**Problem:** `query.c` compares the current analysis state with the
previous analysis state to see if they are equal, so that it can return
early if so. This prevents redundant work. However, the comparison
function here differs from the one used for sorted insertion/lookup in
that it does not check any state data other than the child index. This
is problematic because it leads to infinite analysis when hidden nodes
have cycles.
**Solution:** Remove the custom comparison function, and apply the
insertion/lookup comparison function in place of it.
**NOTE:** This commit also changes the comparison function slightly, so
that some comparisons are reordered. Namely, for performance, it returns
early if the lhs depth is less than the rhs depth. Is this acceptable?
Tests still pass and nothing hangs in my testing, but it still seems
sketchy. Returning early if the lhs depth is greater than the rhs depth
does seem to make query analysis hang, weirdly enough... Keeping the
depth checks at the end of the loop also works, but it introduces a
noticeable performance regression (for queries that otherwise wouldn't
have had analysis cycles, of course).
Predicates/directives are documented to end in either `!` or `?`.
However, `query.c` allows them to be any valid identifier, and also
allows `?` or `!` characters anywhere inside an identifier.
This commit removes `?` and `!` from the list of valid identifier
characters, and asserts that predicates/directives only *end* in `?` or
`!`, respectively.
This commit is breaking because you can no longer do something like
`(#eq? @capture foo!bar)` (`foo!bar` must now be quoted).
**Problem:** After `ts_parser_parse_with_options()`, the parser options
are still stored in the parser object, meaning that a successive call to
`ts_parser_parse()` will actually behave like
`ts_parser_parse_with_options()`, which is not obvious and can have
unintended consequences.
**Solution:** Reset to empty options state after
`ts_parser_parse_with_options()`.
* Rename corpus test functions to allow easy filtering by language
* Use usize for seed argument
* Avoid retaining useless stack versions when reductions merge
We found this problem when debugging an infinite loop that happened
during error recovery when using the Zig grammar. The large number of
unnecessary paused stack versions were preventing the correct recovery
strategy from being tried.
* Fix leaked lookahead token when reduction results in a merged stack
* Enable running PHP tests in CI
* Fix possible infinite loop during error recovery at EOF
* Account for external scanner state changes when detecting changed ranges in subtrees
**Problem:** When resetting the parser during subtree balancing, an
error is thrown:
```
parser.c:2198: ts_parser_parse: Assertion `self->finished_tree.ptr' failed.
```
**Solution:** Reset `canceled_balancing` to false in
`ts_parser_reset()`.
Problem: Macros (re)defined in `endian.h` conflict with system headers
on FreeBSD (at least).
Solution: Rely on system `endian.h` on OpenBSD, FreeBSD, NetBSD, and
DragonFly
Ref. https://github.com/mikepb/endian.h/issues/4
This allows users to bail parsing if an error was *definitely* detected
using the progress callback, as all possible stack versions have a
non-zero error cost.
Co-authored-by: Amaan Qureshi <amaanq12@gmail.com>
