Ben Wing: almost completely rewritten for Mule, 19.12.
*/
-#ifndef _XEMACS_BUFFER_H_
-#define _XEMACS_BUFFER_H_
+#ifndef INCLUDED_buffer_h_
+#define INCLUDED_buffer_h_
#ifdef MULE
#include "mule-charset.h"
Bytind mule_bytind_cache[16];
#endif
+ /* Similar to the above, we keep track of positions for which line
+ number has last been calculated. See line-number.c. */
+ Lisp_Object line_number_cache;
+
/* Change data that goes with the text. */
struct buffer_text_change_data *changes;
/* The markers that refer to this buffer. This is actually a single
marker -- successive elements in its marker `chain' are the other
markers referring to this buffer */
- struct Lisp_Marker *markers;
+ Lisp_Marker *markers;
/* The buffer's extent info. This is its own type, an extent-info
object (done this way for ease in marking / finalizing). */
#define XBUFFER(x) XRECORD (x, buffer, struct buffer)
#define XSETBUFFER(x, p) XSETRECORD (x, p, buffer)
#define BUFFERP(x) RECORDP (x, buffer)
-#define GC_BUFFERP(x) GC_RECORDP (x, buffer)
#define CHECK_BUFFER(x) CHECK_RECORD (x, buffer)
#define CONCHECK_BUFFER(x) CONCHECK_RECORD (x, buffer)
} while (0)
\f
+#define BUFFER_BASE_BUFFER(b) ((b)->base_buffer ? (b)->base_buffer : (b))
+
+/* Map over buffers sharing the same text as MPS_BUF. MPS_BUFVAR is a
+ variable that gets the buffer values (beginning with the base
+ buffer, then the children), and MPS_BUFCONS should be a temporary
+ Lisp_Object variable. */
+#define MAP_INDIRECT_BUFFERS(mps_buf, mps_bufvar, mps_bufcons) \
+for (mps_bufcons = Qunbound, \
+ mps_bufvar = BUFFER_BASE_BUFFER (mps_buf); \
+ UNBOUNDP (mps_bufcons) ? \
+ (mps_bufcons = mps_bufvar->indirect_children, \
+ 1) \
+ : (!NILP (mps_bufcons) \
+ && (mps_bufvar = XBUFFER (XCAR (mps_bufcons)), 1) \
+ && (mps_bufcons = XCDR (mps_bufcons), 1)); \
+ )
+
+\f
+
+/************************************************************************/
+/* */
+/* working with raw internal-format data */
+/* */
+/************************************************************************/
+
/* NOTE: In all the following macros, we follow these rules concerning
multiple evaluation of the arguments:
denoted with the word "unsafe" in their name and are generally
meant to be called only by other macros that have already
stored the calling values in temporary variables.
- */
-/************************************************************************/
-/* */
-/* working with raw internal-format data */
-/* */
-/************************************************************************/
-
-/* Use these on contiguous strings of data. If the text you're
- operating on is known to come from a buffer, use the buffer-level
- functions below -- they know about the gap and may be more
- efficient. */
-/* Functions are as follows:
+ Use the following functions/macros on contiguous strings of data.
+ If the text you're operating on is known to come from a buffer, use
+ the buffer-level functions below -- they know about the gap and may
+ be more efficient.
- (A) For working with charptr's (pointers to internally-formatted text):
- -----------------------------------------------------------------------
+ (A) For working with charptr's (pointers to internally-formatted text):
+ -----------------------------------------------------------------------
- VALID_CHARPTR_P(ptr):
+ VALID_CHARPTR_P (ptr):
Given a charptr, does it point to the beginning of a character?
- ASSERT_VALID_CHARPTR(ptr):
+ ASSERT_VALID_CHARPTR (ptr):
If error-checking is enabled, assert that the given charptr
- points to the beginning of a character. Otherwise, do nothing.
+ points to the beginning of a character. Otherwise, do nothing.
- INC_CHARPTR(ptr):
+ INC_CHARPTR (ptr):
Given a charptr (assumed to point at the beginning of a character),
modify that pointer so it points to the beginning of the next
character.
- DEC_CHARPTR(ptr):
+ DEC_CHARPTR (ptr):
Given a charptr (assumed to point at the beginning of a
character or at the very end of the text), modify that pointer
so it points to the beginning of the previous character.
- VALIDATE_CHARPTR_BACKWARD(ptr):
+ VALIDATE_CHARPTR_BACKWARD (ptr):
Make sure that PTR is pointing to the beginning of a character.
- If not, back up until this is the case. Note that there are not
+ If not, back up until this is the case. Note that there are not
too many places where it is legitimate to do this sort of thing.
It's an error if you're passed an "invalid" char * pointer.
NOTE: PTR *must* be pointing to a valid part of the string (i.e.
not the very end, unless the string is zero-terminated or
something) in order for this function to not cause crashes.
- VALIDATE_CHARPTR_FORWARD(ptr):
+ VALIDATE_CHARPTR_FORWARD (ptr):
Make sure that PTR is pointing to the beginning of a character.
If not, move forward until this is the case. Note that there
are not too many places where it is legitimate to do this sort
section of internally-formatted text:
--------------------------------------------------------------
- bytecount_to_charcount(ptr, nbi):
+ bytecount_to_charcount (ptr, nbi):
Given a pointer to a text string and a length in bytes,
return the equivalent length in characters.
- charcount_to_bytecount(ptr, nch):
+ charcount_to_bytecount (ptr, nch):
Given a pointer to a text string and a length in characters,
return the equivalent length in bytes.
- charptr_n_addr(ptr, n):
+ charptr_n_addr (ptr, n):
Return a pointer to the beginning of the character offset N
(in characters) from PTR.
- charptr_length(ptr):
- Given a zero-terminated pointer to Emacs characters,
- return the number of Emacs characters contained within.
-
(C) For retrieving or changing the character pointed to by a charptr:
---------------------------------------------------------------------
- charptr_emchar(ptr):
+ charptr_emchar (ptr):
Retrieve the character pointed to by PTR as an Emchar.
- charptr_emchar_n(ptr, n):
+ charptr_emchar_n (ptr, n):
Retrieve the character at offset N (in characters) from PTR,
as an Emchar.
- set_charptr_emchar(ptr, ch):
+ set_charptr_emchar (ptr, ch):
Store the character CH (an Emchar) as internally-formatted
text starting at PTR. Return the number of bytes stored.
- charptr_copy_char(ptr, ptr2):
+ charptr_copy_char (ptr, ptr2):
Retrieve the character pointed to by PTR and store it as
internally-formatted text in PTR2.
in mule-charset.h, for retrieving the charset of an Emchar
and such. These are only valid when MULE is defined.]
- valid_char_p(ch):
+ valid_char_p (ch):
Return whether the given Emchar is valid.
- CHARP(ch):
- Return whether the given Lisp_Object is a valid character.
- This is approximately the same as saying the Lisp_Object is
- an int whose value is a valid Emchar. (But not exactly
- because when MULE is not defined, we allow arbitrary values
- in all but the lowest 8 bits and mask them off, for backward
- compatibility.)
+ CHARP (ch):
+ Return whether the given Lisp_Object is a character.
- CHECK_CHAR_COERCE_INT(ch):
- Signal an error if CH is not a valid character as per CHARP().
- Also canonicalize the value into a valid Emchar, as necessary.
- (This only means anything when MULE is not defined.)
-
- COERCE_CHAR(ch):
- Coerce an object that is known to satisfy CHARP() into a
- valid Emchar.
+ CHECK_CHAR_COERCE_INT (ch):
+ Signal an error if CH is not a valid character or integer Lisp_Object.
+ If CH is an integer Lisp_Object, convert it to a character Lisp_Object,
+ but merely by repackaging, without performing tests for char validity.
MAX_EMCHAR_LEN:
Maximum number of buffer bytes per Emacs character.
method because it doesn't have easy access to the first byte of
the character it's moving over. */
-#define real_inc_charptr_fun(ptr) \
- ((ptr) += REP_BYTES_BY_FIRST_BYTE (* (unsigned char *) (ptr)))
-#ifdef ERROR_CHECK_BUFPOS
-#define inc_charptr_fun(ptr) (ASSERT_VALID_CHARPTR (ptr), \
- real_inc_charptr_fun (ptr))
-#else
-#define inc_charptr_fun(ptr) real_inc_charptr_fun (ptr)
-#endif
+#define REAL_INC_CHARPTR(ptr) \
+ ((void) ((ptr) += REP_BYTES_BY_FIRST_BYTE (* (unsigned char *) (ptr))))
-#define REAL_INC_CHARPTR(ptr) ((void) (real_inc_charptr_fun (ptr)))
+#define REAL_INC_CHARBYTIND(ptr,pos) \
+ (pos += REP_BYTES_BY_FIRST_BYTE (* (unsigned char *) (ptr)))
+#define REAL_DEC_CHARPTR(ptr) do { \
+ (ptr)--; \
+} while (!VALID_CHARPTR_P (ptr))
+
+#ifdef ERROR_CHECK_BUFPOS
#define INC_CHARPTR(ptr) do { \
ASSERT_VALID_CHARPTR (ptr); \
REAL_INC_CHARPTR (ptr); \
} while (0)
-#define REAL_DEC_CHARPTR(ptr) do { \
- (ptr)--; \
-} while (!VALID_CHARPTR_P (ptr))
+#define INC_CHARBYTIND(ptr,pos) do { \
+ ASSERT_VALID_CHARPTR (ptr); \
+ REAL_INC_CHARBYTIND (ptr,pos); \
+} while (0)
-#ifdef ERROR_CHECK_BUFPOS
-#define DEC_CHARPTR(ptr) do { \
- CONST Bufbyte *__dcptr__ = (ptr); \
- CONST Bufbyte *__dcptr2__ = __dcptr__; \
- REAL_DEC_CHARPTR (__dcptr2__); \
- assert (__dcptr__ - __dcptr2__ == \
- REP_BYTES_BY_FIRST_BYTE (*__dcptr2__)); \
- (ptr) = __dcptr2__; \
+#define DEC_CHARPTR(ptr) do { \
+ CONST Bufbyte *dc_ptr1 = (ptr); \
+ CONST Bufbyte *dc_ptr2 = dc_ptr1; \
+ REAL_DEC_CHARPTR (dc_ptr2); \
+ assert (dc_ptr1 - dc_ptr2 == \
+ REP_BYTES_BY_FIRST_BYTE (*dc_ptr2)); \
+ (ptr) = dc_ptr2; \
} while (0)
-#else
+
+#else /* ! ERROR_CHECK_BUFPOS */
+#define INC_CHARBYTIND(ptr,pos) REAL_INC_CHARBYTIND (ptr,pos)
+#define INC_CHARPTR(ptr) REAL_INC_CHARPTR (ptr)
#define DEC_CHARPTR(ptr) REAL_DEC_CHARPTR (ptr)
-#endif
+#endif /* ! ERROR_CHECK_BUFPOS */
#ifdef MULE
the end of the string. */
#define VALIDATE_CHARPTR_FORWARD(ptr) do { \
- Bufbyte *__vcfptr__ = (ptr); \
- VALIDATE_CHARPTR_BACKWARD (__vcfptr__); \
- if (__vcfptr__ != (ptr)) \
+ Bufbyte *vcf_ptr = (ptr); \
+ VALIDATE_CHARPTR_BACKWARD (vcf_ptr); \
+ if (vcf_ptr != (ptr)) \
{ \
- (ptr) = __vcfptr__; \
+ (ptr) = vcf_ptr; \
INC_CHARPTR (ptr); \
} \
} while (0)
return ptr + charcount_to_bytecount (ptr, offset);
}
-INLINE Charcount charptr_length (CONST Bufbyte *ptr);
-INLINE Charcount
-charptr_length (CONST Bufbyte *ptr)
-{
- return bytecount_to_charcount (ptr, strlen ((CONST char *) ptr));
-}
-
-
/* -------------------------------------------------------------------- */
/* (C) For retrieving or changing the character pointed to by a charptr */
/* -------------------------------------------------------------------- */
INLINE int
valid_char_p (Emchar ch)
{
- return (ch >= 0 && ch <= 255) || non_ascii_valid_char_p (ch);
+ return ((unsigned int) (ch) <= 0xff) || non_ascii_valid_char_p (ch);
}
#else /* not MULE */
-#define valid_char_p(ch) ((unsigned int) (ch) <= 255)
+#define valid_char_p(ch) ((unsigned int) (ch) <= 0xff)
#endif /* not MULE */
#else
-#define XCHAR_OR_CHAR_INT(obj) (CHARP ((obj)) ? XCHAR ((obj)) : XINT ((obj)))
+#define XCHAR_OR_CHAR_INT(obj) (CHARP (obj) ? XCHAR (obj) : XINT (obj))
#endif
INLINE Bytind
BI_BUF_PTR_BYTE_POS (struct buffer *buf, Bufbyte *ptr)
{
- return ((ptr) - (buf)->text->beg + 1
- - ((ptr - (buf)->text->beg + 1) > (buf)->text->gpt
- ? (buf)->text->gap_size : 0));
+ return (ptr - buf->text->beg + 1
+ - ((ptr - buf->text->beg + 1) > buf->text->gpt
+ ? buf->text->gap_size : 0));
}
#define BUF_PTR_BYTE_POS(buf, ptr) \
INLINE Bufbyte *
BI_BUF_BYTE_ADDRESS (struct buffer *buf, Bytind pos)
{
- return ((buf)->text->beg +
- ((pos >= (buf)->text->gpt ? (pos + (buf)->text->gap_size) : pos)
+ return (buf->text->beg +
+ ((pos >= buf->text->gpt ? (pos + buf->text->gap_size) : pos)
- 1));
}
INLINE Bufbyte *
BI_BUF_BYTE_ADDRESS_BEFORE (struct buffer *buf, Bytind pos)
{
- return ((buf)->text->beg +
- ((pos > (buf)->text->gpt ? (pos + (buf)->text->gap_size) : pos)
+ return (buf->text->beg +
+ ((pos > buf->text->gpt ? (pos + buf->text->gap_size) : pos)
- 2));
}
INLINE int
valid_memind_p (struct buffer *buf, Memind x)
{
- return ((x >= 1 && x <= (Memind) (buf)->text->gpt) ||
- (x > (Memind) ((buf)->text->gpt + (buf)->text->gap_size) &&
- x <= (Memind) ((buf)->text->z + (buf)->text->gap_size)));
+ return ((x >= 1 && x <= (Memind) buf->text->gpt) ||
+ (x > (Memind) (buf->text->gpt + buf->text->gap_size) &&
+ x <= (Memind) (buf->text->z + buf->text->gap_size)));
}
INLINE Memind bytind_to_memind (struct buffer *buf, Bytind x);
INLINE Memind
bytind_to_memind (struct buffer *buf, Bytind x)
{
- return (Memind) ((x > (buf)->text->gpt) ? (x + (buf)->text->gap_size) : x);
+ return (Memind) ((x > buf->text->gpt) ? (x + buf->text->gap_size) : x);
}
#ifdef ERROR_CHECK_BUFPOS
assert (valid_memind_p (buf, x));
#endif
- return (Bytind) ((x > (Memind) (buf)->text->gpt) ?
- x - (buf)->text->gap_size :
+ return (Bytind) ((x > (Memind) buf->text->gpt) ?
+ x - buf->text->gap_size :
x);
}
results with stupid compilers. */
#ifdef MULE
-# define VALIDATE_BYTIND_BACKWARD(buf, x) do \
-{ \
- Bufbyte *__ibptr = BI_BUF_BYTE_ADDRESS (buf, x); \
- while (!BUFBYTE_FIRST_BYTE_P (*__ibptr)) \
- __ibptr--, (x)--; \
+# define VALIDATE_BYTIND_BACKWARD(buf, x) do { \
+ Bufbyte *VBB_ptr = BI_BUF_BYTE_ADDRESS (buf, x); \
+ while (!BUFBYTE_FIRST_BYTE_P (*VBB_ptr)) \
+ VBB_ptr--, (x)--; \
} while (0)
#else
# define VALIDATE_BYTIND_BACKWARD(buf, x)
results with stupid compilers. */
#ifdef MULE
-# define VALIDATE_BYTIND_FORWARD(buf, x) do \
-{ \
- Bufbyte *__ibptr = BI_BUF_BYTE_ADDRESS (buf, x); \
- while (!BUFBYTE_FIRST_BYTE_P (*__ibptr)) \
- __ibptr++, (x)++; \
+# define VALIDATE_BYTIND_FORWARD(buf, x) do { \
+ Bufbyte *VBF_ptr = BI_BUF_BYTE_ADDRESS (buf, x); \
+ while (!BUFBYTE_FIRST_BYTE_P (*VBF_ptr)) \
+ VBF_ptr++, (x)++; \
} while (0)
#else
# define VALIDATE_BYTIND_FORWARD(buf, x)
#define BUF_CHARPTR_COPY_CHAR(buf, pos, str) \
BI_BUF_CHARPTR_COPY_CHAR (buf, bufpos_to_bytind (buf, pos), str)
-
-
\f
/************************************************************************/
-/* */
-/* working with externally-formatted data */
-/* */
+/* */
+/* Converting between internal and external format */
+/* */
/************************************************************************/
+/*
+ All client code should use only the two macros
+
+ TO_EXTERNAL_FORMAT (source_type, source, sink_type, sink, coding_system)
+ TO_INTERNAL_FORMAT (source_type, source, sink_type, sink, coding_system)
+
+ Typical use is
+
+ TO_EXTERNAL_FORMAT (DATA, (ptr, len),
+ LISP_BUFFER, buffer,
+ Qfile_name);
+
+ The source or sink can be specified in one of these ways:
+
+ DATA, (ptr, len), // input data is a fixed buffer of size len
+ ALLOCA, (ptr, len), // output data is in a alloca()ed buffer of size len
+ MALLOC, (ptr, len), // output data is in a malloc()ed buffer of size len
+ C_STRING_ALLOCA, ptr, // equivalent to ALLOCA (ptr, len_ignored) on output.
+ C_STRING_MALLOC, ptr, // equivalent to MALLOC (ptr, len_ignored) on output.
+ C_STRING, ptr, // equivalent to DATA, (ptr, strlen (ptr) + 1) on input
+ LISP_STRING, string, // input or output is a Lisp_Object of type string
+ LISP_BUFFER, buffer, // output is written to (point) in lisp buffer
+ LISP_LSTREAM, lstream, // input or output is a Lisp_Object of type lstream
+ LISP_OPAQUE, object, // input or output is a Lisp_Object of type opaque
+
+ When specifying the sink, use lvalues, since the macro will assign to them,
+ except when the sink is an lstream or a lisp buffer.
+
+ The macros accept the kinds of sources and sinks appropriate for
+ internal and external data representation. See the type_checking_assert
+ macros below for the actual allowed types.
+
+ Since some sources and sinks use one argument (a Lisp_Object) to
+ specify them, while others take a (pointer, length) pair, we use
+ some C preprocessor trickery to allow pair arguments to be specified
+ by parenthesizing them, as in the examples above.
+
+ Anything prefixed by dfc_ (`data format conversion') is private.
+ They are only used to implement these macros.
+
+ Using C_STRING* is appropriate for using with external APIs that take
+ null-terminated strings. For internal data, we should try to be
+ '\0'-clean - i.e. allow arbitrary data to contain embedded '\0'.
+
+ Sometime in the future we might allow output to C_STRING_ALLOCA or
+ C_STRING_MALLOC _only_ with TO_EXTERNAL_FORMAT(), not
+ TO_INTERNAL_FORMAT(). */
+
+#define TO_EXTERNAL_FORMAT(source_type, source, sink_type, sink, coding_system) \
+do { \
+ dfc_conversion_type dfc_simplified_source_type; \
+ dfc_conversion_type dfc_simplified_sink_type; \
+ dfc_conversion_data dfc_source; \
+ dfc_conversion_data dfc_sink; \
+ \
+ type_checking_assert \
+ ((DFC_TYPE_##source_type == DFC_TYPE_DATA || \
+ DFC_TYPE_##source_type == DFC_TYPE_C_STRING || \
+ DFC_TYPE_##source_type == DFC_TYPE_LISP_STRING || \
+ DFC_TYPE_##source_type == DFC_TYPE_LISP_OPAQUE || \
+ DFC_TYPE_##source_type == DFC_TYPE_LISP_LSTREAM) \
+ && \
+ (DFC_TYPE_##sink_type == DFC_TYPE_ALLOCA || \
+ DFC_TYPE_##sink_type == DFC_TYPE_MALLOC || \
+ DFC_TYPE_##sink_type == DFC_TYPE_C_STRING_ALLOCA || \
+ DFC_TYPE_##sink_type == DFC_TYPE_C_STRING_MALLOC || \
+ DFC_TYPE_##sink_type == DFC_TYPE_LISP_LSTREAM || \
+ DFC_TYPE_##sink_type == DFC_TYPE_LISP_OPAQUE)); \
+ \
+ DFC_SOURCE_##source_type##_TO_ARGS (source); \
+ DFC_SINK_##sink_type##_TO_ARGS (sink); \
+ \
+ DFC_CONVERT_TO_EXTERNAL_FORMAT (dfc_simplified_source_type, &dfc_source, \
+ coding_system, \
+ dfc_simplified_sink_type, &dfc_sink); \
+ \
+ DFC_##sink_type##_USE_CONVERTED_DATA (sink); \
+} while (0)
-/* Sometimes strings need to be converted into one or another
- external format, for passing to a library function. (Note
- that we encapsulate and automatically convert the arguments
- of some functions, but not others.) At times this conversion
- also has to go the other way -- i.e. when we get external-
- format strings back from a library function.
-*/
+#define TO_INTERNAL_FORMAT(source_type, source, sink_type, sink, coding_system) \
+do { \
+ dfc_conversion_type dfc_simplified_source_type; \
+ dfc_conversion_type dfc_simplified_sink_type; \
+ dfc_conversion_data dfc_source; \
+ dfc_conversion_data dfc_sink; \
+ \
+ type_checking_assert \
+ ((DFC_TYPE_##source_type == DFC_TYPE_DATA || \
+ DFC_TYPE_##source_type == DFC_TYPE_C_STRING || \
+ DFC_TYPE_##source_type == DFC_TYPE_LISP_OPAQUE || \
+ DFC_TYPE_##source_type == DFC_TYPE_LISP_LSTREAM) \
+ && \
+ (DFC_TYPE_##sink_type == DFC_TYPE_ALLOCA || \
+ DFC_TYPE_##sink_type == DFC_TYPE_MALLOC || \
+ DFC_TYPE_##sink_type == DFC_TYPE_C_STRING_ALLOCA || \
+ DFC_TYPE_##sink_type == DFC_TYPE_C_STRING_MALLOC || \
+ DFC_TYPE_##sink_type == DFC_TYPE_LISP_STRING || \
+ DFC_TYPE_##sink_type == DFC_TYPE_LISP_LSTREAM || \
+ DFC_TYPE_##sink_type == DFC_TYPE_LISP_BUFFER)); \
+ \
+ DFC_SOURCE_##source_type##_TO_ARGS (source); \
+ DFC_SINK_##sink_type##_TO_ARGS (sink); \
+ \
+ DFC_CONVERT_TO_INTERNAL_FORMAT (dfc_simplified_source_type, &dfc_source, \
+ coding_system, \
+ dfc_simplified_sink_type, &dfc_sink); \
+ \
+ DFC_##sink_type##_USE_CONVERTED_DATA (sink); \
+} while (0)
#ifdef FILE_CODING
+#define DFC_CONVERT_TO_EXTERNAL_FORMAT dfc_convert_to_external_format
+#define DFC_CONVERT_TO_INTERNAL_FORMAT dfc_convert_to_internal_format
+#else
+/* ignore coding_system argument */
+#define DFC_CONVERT_TO_EXTERNAL_FORMAT(a, b, coding_system, c, d) \
+ dfc_convert_to_external_format (a, b, c, d)
+#define DFC_CONVERT_TO_INTERNAL_FORMAT(a, b, coding_system, c, d) \
+ dfc_convert_to_internal_format (a, b, c, d)
+#endif
-/* WARNING: These use a static buffer. This can lead to disaster if
- these functions are not used *very* carefully. Under normal
- circumstances, do not call these functions; call the front ends
- below. */
-
-Extbyte *convert_to_external_format (CONST Bufbyte *ptr,
- Bytecount len,
- Extcount *len_out,
- enum external_data_format fmt);
-Bufbyte *convert_from_external_format (CONST Extbyte *ptr,
- Extcount len,
- Bytecount *len_out,
- enum external_data_format fmt);
-
-#else /* ! MULE */
-
-#define convert_to_external_format(ptr, len, len_out, fmt) \
- (*(len_out) = (int) (len), (Extbyte *) (ptr))
-#define convert_from_external_format(ptr, len, len_out, fmt) \
- (*(len_out) = (Bytecount) (len), (Bufbyte *) (ptr))
-
-#endif /* ! MULE */
-
-/* In all of the following macros we use the following general principles:
-
- -- Functions that work with charptr's accept two sorts of charptr's:
-
- a) Pointers to memory with a length specified. The pointer will be
- fundamentally of type `unsigned char *' (although labelled
- as `Bufbyte *' for internal-format data and `Extbyte *' for
- external-format data) and the length will be fundamentally of
- type `int' (although labelled as `Bytecount' for internal-format
- data and `Extcount' for external-format data). The length is
- always a count in bytes.
- b) Zero-terminated pointers; no length specified. The pointer
- is of type `char *', whether the data pointed to is internal-format
- or external-format. These sorts of pointers are available for
- convenience in working with C library functions and literal
- strings. In general you should use these sorts of pointers only
- to interface to library routines and not for general manipulation,
- as you are liable to lose embedded nulls and such. This could
- be a big problem for routines that want Unicode-formatted data,
- which is likely to have lots of embedded nulls in it.
- (In the real world, though, external Unicode data will be UTF-8,
- which will not have embedded nulls and is ASCII-compatible - martin)
-
- -- Functions that work with Lisp strings accept strings as Lisp Objects
- (as opposed to the `struct Lisp_String *' for some of the other
- string accessors). This is for convenience in working with the
- functions, as otherwise you will almost always have to call
- XSTRING() on the object.
-
- -- Functions that work with charptr's are not guaranteed to copy
- their data into alloca()ed space. Functions that work with
- Lisp strings are, however. The reason is that Lisp strings can
- be relocated any time a GC happens, and it could happen at some
- rather unexpected times. The internal-external conversion is
- rarely done in time-critical functions, and so the slight
- extra time required for alloca() and copy is well-worth the
- safety of knowing your string data won't be relocated out from
- under you.
- */
-
-
-/* Maybe convert charptr's data into ext-format and store the result in
- alloca()'ed space.
-
- You may wonder why this is written in this fashion and not as a
- function call. With a little trickery it could certainly be
- written this way, but it won't work because of those DAMN GCC WANKERS
- who couldn't be bothered to handle alloca() properly on the x86
- architecture. (If you put a call to alloca() in the argument to
- a function call, the stack space gets allocated right in the
- middle of the arguments to the function call and you are unbelievably
- hosed.) */
+typedef union
+{
+ struct { const void *ptr; size_t len; } data;
+ Lisp_Object lisp_object;
+} dfc_conversion_data;
-#ifdef MULE
+enum dfc_conversion_type
+{
+ DFC_TYPE_DATA,
+ DFC_TYPE_ALLOCA,
+ DFC_TYPE_MALLOC,
+ DFC_TYPE_C_STRING,
+ DFC_TYPE_C_STRING_ALLOCA,
+ DFC_TYPE_C_STRING_MALLOC,
+ DFC_TYPE_LISP_STRING,
+ DFC_TYPE_LISP_LSTREAM,
+ DFC_TYPE_LISP_OPAQUE,
+ DFC_TYPE_LISP_BUFFER
+};
+typedef enum dfc_conversion_type dfc_conversion_type;
-#define GET_CHARPTR_EXT_DATA_ALLOCA(ptr, len, fmt, ptr_out, len_out) do \
-{ \
- Bytecount gceda_len_in = (Bytecount) (len); \
- Extcount gceda_len_out; \
- CONST Bufbyte *gceda_ptr_in = (ptr); \
- Extbyte *gceda_ptr_out = \
- convert_to_external_format (gceda_ptr_in, gceda_len_in, \
- &gceda_len_out, fmt); \
- /* If the new string is identical to the old (will be the case most \
- of the time), just return the same string back. This saves \
- on alloca()ing, which can be useful on C alloca() machines and \
- on stack-space-challenged environments. */ \
- \
- if (gceda_len_in == gceda_len_out && \
- !memcmp (gceda_ptr_in, gceda_ptr_out, gceda_len_out)) \
- { \
- (ptr_out) = (Extbyte *) gceda_ptr_in; \
- (len_out) = (Extcount) gceda_len_in; \
- } \
- else \
- { \
- (ptr_out) = (Extbyte *) alloca (1 + gceda_len_out); \
- memcpy ((void *) ptr_out, gceda_ptr_out, 1 + gceda_len_out); \
- (len_out) = (Extcount) gceda_len_out; \
- } \
+/* WARNING: These use a static buffer. This can lead to disaster if
+ these functions are not used *very* carefully. Another reason to only use
+ TO_EXTERNAL_FORMATf() and TO_INTERNAL_FORMAT(). */
+void
+dfc_convert_to_external_format (dfc_conversion_type source_type,
+ dfc_conversion_data *source,
+#ifdef FILE_CODING
+ Lisp_Object coding_system,
+#endif
+ dfc_conversion_type sink_type,
+ dfc_conversion_data *sink);
+void
+dfc_convert_to_internal_format (dfc_conversion_type source_type,
+ dfc_conversion_data *source,
+#ifdef FILE_CODING
+ Lisp_Object coding_system,
+#endif
+ dfc_conversion_type sink_type,
+ dfc_conversion_data *sink);
+/* CPP Trickery */
+#define DFC_CPP_CAR(x,y) (x)
+#define DFC_CPP_CDR(x,y) (y)
+
+/* Convert `source' to args for dfc_convert_to_*_format() */
+#define DFC_SOURCE_DATA_TO_ARGS(val) do { \
+ dfc_source.data.ptr = DFC_CPP_CAR val; \
+ dfc_source.data.len = DFC_CPP_CDR val; \
+ dfc_simplified_source_type = DFC_TYPE_DATA; \
} while (0)
-
-#else /* ! MULE */
-
-#define GET_CHARPTR_EXT_DATA_ALLOCA(ptr, len, fmt, ptr_out, len_out) do \
-{ \
- (ptr_out) = (Extbyte *) (ptr); \
- (len_out) = (Extcount) (len); \
+#define DFC_SOURCE_C_STRING_TO_ARGS(val) do { \
+ dfc_source.data.len = \
+ strlen ((char *) (dfc_source.data.ptr = (val))); \
+ dfc_simplified_source_type = DFC_TYPE_DATA; \
} while (0)
-
-#endif /* ! MULE */
-
-#define GET_C_CHARPTR_EXT_DATA_ALLOCA(ptr, fmt, ptr_out) do \
-{ \
- Extcount gcceda_ignored_len; \
- CONST Bufbyte *gcceda_ptr_in = (CONST Bufbyte *) (ptr); \
- Extbyte *gcceda_ptr_out; \
- \
- GET_CHARPTR_EXT_DATA_ALLOCA (gcceda_ptr_in, \
- strlen ((char *) gcceda_ptr_in), \
- fmt, \
- gcceda_ptr_out, \
- gcceda_ignored_len); \
- (ptr_out) = (char *) gcceda_ptr_out; \
+#define DFC_SOURCE_LISP_STRING_TO_ARGS(val) do { \
+ Lisp_Object dfc_slsta = (val); \
+ type_checking_assert (STRINGP (dfc_slsta)); \
+ dfc_source.lisp_object = dfc_slsta; \
+ dfc_simplified_source_type = DFC_TYPE_LISP_STRING; \
} while (0)
-
-#define GET_C_CHARPTR_EXT_BINARY_DATA_ALLOCA(ptr, ptr_out) \
- GET_C_CHARPTR_EXT_DATA_ALLOCA (ptr, FORMAT_BINARY, ptr_out)
-#define GET_CHARPTR_EXT_BINARY_DATA_ALLOCA(ptr, len, ptr_out, len_out) \
- GET_CHARPTR_EXT_DATA_ALLOCA (ptr, len, FORMAT_BINARY, ptr_out, len_out)
-
-#define GET_C_CHARPTR_EXT_FILENAME_DATA_ALLOCA(ptr, ptr_out) \
- GET_C_CHARPTR_EXT_DATA_ALLOCA (ptr, FORMAT_FILENAME, ptr_out)
-#define GET_CHARPTR_EXT_FILENAME_DATA_ALLOCA(ptr, len, ptr_out, len_out) \
- GET_CHARPTR_EXT_DATA_ALLOCA (ptr, len, FORMAT_FILENAME, ptr_out, len_out)
-
-#define GET_C_CHARPTR_EXT_CTEXT_DATA_ALLOCA(ptr, ptr_out) \
- GET_C_CHARPTR_EXT_DATA_ALLOCA (ptr, FORMAT_CTEXT, ptr_out)
-#define GET_CHARPTR_EXT_CTEXT_DATA_ALLOCA(ptr, len, ptr_out, len_out) \
- GET_CHARPTR_EXT_DATA_ALLOCA (ptr, len, FORMAT_CTEXT, ptr_out, len_out)
-
-/* Maybe convert external charptr's data into internal format and store
- the result in alloca()'ed space.
-
- You may wonder why this is written in this fashion and not as a
- function call. With a little trickery it could certainly be
- written this way, but it won't work because of those DAMN GCC WANKERS
- who couldn't be bothered to handle alloca() properly on the x86
- architecture. (If you put a call to alloca() in the argument to
- a function call, the stack space gets allocated right in the
- middle of the arguments to the function call and you are unbelievably
- hosed.) */
-
-#ifdef MULE
-
-#define GET_CHARPTR_INT_DATA_ALLOCA(ptr, len, fmt, ptr_out, len_out) do \
-{ \
- Extcount gcida_len_in = (Extcount) (len); \
- Bytecount gcida_len_out; \
- CONST Extbyte *gcida_ptr_in = (ptr); \
- Bufbyte *gcida_ptr_out = \
- convert_from_external_format (gcida_ptr_in, gcida_len_in, \
- &gcida_len_out, fmt); \
- /* If the new string is identical to the old (will be the case most \
- of the time), just return the same string back. This saves \
- on alloca()ing, which can be useful on C alloca() machines and \
- on stack-space-challenged environments. */ \
- \
- if (gcida_len_in == gcida_len_out && \
- !memcmp (gcida_ptr_in, gcida_ptr_out, gcida_len_out)) \
- { \
- (ptr_out) = (Bufbyte *) gcida_ptr_in; \
- (len_out) = (Bytecount) gcida_len_in; \
- } \
- else \
- { \
- (ptr_out) = (Extbyte *) alloca (1 + gcida_len_out); \
- memcpy ((void *) ptr_out, gcida_ptr_out, 1 + gcida_len_out); \
- (len_out) = gcida_len_out; \
- } \
+#define DFC_SOURCE_LISP_LSTREAM_TO_ARGS(val) do { \
+ Lisp_Object dfc_sllta = (val); \
+ type_checking_assert (LSTREAMP (dfc_sllta)); \
+ dfc_source.lisp_object = dfc_sllta; \
+ dfc_simplified_source_type = DFC_TYPE_LISP_LSTREAM; \
} while (0)
-
-#else /* ! MULE */
-
-#define GET_CHARPTR_INT_DATA_ALLOCA(ptr, len, fmt, ptr_out, len_out) do \
-{ \
- (ptr_out) = (Bufbyte *) (ptr); \
- (len_out) = (Bytecount) (len); \
+#define DFC_SOURCE_LISP_OPAQUE_TO_ARGS(val) do { \
+ Lisp_Opaque *dfc_slota = XOPAQUE (val); \
+ dfc_source.data.ptr = OPAQUE_DATA (dfc_slota); \
+ dfc_source.data.len = OPAQUE_SIZE (dfc_slota); \
+ dfc_simplified_source_type = DFC_TYPE_DATA; \
} while (0)
-#endif /* ! MULE */
-
-#define GET_C_CHARPTR_INT_DATA_ALLOCA(ptr, fmt, ptr_out) do \
-{ \
- Bytecount gccida_ignored_len; \
- CONST Extbyte *gccida_ptr_in = (CONST Extbyte *) (ptr); \
- Bufbyte *gccida_ptr_out; \
- \
- GET_CHARPTR_INT_DATA_ALLOCA (gccida_ptr_in, \
- strlen ((char *) gccida_ptr_in), \
- fmt, \
- gccida_ptr_out, \
- gccida_ignored_len); \
- (ptr_out) = gccida_ptr_out; \
+/* Convert `sink' to args for dfc_convert_to_*_format() */
+#define DFC_SINK_ALLOCA_TO_ARGS(val) \
+ dfc_simplified_sink_type = DFC_TYPE_DATA
+#define DFC_SINK_C_STRING_ALLOCA_TO_ARGS(val) \
+ dfc_simplified_sink_type = DFC_TYPE_DATA
+#define DFC_SINK_MALLOC_TO_ARGS(val) \
+ dfc_simplified_sink_type = DFC_TYPE_DATA
+#define DFC_SINK_C_STRING_MALLOC_TO_ARGS(val) \
+ dfc_simplified_sink_type = DFC_TYPE_DATA
+#define DFC_SINK_LISP_STRING_TO_ARGS(val) \
+ dfc_simplified_sink_type = DFC_TYPE_DATA
+#define DFC_SINK_LISP_OPAQUE_TO_ARGS(val) \
+ dfc_simplified_sink_type = DFC_TYPE_DATA
+#define DFC_SINK_LISP_LSTREAM_TO_ARGS(val) do { \
+ Lisp_Object dfc_sllta = (val); \
+ type_checking_assert (LSTREAMP (dfc_sllta)); \
+ dfc_sink.lisp_object = dfc_sllta; \
+ dfc_simplified_sink_type = DFC_TYPE_LISP_LSTREAM; \
} while (0)
-
-#define GET_C_CHARPTR_INT_BINARY_DATA_ALLOCA(ptr, ptr_out) \
- GET_C_CHARPTR_INT_DATA_ALLOCA (ptr, FORMAT_BINARY, ptr_out)
-#define GET_CHARPTR_INT_BINARY_DATA_ALLOCA(ptr, len, ptr_out, len_out) \
- GET_CHARPTR_INT_DATA_ALLOCA (ptr, len, FORMAT_BINARY, ptr_out, len_out)
-
-#define GET_C_CHARPTR_INT_FILENAME_DATA_ALLOCA(ptr, ptr_out) \
- GET_C_CHARPTR_INT_DATA_ALLOCA (ptr, FORMAT_FILENAME, ptr_out)
-#define GET_CHARPTR_INT_FILENAME_DATA_ALLOCA(ptr, len, ptr_out, len_out) \
- GET_CHARPTR_INT_DATA_ALLOCA (ptr, len, FORMAT_FILENAME, ptr_out, len_out)
-
-#define GET_C_CHARPTR_INT_CTEXT_DATA_ALLOCA(ptr, ptr_out) \
- GET_C_CHARPTR_INT_DATA_ALLOCA (ptr, FORMAT_CTEXT, ptr_out)
-#define GET_CHARPTR_INT_CTEXT_DATA_ALLOCA(ptr, len, ptr_out, len_out) \
- GET_CHARPTR_INT_DATA_ALLOCA (ptr, len, FORMAT_CTEXT, ptr_out, len_out)
-
-
-/* Maybe convert Lisp string's data into ext-format and store the result in
- alloca()'ed space.
-
- You may wonder why this is written in this fashion and not as a
- function call. With a little trickery it could certainly be
- written this way, but it won't work because of those DAMN GCC WANKERS
- who couldn't be bothered to handle alloca() properly on the x86
- architecture. (If you put a call to alloca() in the argument to
- a function call, the stack space gets allocated right in the
- middle of the arguments to the function call and you are unbelievably
- hosed.) */
-
-#define GET_STRING_EXT_DATA_ALLOCA(s, fmt, ptr_out, len_out) do \
-{ \
- Extcount gseda_len_out; \
- struct Lisp_String *gseda_s = XSTRING (s); \
- Extbyte * gseda_ptr_out = \
- convert_to_external_format (string_data (gseda_s), \
- string_length (gseda_s), \
- &gseda_len_out, fmt); \
- (ptr_out) = (Extbyte *) alloca (1 + gseda_len_out); \
- memcpy ((void *) ptr_out, gseda_ptr_out, 1 + gseda_len_out); \
- (len_out) = gseda_len_out; \
+#define DFC_SINK_LISP_BUFFER_TO_ARGS(val) do { \
+ struct buffer *dfc_slbta = XBUFFER (val); \
+ dfc_sink.lisp_object = \
+ make_lisp_buffer_output_stream \
+ (dfc_slbta, BUF_PT (dfc_slbta), 0); \
+ dfc_simplified_sink_type = DFC_TYPE_LISP_LSTREAM; \
} while (0)
-
-#define GET_C_STRING_EXT_DATA_ALLOCA(s, fmt, ptr_out) do \
-{ \
- Extcount gcseda_ignored_len; \
- Extbyte *gcseda_ptr_out; \
- \
- GET_STRING_EXT_DATA_ALLOCA (s, fmt, gcseda_ptr_out, \
- gcseda_ignored_len); \
- (ptr_out) = (char *) gcseda_ptr_out; \
+/* Assign to the `sink' lvalue(s) using the converted data. */
+#define DFC_ALLOCA_USE_CONVERTED_DATA(sink) do { \
+ void * dfc_sink_ret = alloca (dfc_sink.data.len + 1); \
+ memcpy (dfc_sink_ret, dfc_sink.data.ptr, dfc_sink.data.len + 1); \
+ (DFC_CPP_CAR sink) = (unsigned char *) dfc_sink_ret; \
+ (DFC_CPP_CDR sink) = dfc_sink.data.len; \
} while (0)
-
-#define GET_STRING_BINARY_DATA_ALLOCA(s, ptr_out, len_out) \
- GET_STRING_EXT_DATA_ALLOCA (s, FORMAT_BINARY, ptr_out, len_out)
-#define GET_C_STRING_BINARY_DATA_ALLOCA(s, ptr_out) \
- GET_C_STRING_EXT_DATA_ALLOCA (s, FORMAT_BINARY, ptr_out)
-
-#define GET_STRING_FILENAME_DATA_ALLOCA(s, ptr_out, len_out) \
- GET_STRING_EXT_DATA_ALLOCA (s, FORMAT_FILENAME, ptr_out, len_out)
-#define GET_C_STRING_FILENAME_DATA_ALLOCA(s, ptr_out) \
- GET_C_STRING_EXT_DATA_ALLOCA (s, FORMAT_FILENAME, ptr_out)
-
-#define GET_STRING_OS_DATA_ALLOCA(s, ptr_out, len_out) \
- GET_STRING_EXT_DATA_ALLOCA (s, FORMAT_OS, ptr_out, len_out)
-#define GET_C_STRING_OS_DATA_ALLOCA(s, ptr_out) \
- GET_C_STRING_EXT_DATA_ALLOCA (s, FORMAT_OS, ptr_out)
-
-#define GET_STRING_CTEXT_DATA_ALLOCA(s, ptr_out, len_out) \
- GET_STRING_EXT_DATA_ALLOCA (s, FORMAT_CTEXT, ptr_out, len_out)
-#define GET_C_STRING_CTEXT_DATA_ALLOCA(s, ptr_out) \
- GET_C_STRING_EXT_DATA_ALLOCA (s, FORMAT_CTEXT, ptr_out)
-
+#define DFC_MALLOC_USE_CONVERTED_DATA(sink) do { \
+ void * dfc_sink_ret = xmalloc (dfc_sink.data.len + 1); \
+ memcpy (dfc_sink_ret, dfc_sink.data.ptr, dfc_sink.data.len + 1); \
+ (DFC_CPP_CAR sink) = (unsigned char *) dfc_sink_ret; \
+ (DFC_CPP_CDR sink) = dfc_sink.data.len; \
+} while (0)
+#define DFC_C_STRING_ALLOCA_USE_CONVERTED_DATA(sink) do { \
+ void * dfc_sink_ret = alloca (dfc_sink.data.len + 1); \
+ memcpy (dfc_sink_ret, dfc_sink.data.ptr, dfc_sink.data.len + 1); \
+ (sink) = (char *) dfc_sink_ret; \
+} while (0)
+#define DFC_C_STRING_MALLOC_USE_CONVERTED_DATA(sink) do { \
+ void * dfc_sink_ret = xmalloc (dfc_sink.data.len + 1); \
+ memcpy (dfc_sink_ret, dfc_sink.data.ptr, dfc_sink.data.len + 1); \
+ (sink) = (char *) dfc_sink_ret; \
+} while (0)
+#define DFC_LISP_STRING_USE_CONVERTED_DATA(sink) \
+ sink = make_string ((Bufbyte *) dfc_sink.data.ptr, dfc_sink.data.len)
+#define DFC_LISP_OPAQUE_USE_CONVERTED_DATA(sink) \
+ sink = make_opaque (dfc_sink.data.ptr, dfc_sink.data.len)
+#define DFC_LISP_LSTREAM_USE_CONVERTED_DATA(sink) /* data already used */
+#define DFC_LISP_BUFFER_USE_CONVERTED_DATA(sink) \
+ Lstream_delete (XLSTREAM (dfc_sink.lisp_object))
+
+/* Someday we might want to distinguish between Qnative and Qfile_name
+ by using coding-system aliases, but for now it suffices to have
+ these be identical. Qnative can be used as the coding_system
+ argument to TO_EXTERNAL_FORMAT() and TO_INTERNAL_FORMAT(). */
+#define Qnative Qfile_name
\f
/************************************************************************/
#define POINT_MARKER_P(marker) \
(XMARKER (marker)->buffer != 0 && \
- EQ ((marker), XMARKER (marker)->buffer->point_marker))
+ EQ (marker, XMARKER (marker)->buffer->point_marker))
#define BUF_MARKERS(buf) ((buf)->markers)
/* This is the initial (startup) directory, as used for the *scratch* buffer.
We're making this a global to make others aware of the startup directory.
+ `initial_directory' is stored in external format.
*/
extern char initial_directory[];
extern void init_initial_directory (void); /* initialize initial_directory */
#ifdef REL_ALLOC
-char *r_alloc (unsigned char **, unsigned long);
-char *r_re_alloc (unsigned char **, unsigned long);
+char *r_alloc (unsigned char **, size_t);
+char *r_re_alloc (unsigned char **, size_t);
void r_alloc_free (unsigned char **);
#define BUFFER_ALLOC(data, size) \
#else /* !REL_ALLOC */
#define BUFFER_ALLOC(data,size)\
- ((void) (data = xnew_array (Bufbyte, size)))
+ (data = xnew_array (Bufbyte, size))
#define BUFFER_REALLOC(data,size)\
((Bufbyte *) xrealloc (data, (size) * sizeof(Bufbyte)))
/* Avoid excess parentheses, or syntax errors may rear their heads. */
void convert_bufbyte_string_into_emchar_dynarr (CONST Bufbyte *str,
Bytecount len,
Emchar_dynarr *dyn);
-int convert_bufbyte_string_into_emchar_string (CONST Bufbyte *str,
- Bytecount len,
- Emchar *arr);
+Charcount convert_bufbyte_string_into_emchar_string (CONST Bufbyte *str,
+ Bytecount len,
+ Emchar *arr);
void convert_emchar_string_into_bufbyte_dynarr (Emchar *arr, int nels,
Bufbyte_dynarr *dyn);
Bufbyte *convert_emchar_string_into_malloced_string (Emchar *arr, int nels,
typically used to convert between uppercase and lowercase. For
compatibility reasons, trt tables are currently in the form of
a Lisp string of 256 characters, specifying the conversion for each
- of the first 256 Emacs characters (i.e. the 256 extended-ASCII
- characters). This should be generalized at some point to support
- conversions for all of the allowable Mule characters.
+ of the first 256 Emacs characters (i.e. the 256 Latin-1 characters).
+ This should be generalized at some point to support conversions for
+ all of the allowable Mule characters.
*/
/* The _1 macros are named as such because they assume that you have
set_string_char (XSTRING (table), (Charcount) ch1, ch2)
#ifdef MULE
-# define MAKE_MIRROR_TRT_TABLE() make_opaque (256, 0)
+# define MAKE_MIRROR_TRT_TABLE() make_opaque (OPAQUE_CLEAR, 256)
# define MIRROR_TRT_TABLE_AS_STRING(table) ((Bufbyte *) XOPAQUE_DATA (table))
# define MIRROR_TRT_TABLE_CHAR_1(table, ch) \
((Emchar) (MIRROR_TRT_TABLE_AS_STRING (table)[ch]))
return (DOWNCASE_TABLE_OF (buf, ch) == ch) ? UPCASE_TABLE_OF (buf, ch) : ch;
}
-/* Upcase a character known to be not upper case. */
+/* Upcase a character known to be not upper case. Unused. */
#define UPCASE1(buf, ch) UPCASE_TABLE_OF (buf, ch)
#define DOWNCASE(buf, ch) DOWNCASE_TABLE_OF (buf, ch)
-#endif /* _XEMACS_BUFFER_H_ */
+#endif /* INCLUDED_buffer_h_ */
projects / chise / xemacs-chise.git.1 / blobdiff