Update english_stenotype.py

[JorWat25]

Summary of changes

Added 'in'' as a suffix to the orthography rules. May not be formatted completely right.

Closes

Pull Request Checklist

• [ ] Changes have tests
• [ ] News fragment added in news.d. See documentation for details

[JorWat25]

The later changes might not actually be necessary, I now realize. But the first one is, as otherwise TKAOEU/*G writes 'diein'' rather than 'dyin''.

[benoit-pierre]

Can you add a couple of tests here: https://github.com/openstenoproject/plover/blob/master/test/test_orthography.py#L9.

[JorWat25]

Can you add a couple of tests here: https://github.com/openstenoproject/plover/blob/master/test/test_orthography.py#L9.

Have added some tests.

[benoit-pierre]

This need a news fragment (in news.d/feature/1343.core.md).

And you picked the wrong word (interpret) for testing (since interpretin' is not in the wordlist).

[JorWat25]

I'm sorry, but I don't know enough about how this code works to do this properly. Apologies for testing the wrong thing, I just took some of the words in the current test, and replaced 'ing' with 'in''.

[benoit-pierre]

The only reason interpret+ing -> interpreting works is because interpreting is in the wordlist, so without interpretin' being in the wordlist the orthography rules will yield interprettin'. Call it an irregular, that would need a dedicated dictionary entry.

So the question is: is the use case frequent enough, and are the orthography rules going to be right more often than not for this PR to be really useful.

[user202729]

If there were something like {delete_last_char} then {^in'} could be implemented as {^ing}{delete_last_char}{^'}.

(that having said, *G is defined as {^in'} in the dictionary in the current version)

Adding all the in' words into the word list sounds bad.

[JorWat25]

If there were something like {delete_last_char} then {^in'} could be implemented as {^ing}{delete_last_char}{^'}.

(that having said, *G is defined as {^in'} in the dictionary in the current version)

Adding all the in' words into the word list sounds bad.

There's {#BackSpace}, but that does have the problem of not being undoable.

[benoit-pierre]

How about this instead:

 plover/assets/main.json            |  2 +-
 plover/system/english_stenotype.py |  3 +++
 test/test_blackbox.py              | 18 +++++++++++++++++-
 3 files changed, 21 insertions(+), 2 deletions(-)

diff --git i/plover/assets/main.json w/plover/assets/main.json
index be14ac0d..8561e4e8 100644
--- i/plover/assets/main.json
+++ w/plover/assets/main.json
@@ -138064,7 +138064,7 @@
 "*BS": "action",
 "*L": "{^le}",
 "*LT": "{^let}",
-"*G": "{^in'}",
+"*G": "{^ing}{^'}",
 "*T": "{^th}",
 "*TS": "it's",
 "*S": "{^s}",
diff --git i/plover/system/english_stenotype.py w/plover/system/english_stenotype.py
index ac1cb918..5e26981b 100644
--- i/plover/system/english_stenotype.py
+++ w/plover/system/english_stenotype.py
@@ -138,6 +138,9 @@
     # == misc ==
     # defer + ed = deferred (consonant doubling)   XXX monitor(stress not on last syllable)
     (r'^(.*(?:[bcdfghjklmnprstvwxyz]|qu)[aeiou])([bcdfgklmnprtvz]) \^ ([aeiouy].*)$', r'\1\2\2\3'),
+
+    # ing + ' = in'
+    (r"(.+)ing \^ '$", r"\1in'",),
 ]
 
 ORTHOGRAPHY_RULES_ALIASES = {
diff --git i/test/test_blackbox.py w/test/test_blackbox.py
index e0644f4d..3e4d3a0c 100644
--- i/test/test_blackbox.py
+++ w/test/test_blackbox.py
@@ -1834,3 +1834,19 @@ def test_bug_1448_2(self, with_korean_system):
         :system 'English Stenotype'
         -G  ' 12345 -G'
         '''
+
+    def test_ing_contraction_1(self):
+        r'''
+        "TEFT": "test",
+        "*G": "{^ing}{^'}",
+
+        TEFT/*G  " testin'"
+        '''
+
+    def test_ing_contraction_2(self):
+        r'''
+        "TERP": "interpret",
+        "*G": "{^ing}{^'}",
+
+        TERP/*G  " interpretin'"
+        '''

This way, we take advantage of the existing rules / wordlist so we get the correct output, even for irregulars like “interpretin'”.

[TheTabletopGirl]

A new solution to this has been offered and seems like it would be a very simple way to add in' endings. Instead of it's own orthography rule, a single new dictionary entry of:

"*G": "{^ing}{^}\b'"

The in' ending simply adds ing (using the existing rules), backspaces one character and types ', effectively just replacing g with '.

The main.json dictionary itself seems to need a lot of updating and fixing, but this should be fine to add in the meantime.