:
+            warnings.warn('On Windows, encoding errors may arise when displaying the data.\n'
+                          'If such errors occur, please install `win_unicode_consolde` via \n'
+                          'the command `pip install win-unicode-console`.')
+
+    sotawhat.main(args)

sotawhat.py → sotawhat/sotawhat.py

@@ -1,11 +1,23 @@
 import re
-import sys
+import os
 import urllib.error
 import urllib.request
-import enchant
-from nltk.tokenize import word_tokenize
 from six.moves.html_parser import HTMLParser
 
+import nltk
+from nltk.tokenize import word_tokenize
+
+try:
+    import enchant
+    _ENCHANT_AVAILABLE = True
+except ImportError:
+    _ENCHANT_AVAILABLE = False
+
+try:
+    nltk.data.find('tokenizers/punkt')
+except LookupError:
+    nltk.download('punkt')
+
 h = HTMLParser()
 
 AUTHOR_TAG = '<a href="/search/?searchtype=author'
@@ -129,7 +141,7 @@ def has_number(sent):
             return True
         try:
             value = int(token)
-        except:
+        except Exception:
             continue
         if (not is_citation_year(tokens, i)) and (not is_list_numer(tokens, i, value)):
             return True
@@ -175,7 +187,7 @@ def extract_line(abstract, keyword, limit):
 def get_report(paper, keyword):
     if keyword in paper['abstract'].lower():
         title = h.unescape(paper['title'])
-        headline = '{} ({} - {})\n'.format(title, paper['authors'][0], paper['date'])
+        headline = '{} ({} - {})\n'.format(title, h.unescape(paper['authors'][0]), paper['date'])
         abstract = h.unescape(paper['abstract'])
         extract, has_number = extract_line(abstract, keyword, 280 - len(headline))
         if extract:
@@ -214,21 +226,28 @@ def txt2reports(txt, keyword, num_to_show):
     return unshown, num_to_show, found
 
 
-def get_papers(keyword, num_results=5):
+def get_papers(keyword, num_results=5, args=None):
     """
     If keyword is an English word, then search in CS category only to avoid papers from other categories, resulted from the ambiguity
     """
 
-    if keyword in set(['GAN', 'bpc']):
+    if keyword in {'GAN', 'bpc'}:
         query_temp = 'https://arxiv.org/search/advanced?advanced=&terms-0-operator=AND&terms-0-term={}&terms-0-field=all&classification-computer_science=y&classification-physics_archives=all&date-filter_by=all_dates&date-year=&date-from_date=&date-to_date=&date-date_type=submitted_date&abstracts=show&size={}&order=-announced_date_first&start={}'
         keyword = keyword.lower()
     else:
         keyword = keyword.lower()
-        d = enchant.Dict('en_US')
-        if d.check(keyword):
-            query_temp = 'https://arxiv.org/search/advanced?advanced=&terms-0-operator=AND&terms-0-term={}&terms-0-field=all&classification-computer_science=y&classification-physics_archives=all&date-filter_by=all_dates&date-year=&date-from_date=&date-to_date=&date-date_type=submitted_date&abstracts=show&size={}&order=-announced_date_first&start={}'
+        if _ENCHANT_AVAILABLE:
+            d = enchant.Dict('en_US')
+            if d.check(keyword):
+                query_temp = 'https://arxiv.org/search/advanced?advanced=&terms-0-operator=AND&terms-0-term={}&terms-0-field=all&classification-computer_science=y&classification-physics_archives=all&date-filter_by=all_dates&date-year=&date-from_date=&date-to_date=&date-date_type=submitted_date&abstracts=show&size={}&order=-announced_date_first&start={}'
+            else:
+                query_temp = 'https://arxiv.org/search/?searchtype=all&query={}&abstracts=show&size={}&order=-announced_date_first&start={}'
         else:
-            query_temp = 'https://arxiv.org/search/?searchtype=all&query={}&abstracts=show&size={}&order=-announced_date_first&start={}'
+            if args.exact:
+                query_temp = 'https://arxiv.org/search/advanced?advanced=&terms-0-operator=AND&terms-0-term={}&terms-0-field=all&classification-computer_science=y&classification-physics_archives=all&date-filter_by=all_dates&date-year=&date-from_date=&date-to_date=&date-date_type=submitted_date&abstracts=show&size={}&order=-announced_date_first&start={}'
+            else:
+                query_temp = 'https://arxiv.org/search/?searchtype=all&query={}&abstracts=show&size={}&order=-announced_date_first&start={}'
+
     keyword_q = keyword.replace(' ', '+')
     page = 0
     per_page = 200
@@ -263,28 +282,12 @@ def get_papers(keyword, num_results=5):
         page += 1
 
 
-def main():
-    if len(sys.argv) < 2:
-        raise ValueError('You must specify a keyword')
-    if len(sys.argv) > 3:
-        raise ValueError("Too many arguments")
-
-    keyword = sys.argv[1]
-
-    if len(sys.argv) == 3:
-        try:
-            num_results = int(sys.argv[2])
-        except:
-            print('The second argument must be an integer')
-            return
-        if num_results <= 0:
-            raise ValueError('You must choose to show a positive number of results')
-
-    else:
-        num_results = 5
+def main(args):
+    keyword = str(args.query)
+    num_results = int(args.count)
 
-    get_papers(keyword, num_results)
+    if num_results <= 0:
+        raise ValueError('You must choose to show a positive number of results')
 
+    get_papers(keyword, num_results, args)
 
-if __name__ == '__main__':
-    main()