Added reserved keywords#1741
Conversation
Codecov Report
@@ Coverage Diff @@
## master #1741 +/- ##
==========================================
+ Coverage 87.3% 87.32% +0.02%
==========================================
Files 48 48
Lines 5709 5720 +11
Branches 1512 1512
==========================================
+ Hits 4984 4995 +11
Misses 448 448
Partials 277 277
Continue to review full report at Codecov.
|
|
@charles-cooper 👍 (and unit tests pass) => can merge |
charles-cooper
left a comment
charles-cooper
left a comment
There was a problem hiding this comment.
can you add a test to validate that #1710 has been fixed? and then I think this is good to go
|
@charles-cooper went to go add tests, but ran into a ton of issues. And then I got into a refactoring rut, so #1746 came out. TL;DR is that we use inconsistent methods for representing constants |
fubuloubu
force-pushed
the
fubuloubu/reserved-keywords
branch
3 times, most recently
from
5d35c71 to
eeac89b
Compare
|
Found a bunch of uncaught |
yay for code quality :) |
fubuloubu
force-pushed
the
fubuloubu/reserved-keywords
branch
2 times, most recently
from
01bd2af to
5a9460d
Compare
|
A bunch of force pushing for linting issues lol |
| Conversion from error using ast.parse() | ||
| """ | ||
| def __init__(self, syntax_error: SyntaxError, source_code: str): | ||
| item = types.SimpleNamespace() # TODO: Create an actual object for this |
There was a problem hiding this comment.
TODO is here in case this object is more broadly useful (which I don't think it is)
Would also just be better if it were a static object
| try: | ||
| py_ast = python_ast.parse(reformatted_code) | ||
| except SyntaxError as e: | ||
| # TODO: Ensure 1-to-1 match of source_code:reformatted_code SyntaxErrors |
There was a problem hiding this comment.
TODO is here because theorhetically there is a scenario where this code reformatting could affect the location of offsets used to display an error. Would be best if the conversion was accounted for.
There was a problem hiding this comment.
This is interesting because syntax error reporting actually happens on the reformatted code, but other error reporting can happen on the untransformed code. Look at for instance how the struct -> class preparsing rule affects error reporting:
# syntax error
$ vyper -f ir /dev/stdin <<EOF
struct Foo: aoijewofijwef098 \`\`
EOF
File "<unknown>", line 1
class Foo: aoijewofijwef098 ``
^
SyntaxError: invalid syntax
During handling of the above exception, another exception occurred:
vyper.exceptions.PythonSyntaxException: line 1:29 SyntaxError: invalid syntax
---> 1 struct Foo: aoijewofijwef098 ``
------------------------------------^
# vs plain structure exception
$ vyper -f ir /dev/stdin <<EOF
struct Foo: aoijewofijwef098
EOF
Error compiling: /dev/stdin
vyper.exceptions.StructureException: line 1:11 Structs can only contain variables
---> 1 struct Foo: aoijewofijwef098
------------------^
2There was a problem hiding this comment.
Does this add buggy behavior now?
There was a problem hiding this comment.
no, the behavior is as buggy as before
- Changed filename to be more correct - Added built-in functions and reserved keywords to tests - Removed redundant testing
Note: Sort in pytest decorator instead
Note: RLPList in built-in functions was mixed case and caused an error
Co-Authored-By: Charles Cooper <cooper.charles.m@gmail.com>
fubuloubu
force-pushed
the
fubuloubu/reserved-keywords
branch
from
146c4f6 to
c5160de
Compare
Co-Authored-By: Charles Cooper <cooper.charles.m@gmail.com>
charles-cooper
left a comment
charles-cooper
left a comment
There was a problem hiding this comment.
I'm pretty happy with this. Thanks!
4 participants
What I did
Added some more things to the internal reserved keyword list
Fixes: #1710
How I did it
grep-foo
How to verify it
Added all reserved keywords to test. Added test for function names.
Description for the changelog
N/A
Cute Animal Picture