Mercurial > hg > xemacs-beta
view man/lispref/errors.texi @ 4995:8431b52e43b1
Move the various map* functions to C; add #'map-into.
src/ChangeLog addition:
2010-01-31 Aidan Kehoe <kehoea@parhasard.net>
Move #'mapcar*, #'mapcan, #'mapc, #'map, #'mapl, #'mapcon to C;
extend #'mapvector, #'mapconcat, #'mapcar to support more
SEQUENCES; have them all error with circular lists.
* fns.c (Fsubseq): Call CHECK_SEQUENCE here; Flength can return
from the debugger if it errors with a non-sequence, leading to a
crash in Fsubseq if sequence really is *not* a sequence.
(mapcarX): Rename mapcar1 to mapcarX; rework it comprehensively to
take an optional lisp output argument, and a varying number of
sequences.
Special-case a single list argument, as we used to, saving its
elements in the stack space for the results before calling
FUNCTION, so FUNCTION can corrupt the list all it
wants. dead_wrong_type_argument() in the other cases if we
encounter a non-cons where we expected a cons.
(Fmapconcat):
Accept further SEQUENCES after separator here. Special-case
the idiom (mapconcat 'identity SEQUENCE), don't even funcall.
(FmapcarX): Rename this from Fmapcar. Accept optional SEQUENCES.
(Fmapvector): Accept optional SEQUENCES.
(Fmapcan, Fmapc, Fmap): Move these here from cl-extra.el.
(Fmap_into): New function, as specified by Common Lisp.
(maplist): New function, the guts of the implementation of
Fmaplist and Fmapl.
(Fmaplist, Fmapl, Fmapcon): Move these from cl-extra.el.
(syms_of_fns):
Add a few needed symbols here, for the type tests
used by #'map. Add the new subrs, with aliases for #'mapc-internal
and #'mapcar.
* general-slots.h: Declare Qcoerce here, now it's used in both
indent.c and fns.c
* indent.c (syms_of_indent): Qcoerce is gone from here.
* lisp.h: Add ARRAYP(), SEQUENCEP(), and the corresponding CHECK_*
macros. Declare Fbit_vector, Fstring, FmapcarX, now other files
need to use them.
* data.c (Farrayp, Fsequencep): Use ARRAYP and SEQUENCEP, just
added to lisp.h
* buffer.c (Fbuffer_list): Now Fmapcar has been renamed FmapcarX
and takes MANY arguments, update this function to reflect that.
lisp/ChangeLog addition:
2010-01-31 Aidan Kehoe <kehoea@parhasard.net>
* cl.el (mapcar*): Delete; this is now in fns.c.
Use #'mapc, not #'mapc-internal in a couple of places.
* cl-macs.el (mapc, mapcar*, map): Delete these compiler macros
now the corresponding functions are in fns.c; there's no run-time
advantage to the macros.
* cl-extra.el (coerce): Extend the possible conversions here a
little; it's not remotely comprehensive yet, though it does allow
running slightly more Common Lisp code than previously.
(cl-mapcar-many): Delete.
(map, maplist, mapc, mapl, mapcan, mapcon): Move these to fns.c.
* bytecomp.el (byte-compile-maybe-mapc):
Use #'mapc itself, not #'mapc-internal, now the former is in C.
(mapcar*): Use #'byte-compile-maybe-mapc as this function's
byte-compile method, now a #'mapc that can take more than one
sequence is in C.
* obsolete.el (cl-mapc): Move this compatibility alias to this file.
* update-elc.el (do-autoload-commands): Use #'mapc, not
#'mapc-internal here.
author | Aidan Kehoe <kehoea@parhasard.net> |
---|---|
date | Sun, 31 Jan 2010 18:29:48 +0000 |
parents | 576fb035e263 |
children |
line wrap: on
line source
@c -*-texinfo-*- @c This is part of the XEmacs Lisp Reference Manual. @c Copyright (C) 1990, 1991, 1992, 1993 Free Software Foundation, Inc. @c See the file lispref.texi for copying conditions. @setfilename ../../info/errors.info @node Standard Errors, Standard Buffer-Local Variables, Building XEmacs and Object Allocation, Top @appendix Standard Errors Here is the complete list of the error symbols in standard Emacs, grouped by concept. The list includes each symbol's message (on the @code{error-message} property of the symbol) and a cross reference to a description of how the error can occur. Each error symbol has an @code{error-conditions} property that is a list of symbols. Normally this list includes the error symbol itself and the symbol @code{error}. Occasionally it includes additional symbols, which are intermediate classifications, narrower than @code{error} but broader than a single error symbol. For example, all the errors in accessing files have the condition @code{file-error}. As a special exception, the error symbol @code{quit} does not have the condition @code{error}, because quitting is not considered an error. @xref{Errors}, for an explanation of how errors are generated and handled. @table @code @item @var{symbol} @var{string}; @var{reference}. @item error @code{"error"}@* @xref{Errors}. @item quit @code{"Quit"}@* @xref{Quitting}. @item args-out-of-range @code{"Args out of range"}@* @xref{Sequences Arrays Vectors}. @item arith-error @code{"Arithmetic error"}@* See @code{/} and @code{%} in @ref{Numbers}. @item beginning-of-buffer @code{"Beginning of buffer"}@* @xref{Motion}. @item buffer-read-only @code{"Buffer is read-only"}@* @xref{Read Only Buffers}. @item cyclic-function-indirection @code{"Symbol's chain of function indirections contains a loop"}@* @xref{Function Indirection}. @c XEmacs feature @item domain-error @code{"Arithmetic domain error"}@* @item end-of-buffer @code{"End of buffer"}@* @xref{Motion}. @item end-of-file @code{"End of file during parsing"}@* This is not a @code{file-error}.@* @xref{Input Functions}. @item file-error This error and its subcategories do not have error-strings, because the error message is constructed from the data items alone when the error condition @code{file-error} is present.@* @xref{Files}. @item file-locked This is a @code{file-error}.@* @xref{File Locks}. @item file-already-exists This is a @code{file-error}.@* @xref{Writing to Files}. @item file-supersession This is a @code{file-error}.@* @xref{Modification Time}. @item invalid-byte-code @code{"Invalid byte code"}@* @xref{Byte Compilation}. @item invalid-function @code{"Invalid function"}@* @xref{Classifying Lists}. @item invalid-read-syntax @code{"Invalid read syntax"}@* @xref{Input Functions}. @item invalid-regexp @code{"Invalid regexp"}@* @xref{Regular Expressions}. @c XEmacs feature @item mark-inactive @code{"The mark is not active now"}@* @item no-catch @code{"No catch for tag"}@* @xref{Catch and Throw}. @c XEmacs feature @item overflow-error @code{"Arithmetic overflow error"}@* @c XEmacs feature @item protected-field @code{"Attempt to modify a protected field"}@* @c XEmacs feature @item range-error @code{"Arithmetic range error"}@* @item search-failed @code{"Search failed"}@* @xref{Searching and Matching}. @item setting-constant @code{"Attempt to set a constant symbol"}@* @xref{Constant Variables, , Variables that Never Change}. @c XEmacs feature @item singularity-error @code{"Arithmetic singularity error"}@* @c XEmacs feature @item tooltalk-error @code{"ToolTalk error"}@* @xref{ToolTalk Support}. @c XEmacs feature @item undefined-keystroke-sequence @code{"Undefined keystroke sequence"}@* @ignore FSF Emacs only @item undefined-color @code{"Undefined color"}@* @xref{Color Names}. @end ignore @item void-function @code{"Symbol's function definition is void"}@* @xref{Function Cells}. @item void-variable @code{"Symbol's value as variable is void"}@* @xref{Accessing Variables}. @item wrong-number-of-arguments @code{"Wrong number of arguments"}@* @xref{Classifying Lists}. @item wrong-type-argument @code{"Wrong type argument"}@* @xref{Type Predicates}. @end table These error types, which are all classified as special cases of @code{arith-error}, can occur on certain systems for invalid use of mathematical functions. @table @code @item domain-error @code{"Arithmetic domain error"}@* @xref{Math Functions}. @item overflow-error @code{"Arithmetic overflow error"}@* @xref{Math Functions}. @item range-error @code{"Arithmetic range error"}@* @xref{Math Functions}. @item singularity-error @code{"Arithmetic singularity error"}@* @xref{Math Functions}. @item underflow-error @code{"Arithmetic underflow error"}@* @xref{Math Functions}. @end table