Add WritingGoal class (#19)

* Add WritingGoal class

 - add WG<=>TT mapping to constants;
 - add test to make sure we can get the WGs correctly from the TTs;
 - add WG column to master CSV.

* Update readme with writing goal info

README.md

@@ -100,11 +100,11 @@ ABSAGE_SUBST
 This script is meant to provide an all-encompassing view of **all** rules created by the lingu team in **one single file** – this means *all locales* are present in the same file. For example:
 
 ```csv
-row,id,locale,source_repo,type,source_file,tone_tags,is_goal_specific
-0,IN_SHANGHAI,en,os,grammar,grammar.xml,,false
-1,IN_SHANGHAI,en,os,grammar,grammar.xml,,false
-2,LOWERCASE_NAMES,fr,premium,grammar,grammar.xml,,false
-3,SOME_EXAMPLE,es,os,style,grammar.xml,professional,true
+row,id,subId,locale,source_repo,type,source_file,tone_tags,writing_goals,is_goal_specific
+0,IN_SHANGHAI,1,en,os,grammar,grammar.xml,,,false
+1,IN_SHANGHAI,2,en,os,grammar,grammar.xml,,,false
+2,LOWERCASE_NAMES,1,fr,premium,grammar,grammar.xml,,,false
+3,SOME_EXAMPLE,1,es,os,style,grammar.xml,professional,serious,true
 ```
 
 #### Table headers
@@ -118,6 +118,7 @@ row,id,locale,source_repo,type,source_file,tone_tags,is_goal_specific
 |type|'grammar', 'style', or 'unknown'|
 |source_file|the path to the file where this rule comes from|
 |tone_tags|comma-separated list of `tone_tags` applied to the the rule, including those inherited from rulegroups and categories|
+|writing_goals|comma-separate list of writing goals served by the tone tags in the previous column|
 |is_goal_specific|boolean value of `is_goal_specific` rule attribute|
 
 

scripts/lib/constants.py

@@ -4,3 +4,11 @@
 LOCALES = {'de', 'es', 'en', 'fr', 'nl', 'pt'}
 REPOS = {'os': 'languagetool',
          'premium': 'languagetool-premium-modules'}
+
+WRITING_GOAL_MAPPING = {
+    "serious": ['clarity', 'confident', 'formal', 'general', 'positive', 'professional'],
+    "objective": ['academic', 'clarity', 'formal', 'general', 'objective', 'povrem', 'scientific'],
+    "confident": ['clarity', 'confident', 'general', 'persuasive', 'positive'],
+    "personal": ['clarity', 'general', 'informal', 'positive', 'povadd'],
+    "expressive": ['clarity', 'general']
+}

scripts/lib/writing_goals.py

@@ -0,0 +1,20 @@
+from typing import List
+from elements import ToneTag
+from constants import *
+
+
+class WritingGoal:
+    # This constructor isn't needed for now tbh
+    def __init__(self, name: str):
+        self.name = name
+        self.tags = WRITING_GOAL_MAPPING[name]
+
+    @staticmethod
+    # From a list of tone tags, return all the writing goals serviced by that tag
+    def list_from_tags(tags: List[ToneTag]):
+        goals = []
+        for tag in tags:
+            for goal, tag_list in WRITING_GOAL_MAPPING.items():
+                if tag.tag in tag_list and goal not in goals:
+                    goals.append(goal)
+        return goals

scripts/master_csv.py

@@ -4,6 +4,7 @@
 from lib.logger import logger_wrapper
 import os
 from pandas import DataFrame
+from lib.writing_goals import WritingGoal
 
 
 class CLI:
@@ -26,13 +27,15 @@ def __init__(self):
 # type (grammar, style, unknown)
 # source file
 # tone_tags
+# writing goals
 # is_goal_specific
 def __main__():
     cli = CLI()
     logger = logger_wrapper(cli.parser.prog, cli.args.verbosity)
     logger.debug(f"Starting script...\nInvoked with options: {cli.args}")
     out_path = path.join(cli.args.out_dir, 'all_rules.csv')
-    headers = ['id', 'subId', 'locale', 'source_repo', 'type', 'source_file', 'tone_tags', 'is_goal_specific']
+    headers = ['id', 'subId', 'locale', 'source_repo', 'type', 'source_file', 'tone_tags', 'writing_goals',
+               'is_goal_specific']
     rows = []
     for locale in LOCALES:
         for repo_name, repo_dir in REPOS.items():
@@ -43,6 +46,7 @@ def __main__():
                     rows.append([
                         rule.id, rule.sub_id, locale, repo_name, file.type, file.rel_path,
                         ','.join([tt.tag for tt in rule.tone_tags]),
+                        ','.join(WritingGoal.list_from_tags(rule.tone_tags)),
                         rule.is_goal_specific
                     ])
     df = DataFrame(rows, columns=headers)

tests/test_tone_tags.py

@@ -1,6 +1,7 @@
 import unittest
 from tests.mock import mock_rule
 from scripts.lib.elements import ToneTag
+from scripts.lib.writing_goals import WritingGoal
 
 
 class TestToneTags(unittest.TestCase):
@@ -26,6 +27,10 @@ def test_tone_tag_inheritance_repeated(self):
                          {'tone_tags': 'formal'})
         self.assertEquals(len(rule.tone_tags), 1)
 
+    def test_writing_goal(self):
+        self.assertEquals(len(WritingGoal.list_from_tags([ToneTag('clarity')])), 5)
+        self.assertEquals(len(WritingGoal.list_from_tags([ToneTag('formal'), ToneTag('informal')])), 3)
+
 
 if __name__ == '__main__':
     unittest.main()