Related issue: #87
The tokenizer incorrectly handles plain scalars in cases like the following:
foo: some "text" more text
bar: some [text] more text
The second case fails because the tokenizer only checks in_flow when encountering opening flow indicators ([ and {), but not when encountering the corresponding closing tokens (] and }).
(
|
'\r', '\n', ' ', '\'', '"', ']', '}' => { |
)
As a result, the tokenizer breaks the literal flow even when it is not actually inside a flow sequence.
I noticed that #91 attempted to address this issue, but it appears to have been closed without much explanation.
The first case involving " is more complex due to the different contexts in which it can appear. However, introducing a state similar to in_flow (or extending the existing state handling) might help resolve this case as well.
Related issue: #87
The tokenizer incorrectly handles plain scalars in cases like the following:
The second case fails because the tokenizer only checks in_flow when encountering opening flow indicators ([ and {), but not when encountering the corresponding closing tokens (] and }).
(
zig-yaml/src/Tokenizer.zig
Line 288 in 84d747b
As a result, the tokenizer breaks the literal flow even when it is not actually inside a flow sequence.
I noticed that #91 attempted to address this issue, but it appears to have been closed without much explanation.
The first case involving " is more complex due to the different contexts in which it can appear. However, introducing a state similar to in_flow (or extending the existing state handling) might help resolve this case as well.