You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The changes propose a better (but still not flawless) approach to
parsing the Label attribute.
Re ECFLOW-1890
Explanation of the Problem:
A Label attribute is stores in the .def file in the following format:
> label <name> "<value>" # "<new-value>"
with the double quotes and hash symbol used as separators.
This approach works fine while <value> or <new-value> do not contain
any double quotes or hash symbols.
When the values contain double quotes or hash symbols, the parsing
becomes ambiguous. In the following example
> label <name> "value X " # " and Y " # " value Z"
Should the parser consider
<value> = "value X " # " and Y "
<new-value> = " value Z"
or
<value> = "value X "
<new-value> " and Y " # " value Z"
?
Both might be correct depending on the user's intents!
This becomes even more convoluted when the value contains double quotes,
as in this example:
label <name> " "value X" " # " # " # " "value Y" "
0 commit comments