[Bug][Python Migration] Added support for parsing inside list of lists #2670
Conversation
TejaX-Alaghari
force-pushed
the
py_mig
branch
from
936c9eb to
7b4fdae
Compare
There was a problem hiding this comment.
From an overall view, I think this PR is a little strange.
There are too many special processes in the parser.
For example: can we just pass MatchPattern into parseBlock and use it in the parseCodeElement call?
@tomflinda
| if (Character == '{') { | ||
| Index = parseBlock('{', '}', Input, Index, Mode); | ||
| continue; | ||
| } | ||
| if (isLeftDelimiter(Character)) { | ||
| char rightDelimiter = getRightDelimiter(Character); | ||
|
|
||
| if (Character == '[') { | ||
| Index = parseBlock('[', ']', Input, Index, Mode); | ||
| continue; | ||
| } | ||
|
|
||
| if (Character == '(') { | ||
| Index = parseBlock('(', ')', Input, Index, Mode); | ||
| continue; |
There was a problem hiding this comment.
And I think to parse the content inside, Index should be ++Index
There was a problem hiding this comment.
Index is getting incremented inside parseBlock here
Did you mean the same or more additional increment?
There was a problem hiding this comment.
Index is getting incremented inside parseBlock here
You are right.
@zhiweij1 yes, I think so. |
| if (Input[Pos] == '[') | ||
| Pos++; |
There was a problem hiding this comment.
Take the "out = func([["foo.cpp", "bar.cu", "baz.cpp"]])" for example.
Instead of skipping '{', we should locate the root cause of why the target "foo.cpp" could not be matched.
This PR adds support for parsing Python syntax inside list of lists like '[[a.cu]]'