robust http/https parsing

Author: jxmorris12

README.md

@@ -1,6 +1,8 @@
 # `language_tool_python`: a grammar checker for Python 📝
 
-<img src="https://badge.fury.io/py/language-tool-python.svg" alt="language_tool_python pypi badge" />
+![language tool python on pypi](https://badge.fury.io/py/language-tool-python.svg)
+
+![Test with PyTest](https://github.com/jxmorris12/language_tool_python/workflows/Test%20with%20PyTest/badge.svg)
 
 This is a Python wrapper for [LanguageTool](languagetool.org). LanguageTool is open-source grammar tool, also known as the spellchecker for OpenOffice. This library allows you to make to detect grammar errors and spelling mistakes through a Python script or through a command-line interface.
 

language_tool_python/download_lt.py

@@ -16,7 +16,7 @@
 
 from distutils.spawn import find_executable
 from urllib.parse import urljoin
-from .utils import get_download_directory
+from .utils import get_language_tool_download_path
 
 # Create logger for this file.
 logging.basicConfig(format='%(message)s')
@@ -131,7 +131,7 @@ def download_zip(url, directory):
     logger.info('Downloaded {} to {}.'.format(url, directory))
 
 def download_lt(update=False):
-    download_folder = get_download_directory()
+    download_folder = get_language_tool_download_path()
     assert os.path.isdir(download_folder)
     old_path_list = [
         path for path in

language_tool_python/server.py

@@ -14,7 +14,6 @@
 
 DEBUG_MODE = False
 
-
 # Keep track of running server PIDs in a global list. This way,
 # we can ensure they're killed on exit.
 RUNNING_SERVER_PROCESSES = []
@@ -37,8 +36,8 @@ class LanguageTool:
     def __init__(self, language=None, motherTongue=None, remote_server=None):
         if remote_server is not None:
             self._remote = True
-            if remote_server[-1] == '/': remote_server = remote_server[:-1]
-            self._url = '{}/v2/'.format(remote_server)
+            self._url = parse_url(remote_server)
+            self._url = urllib.parse.urljoin(self._url, 'v2/')
             self._update_remote_server_config(self._url)
         elif not self._server_is_alive():
             self._start_server_on_free_port()
@@ -47,6 +46,9 @@ def __init__(self, language=None, motherTongue=None, remote_server=None):
                 language = get_locale_language()
             except ValueError:
                 language = FAILSAFE_LANGUAGE
+        # if self._url[-1] != '/':
+            # self._url += '/'
+        print('self._url:', self._url)
         self._language = LanguageTag(language, self._get_languages())
         self.motherTongue = motherTongue
         self.disabled_rules = set()

language_tool_python/utils.py

@@ -4,6 +4,7 @@
 import os
 import re
 import sys
+import urllib.parse
 import urllib.request
 
 from .backports import subprocess
@@ -41,6 +42,13 @@ class PathError(LanguageToolError):
 
 cache = {}
 
+def parse_url(url_str):
+    """ Parses a URL string, and adds 'http' if necessary. """
+    if 'http' not in url_str:
+        url_str = 'http://' + url_str   
+
+    return urllib.parse.urlparse(url_str).geturl()
+
 def correct(text: str, matches: [Match]) -> str:
     """Automatically apply suggestions to the text."""
     ltext = list(text)
@@ -58,7 +66,7 @@ def correct(text: str, matches: [Match]) -> str:
         correct_offset += len(repl) - len(errors[n])
     return ''.join(ltext)
 
-def get_download_directory():
+def get_language_tool_download_path():
     # Get download path from environment or use default.
     download_path = os.environ.get('LTP_PATH', '~/.cache/language_tool_python/')
     download_path = os.path.expanduser(download_path)
@@ -68,7 +76,7 @@ def get_download_directory():
 
 def get_language_tool_directory():
     """Get LanguageTool directory."""
-    download_folder = get_download_directory()
+    download_folder = get_language_tool_download_path()
     assert os.path.isdir(download_folder)
     language_tool_path_list = [
         path for path in

tests/test_major_functionality.py

@@ -5,6 +5,11 @@ def test_langtool_load():
 	matches = lang_tool.check('ain\'t nothin but a thang')
 	assert str(matches) == """[Match({'ruleId': 'UPPERCASE_SENTENCE_START', 'message': 'This sentence does not start with an uppercase letter', 'replacements': ['Ain'], 'context': "ain't nothin but a thang", 'offset': 0, 'errorLength': 3, 'category': 'CASING', 'ruleIssueType': 'typographical'}), Match({'ruleId': 'MORFOLOGIK_RULE_EN_US', 'message': 'Possible spelling mistake found.', 'replacements': ['nothing', 'no thin'], 'context': "ain't nothin but a thang", 'offset': 6, 'errorLength': 6, 'category': 'TYPOS', 'ruleIssueType': 'misspelling'}), Match({'ruleId': 'MORFOLOGIK_RULE_EN_US', 'message': 'Possible spelling mistake found.', 'replacements': ['than', 'thing', 'hang', 'thank', 'Chang', 'tang', 'thong', 'twang', 'Thant', 'thane', 'Thanh', 't hang', 'than g', 'Shang', 'Zhang'], 'context': "ain't nothin but a thang", 'offset': 19, 'errorLength': 5, 'category': 'TYPOS', 'ruleIssueType': 'misspelling'})]"""
 
+def test_langtool_languages():
+	import language_tool_python
+	lang_tool = language_tool_python.LanguageTool("en-US")
+	assert lang_tool._get_languages() == {'ca', 'de-CH', 'ja', 'pt-BR', 'sk', 'sl', 'uk', 'ru-RU', 'tl-PH', 'ta', 'pl-PL', 'pt', 'sl-SI', 'eo', 'it', 'ar', 'km', 'uk-UA', 'de-DE', 'da', 'en-GB', 'ta-IN', 'ca-ES', 'en-NZ', 'ga', 'ro-RO', 'pt-AO', 'br', 'ast-ES', 'pt-MZ', 'en', 'da-DK', 'ca-ES-valencia', 'en-AU', 'zh-CN', 'gl-ES', 'ga-IE', 'zh', 'de-DE-x-simple-language', 'be', 'be-BY', 'ar-DZ', 'ru', 'nl', 'en-CA', 'sv', 'fr', 'br-FR', 'en-ZA', 'ja-JP', 'tl', 'pt-PT', 'ro', 'ast', 'el-GR', 'en-US', 'de-AT', 'de', 'el', 'es', 'fa', 'pl', 'gl', 'sk-SK', 'km-KH'}
+
 def test_match():
 	import language_tool_python
 	tool = language_tool_python.LanguageTool('en-US')

tests/test_remote.bash

@@ -3,7 +3,8 @@
 set -eux
 
 readonly port=8081
-readonly jar='./language_tool_python/LanguageTool-*/languagetool-server.jar'
+readonly ltp_path=$(printf "import language_tool_python\nprint(language_tool_python.utils.get_language_tool_download_path())\n" | python)
+readonly jar="${ltp_path}LanguageTool-*/languagetool-server.jar"
 
 java -cp $jar org.languagetool.server.HTTPServer --port "$port" &
 java_pid=$!