We prevent signed overflow by making the count unsigned. The actual
interpretation of the count doesn't matter as it's just used to denote a
limit.
The test output for some limit values looks strange though, so that may
need extra investigation. However, that's orthogonal to this fix.
Closes GH-18906.
This API can't handle references, yet everyone keeps forgetting that it
can't and that you should DEREF upfront. Fix every type of this issue
once and for all by moving the reference handling to this Zend API.
Closes GH-18761.
Conversion of floating point to integer values is undefined if the
integral part of the float value cannot be represented by the integer
type. We need to cater to that explicitly (in a manner similar to
`zend_dval_to_lval_cap()`).
Closes GH-17689.
The behaviour is weird in the sense that the reference must get
unwrapped. What ended up happening is that when destroying the old
reference the sources list was not cleaned properly. We add handling for
that. Normally we would use use ZEND_TRY_ASSIGN_STRINGL but that doesn't
work here as it would keep the reference and change values through
references (see bug #26639).
Closes GH-16272.
Updates UCD to Unicode 16.0 (released 2024 Sept).
Previously: 0fdffc18, #7502, #14680
Unicode 16 adds several new character sets and case folding rules.
However, the existing ucgendat script can still parse them.
This also adds a couple test cases to make sure the new rules for
East Asian Wide characters and case folding work correctly. These
tests fail on Unicode 15.1 and older because those verisons do not
contain those rules.
I fixed from strcasecmp to strncasecmp.
However, strncasecmp is specify size to #3 parameter.
Hence, Add check length to mime and aliases.
Co-authored-by: Niels Dossche <7771979+nielsdos@users.noreply.github.com>
Updates UCD to Unicode 15.1 (released 2023 Sept). The upcoming
Unicode 16 version will be released roughly on 2024 Sept.
Previously: 0fdffc18, #7502
UCD 15.1 `DerivedNormalizationProps` contains multiple properties in
the same line, which breaks the parser. This also updates the
`ucgendat.php` script to allow 2 or three fields in each line, and to
look for the `Cased` and `Case_Ignorable` properties in either of the
fields to mimic the previous behavior.
Because the default characters are defined in the stub file, and the
stub file is UTF-8 (typically), the characters are encoded in the string
as UTF-8. When using a different character encoding, there is a mismatch
between what mb_trim expects and the UTF-8 encoded string it gets.
One way of solving this is by making the characters argument nullable,
which would mean that it always uses the internal code path that has the
unicode codepoints that are defaulted actually stored as codepoint
numbers instead of in a string.
Co-authored-by: @ranvis
This also fixes skipped tests due to different naming "zend-test"
instead of "zend_test" and "PDO" instead of "pdo":
- ext/dom/tests/libxml_global_state_entity_loader_bypass.phpt
- ext/simplexml/tests/libxml_global_state_entity_loader_bypass.phpt
- ext/xmlreader/tests/libxml_global_state_entity_loader_bypass.phpt
- ext/zend_test/tests/observer_sqlite_create_function.phpt
EXTENSIONS section is used for the Windows build to load the non-static
extensions.
Closes GH-13276
When mbstring.encoding_translation=1, and PHP receives an (RFC1867)
form-based file upload, and the Content-Disposition HTTP header contains
a filename for the uploaded file, PHP will internally invoke mbstring
code to 1) try to auto-detect the text encoding of the filename, and if
that succeeds, 2) convert the filename to internal text encoding.
In such cases, the candidate text encodings which are considered during
"auto-detection" are those listed in the INI parameter
mbstring.http_input. Further, mbstring.http_input is one of the few
contexts where mbstring allows the magic string "pass" to appear in
place of an actual text encoding name.
Before mbstring's encoding auto-detection function was reimplemented,
the old implementation would never return "pass", even if "pass" was the
only candidate it was given to choose from. It is not clear if this was
intended by the original developers or not. This behavior was the result
of some rather subtle details of the implementation.
After mbstring's auto-detection function was reimplemented, if the new
implementation was given only one candidate to choose, and it was not
running in 'strict' mode, it would always return that candidate, even
if the candidate was the non-encoding "pass".
The upshot of all of this: Previously, if
mbstring.encoding_translation=1 and mbstring.http_input=pass, encoding
conversion of RFC1867 filenames would never be attempted. But after
the reimplementation, encoding 'conversion' would occur (uselessly).
Further, in December 2022, I reimplemented the relevant bit of
encoding conversion code. When doing this, I never bothered to
implement encoding/decoding routines for the non-encoding "pass",
because I thought that they would never be used. Well, in the one case
described above, those routines *would* have been used, had they
actually existed. Because they didn't exist, we get a nice NULL pointer
dereference and ensuing segfault instead.
Instead of 'fixing' this by adding encoding/decoding routines for the
non-encoding "pass", I have modified the function which the RFC1867
form-handling code invokes to auto-detect input encoding. This function
will never return "pass" now, just like the previous implementation.
Thanks to the GitHub user 'tstangner' for reporting this bug.
The previous version of the GB-18030 standard was published in 2005.
This commit adds support for the updated (2022) version of this text
encoding. The existing GB18030 implementation has been left unchanged
for backwards compatibility; users who want to use the new standard
must explicitly indicate the desired text encoding is 'GB18030-2022'.
The document which defines GB18030-2022, published by the government
of the People's Republic of China, defines three levels of standards
compliance. This implementation is intended to achieve Implementation
Level 3, which is the highest level of compliance.
Experts in the GB18030 standard are requested to assess this
implementation and report any deviation from the standard.
Thanks to Maurício Fauth for finding and reporting this bug.
The bug was introduced in October 2022. It originally only affected
text encodings which do not have a fixed byte width per characters
and for which mbstring does not have an mblen_table. However, I recently
made another change to mbstring, such that mb_substr no longer relies
on the mblen_table even if one is available. Because of this change,
the bug earlier introduced in October 2022 now affected a greater
number of text encodings, including UTF-8.
For GB18030, it is not generally possible to identify character
boundaries without scanning through the entire string. Therefore,
implement mb_strcut using a similar strategy as the mblen_table based
implementation in mbstring.c. The difference is that for GB18030, we
need to look at two leading bytes to determine the byte length of a
multi-byte character.
The new implementation is 4-5x faster for short strings, and more than
10x faster for long strings. (Part of the reason why this new code has
such a great performance advantage is because it is replacing code
based on the older text conversion filters provided by libmbfl, which
were quite slow.)
The behavior is the same as before for valid GB18030 strings; for
some invalid strings, mb_strcut will choose different 'cut' points
as compared to before. (Clang's libFuzzer was used to compare the
old and new implementations, searching for test cases where they had
different behavior; no such cases were found.)
Starting many years ago, libmbfl included a 'mblen_table' for selected
text encodings. This table allows looking up the byte length of a
(possibly multi-byte) character from the value of the first byte.
libmbfl uses these tables to optimize certain operations; if a
text-processing operation can be performed using an mblen_table,
it may not be necessary to decode the text to codepoints. Since
libmbfl's decoding filters are generally slow, this improves
performance.
Since mbstring is (or was) based on libmbfl, it has always used
these mblen_tables to implement some functions. This design has a
significant downside. Let me explain:
While some mbstring functions are implemented by converting input text
to codepoints and operating on the codepoints, others operate directly
on the original input bytes (using an mblen_table to identify character
boundaries). Both of these implementation styles, if correctly coded,
yield equivalent results on valid strings. However, on strings which
contain encoding errors, the results are often different.
When decoding byte strings to codepoints using some text encoding,
mbstring uses the non-existent codepoint 0xFFFFFFFF to represent a
byte sequence which cannot be decoded. Then, when mbstring indexes into
the resulting sequence of codepoints, the index of any particular
character depends on the number of such 'error markers' which were
produced during the decoding process. In contrast, when an mblen_table
is used to split a byte sequence into characters, there is no question
of counting encoding errors; rather, table lookups into the mblen_table
are used to repeatedly 'bite off' some number of bytes (which are
treated as one 'character'). In the presence of encoding errors, these
two methods of mapping between byte indices and character indices are
inherently different and will rarely agree.
(For completeness, it must be said that some internal mbstring code
which operates only on UTF-8 text uses a third method for mapping
between byte indices and character indices, that is: counting
non-continuation UTF-8 bytes, which are all bytes whose binary
representation is NOT like 0b10xxxxxx. This method happens to agree with
the method which involves decoding the input text to codepoints and then
counting the codepoints.)
I have been aware of this issue for years, but only recently became
aware that in the case of mb_strstr, mb_strpos, and mb_substr,
this issue can cause seriously unintuitive behavior (and even security
vulnerabilities). This was reported by Stefan Schiller.
Stefan Schiller shared the following example for mb_strstr:
var_dump(mb_strstr("\xf0start", "start", false, "UTF-8"));
// string(2) "rt"
Similarly, when mb_strpos and mb_substr are used to identify and
extract a substring from a string with encoding errors, Stefan Schiller
pointed out that the extracted portion may be completely different than
desired. This is because (for UTF-8 strings) mb_strpos works by counting
non-continuation bytes, but mb_substr uses an mblen_table.
Since some mbstring functions *cannot* be implemented using an
mblen_table, as long as mblen_tables are used, similar inconsistencies
cannot be totally avoided. But the mblen_tables are critical to
mbstring's performance. Or are they? Benchmarking mb_substr on various
UTF-8, SJIS, and EUC-JP strings revealed something interesting.
On all SJIS and EUC-JP test cases, mb_substr was slightly faster when
the mblen_table based code was deleted. For some UTF-8 test cases, the
mblen_table-based code was a tiny bit faster, while for others the
fallback code was a touch faster; in no case was the difference
significant.
Therefore, the simple fix is to delete the mblen_table-based
implementation of mb_substr.
Aside from making the function behave consistently with other mbstring
functions on invalid strings, there is ONE case where behavior is now
different on valid strings: that is, on SJIS-Mac (MacJapanese) strings
which contain any of the following code units:
0x85AB-0x85AD, 0x85BF, 0x85C0, 0x85C1, 0x8645, 0x864B, 0x865D, 0x869E,
0x86CE, 0x86D3-0x86D5, 0x86D6, 0x8971, 0x8792, 0x879D, 0x87FB, 0x87FC,
0xEB41, 0xEB42, 0xEB50, 0xEB5B, 0xEB5D, 0xEB60-0xEB6E, and all from
0xEB81 and above.
All of these SJIS-Mac code units share the (very unusual) property that
they do not correspond to any one Unicode codepoint. When converting
from SJIS-Mac to Unicode, these must be converted to 2, 3, 4, or 5
codepoints each.
The previous, mblen_table-based implementation of mb_substr would treat
all of these SJIS-Mac byte sequences as 'one character'. Now, they are
treated as multiple characters (one for each of the Unicode codepoints
which they decode to). The new behavior is more consistent with other
mbstring functions.
I don't know if SJIS-Mac users will like this change or not (probably
most will never notice), but the BC break is justified by the very
real security impact of the previous, inconsistent behavior.
Finally, I should comment on whether similar changes are needed
elsewhere. The remaining functions which use an mblen_table are:
mb_str_split, mb_strcut, and various search functions (such as
mb_strpos). The search functions are only affected now when they
receive a positive 'offset' parameter specifying where to start
searching from.
The search functions should definitely be fixed so they do not use
an mblen_table to implement the 'offset' parameter. I am not convinced
that there is any good reason to change mb_str_split and mb_strcut.
On microbenchmarks run on my dev machine, mb_strcut is now ~50% faster
for fixed-byte-length text encodings like ASCII. (This is because the
previous code did an extra, unnecessary copy operation on the
resulting output string.)
* Fast path for when there is nothing to trim in mb_trim
* Make mb_trim decide between linear search vs hash table lookup
Using empirical experiments I noticed that on my i7-4790 the hash table
approach becomes faster once we have more than 4 code points in the trim
characters, when evaluated on the worst case.
This patch changes the logic so that a hash table is used for a large
number of trim characters, and linear search when the number of trim
characters is <= 4.
