We address the `::setSize(0)` case by setting `array->element = NULL`
and `array->size = 0` before we destroy the elements.
Co-authored-by: Tyson Andre <tyson.andre@uwaterloo.ca>
Closes GH-7503.
Same as with the IteratorAggregate case, allow reusing get_iterator
if none of the Iterator methods are overridden. Drop the
REUSE_GET_ITERATOR flag that previously allowed ArrayIterator to
opt-in to unconditional get_iterator reuse, and drop the override
handling it did, in favor of the automated approach.
There were two separate bugs here:
* The get_iterator implementation did not match the Iterator
implementation. In particular, get_iterator did not respect
SKIP_DOTS.
* The constructor did not honor an explicitly omitted SKIP_DOTS
flag. It could still be unset through setFlags() though.
offsetSet did not account for the fact that the array may no longer exist after
the field is overwritten. This fixes that.
Add test of resizing both to the empty array and a smaller array - there should
be no valgrind warnings with a proper fix.
Alternate approach to #7486 (described in https://bugs.php.net/bug.php?id=81429)
Both the documentation and the stubs state that this method is
supposed to return false on failure. However, if the line read
(rather than the getcsv operation) fails, it would incorrectly
return null instead.
* Move spl_offset_convert_to_long() to spl_fixedarray.c
It is only used there, which explains its weird offset semantics
* Refactor SplFixedArray offset handling
- Implement warning for resource type
- Throw a proper TypeError instead of a RuntimeException
* Use a proper Error to signal that [] cannot be used with SplFixedArray
* Refactor SplFixedArray has_dimension helper
* Drop some ZPP tests
Updates the deprecation message for implicit incompatible float to int conversion from:
```
Implicit conversion from non-compatible float %.*H to int in %s on line %d
```
to
```
Implicit conversion from float %.*H to int loses precision in %s on line %d
```
Related: #6661
This format matches against null bytes, and prevents the test
expectation from being interpreted as binary data.
bless_tests.php will automatically replace \0 with %0 as well.
The object hash is not particularly useful (anymore) and just
clutters the output. It encodes the same information as the
object ID, which is already part of the output.
open_mode was initialized to "wb", but the length set only to 1.
The effect of this was that the stream is opened using "wb", but
we only report "w".
Fix the length to report the actually used open_mode.
`buf` may contain NUL bytes, so we must not use `strcspn()` but rather
a binary safe variant. However, we also must not detect a stray CR as
line ending, and since we only need to check line endings at the end
of the buffer, we can nicely optimize.
Co-authored-by: Nikita Popov <nikita.ppv@gmail.com>
Closes GH-6836.
For rationale, see #6787
Extensions migrated in part 4:
* simplexml
* skeleton
* soap
* spl
* sqlite3
* sysvmsg
* sysvsem
* tidy - also removed a check for an ancient dependency version
Christoph M. Becker