Add special handling in PycodeSerializer for bytes

[SrirachaHorse]

📒 Description

Adds special handling to PycodeSerializer to ensure that byte strings are not modified and that their representation is handled literally.

Resolves #878

🔗 What I've Done

Update literal_value() to add special handling for byte strings, so that the literal repr() of the string is used, instead of falling back to the default call to repr() which subsequently replaces single quotes in the resulting string.

Also adds a quick unit test to exhibit the fix.

💬 Comments

🛫 Checklist

• Updated docs
• Added unit-tests
• Sample tests pass
• W3C tests pass

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (0b2fe8b) 99.89% compared to head (55fad70) 99.89%.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #879   +/-   ##
=======================================
  Coverage   99.89%   99.89%           
=======================================
  Files         105      105           
  Lines        9304     9307    +3     
  Branches     2079     2080    +1     
=======================================
+ Hits         9294     9297    +3     
  Misses          3        3           
  Partials        7        7           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[tefra]

Can you maintain the double quotes, please?

[SrirachaHorse]

Can you maintain the double quotes, please?

Bit confused - based on the unit test I think the repr() of a bytes object is printing as double quotes already. e.g. repr(b'a') will be b"a". Is there something I'm missing?

[tefra]

Can you maintain the double quotes, please?

Bit confused - based on the unit test I think the repr() of a bytes object is printing as double quotes already. e.g. repr(b'a') will be b"a". Is there something I'm missing?

Python 3.11.4+ (heads/3.11:ff5dd9d, Jul 30 2023, 17:40:33) [GCC 11.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> foo = b'a'
>>> repr(foo)
"b'a'"
>>>

from the w3c-tests

--- a/output/instances/ibmData/valid/D3_3_16/d3_3_16v01.py
+++ b/output/instances/ibmData/valid/D3_3_16/d3_3_16v01.py
@@ -3,6 +3,6 @@ from output.models.ibm_data.valid.d3_3_16.d3_3_16v01_xsd.d3_3_16v01 import Root

 obj = Root(
     ele=[
-        b"\x124V",
+        b'\x124V',
     ]
 )
diff --git a/output/instances/ibmData/valid/D3_3_17/d3_3_17v01.py b/output/instances/ibmData/valid/D3_3_17/d3_3_17v01.py
index 1d9122686..1db8afcf3 100644
--- a/output/instances/ibmData/valid/D3_3_17/d3_3_17v01.py
+++ b/output/instances/ibmData/valid/D3_3_17/d3_3_17v01.py
@@ -3,6 +3,6 @@ from output.models.ibm_data.valid.d3_3_17.d3_3_17v01_xsd.d3_3_17v01 import Root

 obj = Root(
     ele=[
-        b"\xd7m\xf8\xe7\xae\xfc",
+        b'\xd7m\xf8\xe7\xae\xfc',
     ]
 )

[SrirachaHorse]

Ah whoops, got it.

I've added a fix that should maintain the double quotes for bytes. It's a bit messier than I was hoping, just because there's the chance that a byte string could contain both single quotes and double quotes, which makes a blind replace() call much more difficult. I've expanded out the unit test to validate that case.

[sonarqubecloud]

Quality Gate passed

Kudos, no new issues were introduced!

0 New issues
0 Security Hotspots
No data about Coverage
0.0% Duplication on New Code

See analysis details on SonarCloud

[tefra]

Thank you1