Unlike iconv or the internal converters, ICU consumes truncated multi-
byte sequences at the end of an input buffer. We currently check for a
non-empty raw input buffer to detect truncated sequences, so this fails
with ICU.
It might be possible to inspect the pivot buffer pointers, but it seems
cleaner to implement a `flush` flag for some encoding and I/O functions.
After flushing, we can check for U_TRUNCATED_CHAR_FOUND with ICU, or
detect remaining input with other converters.
Also fix detection of truncated sequences for HTML, XML content and
DTDs with iconv.
Another regression related to reading from stdin.
Making a "-" filename read from stdin was deeply baked into the core
IO code but is inherently insecure. I really want to reenable this
dangerous feature as sparingly as possible.
This now enables compressed input when using the "Fd" API functions
which wan't supported before. But XML_PARSE_NO_UNZIP will be
inverted later.
Allow compressed stdin in xmlReadFile to support xmlstarlet and older
versions of xsltproc. So far, these are the only known command-line
tools that rely on "-" meaning stdin.
Commit 7e14c05d removed unnecessary copying of uncompressed input
through zlib or xzlib. This broke input from non-regular files like
pipes which can't be reopened. Try to detect such files by checking
whether they're seekable and always pipe them through zlib or xzlib.
Also remove seemingly unnecessary calls to gzread and gzrewind to
support unseekable files.
Fixes https://gitlab.gnome.org/GNOME/libxslt/-/issues/124.
When push parsing, we want to convert as much of the input as possible.
When pull parsing memory buffers, we want to convert data chunk by chunk
to save memory.
Always use what the old implementation called the "IO" allocation
scheme, allowing to move the content pointer past the initial
allocation. This is inexpensive and allows efficient shrinking.
Optimize xmlBufGrow, reusing shrunken memory as much as possible.
Simplify xmlBufAdd.
Make xmlBufBackToBuffer return an error on overflow.
Make "size" exclude the terminating NULL byte.
Always provide an initial size.
Reintroduce static buffers.
Remove xmlBufResize and several other functions.
Stop using xmlOutputBufferWriteEscape except when using deprecated
xmlSaveSetEscape. Rewrite xmlOutputBufferWriteEscape to use an extra
buffer and call xmlOutputBufferWrite.
Introduce xmlSerializeText to serialize both text and attribute content.
Don't read encoding from document when serializing and remove all hacks
that temporarily changed the document's encoding.
Unless we are on Windows, the following POSIX headers are required.
They're part of the earliest POSIX specs and it doesn't make sense to
check for them.
- fcntl.h
- unistd.h
- sys/stat.h
- sys/time.h
On Windows, io.h, fcntl.h and sys/stat.h are always available.
- xmlInputCreateUrl
- xmlInputCreateMemory
- xmlInputCreateString
- xmlInputCreateFd
- xmlInputCreateIO
- xmlInputSetEncoding
These functions don't take a parser context and work on xmlParserInputs,
replacing functions working on xmlParserInputBuffers.
xmlInputCreateUrl and xmlInputSetEncoding offer fine-grained error
handling.
Several XML_INPUT_* flags offer additional control.
Provide a new set of functions to create xmlParserInputs. These can be
used for the document entity or from external entity loaders.
- Don't require xmlParserInputBuffer.
- All functions take a base URI.
- All functions take an encoding as string.
- xmlNewInputURL also takes a public ID.
- xmlNewInputMemory takes a size_t.
- Optimization hints for memory buffers.
Improve documentation.
Only call xmlInitParser before allocating a new parser context.
Call xmlCtxtUseOptions as early as possible.
