In particular, if input has no newline at the end, we need to treat it
specially, because otherwise we will get confusing “incorrect
indentation” message.
Close#75.
Now accumulated hints are not used with ‘ParseError’ records that have
only custom messages in them (created with ‘Message’ constructor, as
opposed to ‘Unexpected’ or ‘Expected’). This strips “expected” line from
custom error messages where it's unlikely to be relevant anyway.
Arbitrary messages created with ‘Message’ constructor should not be
rendered as “or”-separated list. This commit makes every such message be
displayed on new line.
After some thinking I decided that this may be not desirable in some
cases, so we should not enable it by default. I've edited documentation
of ‘makeExprParser’ to explain why this doesn't work by default and how
to make it work.
Close#64.
‘makeExprParser’ now generates parser that can handle several
occurrences of the same prefix or postfix operator in a row. This allows
to parse something like C pointers (for example ‘**i’) without resorting
to hacks.
The feature is experimental, I'm not entirely sure it's not
buggy. Upcoming additional tests for ‘Text.Megaparsec.Expr’ will show
whether it behaves correctly in all cases and doesn't have adverse
effects. For now, I've edited existing test to generate data with
repeating prefix negations and postfix factorials. Current code-base
passes the test.
Close#69.
Although previously used syntax is correct Haskell syntax for multi-line
string literals, CPP extension that we need to use for compatibility
reasons obviously makes ‘\’ symbol escape following newline character
that leads to ‘\t’ being interpreted as tab character.
The proposed solution just concatenates result error message from list
of strings — the most lightweight and reliable solution in our case.
What Parsec used is called “FreeBSD” or “BSD 2 clause”. Addition of the
third clause may require contacting all the authors. To hell with it,
let it be “FreeBSD” (which is anyway better than “BSD-like”), I'm a
hacker, not a lawyer (tm).