@ifinfo
@dircategory XEmacs Editor
@direntry
-* Internals: (internals). XEmacs Internals Manual.
+* Internals: (internals). XEmacs Internals Manual.
@end direntry
Copyright @copyright{} 1992 - 1996 Ben Wing.
Copyright @copyright{} 1996, 1997 Sun Microsystems.
-Copyright @copyright{} 1994 - 1998 Free Software Foundation.
+Copyright @copyright{} 1994 - 1998, 2002, 2003 Free Software Foundation.
Copyright @copyright{} 1994, 1995 Board of Trustees, University of Illinois.
@titlepage
@title XEmacs Internals Manual
-@subtitle Version 1.3, August 1999
+@subtitle Version 1.4, March 2001
@author Ben Wing
@author Martin Buchholz
@author Hrvoje Niksic
@author Matthias Neubauer
+@author Olivier Galibert
@page
@vskip 0pt plus 1fill
@noindent
-Copyright @copyright{} 1992 - 1996 Ben Wing. @*
+Copyright @copyright{} 1992 - 1996, 2001 Ben Wing. @*
Copyright @copyright{} 1996, 1997 Sun Microsystems, Inc. @*
Copyright @copyright{} 1994 - 1998 Free Software Foundation. @*
Copyright @copyright{} 1994, 1995 Board of Trustees, University of Illinois.
@sp 2
-Version 1.3 @*
-August 1999.@*
+Version 1.4 @*
+March 2001.@*
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
@node Top, A History of Emacs, (dir), (dir)
@ifinfo
-This Info file contains v1.0 of the XEmacs Internals Manual.
+This Info file contains v1.4 of the XEmacs Internals Manual, March 2001.
@end ifinfo
@menu
* The XEmacs Object System (Abstractly Speaking)::
* How Lisp Objects Are Represented in C::
* Rules When Writing New C Code::
+* Regression Testing XEmacs::
* A Summary of the Various XEmacs Modules::
* Allocation of Objects in XEmacs Lisp::
+* Dumping::
* Events and the Event Loop::
* Evaluation; Stack Frames; Bindings::
* Symbols and Variables::
* Specifiers::
* Menus::
* Subprocesses::
-* Interface to X Windows::
-* Index:: Index including concepts, functions, variables,
- and other terms.
+* Interface to the X Window System::
+* Index::
- --- The Detailed Node Listing ---
+@detailmenu
-Here are other nodes that are inferiors of those already listed,
-mentioned here so you can get to them in one step:
+--- The Detailed Node Listing ---
A History of Emacs
* Through Version 18:: Unification prevails.
* Lucid Emacs:: One version 19 Emacs.
* GNU Emacs 19:: The other version 19 Emacs.
+* GNU Emacs 20:: The other version 20 Emacs.
* XEmacs:: The continuation of Lucid Emacs.
Rules When Writing New C Code
* General Coding Rules::
* Writing Lisp Primitives::
* Adding Global Lisp Variables::
+* Coding for Mule::
* Techniques for XEmacs Developers::
+Coding for Mule
+
+* Character-Related Data Types::
+* Working With Character and Byte Positions::
+* Conversion to and from External Data::
+* General Guidelines for Writing Mule-Aware Code::
+* An Example of Mule-Aware Code::
+
+Regression Testing XEmacs
+
A Summary of the Various XEmacs Modules
* Low-Level Modules::
* Modules for Interfacing with the Operating System::
* Modules for Interfacing with X Windows::
* Modules for Internationalization::
+* Modules for Regression Testing::
Allocation of Objects in XEmacs Lisp
* Allocation from Frob Blocks::
* lrecords::
* Low-level allocation::
-* Pure Space::
* Cons::
* Vector::
* Bit Vector::
* String::
* Compiled Function::
+Garbage Collection - Step by Step
+
+* Invocation::
+* garbage_collect_1::
+* mark_object::
+* gc_sweep::
+* sweep_lcrecords_1::
+* compact_string_chars::
+* sweep_strings::
+* sweep_bit_vectors_1::
+
+Dumping
+
+* Overview::
+* Data descriptions::
+* Dumping phase::
+* Reloading phase::
+
+Dumping phase
+
+* Object inventory::
+* Address allocation::
+* The header::
+* Data dumping::
+* Pointers dumping::
+
Events and the Event Loop
* Introduction to Events::
* Character Sets::
* Encodings::
* Internal Mule Encodings::
+* CCL::
Encodings
* Internal String Encoding::
* Internal Character Encoding::
-The Lisp Reader and Compiler
-
Lstreams
+* Creating an Lstream:: Creating an lstream object.
+* Lstream Types:: Different sorts of things that are streamed.
+* Lstream Functions:: Functions for working with lstreams.
+* Lstream Methods:: Creating new lstream types.
+
Consoles; Devices; Frames; Windows
* Introduction to Consoles; Devices; Frames; Windows::
* Point::
* Window Hierarchy::
+* The Window Object::
The Redisplay Mechanism
* Critical Redisplay Sections::
* Line Start Cache::
+* Redisplay Piece by Piece::
Extents
* Extent Ordering:: How extents are ordered internally.
* Format of the Extent Info:: The extent information in a buffer or string.
* Zero-Length Extents:: A weird special case.
-* Mathematics of Extent Ordering:: A rigorous foundation.
+* Mathematics of Extent Ordering:: A rigorous foundation.
* Extent Fragments:: Cached information useful for redisplay.
-Faces
-
-Glyphs
-
-Specifiers
-
-Menus
-
-Subprocesses
-
-Interface to X Windows
-
+@end detailmenu
@end menu
@node A History of Emacs, XEmacs From the Outside, Top, Top
@chapter A History of Emacs
-@cindex history of Emacs
+@cindex history of Emacs, a
+@cindex Emacs, a history of
@cindex Hackers (Steven Levy)
@cindex Levy, Steven
@cindex ITS (Incompatible Timesharing System)
@node Through Version 18
@section Through Version 18
+@cindex version 18, through
@cindex Gosling, James
@cindex Great Usenet Renaming
and Eric Benson, and the work was later taken over by Jamie Zawinski,
who became ``Mr. Lucid Emacs'' for many releases.
- A time line for Lucid Emacs/XEmacs is
+ A time line for Lucid Emacs is
@itemize @bullet
@item
@item
version 20.3 (the first stable version of XEmacs 20.x) released November 30,
1997.
+@item
version 20.4 released February 28, 1998.
+@item
+version 21.1.2 released May 14, 1999. (The version naming scheme was
+changed at this point: [a] the second version number is odd for stable
+versions, even for beta versions; [b] a third version number is added,
+replacing the "beta xxx" ending for beta versions and allowing for
+periodic maintenance releases for stable versions. Therefore, 21.0 was
+never "officially" released; similarly for 21.2, etc.)
+@item
+version 21.1.3 released June 26, 1999.
+@item
+version 21.1.4 released July 8, 1999.
+@item
+version 21.1.6 released August 14, 1999. (There was no 21.1.5.)
+@item
+version 21.1.7 released September 26, 1999.
+@item
+version 21.1.8 released November 2, 1999.
+@item
+version 21.1.9 released February 13, 2000.
+@item
+version 21.1.10 released May 7, 2000.
+@item
+version 21.1.10a released June 24, 2000.
+@item
+version 21.1.11 released July 18, 2000.
+@item
+version 21.1.12 released August 5, 2000.
+@item
+version 21.1.13 released January 7, 2001.
+@item
+version 21.1.14 released January 27, 2001.
@end itemize
@node GNU Emacs 19
@section GNU Emacs 19
@cindex GNU Emacs 19
+@cindex Emacs 19, GNU
+@cindex version 19, GNU Emacs
@cindex FSF Emacs
About a year after the initial release of Lucid Emacs, the FSF
@node GNU Emacs 20
@section GNU Emacs 20
@cindex GNU Emacs 20
+@cindex Emacs 20, GNU
+@cindex version 20, GNU Emacs
@cindex FSF Emacs
On February 2, 1997 work began on GNU Emacs to integrate Mule. The first
A more detailed history is contained in the XEmacs About page.
+ A time line for XEmacs is
+
+@itemize @bullet
+@item
+version 19.11 (first XEmacs) released September 13, 1994.
+@item
+version 19.12 released June 23, 1995.
+@item
+version 19.13 released September 1, 1995.
+@item
+version 19.14 released June 23, 1996.
+@item
+version 20.0 released February 9, 1997.
+@item
+version 19.15 released March 28, 1997.
+@item
+version 20.1 (not released to the net) April 15, 1997.
+@item
+version 20.2 released May 16, 1997.
+@item
+version 19.16 released October 31, 1997.
+@item
+version 20.3 (the first stable version of XEmacs 20.x) released November 30,
+1997.
+@item
+version 20.4 released February 28, 1998.
+@item
+version 21.0.60 released December 10, 1998. (The version naming scheme was
+changed at this point: [a] the second version number is odd for stable
+versions, even for beta versions; [b] a third version number is added,
+replacing the "beta xxx" ending for beta versions and allowing for
+periodic maintenance releases for stable versions. Therefore, 21.0 was
+never "officially" released; similarly for 21.2, etc.)
+@item
+version 21.0.61 released January 4, 1999.
+@item
+version 21.0.63 released February 3, 1999.
+@item
+version 21.0.64 released March 1, 1999.
+@item
+version 21.0.65 released March 5, 1999.
+@item
+version 21.0.66 released March 12, 1999.
+@item
+version 21.0.67 released March 25, 1999.
+@item
+version 21.1.2 released May 14, 1999. (This is the followup to 21.0.67.
+The second version number was bumped to indicate the beginning of the
+"stable" series.)
+@item
+version 21.1.3 released June 26, 1999.
+@item
+version 21.1.4 released July 8, 1999.
+@item
+version 21.1.6 released August 14, 1999. (There was no 21.1.5.)
+@item
+version 21.1.7 released September 26, 1999.
+@item
+version 21.1.8 released November 2, 1999.
+@item
+version 21.1.9 released February 13, 2000.
+@item
+version 21.1.10 released May 7, 2000.
+@item
+version 21.1.10a released June 24, 2000.
+@item
+version 21.1.11 released July 18, 2000.
+@item
+version 21.1.12 released August 5, 2000.
+@item
+version 21.1.13 released January 7, 2001.
+@item
+version 21.1.14 released January 27, 2001.
+@item
+version 21.2.9 released February 3, 1999.
+@item
+version 21.2.10 released February 5, 1999.
+@item
+version 21.2.11 released March 1, 1999.
+@item
+version 21.2.12 released March 5, 1999.
+@item
+version 21.2.13 released March 12, 1999.
+@item
+version 21.2.14 released May 14, 1999.
+@item
+version 21.2.15 released June 4, 1999.
+@item
+version 21.2.16 released June 11, 1999.
+@item
+version 21.2.17 released June 22, 1999.
+@item
+version 21.2.18 released July 14, 1999.
+@item
+version 21.2.19 released July 30, 1999.
+@item
+version 21.2.20 released November 10, 1999.
+@item
+version 21.2.21 released November 28, 1999.
+@item
+version 21.2.22 released November 29, 1999.
+@item
+version 21.2.23 released December 7, 1999.
+@item
+version 21.2.24 released December 14, 1999.
+@item
+version 21.2.25 released December 24, 1999.
+@item
+version 21.2.26 released December 31, 1999.
+@item
+version 21.2.27 released January 18, 2000.
+@item
+version 21.2.28 released February 7, 2000.
+@item
+version 21.2.29 released February 16, 2000.
+@item
+version 21.2.30 released February 21, 2000.
+@item
+version 21.2.31 released February 23, 2000.
+@item
+version 21.2.32 released March 20, 2000.
+@item
+version 21.2.33 released May 1, 2000.
+@item
+version 21.2.34 released May 28, 2000.
+@item
+version 21.2.35 released July 19, 2000.
+@item
+version 21.2.36 released October 4, 2000.
+@item
+version 21.2.37 released November 14, 2000.
+@item
+version 21.2.38 released December 5, 2000.
+@item
+version 21.2.39 released December 31, 2000.
+@item
+version 21.2.40 released January 8, 2001.
+@item
+version 21.2.41 released January 17, 2001.
+@item
+version 21.2.42 released January 20, 2001.
+@item
+version 21.2.43 released January 26, 2001.
+@item
+version 21.2.44 released February 8, 2001.
+@item
+version 21.2.45 released February 23, 2001.
+@item
+version 21.2.46 released March 21, 2001.
+@end itemize
+
@node XEmacs From the Outside, The Lisp Language, A History of Emacs, Top
@chapter XEmacs From the Outside
+@cindex XEmacs from the outside
+@cindex outside, XEmacs from the
@cindex read-eval-print
XEmacs appears to the outside world as an editor, but it is really a
displayable representations, and XEmacs provides a function
@code{redisplay()} that ensures that the display of all such objects
matches their internal state. Most of the time, a standard Lisp
-environment is in a @dfn{read-eval-print} loop -- i.e. ``read some Lisp
+environment is in a @dfn{read-eval-print} loop---i.e. ``read some Lisp
code, execute it, and print the results''. XEmacs has a similar loop:
@itemize @bullet
@node The Lisp Language, XEmacs From the Perspective of Building, XEmacs From the Outside, Top
@chapter The Lisp Language
+@cindex Lisp language, the
@cindex Lisp vs. C
@cindex C vs. Lisp
@cindex Lisp vs. Java
executed; this prints out the error and continues.) Routines can also
specify cleanup code (called an @dfn{unwind-protect}) that will be
called when control exits from a block of code, no matter how that exit
-occurs -- i.e. even if a function deeply nested below it causes a
+occurs---i.e. even if a function deeply nested below it causes a
non-local exit back to the top level.
Note that this facility has appeared in some recent vintages of C, in
you declared. This is actually considered a bug in Emacs Lisp and in
all other early dialects of Lisp, and was corrected in Common Lisp. (In
Common Lisp, you can still declare dynamically scoped variables if you
-want to -- they are sometimes useful -- but variables by default are
+want to---they are sometimes useful---but variables by default are
@dfn{lexically scoped} as in C.)
@end enumerate
Unfortunately, there is no perfect language. Static typing allows a
compiler to catch programmer errors and produce more efficient code, but
-makes programming more tedious and less fun. For the forseeable future,
+makes programming more tedious and less fun. For the foreseeable future,
an Ideal Editing and Programming Environment (and that is what XEmacs
aspires to) will be programmable in multiple languages: high level ones
like Lisp for user customization and prototyping, and lower level ones
@node XEmacs From the Perspective of Building, XEmacs From the Inside, The Lisp Language, Top
@chapter XEmacs From the Perspective of Building
+@cindex XEmacs from the perspective of building
+@cindex building, XEmacs from the perspective of
The heart of XEmacs is the Lisp environment, which is written in C.
This is contained in the @file{src/} subdirectory. Underneath
@node XEmacs From the Inside, The XEmacs Object System (Abstractly Speaking), XEmacs From the Perspective of Building, Top
@chapter XEmacs From the Inside
+@cindex XEmacs from the inside
+@cindex inside, XEmacs from the
Internally, XEmacs is quite complex, and can be very confusing. To
simplify things, it can be useful to think of XEmacs as containing an
frames are displayed correctly. It is periodically told (by the event
loop) to actually ``do its job'', i.e. snoop around and see what the
current state of the environment (mostly of the currently-existing
-windows, frames, and buffers) is, and make sure that that state matches
+windows, frames, and buffers) is, and make sure that state matches
what's actually displayed. It keeps lots and lots of information around
(such as what is actually being displayed currently, and what the
environment was last time it checked) so that it can minimize the work
When the Lisp initialization code is done, the C code enters the event
loop, and stays there for the duration of the XEmacs process. The code
-for the event loop is contained in @file{keyboard.c}, and is called
+for the event loop is contained in @file{cmdloop.c}, and is called
@code{Fcommand_loop_1()}. Note that this event loop could very well be
written in Lisp, and in fact a Lisp version exists; but apparently,
doing this makes XEmacs run noticeably slower.
@node The XEmacs Object System (Abstractly Speaking), How Lisp Objects Are Represented in C, XEmacs From the Inside, Top
@chapter The XEmacs Object System (Abstractly Speaking)
+@cindex XEmacs object system (abstractly speaking), the
+@cindex object system (abstractly speaking), the XEmacs
At the heart of the Lisp interpreter is its management of objects.
XEmacs Lisp contains many built-in objects, some of which are
An object representing a single character of text; chars behave like
integers in many ways but are logically considered text rather than
numbers and have a different read syntax. (the read syntax for a char
-contains the char itself or some textual encoding of it -- for example,
+contains the char itself or some textual encoding of it---for example,
a Japanese Kanji character might be encoded as @samp{^[$(B#&^[(B} using the
-ISO-2022 encoding standard -- rather than the numerical representation
+ISO-2022 encoding standard---rather than the numerical representation
of the char; this way, if the mapping between chars and integers
changes, which is quite possible for Kanji characters and other extended
characters, the same character will still be created. Note that some
@node How Lisp Objects Are Represented in C, Rules When Writing New C Code, The XEmacs Object System (Abstractly Speaking), Top
@chapter How Lisp Objects Are Represented in C
+@cindex Lisp objects are represented in C, how
+@cindex objects are represented in C, how Lisp
+@cindex represented in C, how Lisp objects are
Lisp objects are represented in C using a 32-bit or 64-bit machine word
(depending on the processor; i.e. DEC Alphas use 64-bit Lisp objects and
[ 3 3 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 ]
[ 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 ]
- <---> ^ <------------------------------------------------------>
- tag | a pointer to a structure, or an integer
- |
- mark bit
-@end example
-
-The tag describes the type of the Lisp object. For integers and chars,
-the lower 28 bits contain the value of the integer or char; for all
-others, the lower 28 bits contain a pointer. The mark bit is used
-during garbage-collection, and is always 0 when garbage collection is
-not happening. (The way that garbage collection works, basically, is that it
-loops over all places where Lisp objects could exist -- this includes
-all global variables in C that contain Lisp objects [including
-@code{Vobarray}, the C equivalent of @code{obarray}; through this, all
-Lisp variables will get marked], plus various other places -- and
-recursively scans through the Lisp objects, marking each object it finds
-by setting the mark bit. Then it goes through the lists of all objects
-allocated, freeing the ones that are not marked and turning off the mark
-bit of the ones that are marked.)
-
-Lisp objects use the typedef @code{Lisp_Object}, but the actual C type
-used for the Lisp object can vary. It can be either a simple type
-(@code{long} on the DEC Alpha, @code{int} on other machines) or a
-structure whose fields are bit fields that line up properly (actually, a
-union of structures is used). Generally the simple integral type is
-preferable because it ensures that the compiler will actually use a
-machine word to represent the object (some compilers will use more
-general and less efficient code for unions and structs even if they can
-fit in a machine word). The union type, however, has the advantage of
-stricter type checking (if you accidentally pass an integer where a Lisp
-object is desired, you get a compile error), and it makes it easier to
-decode Lisp objects when debugging. The choice of which type to use is
-determined by the preprocessor constant @code{USE_UNION_TYPE} which is
-defined via the @code{--use-union-type} option to @code{configure}.
-
-@cindex record type
-
-Note that there are only eight types that the tag can represent, but
-many more actual types than this. This is handled by having one of the
-tag types specify a meta-type called a @dfn{record}; for all such
-objects, the first four bytes of the pointed-to structure indicate what
-the actual type is.
-
-Note also that having 28 bits for pointers and integers restricts a lot
-of things to 256 megabytes of memory. (Basically, enough pointers and
-indices and whatnot get stuffed into Lisp objects that the total amount
-of memory used by XEmacs can't grow above 256 megabytes. In older
-versions of XEmacs and GNU Emacs, the tag was 5 bits wide, allowing for
-32 types, which was more than the actual number of types that existed at
-the time, and no ``record'' type was necessary. However, this limited
-the editor to 64 megabytes total, which some users who edited large
-files might conceivably exceed.)
-
-Also, note that there is an implicit assumption here that all pointers
-are low enough that the top bits are all zero and can just be chopped
-off. On standard machines that allocate memory from the bottom up (and
-give each process its own address space), this works fine. Some
-machines, however, put the data space somewhere else in memory
-(e.g. beginning at 0x80000000). Those machines cope by defining
-@code{DATA_SEG_BITS} in the corresponding @file{m/} or @file{s/} file to
-the proper mask. Then, pointers retrieved from Lisp objects are
-automatically OR'ed with this value prior to being used.
-
-A corollary of the previous paragraph is that @strong{(pointers to)
-stack-allocated structures cannot be put into Lisp objects}. The stack
-is generally located near the top of memory; if you put such a pointer
-into a Lisp object, it will get its top bits chopped off, and you will
-lose.
-
-Actually, there's an alternative representation of a @code{Lisp_Object},
-invented by Kyle Jones, that is used when the
-@code{--use-minimal-tagbits} option to @code{configure} is used. In
-this case the 2 lower bits are used for the tag bits. This
-representation assumes that pointers to structs are always aligned to
-multiples of 4, so the lower 2 bits are always zero.
-
-@example
- [ 3 3 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 ]
- [ 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 ]
-
<---------------------------------------------------------> <->
a pointer to a structure, or an integer tag
@end example
A tag of 00 is used for all pointer object types, a tag of 10 is used
for characters, and the other two tags 01 and 11 are joined together to
-form the integer object type. The markbit is moved to part of the
-structure being pointed at (integers and chars do not need to be marked,
-since no memory is allocated). This representation has these
-advantages:
-
-@enumerate
-@item
-31 bits can be used for Lisp Integers.
-@item
-@emph{Any} pointer can be represented directly, and no bit masking
-operations are necessary.
-@end enumerate
+form the integer object type. This representation gives us 31 bit
+integers and 30 bit characters, while pointers are represented directly
+without any bit masking or shifting. This representation, though,
+assumes that pointers to structs are always aligned to multiples of 4,
+so the lower 2 bits are always zero.
-The disadvantages are:
-
-@enumerate
-@item
-An extra level of indirection is needed when accessing the object types
-that were not record types. So checking whether a Lisp object is a cons
-cell becomes a slower operation.
-@item
-Mark bits can no longer be stored directly in Lisp objects, so another
-place for them must be found. This means that a cons cell requires more
-memory than merely room for 2 lisp objects, leading to extra memory use.
-@end enumerate
+Lisp objects use the typedef @code{Lisp_Object}, but the actual C type
+used for the Lisp object can vary. It can be either a simple type
+(@code{long} on the DEC Alpha, @code{int} on other machines) or a
+structure whose fields are bit fields that line up properly (actually, a
+union of structures is used). The choice of which type to use is
+determined by the preprocessor constant @code{USE_UNION_TYPE} which is
+defined via the @code{--use-union-type} option to @code{configure}.
-Various macros are used to construct Lisp objects and extract the
-components. Macros of the form @code{XINT()}, @code{XCHAR()},
-@code{XSTRING()}, @code{XSYMBOL()}, etc. mask out the pointer/integer
-field and cast it to the appropriate type. All of the macros that
-construct pointers will @code{OR} with @code{DATA_SEG_BITS} if
-necessary. @code{XINT()} needs to be a bit tricky so that negative
-numbers are properly sign-extended: Usually it does this by shifting the
-number four bits to the left and then four bits to the right. This
-assumes that the right-shift operator does an arithmetic shift (i.e. it
-leaves the most-significant bit as-is rather than shifting in a zero, so
-that it mimics a divide-by-two even for negative numbers). Not all
-machines/compilers do this, and on the ones that don't, a more
-complicated definition is selected by defining
-@code{EXPLICIT_SIGN_EXTEND}.
-
-Note that when @code{ERROR_CHECK_TYPECHECK} is defined, the extractor
-macros become more complicated -- they check the tag bits and/or the
+Generally the simple integral type is preferable because it ensures that
+the compiler will actually use a machine word to represent the object
+(some compilers will use more general and less efficient code for unions
+and structs even if they can fit in a machine word). The union type,
+however, has the advantage of stricter @emph{static} type checking.
+Places where a @code{Lisp_Object} is mistakenly passed to a routine
+expecting an @code{int} (or vice-versa), or a check is written @samp{if
+(foo)} (instead of @samp{if (!NILP (foo))}, will be flagged as errors.
+None of these lead to the expected results! @code{Qnil} is not
+represented as 0 (so @samp{if (foo)} will *ALWAYS* be true for a
+@code{Lisp_Object}), and the representation of an integer as a
+@code{Lisp_Object} is not just the integer's numeric value, but usually
+2x the integer +/- 1.)
+
+There used to be a claim that the union type simplified debugging.
+There may have been a grain of truth to this pre-19.8, when there was no
+@samp{lrecord} type and all objects had a separate type appearing in the
+tag. Nowadays, however, there is no debugging gain, and in fact
+frequent debugging *@emph{loss}*, since many debuggers don't handle
+unions very well, and usually there is no way to directly specify a
+union from a debugging prompt.
+
+Furthermore, release builds should *@emph{not}* be done with union type
+because (a) you may get less efficiency, with compilers that can't
+figure out how to optimize the union into a machine word; (b) even
+worse, the union type often triggers miscompilation, especially when
+combined with Mule and error-checking. This has been the case at
+various times when using GCC and MS VC, at least with @samp{--pdump}.
+Therefore, be warned!
+
+As of 2002 4Q, miscompilation is known to happen with current versions
+of @strong{Microsoft VC++} and @strong{GCC in combination with Mule,
+pdump, and KKCC} (no error checking).
+
+Various macros are used to convert between Lisp_Objects and the
+corresponding C type. Macros of the form @code{XINT()}, @code{XCHAR()},
+@code{XSTRING()}, @code{XSYMBOL()}, do any required bit shifting and/or
+masking and cast it to the appropriate type. @code{XINT()} needs to be
+a bit tricky so that negative numbers are properly sign-extended. Since
+integers are stored left-shifted, if the right-shift operator does an
+arithmetic shift (i.e. it leaves the most-significant bit as-is rather
+than shifting in a zero, so that it mimics a divide-by-two even for
+negative numbers) the shift to remove the tag bit is enough. This is
+the case on all the systems we support.
+
+Note that when @code{ERROR_CHECK_TYPECHECK} is defined, the converter
+macros become more complicated---they check the tag bits and/or the
type field in the first four bytes of a record type to ensure that the
object is really of the correct type. This is great for catching places
-where an incorrect type is being dereferenced -- this typically results
+where an incorrect type is being dereferenced---this typically results
in a pointer being dereferenced as the wrong type of structure, with
unpredictable (and sometimes not easily traceable) results.
There are similar @code{XSET@var{TYPE}()} macros that construct a Lisp
object. These macros are of the form @code{XSET@var{TYPE}
-(@var{lvalue}, @var{result})},
-i.e. they have to be a statement rather than just used in an expression.
-The reason for this is that standard C doesn't let you ``construct'' a
-structure (but GCC does). Granted, this sometimes isn't too convenient;
-for the case of integers, at least, you can use the function
-@code{make_int()}, which constructs and @emph{returns} an integer
-Lisp object. Note that the @code{XSET@var{TYPE}()} macros are also
-affected by @code{ERROR_CHECK_TYPECHECK} and make sure that the
-structure is of the right type in the case of record types, where the
-type is contained in the structure.
+(@var{lvalue}, @var{result})}, i.e. they have to be a statement rather
+than just used in an expression. The reason for this is that standard C
+doesn't let you ``construct'' a structure (but GCC does). Granted, this
+sometimes isn't too convenient; for the case of integers, at least, you
+can use the function @code{make_int()}, which constructs and
+@emph{returns} an integer Lisp object. Note that the
+@code{XSET@var{TYPE}()} macros are also affected by
+@code{ERROR_CHECK_TYPECHECK} and make sure that the structure is of the
+right type in the case of record types, where the type is contained in
+the structure.
The C programmer is responsible for @strong{guaranteeing} that a
-Lisp_Object is is the correct type before using the @code{X@var{TYPE}}
+Lisp_Object is the correct type before using the @code{X@var{TYPE}}
macros. This is especially important in the case of lists. Use
@code{XCAR} and @code{XCDR} if a Lisp_Object is certainly a cons cell,
else use @code{Fcar()} and @code{Fcdr()}. Trust other C code, but not
Lisp code. On the other hand, if XEmacs has an internal logic error,
-it's better to crash immediately, so sprinkle ``unreachable''
-@code{abort()}s liberally about the source code.
-
-@node Rules When Writing New C Code, A Summary of the Various XEmacs Modules, How Lisp Objects Are Represented in C, Top
+it's better to crash immediately, so sprinkle @code{assert()}s and
+``unreachable'' @code{abort()}s liberally about the source code. Where
+performance is an issue, use @code{type_checking_assert},
+@code{bufpos_checking_assert}, and @code{gc_checking_assert}, which do
+nothing unless the corresponding configure error checking flag was
+specified.
+
+@node Rules When Writing New C Code, Regression Testing XEmacs, How Lisp Objects Are Represented in C, Top
@chapter Rules When Writing New C Code
+@cindex writing new C code, rules when
+@cindex C code, rules when writing new
+@cindex code, rules when writing new C
The XEmacs C Code is extremely complex and intricate, and there are many
rules that are more or less consistently followed throughout the code.
situations, often in code far away from where the actual breakage is.
@menu
+* A Reader's Guide to XEmacs Coding Conventions::
* General Coding Rules::
* Writing Lisp Primitives::
+* Writing Good Comments::
* Adding Global Lisp Variables::
+* Proper Use of Unsigned Types::
* Coding for Mule::
* Techniques for XEmacs Developers::
@end menu
+@node A Reader's Guide to XEmacs Coding Conventions
+@section A Reader's Guide to XEmacs Coding Conventions
+@cindex coding conventions
+@cindex reader's guide
+@cindex coding rules, naming
+
+Of course the low-level implementation language of XEmacs is C, but much
+of that uses the Lisp engine to do its work. However, because the code
+is ``inside'' of the protective containment shell around the ``reactor
+core,'' you'll see lots of complex ``plumbing'' needed to do the work
+and ``safety mechanisms,'' whose failure results in a meltdown. This
+section provides a quick overview (or review) of the various components
+of the implementation of Lisp objects.
+
+ Two typographic conventions help to identify C objects that implement
+Lisp objects. The first is that capitalized identifiers, especially
+beginning with the letters @samp{Q}, @samp{V}, @samp{F}, and @samp{S},
+for C variables and functions, and C macros with beginning with the
+letter @samp{X}, are used to implement Lisp. The second is that where
+Lisp uses the hyphen @samp{-} in symbol names, the corresponding C
+identifiers use the underscore @samp{_}. Of course, since XEmacs Lisp
+contains interfaces to many external libraries, those external names
+will follow the coding conventions their authors chose, and may overlap
+the ``XEmacs name space.'' However these cases are usually pretty
+obvious.
+
+ All Lisp objects are handled indirectly. The @code{Lisp_Object}
+type is usually a pointer to a structure, except for a very small number
+of types with immediate representations (currently characters and
+integers). However, these types cannot be directly operated on in C
+code, either, so they can also be considered indirect. Types that do
+not have an immediate representation always have a C typedef
+@code{Lisp_@var{type}} for a corresponding structure.
+@c #### mention l(c)records here?
+
+ In older code, it was common practice to pass around pointers to
+@code{Lisp_@var{type}}, but this is now deprecated in favor of using
+@code{Lisp_Object} for all function arguments and return values that are
+Lisp objects. The @code{X@var{type}} macro is used to extract the
+pointer and cast it to @code{(Lisp_@var{type} *)} for the desired type.
+
+ @strong{Convention}: macros whose names begin with @samp{X} operate on
+@code{Lisp_Object}s and do no type-checking. Many such macros are type
+extractors, but others implement Lisp operations in C (@emph{e.g.},
+@code{XCAR} implements the Lisp @code{car} function). These are unsafe,
+and must only be used where types of all data have already been checked.
+Such macros are only applied to @code{Lisp_Object}s. In internal
+implementations where the pointer has already been converted, the
+structure is operated on directly using the C @code{->} member access
+operator.
+
+ The @code{@var{type}P}, @code{CHECK_@var{type}}, and
+@code{CONCHECK_@var{type}} macros are used to test types. The first
+returns a Boolean value, and the latter signal errors. (The
+@samp{CONCHECK} variety allows execution to be CONtinued under some
+circumstances, thus the name.) Functions which expect to be passed user
+data invariably call @samp{CHECK} macros on arguments.
+
+ There are many types of specialized Lisp objects implemented in C, but
+the most pervasive type is the @dfn{symbol}. Symbols are used as
+identifiers, variables, and functions.
+
+ @strong{Convention}: Global variables whose names begin with @samp{Q}
+are constants whose value is a symbol. The name of the variable should
+be derived from the name of the symbol using the same rules as for Lisp
+primitives. Such variables allow the C code to check whether a
+particular @code{Lisp_Object} is equal to a given symbol. Symbols are
+Lisp objects, so these variables may be passed to Lisp primitives. (An
+alternative to the use of @samp{Q...} variables is to call the
+@code{intern} function at initialization in the
+@code{vars_of_@var{module}} function, which is hardly less efficient.)
+
+ @strong{Convention}: Global variables whose names begin with @samp{V}
+are variables that contain Lisp objects. The convention here is that
+all global variables of type @code{Lisp_Object} begin with @samp{V}, and
+no others do (not even integer and boolean variables that have Lisp
+equivalents). Most of the time, these variables have equivalents in
+Lisp, which are defined via the @samp{DEFVAR} family of macros, but some
+don't. Since the variable's value is a @code{Lisp_Object}, it can be
+passed to Lisp primitives.
+
+ The implementation of Lisp primitives is more complex.
+@strong{Convention}: Global variables with names beginning with @samp{S}
+contain a structure that allows the Lisp engine to identify and call a C
+function. In modern versions of XEmacs, these identifiers are almost
+always completely hidden in the @code{DEFUN} and @code{SUBR} macros, but
+you will encounter them if you look at very old versions of XEmacs or at
+GNU Emacs. @strong{Convention}: Functions with names beginning with
+@samp{F} implement Lisp primitives. Of course all their arguments and
+their return values must be Lisp_Objects. (This is hidden in the
+@code{DEFUN} macro.)
+
+
@node General Coding Rules
@section General Coding Rules
+@cindex coding rules, general
The C code is actually written in a dialect of C called @dfn{Clean C},
meaning that it can be compiled, mostly warning-free, with either a C or
tools means that a greater variety of development tools are available to
the developer.
-Almost every module contains a @code{syms_of_*()} function and a
-@code{vars_of_*()} function. The former declares any Lisp primitives
-you have defined and defines any symbols you will be using. The latter
-declares any global Lisp variables you have added and initializes global
-C variables in the module. For each such function, declare it in
-@file{symsinit.h} and make sure it's called in the appropriate place in
-@file{emacs.c}. @strong{Important}: There are stringent requirements on
-exactly what can go into these functions. See the comment in
-@file{emacs.c}. The reason for this is to avoid obscure unwanted
-interactions during initialization. If you don't follow these rules,
-you'll be sorry! If you want to do anything that isn't allowed, create
-a @code{complex_vars_of_*()} function for it. Doing this is tricky,
-though: You have to make sure your function is called at the right time
-so that all the initialization dependencies work out.
-
Every module includes @file{<config.h>} (angle brackets so that
@samp{--srcdir} works correctly; @file{config.h} may or may not be in
the same directory as the C sources) and @file{lisp.h}. @file{config.h}
system header files) to ensure that certain tricks played by various
@file{s/} and @file{m/} files work out correctly.
+When including header files, always use angle brackets, not double
+quotes, except when the file to be included is always in the same
+directory as the including file. If either file is a generated file,
+then that is not likely to be the case. In order to understand why we
+have this rule, imagine what happens when you do a build in the source
+directory using @samp{./configure} and another build in another
+directory using @samp{../work/configure}. There will be two different
+@file{config.h} files. Which one will be used if you @samp{#include
+"config.h"}?
+
+Almost every module contains a @code{syms_of_*()} function and a
+@code{vars_of_*()} function. The former declares any Lisp primitives
+you have defined and defines any symbols you will be using. The latter
+declares any global Lisp variables you have added and initializes global
+C variables in the module. @strong{Important}: There are stringent
+requirements on exactly what can go into these functions. See the
+comment in @file{emacs.c}. The reason for this is to avoid obscure
+unwanted interactions during initialization. If you don't follow these
+rules, you'll be sorry! If you want to do anything that isn't allowed,
+create a @code{complex_vars_of_*()} function for it. Doing this is
+tricky, though: you have to make sure your function is called at the
+right time so that all the initialization dependencies work out.
+
+Declare each function of these kinds in @file{symsinit.h}. Make sure
+it's called in the appropriate place in @file{emacs.c}. You never need
+to include @file{symsinit.h} directly, because it is included by
+@file{lisp.h}.
+
@strong{All global and static variables that are to be modifiable must
be declared uninitialized.} This means that you may not use the
``declare with initializer'' form for these variables, such as @code{int
segment is re-mapped so that it becomes part of the (unmodifiable) code
segment in the dumped executable. This allows this memory to be shared
among multiple running XEmacs processes. XEmacs is careful to place as
-much constant data as possible into initialized variables (in
-particular, into what's called the @dfn{pure space} -- see below) during
-the @file{temacs} phase.
+much constant data as possible into initialized variables during the
+@file{temacs} phase.
@cindex copy-on-write
@strong{Please note:} This kludge only works on a few systems nowadays,
macro style is:
@example
-#define FOO(var, value) do @{ \
- Lisp_Object FOO_value = (value); \
- ... /* compute using FOO_value */ \
- (var) = bar; \
+#define FOO(var, value) do @{ \
+ Lisp_Object FOO_value = (value); \
+ ... /* compute using FOO_value */ \
+ (var) = bar; \
@} while (0)
@end example
Elisp. There are two sets of macros that iterate over lists.
@code{EXTERNAL_LIST_LOOP_@var{n}} should be used when the list has been
supplied by the user, and cannot be trusted to be acyclic and
-nil-terminated. A @code{malformed-list} or @code{circular-list} error
+@code{nil}-terminated. A @code{malformed-list} or @code{circular-list} error
will be generated if the list being iterated over is not entirely
kosher. @code{LIST_LOOP_@var{n}}, on the other hand, is faster and less
safe, and can be used only on trusted lists.
@node Writing Lisp Primitives
@section Writing Lisp Primitives
+@cindex writing Lisp primitives
+@cindex Lisp primitives, writing
+@cindex primitives, writing Lisp
Lisp primitives are Lisp functions implemented in C. The details of
interfacing the C function so that Lisp can call it are handled by a few
@file{lisp.h} contains the definitions for important macros and
functions.
+@node Writing Good Comments
+@section Writing Good Comments
+@cindex writing good comments
+@cindex comments, writing good
+
+Comments are a lifeline for programmers trying to understand tricky
+code. In general, the less obvious it is what you are doing, the more
+you need a comment, and the more detailed it needs to be. You should
+always be on guard when you're writing code for stuff that's tricky, and
+should constantly be putting yourself in someone else's shoes and asking
+if that person could figure out without much difficulty what's going
+on. (Assume they are a competent programmer who understands the
+essentials of how the XEmacs code is structured but doesn't know much
+about the module you're working on or any algorithms you're using.) If
+you're not sure whether they would be able to, add a comment. Always
+err on the side of more comments, rather than less.
+
+Generally, when making comments, there is no need to attribute them with
+your name or initials. This especially goes for small,
+easy-to-understand, non-opinionated ones. Also, comments indicating
+where, when, and by whom a file was changed are @emph{strongly}
+discouraged, and in general will be removed as they are discovered.
+This is exactly what @file{ChangeLogs} are there for. However, it can
+occasionally be useful to mark exactly where (but not when or by whom)
+changes are made, particularly when making small changes to a file
+imported from elsewhere. These marks help when later on a newer version
+of the file is imported and the changes need to be merged. (If
+everything were always kept in CVS, there would be no need for this.
+But in practice, this often doesn't happen, or the CVS repository is
+later on lost or unavailable to the person doing the update.)
+
+When putting in an explicit opinion in a comment, you should
+@emph{always} attribute it with your name, and optionally the date.
+This also goes for long, complex comments explaining in detail the
+workings of something -- by putting your name there, you make it
+possible for someone who has questions about how that thing works to
+determine who wrote the comment so they can write to them. Preferably,
+use your actual name and not your initials, unless your initials are
+generally recognized (e.g. @samp{jwz}). You can use only your first
+name if it's obvious who you are; otherwise, give first and last name.
+If you're not a regular contributor, you might consider putting your
+email address in -- it may be in the ChangeLog, but after awhile
+ChangeLogs have a tendency of disappearing or getting
+muddled. (E.g. your comment may get copied somewhere else or even into
+another program, and tracking down the proper ChangeLog may be very
+difficult.)
+
+If you come across an opinion that is not or no longer valid, or you
+come across any comment that no longer applies but you want to keep it
+around, enclose it in @samp{[[ } and @samp{ ]]} marks and add a comment
+afterwards explaining why the preceding comment is no longer valid. Put
+your name on this comment, as explained above.
+
+Just as comments are a lifeline to programmers, incorrect comments are
+death. If you come across an incorrect comment, @strong{immediately}
+correct it or flag it as incorrect, as described in the previous
+paragraph. Whenever you work on a section of code, @emph{always} make
+sure to update any comments to be correct -- or, at the very least, flag
+them as incorrect.
+
+To indicate a "todo" or other problem, use four pound signs --
+i.e. @samp{####}.
+
@node Adding Global Lisp Variables
@section Adding Global Lisp Variables
+@cindex global Lisp variables, adding
+@cindex variables, adding global Lisp
Global variables whose names begin with @samp{Q} are constants whose
value is a symbol of a particular name. The name of the variable should
@code{defsymbol()} are no problem, but some linkers will complain about
multiply-defined symbols. The most insidious aspect of this is that
often the link will succeed anyway, but then the resulting executable
-will sometimes crash in obscure ways during certain operations! To
-avoid this problem, declare any symbols with common names (such as
+will sometimes crash in obscure ways during certain operations!
+
+To avoid this problem, declare any symbols with common names (such as
@code{text}) that are not obviously associated with this particular
-module in the module @file{general.c}.
+module in the file @file{general-slots.h}. The ``-slots'' suffix
+indicates that this is a file that is included multiple times in
+@file{general.c}. Redefinition of preprocessor macros allows the
+effects to be different in each context, so this is actually more
+convenient and less error-prone than doing it in your module.
Global variables whose names begin with @samp{V} are variables that
contain Lisp objects. The convention here is that all global variables
Lisp object, and you will be the one who's unhappy when you can't figure
out how your variable got overwritten.
+@node Proper Use of Unsigned Types
+@section Proper Use of Unsigned Types
+@cindex unsigned types, proper use of
+@cindex types, proper use of unsigned
+
+Avoid using @code{unsigned int} and @code{unsigned long} whenever
+possible. Unsigned types are viral -- any arithmetic or comparisons
+involving mixed signed and unsigned types are automatically converted to
+unsigned, which is almost certainly not what you want. Many subtle and
+hard-to-find bugs are created by careless use of unsigned types. In
+general, you should almost @emph{never} use an unsigned type to hold a
+regular quantity of any sort. The only exceptions are
+
+@enumerate
+@item
+When there's a reasonable possibility you will actually need all 32 or
+64 bits to store the quantity.
+@item
+When calling existing API's that require unsigned types. In this case,
+you should still do all manipulation using signed types, and do the
+conversion at the very threshold of the API call.
+@item
+In existing code that you don't want to modify because you don't
+maintain it.
+@item
+In bit-field structures.
+@end enumerate
+
+Other reasonable uses of @code{unsigned int} and @code{unsigned long}
+are representing non-quantities -- e.g. bit-oriented flags and such.
+
@node Coding for Mule
@section Coding for Mule
-@cindex Coding for Mule
+@cindex coding for Mule
+@cindex Mule, coding for
Although Mule support is not compiled by default in XEmacs, many people
are using it, and we consider it crucial that new code works correctly
@node Character-Related Data Types
@subsection Character-Related Data Types
+@cindex character-related data types
+@cindex data types, character-related
First, let's review the basic character-related datatypes used by
XEmacs. Note that the separate @code{typedef}s are not mandatory in the
The data representing the text in a buffer or string is logically a set
of @code{Bufbyte}s.
-XEmacs does not work with character formats all the time; when reading
-characters from the outside, it decodes them to an internal format, and
-likewise encodes them when writing. @code{Bufbyte} (in fact
+XEmacs does not work with the same character formats all the time; when
+reading characters from the outside, it decodes them to an internal
+format, and likewise encodes them when writing. @code{Bufbyte} (in fact
@code{unsigned char}) is the basic unit of XEmacs internal buffers and
-strings format.
+strings format. A @code{Bufbyte *} is the type that points at text
+encoded in the variable-width internal encoding.
One character can correspond to one or more @code{Bufbyte}s. In the
-current implementation, an ASCII character is represented by the same
-@code{Bufbyte}, and extended characters are represented by a sequence of
-@code{Bufbyte}s.
+current Mule implementation, an ASCII character is represented by the
+same @code{Bufbyte}, and other characters are represented by a sequence
+of two or more @code{Bufbyte}s.
-Without Mule support, a @code{Bufbyte} is equivalent to an
-@code{Emchar}.
+Without Mule support, there are exactly 256 characters, implicitly
+Latin-1, and each character is represented using one @code{Bufbyte}, and
+there is a one-to-one correspondence between @code{Bufbyte}s and
+@code{Emchar}s.
@item Bufpos
@itemx Charcount
A @code{Charcount} represents a number (count) of characters.
Logically, subtracting two @code{Bufpos} values yields a
@code{Charcount} value. Although all of these are @code{typedef}ed to
-@code{int}, we use them in preference to @code{int} to make it clear
-what sort of position is being used.
+@code{EMACS_INT}, we use them in preference to @code{EMACS_INT} to make
+it clear what sort of position is being used.
@code{Bufpos} and @code{Charcount} values are the only ones that are
ever visible to Lisp.
@cindex Bytind
@cindex Bytecount
A @code{Bytind} represents a byte position in a buffer or string. A
-@code{Bytecount} represents the distance between two positions in bytes.
+@code{Bytecount} represents the distance between two positions, in bytes.
The relationship between @code{Bytind} and @code{Bytecount} is the same
as the relationship between @code{Bufpos} and @code{Charcount}.
@node Working With Character and Byte Positions
@subsection Working With Character and Byte Positions
+@cindex character and byte positions, working with
+@cindex byte positions, working with character and
+@cindex positions, working with character and byte
Now that we have defined the basic character-related types, we can look
at the macros and functions designed for work with them and for
@table @code
@item MAX_EMCHAR_LEN
@cindex MAX_EMCHAR_LEN
-This preprocessor constant is the maximum number of buffer bytes per
-Emacs character, i.e. the byte length of an @code{Emchar}. It is useful
-when allocating temporary strings to keep a known number of characters.
-For instance:
+This preprocessor constant is the maximum number of buffer bytes to
+represent an Emacs character in the variable width internal encoding.
+It is useful when allocating temporary strings to keep a known number of
+characters. For instance:
@example
@group
@node Conversion to and from External Data
@subsection Conversion to and from External Data
+@cindex conversion to and from external data
+@cindex external data, conversion to and from
When an external function, such as a C library function, returns a
@code{char} pointer, you should almost never treat it as @code{Bufbyte}.
The interface to conversion between the internal and external
representations of text are the numerous conversion macros defined in
-@file{buffer.h}. Before looking at them, we'll look at the external
-formats supported by these macros.
-
-Currently meaningful formats are @code{FORMAT_BINARY},
-@code{FORMAT_FILENAME}, @code{FORMAT_OS}, and @code{FORMAT_CTEXT}. Here
-is a description of these.
+@file{buffer.h}. There used to be a fixed set of external formats
+supported by these macros, but now any coding system can be used with
+these macros. The coding system alias mechanism is used to create the
+following logical coding systems, which replace the fixed external
+formats. The (dontusethis-set-symbol-value-handler) mechanism was
+enhanced to make this possible (more work on that is needed - like
+remove the @code{dontusethis-} prefix).
@table @code
-@item FORMAT_BINARY
-Binary format. This is the simplest format and is what we use in the
-absence of a more appropriate format. This converts according to the
-@code{binary} coding system:
+@item Qbinary
+This is the simplest format and is what we use in the absence of a more
+appropriate format. This converts according to the @code{binary} coding
+system:
@enumerate a
@item
-On input, bytes 0--255 are converted into characters 0--255.
+On input, bytes 0--255 are converted into (implicitly Latin-1)
+characters 0--255. A non-Mule xemacs doesn't really know about
+different character sets and the fonts to display them, so the bytes can
+be treated as text in different 1-byte encodings by simply setting the
+appropriate fonts. So in a sense, non-Mule xemacs is a multi-lingual
+editor if, for example, different fonts are used to display text in
+different buffers, faces, or windows. The specifier mechanism gives the
+user complete control over this kind of behavior.
@item
On output, characters 0--255 are converted into bytes 0--255 and other
-characters are converted into `X'.
+characters are converted into `~'.
@end enumerate
-@item FORMAT_FILENAME
-Format used for filenames. In the original Mule, this is user-definable
-with the @code{pathname-coding-system} variable. For the moment, we
-just use the @code{binary} coding system.
+@item Qfile_name
+Format used for filenames. This is user-definable via either the
+@code{file-name-coding-system} or @code{pathname-coding-system} (now
+obsolete) variables.
-@item FORMAT_OS
+@item Qnative
Format used for the external Unix environment---@code{argv[]}, stuff
from @code{getenv()}, stuff from the @file{/etc/passwd} file, etc.
+Currently this is the same as Qfile_name. The two should be
+distinguished for clarity and possible future separation.
-Perhaps should be the same as FORMAT_FILENAME.
-
-@item FORMAT_CTEXT
-Compound--text format. This is the standard X format used for data
+@item Qctext
+Compound--text format. This is the standard X11 format used for data
stored in properties, selections, and the like. This is an 8-bit
-no-lock-shift ISO2022 coding system.
+no-lock-shift ISO2022 coding system. This is a real coding system,
+unlike Qfile_name, which is user-definable.
@end table
-The macros to convert between these formats and the internal format, and
-vice versa, follow.
+There are two fundamental macros to convert between external and
+internal format.
+
+@code{TO_INTERNAL_FORMAT} converts external data to internal format, and
+@code{TO_EXTERNAL_FORMAT} converts the other way around. The arguments
+each of these receives are a source type, a source, a sink type, a sink,
+and a coding system (or a symbol naming a coding system).
+
+A typical call looks like
+@example
+TO_EXTERNAL_FORMAT (LISP_STRING, str, C_STRING_MALLOC, ptr, Qfile_name);
+@end example
+
+which means that the contents of the lisp string @code{str} are written
+to a malloc'ed memory area which will be pointed to by @code{ptr}, after
+the function returns. The conversion will be done using the
+@code{file-name} coding system, which will be controlled by the user
+indirectly by setting or binding the variable
+@code{file-name-coding-system}.
+
+Some sources and sinks require two C variables to specify. We use some
+preprocessor magic to allow different source and sink types, and even
+different numbers of arguments to specify different types of sources and
+sinks.
+
+So we can have a call that looks like
+@example
+TO_INTERNAL_FORMAT (DATA, (ptr, len),
+ MALLOC, (ptr, len),
+ coding_system);
+@end example
+
+The parenthesized argument pairs are required to make the preprocessor
+magic work.
+
+Here are the different source and sink types:
@table @code
-@item GET_CHARPTR_INT_DATA_ALLOCA
-@itemx GET_CHARPTR_EXT_DATA_ALLOCA
-These two are the most basic conversion macros.
-@code{GET_CHARPTR_INT_DATA_ALLOCA} converts external data to internal
-format, and @code{GET_CHARPTR_EXT_DATA_ALLOCA} converts the other way
-around. The arguments each of these receives are @var{ptr} (pointer to
-the text in external format), @var{len} (length of texts in bytes),
-@var{fmt} (format of the external text), @var{ptr_out} (lvalue to which
-new text should be copied), and @var{len_out} (lvalue which will be
-assigned the length of the internal text in bytes). The resulting text
-is stored to a stack-allocated buffer. If the text doesn't need
-changing, these macros will do nothing, except for setting
-@var{len_out}.
-
-The macros above take many arguments which makes them unwieldy. For
-this reason, a number of convenience macros are defined with obvious
-functionality, but accepting less arguments. The general rule is that
-macros with @samp{INT} in their name convert text to internal Emacs
-representation, whereas the @samp{EXT} macros convert to external
-representation.
-
-@item GET_C_CHARPTR_INT_DATA_ALLOCA
-@itemx GET_C_CHARPTR_EXT_DATA_ALLOCA
-As their names imply, these macros work on C char pointers, which are
-zero-terminated, and thus do not need @var{len} or @var{len_out}
-parameters.
-
-@item GET_STRING_EXT_DATA_ALLOCA
-@itemx GET_C_STRING_EXT_DATA_ALLOCA
-These two macros convert a Lisp string into an external representation.
-The difference between them is that @code{GET_STRING_EXT_DATA_ALLOCA}
-stores its output to a generic string, providing @var{len_out}, the
-length of the resulting external string. On the other hand,
-@code{GET_C_STRING_EXT_DATA_ALLOCA} assumes that the caller will be
-satisfied with output string being zero-terminated.
-
-Note that for Lisp strings only one conversion direction makes sense.
-
-@item GET_C_CHARPTR_EXT_BINARY_DATA_ALLOCA
-@itemx GET_CHARPTR_EXT_BINARY_DATA_ALLOCA
-@itemx GET_STRING_BINARY_DATA_ALLOCA
-@itemx GET_C_STRING_BINARY_DATA_ALLOCA
-@itemx GET_C_CHARPTR_EXT_FILENAME_DATA_ALLOCA
-@itemx ...
-These macros convert internal text to a specific external
-representation, with the external format being encoded into the name of
-the macro. Note that the @code{GET_STRING_...} and
-@code{GET_C_STRING...} macros lack the @samp{EXT} tag, because they
-only make sense in that direction.
-
-@item GET_C_CHARPTR_INT_BINARY_DATA_ALLOCA
-@itemx GET_CHARPTR_INT_BINARY_DATA_ALLOCA
-@itemx GET_C_CHARPTR_INT_FILENAME_DATA_ALLOCA
-@itemx ...
-These macros convert external text of a specific format to its internal
-representation, with the external format being incoded into the name of
-the macro.
+@item @code{DATA, (ptr, len),}
+input data is a fixed buffer of size @var{len} at address @var{ptr}
+@item @code{ALLOCA, (ptr, len),}
+output data is placed in an alloca()ed buffer of size @var{len} pointed to by @var{ptr}
+@item @code{MALLOC, (ptr, len),}
+output data is in a malloc()ed buffer of size @var{len} pointed to by @var{ptr}
+@item @code{C_STRING_ALLOCA, ptr,}
+equivalent to @code{ALLOCA (ptr, len_ignored)} on output.
+@item @code{C_STRING_MALLOC, ptr,}
+equivalent to @code{MALLOC (ptr, len_ignored)} on output
+@item @code{C_STRING, ptr,}
+equivalent to @code{DATA, (ptr, strlen (ptr) + 1)} on input
+@item @code{LISP_STRING, string,}
+input or output is a Lisp_Object of type string
+@item @code{LISP_BUFFER, buffer,}
+output is written to @code{(point)} in lisp buffer @var{buffer}
+@item @code{LISP_LSTREAM, lstream,}
+input or output is a Lisp_Object of type lstream
+@item @code{LISP_OPAQUE, object,}
+input or output is a Lisp_Object of type opaque
@end table
+Often, the data is being converted to a '\0'-byte-terminated string,
+which is the format required by many external system C APIs. For these
+purposes, a source type of @code{C_STRING} or a sink type of
+@code{C_STRING_ALLOCA} or @code{C_STRING_MALLOC} is appropriate.
+Otherwise, we should try to keep XEmacs '\0'-byte-clean, which means
+using (ptr, len) pairs.
+
+The sinks to be specified must be lvalues, unless they are the lisp
+object types @code{LISP_LSTREAM} or @code{LISP_BUFFER}.
+
+For the sink types @code{ALLOCA} and @code{C_STRING_ALLOCA}, the
+resulting text is stored in a stack-allocated buffer, which is
+automatically freed on returning from the function. However, the sink
+types @code{MALLOC} and @code{C_STRING_MALLOC} return @code{xmalloc()}ed
+memory. The caller is responsible for freeing this memory using
+@code{xfree()}.
+
+Note that it doesn't make sense for @code{LISP_STRING} to be a source
+for @code{TO_INTERNAL_FORMAT} or a sink for @code{TO_EXTERNAL_FORMAT}.
+You'll get an assertion failure if you try.
+
+
@node General Guidelines for Writing Mule-Aware Code
@subsection General Guidelines for Writing Mule-Aware Code
+@cindex writing Mule-aware code, general guidelines for
+@cindex Mule-aware code, general guidelines for writing
+@cindex code, general guidelines for writing Mule-aware
This section contains some general guidance on how to write Mule-aware
code, as well as some pitfalls you should avoid.
buffers literally.
This means that when a system function, such as @code{readdir}, returns
-a string, you need to convert it using one of the conversion macros
+a string, you may need to convert it using one of the conversion macros
described in the previous chapter, before passing it further to Lisp.
-In the case of @code{readdir}, you would use the
-@code{GET_C_CHARPTR_INT_FILENAME_DATA_ALLOCA} macro.
+
+Actually, most of the basic system functions that accept '\0'-terminated
+string arguments, like @code{stat()} and @code{open()}, have been
+@strong{encapsulated} so that they are they @code{always} do internal to
+external conversion themselves. This means you must pass internally
+encoded data, typically the @code{XSTRING_DATA} of a Lisp_String to
+these functions. This is actually a design bug, since it unexpectedly
+changes the semantics of the system functions. A better design would be
+to provide separate versions of these system functions that accepted
+Lisp_Objects which were lisp strings in place of their current
+@code{char *} arguments.
+
+@example
+int stat_lisp (Lisp_Object path, struct stat *buf); /* Implement me */
+@end example
Also note that many internal functions, such as @code{make_string},
accept Bufbytes, which removes the need for them to convert the data
@node An Example of Mule-Aware Code
@subsection An Example of Mule-Aware Code
+@cindex code, an example of Mule-aware
+@cindex Mule-aware code, an example of
-As an example of Mule-aware code, we shall will analyze the
-@code{string} function, which conses up a Lisp string from the character
-arguments it receives. Here is the definition, pasted from
-@code{alloc.c}:
+As an example of Mule-aware code, we will analyze the @code{string}
+function, which conses up a Lisp string from the character arguments it
+receives. Here is the definition, pasted from @code{alloc.c}:
@example
@group
@node Techniques for XEmacs Developers
@section Techniques for XEmacs Developers
+@cindex techniques for XEmacs developers
+@cindex developers, techniques for XEmacs
+@cindex Purify
+@cindex Quantify
+To make a purified XEmacs, do: @code{make puremacs}.
To make a quantified XEmacs, do: @code{make quantmacs}.
-You simply can't dump Quantified and Purified images. Run the image
-like so: @code{quantmacs -batch -l loadup.el run-temacs @var{xemacs-args...}}.
+You simply can't dump Quantified and Purified images (unless using the
+portable dumper). Purify gets confused when xemacs frees memory in one
+process that was allocated in a @emph{different} process on a different
+machine!. Run it like so:
+@example
+temacs -batch -l loadup.el run-temacs @var{xemacs-args...}
+@end example
+@cindex error checking
Before you go through the trouble, are you compiling with all
-debugging and error-checking off? If not try that first. Be warned
+debugging and error-checking off? If not, try that first. Be warned
that while Quantify is directly responsible for quite a few
optimizations which have been made to XEmacs, doing a run which
generates results which can be acted upon is not necessarily a trivial
If you want to make XEmacs faster, target your favorite slow benchmark,
run a profiler like Quantify, @code{gprof}, or @code{tcov}, and figure
-out where the cycles are going. Specific projects:
+out where the cycles are going. In many cases you can localize the
+problem (because a particular new feature or even a single patch
+elicited it). Don't hesitate to use brute force techniques like a
+global counter incremented at strategic places, especially in
+combination with other performance indications (@emph{e.g.}, degree of
+buffer fragmentation into extents).
+
+Specific projects:
@itemize @bullet
@item
@item
Speed up redisplay.
@item
-Speed up syntax highlighting. Maybe moving some of the syntax
-highlighting capabilities into C would make a difference.
+Speed up syntax highlighting. It was suggested that ``maybe moving some
+of the syntax highlighting capabilities into C would make a
+difference.'' Wrong idea, I think. When processing one large file a
+particular low-level routine was being called 40 @emph{million} times
+simply for @emph{one} call to @code{newline-and-indent}. Syntax
+highlighting needs to be rewritten to use a reliable, fast parser, then
+to trust the pre-parsed structure, and only do re-highlighting locally
+to a text change. Modern machines are fast enough to implement such
+parsers in Lisp; but no machine will ever be fast enough to deal with
+quadratic (or worse) algorithms!
@item
Implement tail recursion in Emacs Lisp (hard!).
@end itemize
calls in elisp are especially expensive. Iterating over a long list is
going to be 30 times faster implemented in C than in Elisp.
-To get started debugging XEmacs, take a look at the @file{gdbinit} and
-@file{dbxrc} files in the @file{src} directory.
-@xref{Q2.1.15 - How to Debug an XEmacs problem with a debugger,,,
-xemacs-faq, XEmacs FAQ}.
+Heavily used small code fragments need to be fast. The traditional way
+to implement such code fragments in C is with macros. But macros in C
+are known to be broken.
+
+@cindex macro hygiene
+Macro arguments that are repeatedly evaluated may suffer from repeated
+side effects or suboptimal performance.
+
+Variable names used in macros may collide with caller's variables,
+causing (at least) unwanted compiler warnings.
+
+In order to solve these problems, and maintain statement semantics, one
+should use the @code{do @{ ... @} while (0)} trick while trying to
+reference macro arguments exactly once using local variables.
+
+Let's take a look at this poor macro definition:
+
+@example
+#define MARK_OBJECT(obj) \
+ if (!marked_p (obj)) mark_object (obj), did_mark = 1
+@end example
+
+This macro evaluates its argument twice, and also fails if used like this:
+@example
+ if (flag) MARK_OBJECT (obj); else do_something();
+@end example
+
+A much better definition is
+
+@example
+#define MARK_OBJECT(obj) do @{ \
+ Lisp_Object mo_obj = (obj); \
+ if (!marked_p (mo_obj)) \
+ @{ \
+ mark_object (mo_obj); \
+ did_mark = 1; \
+ @} \
+@} while (0)
+@end example
+
+Notice the elimination of double evaluation by using the local variable
+with the obscure name. Writing safe and efficient macros requires great
+care. The one problem with macros that cannot be portably worked around
+is, since a C block has no value, a macro used as an expression rather
+than a statement cannot use the techniques just described to avoid
+multiple evaluation.
+
+@cindex inline functions
+In most cases where a macro has function semantics, an inline function
+is a better implementation technique. Modern compiler optimizers tend
+to inline functions even if they have no @code{inline} keyword, and
+configure magic ensures that the @code{inline} keyword can be safely
+used as an additional compiler hint. Inline functions used in a single
+.c files are easy. The function must already be defined to be
+@code{static}. Just add another @code{inline} keyword to the
+definition.
+
+@example
+inline static int
+heavily_used_small_function (int arg)
+@{
+ ...
+@}
+@end example
+
+Inline functions in header files are trickier, because we would like to
+make the following optimization if the function is @emph{not} inlined
+(for example, because we're compiling for debugging). We would like the
+function to be defined externally exactly once, and each calling
+translation unit would create an external reference to the function,
+instead of including a definition of the inline function in the object
+code of every translation unit that uses it. This optimization is
+currently only available for gcc. But you don't have to worry about the
+trickiness; just define your inline functions in header files using this
+pattern:
+
+@example
+INLINE_HEADER int
+i_used_to_be_a_crufty_macro_but_look_at_me_now (int arg);
+INLINE_HEADER int
+i_used_to_be_a_crufty_macro_but_look_at_me_now (int arg)
+@{
+ ...
+@}
+@end example
+
+The declaration right before the definition is to prevent warnings when
+compiling with @code{gcc -Wmissing-declarations}. I consider issuing
+this warning for inline functions a gcc bug, but the gcc maintainers disagree.
+
+@cindex inline functions, headers
+@cindex header files, inline functions
+Every header which contains inline functions, either directly by using
+@code{INLINE_HEADER} or indirectly by using @code{DECLARE_LRECORD} must
+be added to @file{inline.c}'s includes to make the optimization
+described above work. (Optimization note: if all INLINE_HEADER
+functions are in fact inlined in all translation units, then the linker
+can just discard @code{inline.o}, since it contains only unreferenced code).
+
+To get started debugging XEmacs, take a look at the @file{.gdbinit} and
+@file{.dbxrc} files in the @file{src} directory. See the section in the
+XEmacs FAQ on How to Debug an XEmacs problem with a debugger.
After making source code changes, run @code{make check} to ensure that
-you haven't introduced any regressions. If you're feeling ambitious,
-you can try to improve the test suite in @file{tests/automated}.
+you haven't introduced any regressions. If you want to make xemacs more
+reliable, please improve the test suite in @file{tests/automated}.
+
+Did you make sure you didn't introduce any new compiler warnings?
+
+Before submitting a patch, please try compiling at least once with
+
+@example
+configure --with-mule --use-union-type --error-checking=all
+@end example
Here are things to know when you create a new source file:
Generated header files should be included using the @code{#include <...>} syntax,
not the @code{#include "..."} syntax. The generated headers are:
-@file{config.h puresize-adjust.h sheap-adjust.h paths.h Emacs.ad.h}
+@file{config.h sheap-adjust.h paths.h Emacs.ad.h}
The basic rule is that you should assume builds using @code{--srcdir}
and the @code{#include <...>} syntax needs to be used when the
@code{"lisp.h"}. It is the responsibility of the @file{.c} files that
use it to do so.
-@item
-If the header uses @code{INLINE}, either directly or through
-@code{DECLARE_LRECORD}, then it must be added to @file{inline.c}'s
-includes.
-
-@item
-Try compiling at least once with
+@end itemize
-@example
-gcc --with-mule --with-union-type --error-checking=all
-@end example
+@cindex Lisp object types, creating
+@cindex creating Lisp object types
+@cindex object types, creating Lisp
+Here is a checklist of things to do when creating a new lisp object type
+named @var{foo}:
+@enumerate
@item
-Did I mention that you should run the test suite?
-@example
-make check
-@end example
-@end itemize
+create @var{foo}.h
+@item
+create @var{foo}.c
+@item
+add definitions of @code{syms_of_@var{foo}}, etc. to @file{@var{foo}.c}
+@item
+add declarations of @code{syms_of_@var{foo}}, etc. to @file{symsinit.h}
+@item
+add calls to @code{syms_of_@var{foo}}, etc. to @file{emacs.c}
+@item
+add definitions of macros like @code{CHECK_@var{FOO}} and
+@code{@var{FOO}P} to @file{@var{foo}.h}
+@item
+add the new type index to @code{enum lrecord_type}
+@item
+add a DEFINE_LRECORD_IMPLEMENTATION call to @file{@var{foo}.c}
+@item
+add an INIT_LRECORD_IMPLEMENTATION call to @code{syms_of_@var{foo}.c}
+@end enumerate
-@node A Summary of the Various XEmacs Modules, Allocation of Objects in XEmacs Lisp, Rules When Writing New C Code, Top
+@node Regression Testing XEmacs, A Summary of the Various XEmacs Modules, Rules When Writing New C Code, Top
+@chapter Regression Testing XEmacs
+@cindex testing, regression
+
+The source directory @file{tests/automated} contains XEmacs' automated
+test suite. The usual way of running all the tests is running
+@code{make check} from the top-level source directory.
+
+The test suite is unfinished and it's still lacking some essential
+features. It is nevertheless recommended that you run the tests to
+confirm that XEmacs behaves correctly.
+
+If you want to run a specific test case, you can do it from the
+command-line like this:
+
+@example
+$ xemacs -batch -l test-harness.elc -f batch-test-emacs TEST-FILE
+@end example
+
+If something goes wrong, you can run the test suite interactively by
+loading @file{test-harness.el} into a running XEmacs and typing
+@kbd{M-x test-emacs-test-file RET <filename> RET}. You will see a log of
+passed and failed tests, which should allow you to investigate the
+source of the error and ultimately fix the bug.
+
+Adding a new test file is trivial: just create a new file here and it
+will be run. There is no need to byte-compile any of the files in
+this directory---the test-harness will take care of any necessary
+byte-compilation.
+
+Look at the existing test cases for the examples of coding test cases.
+It all boils down to your imagination and judicious use of the macros
+@code{Assert}, @code{Check-Error}, @code{Check-Error-Message}, and
+@code{Check-Message}.
+
+Here's a simple example checking case-sensitive and case-insensitive
+comparisons from @file{case-tests.el}.
+
+@example
+(with-temp-buffer
+ (insert "Test Buffer")
+ (let ((case-fold-search t))
+ (goto-char (point-min))
+ (Assert (eq (search-forward "test buffer" nil t) 12))
+ (goto-char (point-min))
+ (Assert (eq (search-forward "Test buffer" nil t) 12))
+ (goto-char (point-min))
+ (Assert (eq (search-forward "Test Buffer" nil t) 12))
+
+ (setq case-fold-search nil)
+ (goto-char (point-min))
+ (Assert (not (search-forward "test buffer" nil t)))
+ (goto-char (point-min))
+ (Assert (not (search-forward "Test buffer" nil t)))
+ (goto-char (point-min))
+ (Assert (eq (search-forward "Test Buffer" nil t) 12))))
+@end example
+
+This example could be inserted in a file in @file{tests/automated}, and
+it would be a complete test, automatically executed when you run
+@kbd{make check} after building XEmacs. More complex tests may require
+substantial temporary scaffolding to create the environment that elicits
+the bugs, but the top-level Makefile and @file{test-harness.el} handle
+the running and collection of results from the @code{Assert},
+@code{Check-Error}, @code{Check-Error-Message}, and @code{Check-Message}
+macros.
+
+In general, you should avoid using functionality from packages in your
+tests, because you can't be sure that everyone will have the required
+package. However, if you've got a test that works, by all means add it.
+Simply wrap the test in an appropriate test, add a notice that the test
+was skipped, and update the @code{skipped-test-reasons} hashtable.
+Here's an example from @file{syntax-tests.el}:
+
+@example
+;; Test forward-comment at buffer boundaries
+(with-temp-buffer
+
+ ;; try to use exactly what you need: featurep, boundp, fboundp
+ (if (not (fboundp 'c-mode))
+
+ ;; We should provide a standard function for this boilerplate,
+ ;; probably called `Skip-Test' -- check for that API with C-h f
+ (let* ((reason "c-mode unavailable")
+ (count (gethash reason skipped-test-reasons)))
+ (puthash reason (if (null count) 1 (1+ count))
+ skipped-test-reasons)
+ (Print-Skip "comment and parse-partial-sexp tests" reason))
+
+ ;; and here's the test code
+ (c-mode)
+ (insert "// comment\n")
+ (forward-comment -2)
+ (Assert (eq (point) (point-min)))
+ (let ((point (point)))
+ (insert "/* comment */")
+ (goto-char point)
+ (forward-comment 2)
+ (Assert (eq (point) (point-max)))
+ (parse-partial-sexp point (point-max)))))
+@end example
+
+@code{Skip-Test} is intended for use with features that are normally
+present in typical configurations. For truly optional features, or
+tests that apply to one of several alternative implementations (eg, to
+GTK widgets, but not Athena, Motif, MS Windows, or Carbon), simply
+silently omit the test.
+
+
+@node A Summary of the Various XEmacs Modules, Allocation of Objects in XEmacs Lisp, Regression Testing XEmacs, Top
@chapter A Summary of the Various XEmacs Modules
+@cindex modules, a summary of the various XEmacs
This is accurate as of XEmacs 20.0.
* Modules for Interfacing with the Operating System::
* Modules for Interfacing with X Windows::
* Modules for Internationalization::
+* Modules for Regression Testing::
@end menu
@node Low-Level Modules
@section Low-Level Modules
+@cindex low-level modules
+@cindex modules, low-level
@example
config.h
@example
-crt0.c
+ecrt0.c
lastfile.c
pre-crt0.c
@end example
@example
-prefix-args.c
-@end example
-
-This is actually the source for a small, self-contained program
-used during building.
-
-
-@example
universe.h
@end example
@node Basic Lisp Modules
@section Basic Lisp Modules
+@cindex Lisp modules, basic
+@cindex modules, basic Lisp
@example
-emacsfns.h
lisp-disunion.h
lisp-union.h
lisp.h
typedefs section as necessary.
@file{lrecord.h} contains the basic structures and macros that implement
-all record-type Lisp objects -- i.e. all objects whose type is a field
+all record-type Lisp objects---i.e. all objects whose type is a field
in their C structure, which includes all objects except the few most
basic ones.
@example
alloc.c
-pure.c
-puresize.h
@end example
The large module @file{alloc.c} implements all of the basic allocation and
type-specific methods. This scheme is a fundamental principle of
object-oriented programming and is heavily used throughout XEmacs. The
great advantage of this is that it allows for a clean separation of
-functionality into different modules -- new classes of Lisp objects, new
+functionality into different modules---new classes of Lisp objects, new
event interfaces, new device types, new stream interfaces, etc. can be
added transparently without affecting code anywhere else in XEmacs.
Because the different subsystems are divided into general and specific
subtypes in the subsystem; this provides a great deal of robustness to
the XEmacs code.
-@cindex pure space
-@file{pure.c} contains the declaration of the @dfn{purespace} array.
-Pure space is a hack used to place some constant Lisp data into the code
-segment of the XEmacs executable, even though the data needs to be
-initialized through function calls. (See above in section VIII for more
-info about this.) During startup, certain sorts of data is
-automatically copied into pure space, and other data is copied manually
-in some of the basic Lisp files by calling the function @code{purecopy},
-which copies the object if possible (this only works in temacs, of
-course) and returns the new object. In particular, while temacs is
-executing, the Lisp reader automatically copies all compiled-function
-objects that it reads into pure space. Since compiled-function objects
-are large, are never modified, and typically comprise the majority of
-the contents of a compiled-Lisp file, this works well. While XEmacs is
-running, any attempt to modify an object that resides in pure space
-causes an error. Objects in pure space are never garbage collected --
-almost all of the time, they're intended to be permanent, and in any
-case you can't write into pure space to set the mark bits.
-
-@file{puresize.h} contains the declaration of the size of the pure space
-array. This depends on the optional features that are compiled in, any
-extra purespace requested by the user at compile time, and certain other
-factors (e.g. 64-bit machines need more pure space because their Lisp
-objects are larger). The smallest size that suffices should be used, so
-that there's no wasted space. If there's not enough pure space, you
-will get an error during the build process, specifying how much more
-pure space is needed.
-
-
@example
eval.c
@file{symbols.c} implements the handling of symbols, obarrays, and
retrieving the values of symbols. Much of the code is devoted to
handling the special @dfn{symbol-value-magic} objects that define
-special types of variables -- this includes buffer-local variables,
+special types of variables---this includes buffer-local variables,
variable aliases, variables that forward into C variables, etc. This
module is initialized extremely early (right after @file{alloc.c}),
because it is here that the basic symbols @code{t} and @code{nil} are
@node Modules for Standard Editing Operations
@section Modules for Standard Editing Operations
+@cindex modules for standard editing operations
+@cindex editing operations, modules for standard
@example
buffer.c
@node Editor-Level Control Flow Modules
@section Editor-Level Control Flow Modules
+@cindex control flow modules, editor-level
+@cindex modules, editor-level control flow
@example
event-Xt.c
+event-msw.c
event-stream.c
event-tty.c
+events-mod.h
+gpmevent.c
+gpmevent.h
events.c
events.h
@end example
@example
-keyboard.c
+cmdloop.c
@end example
-@file{keyboard.c} contains functions that implement the actual editor
-command loop -- i.e. the event loop that cyclically retrieves and
+@file{cmdloop.c} contains functions that implement the actual editor
+command loop---i.e. the event loop that cyclically retrieves and
dispatches events. This code is also rather tricky, just like
@file{event-stream.c}.
@node Modules for the Basic Displayable Lisp Objects
@section Modules for the Basic Displayable Lisp Objects
+@cindex modules for the basic displayable Lisp objects
+@cindex displayable Lisp objects, modules for the basic
+@cindex Lisp objects, modules for the basic displayable
+@cindex objects, modules for the basic displayable Lisp
+
+@example
+console-msw.c
+console-msw.h
+console-stream.c
+console-stream.h
+console-tty.c
+console-tty.h
+console-x.c
+console-x.h
+console.c
+console.h
+@end example
+
+These modules implement the @dfn{console} Lisp object type. A console
+contains multiple display devices, but only one keyboard and mouse.
+Most of the time, a console will contain exactly one device.
+
+Consoles are the top of a lisp object inclusion hierarchy. Consoles
+contain devices, which contain frames, which contain windows.
+
+
@example
-device-ns.h
-device-stream.c
-device-stream.h
+device-msw.c
device-tty.c
-device-tty.h
device-x.c
-device-x.h
device.c
device.h
@end example
@example
-frame-ns.h
+frame-msw.c
frame-tty.c
frame-x.c
-frame-x.h
frame.c
frame.h
@end example
@node Modules for other Display-Related Lisp Objects
@section Modules for other Display-Related Lisp Objects
+@cindex modules for other display-related Lisp objects
+@cindex display-related Lisp objects, modules for other
+@cindex Lisp objects, modules for other display-related
@example
faces.c
@example
bitmaps.h
-glyphs-ns.h
+glyphs-eimage.c
+glyphs-msw.c
+glyphs-msw.h
+glyphs-widget.c
glyphs-x.c
glyphs-x.h
glyphs.c
@example
-objects-ns.h
+objects-msw.c
+objects-msw.h
objects-tty.c
objects-tty.h
objects-x.c
@example
+menubar-msw.c
+menubar-msw.h
menubar-x.c
menubar.c
+menubar.h
@end example
@example
+scrollbar-msw.c
+scrollbar-msw.h
scrollbar-x.c
scrollbar-x.h
scrollbar.c
@example
+toolbar-msw.c
toolbar-x.c
toolbar.c
toolbar.h
font-lock.c
@end example
-This file provides C support for syntax highlighting -- i.e.
+This file provides C support for syntax highlighting---i.e.
highlighting different syntactic constructs of a source file in
different colors, for easy reading. The C support is provided so that
this is fast.
+As of 21.4.10, bugs introduced at the very end of the 21.2 series in the
+``syntax properties'' code were fixed, and highlighting is acceptably
+quick again. However, presumably more improvements are possible, and
+the places to look are probably here, in the defun-traversing code, and
+in @file{syntax.c}, in the comment-traversing code.
@example
@end example
These modules decode GIF-format image files, for use with glyphs.
+These files were removed due to Unisys patent infringement concerns.
@node Modules for the Redisplay Mechanism
@section Modules for the Redisplay Mechanism
+@cindex modules for the redisplay mechanism
+@cindex redisplay mechanism, modules for the
@example
redisplay-output.c
+redisplay-msw.c
redisplay-tty.c
redisplay-x.c
redisplay.c
@node Modules for Interfacing with the File System
@section Modules for Interfacing with the File System
+@cindex modules for interfacing with the file system
+@cindex interfacing with the file system, modules for
+@cindex file system, modules for interfacing with the
@example
lstream.c
Similar to other subsystems in XEmacs, lstreams are separated into
generic functions and a set of methods for the different types of
lstreams. @file{lstream.c} provides implementations of many different
-types of streams; others are provided, e.g., in @file{mule-coding.c}.
+types of streams; others are provided, e.g., in @file{file-coding.c}.
@node Modules for Other Aspects of the Lisp Interpreter and Object System
@section Modules for Other Aspects of the Lisp Interpreter and Object System
+@cindex modules for other aspects of the Lisp interpreter and object system
+@cindex Lisp interpreter and object system, modules for other aspects of the
+@cindex interpreter and object system, modules for other aspects of the Lisp
+@cindex object system, modules for other aspects of the Lisp interpreter and
@example
elhash.c
@code{forward-sexp}, and by @file{font-lock.c} to locate quoted strings,
comments, etc.
+@c #### Break this out into a separate node somewhere!
+Syntax codes are implemented as bitfields in an int. Bits 0-6 contain
+the syntax code itself, bit 7 is a special prefix flag used for Lisp,
+and bits 16-23 contain comment syntax flags. From the Lisp programmer's
+point of view, there are 11 flags: 2 styles X 2 characters X @{start,
+end@} flags for two-character comment delimiters, 2 style flags for
+one-character comment delimiters, and the prefix flag.
+
+Internally, however, the characters used in multi-character delimiters
+will have non-comment-character syntax classes (@emph{e.g.}, the
+@samp{/} in C's @samp{/*} comment-start delimiter has ``punctuation''
+(here meaning ``operator-like'') class in C modes). Thus in a mixed
+comment style, such as C++'s @samp{//} to end of line, is represented by
+giving @samp{/} the ``punctuation'' class and the ``style b first
+character of start sequence'' and ``style b second character of start
+sequence'' flags. The fact that class is @emph{not} punctuation allows
+the syntax scanner to recognize that this is a multi-character
+delimiter. The @samp{newline} character is given (single-character)
+``comment-end'' @emph{class} and the ``style b first character of end
+sequence'' @emph{flag}. The ``comment-end'' class allows the scanner to
+determine that no second character is needed to terminate the comment.
@example
with them, in case the block of memory contains other Lisp objects that
need to be marked for garbage-collection purposes. (If you need other
object methods, such as a finalize method, you should just go ahead and
-create a new Lisp object type -- it's not hard.)
+create a new Lisp object type---it's not hard.)
@node Modules for Interfacing with the Operating System
@section Modules for Interfacing with the Operating System
+@cindex modules for interfacing with the operating system
+@cindex interfacing with the operating system, modules for
+@cindex operating system, modules for interfacing with the
@example
callproc.c
-@example
-msdos.c
-msdos.h
-@end example
-
-These modules are used for MS-DOS support, which does not work in
-XEmacs.
-
-
-
@node Modules for Interfacing with X Windows
@section Modules for Interfacing with X Windows
+@cindex modules for interfacing with X Windows
+@cindex interfacing with X Windows, modules for
+@cindex X Windows, modules for interfacing with
@example
Emacs.ad.h
@example
-xselect.c
+select-msw.c
+select-x.c
+select.c
+select.h
@end example
@cindex selections
@node Modules for Internationalization
@section Modules for Internationalization
+@cindex modules for internationalization
+@cindex internationalization, modules for
@example
mule-canna.c
mule-ccl.c
mule-charset.c
mule-charset.h
-mule-coding.c
-mule-coding.h
+file-coding.c
+file-coding.h
mule-mcpath.c
mule-mcpath.h
mule-wnnfns.c
just Asian languages (although they are generally the most complicated
to support). This code is still in beta.
-@file{mule-charset.*} and @file{mule-coding.*} provide the heart of the
+@file{mule-charset.*} and @file{file-coding.*} provide the heart of the
XEmacs MULE support. @file{mule-charset.*} implements the @dfn{charset}
Lisp object type, which encapsulates a character set (an ordered one- or
two-dimensional set of characters, such as US ASCII or JISX0208 Japanese
Kanji).
-@file{mule-coding.*} implements the @dfn{coding-system} Lisp object
+@file{file-coding.*} implements the @dfn{coding-system} Lisp object
type, which encapsulates a method of converting between different
encodings. An encoding is a representation of a stream of characters,
possibly from multiple character sets, using a stream of bytes or words,
@file{mule-mcpath.c} provides some functions to allow for pathnames
containing extended characters. This code is fragmentary, obsolete, and
-completely non-working. Instead, @var{pathname-coding-system} is used
+completely non-working. Instead, @code{pathname-coding-system} is used
to specify conversions of names of files and directories. The standard
C I/O functions like @samp{open()} are wrapped so that conversion occurs
automatically.
-@node Allocation of Objects in XEmacs Lisp, Events and the Event Loop, A Summary of the Various XEmacs Modules, Top
+@node Modules for Regression Testing
+@section Modules for Regression Testing
+@cindex modules for regression testing
+@cindex regression testing, modules for
+
+@example
+test-harness.el
+base64-tests.el
+byte-compiler-tests.el
+case-tests.el
+ccl-tests.el
+c-tests.el
+database-tests.el
+extent-tests.el
+hash-table-tests.el
+lisp-tests.el
+md5-tests.el
+mule-tests.el
+regexp-tests.el
+symbol-tests.el
+syntax-tests.el
+@end example
+
+@file{test-harness.el} defines the macros @code{Assert},
+@code{Check-Error}, @code{Check-Error-Message}, and
+@code{Check-Message}. The other files are test files, testing various
+XEmacs modules.
+
+
+
+@node Allocation of Objects in XEmacs Lisp, Dumping, A Summary of the Various XEmacs Modules, Top
@chapter Allocation of Objects in XEmacs Lisp
+@cindex allocation of objects in XEmacs Lisp
+@cindex objects in XEmacs Lisp, allocation of
+@cindex Lisp objects, allocation of in XEmacs
@menu
* Introduction to Allocation::
* Allocation from Frob Blocks::
* lrecords::
* Low-level allocation::
-* Pure Space::
* Cons::
* Vector::
* Bit Vector::
@node Introduction to Allocation
@section Introduction to Allocation
+@cindex allocation, introduction to
Emacs Lisp, like all Lisps, has garbage collection. This means that
the programmer never has to explicitly free (destroy) an object; it
have no corresponding Lisp primitives. Every Lisp object, though,
has at least one C primitive for creating it.
- Recall from section (VII) that a Lisp object, as stored in a 32-bit
-or 64-bit word, has a mark bit, a few tag bits, and a ``value'' that
-occupies the remainder of the bits. We can separate the different
-Lisp object types into four broad categories:
+ Recall from section (VII) that a Lisp object, as stored in a 32-bit or
+64-bit word, has a few tag bits, and a ``value'' that occupies the
+remainder of the bits. We can separate the different Lisp object types
+into three broad categories:
@itemize @bullet
@item
@code{GCPRO}ed.
@end itemize
- In the remaining three categories, the value is a pointer to a
-structure.
-
-@itemize @bullet
-@item
-@cindex frob block
-(b) Those for whom the tag directly specifies the type. Recall that
-there are only three tag bits; this means that at most five types can be
-specified this way. The most commonly-used types are stored in this
-format; this includes conses, strings, vectors, and sometimes symbols.
-With the exception of vectors, objects in this category are allocated in
-@dfn{frob blocks}, i.e. large blocks of memory that are subdivided into
-individual objects. This saves a lot on malloc overhead, since there
-are typically quite a lot of these objects around, and the objects are
-small. (A cons, for example, occupies 8 bytes on 32-bit machines -- 4
-bytes for each of the two objects it contains.) Vectors are individually
-@code{malloc()}ed since they are of variable size. (It would be
-possible, and desirable, to allocate vectors of certain small sizes out
-of frob blocks, but it isn't currently done.) Strings are handled
-specially: Each string is allocated in two parts, a fixed size structure
-containing a length and a data pointer, and the actual data of the
-string. The former structure is allocated in frob blocks as usual, and
-the latter data is stored in @dfn{string chars blocks} and is relocated
-during garbage collection to eliminate holes.
-@end itemize
-
In the remaining two categories, the type is stored in the object
itself. The tag for all such objects is the generic @dfn{lrecord}
-(Lisp_Record) tag. The first four bytes (or eight, for 64-bit machines)
-of the object's structure are a pointer to a structure that describes
-the object's type, which includes method pointers and a pointer to a
-string naming the type. Note that it's possible to save some space by
-using a one- or two-byte tag, rather than a four- or eight-byte pointer
-to store the type, but it's not clear it's worth making the change.
+(Lisp_Type_Record) tag. The first bytes of the object's structure are an
+integer (actually a char) characterising the object's type and some
+flags, in particular the mark bit used for garbage collection. A
+structure describing the type is accessible thru the
+lrecord_implementation_table indexed with said integer. This structure
+includes the method pointers and a pointer to a string naming the type.
@itemize @bullet
@item
-(c) Those lrecords that are allocated in frob blocks (see above). This
+(b) Those lrecords that are allocated in frob blocks (see above). This
includes the objects that are most common and relatively small, and
-includes floats, compiled functions, symbols (when not in category (b)),
+includes conses, strings, subrs, floats, compiled functions, symbols,
extents, events, and markers. With the cleanup of frob blocks done in
19.12, it's not terribly hard to add more objects to this category, but
-it's a bit trickier than adding an object type to type (d) (esp. if the
+it's a bit trickier than adding an object type to type (c) (esp. if the
object needs a finalization method), and is not likely to save much
space unless the object is small and there are many of them. (In fact,
if there are very few of them, it might actually waste space.)
@item
-(d) Those lrecords that are individually @code{malloc()}ed. These are
+(c) Those lrecords that are individually @code{malloc()}ed. These are
called @dfn{lcrecords}. All other types are in this category. Adding a
new type to this category is comparatively easy, and all types added
since 19.8 (when the current allocation scheme was devised, by Richard
@end itemize
Note that bit vectors are a bit of a special case. They are
-simple lrecords as in category (c), but are individually @code{malloc()}ed
+simple lrecords as in category (b), but are individually @code{malloc()}ed
like vectors. You can basically view them as exactly like vectors
except that their type is stored in lrecord fashion rather than
in directly-tagged fashion.
- Note that FSF Emacs redesigned their object system in 19.29 to follow
-a similar scheme. However, given RMS's expressed dislike for data
-abstraction, the FSF scheme is not nearly as clean or as easy to
-extend. (FSF calls items of type (c) @code{Lisp_Misc} and items of type
-(d) @code{Lisp_Vectorlike}, with separate tags for each, although
-@code{Lisp_Vectorlike} is also used for vectors.)
@node Garbage Collection
@section Garbage Collection
all vectors (which are chained in one big list), and all
lcrecords (which are likewise chained).
- Note that, when an object is marked, the mark has to occur
-inside of the object's structure, rather than in the 32-bit
-@code{Lisp_Object} holding the object's pointer; i.e. you can't just
-set the pointer's mark bit. This is because there may be many
-pointers to the same object. This means that the method of
-marking an object can differ depending on the type. The
-different marking methods are approximately as follows:
-
-@enumerate
-@item
-For conses, the mark bit of the car is set.
-@item
-For strings, the mark bit of the string's plist is set.
-@item
-For symbols when not lrecords, the mark bit of the
-symbol's plist is set.
-@item
-For vectors, the length is negated after adding 1.
-@item
-For lrecords, the pointer to the structure describing
-the type is changed (see below).
-@item
-Integers and characters do not need to be marked, since
-no allocation occurs for them.
-@end enumerate
-
- The details of this are in the @code{mark_object()} function.
-
- Note that any code that operates during garbage collection has
-to be especially careful because of the fact that some objects
-may be marked and as such may not look like they normally do.
-In particular:
-
-@itemize @bullet
-Some object pointers may have their mark bit set. This will make
-@code{FOOBARP()} predicates fail. Use @code{GC_FOOBARP()} to deal with
-this.
-@item
-Even if you clear the mark bit, @code{FOOBARP()} will still fail
-for lrecords because the implementation pointer has been
-changed (see below). @code{GC_FOOBARP()} will correctly deal with
-this.
-@item
-Vectors have their size field munged, so anything that
-looks at this field will fail.
-@item
-Note that @code{XFOOBAR()} macros @emph{will} work correctly on object
-pointers with their mark bit set, because the logical shift operations
-that remove the tag also remove the mark bit.
-@end itemize
+ Garbage collection can be invoked explicitly by calling
+@code{garbage-collect} but is also called automatically by @code{eval},
+once a certain amount of memory has been allocated since the last
+garbage collection (according to @code{gc-cons-threshold}).
- Finally, note that garbage collection can be invoked explicitly
-by calling @code{garbage-collect} but is also called automatically
-by @code{eval}, once a certain amount of memory has been allocated
-since the last garbage collection (according to @code{gc-cons-threshold}).
@node GCPROing
@section @code{GCPRO}ing
+@cindex @code{GCPRO}ing
+@cindex garbage collection protection
+@cindex protection, garbage collection
@code{GCPRO}ing is one of the ugliest and trickiest parts of Emacs
internals. The basic idea is that whenever garbage collection
@enumerate
@item
-All objects that have been @code{staticpro()}d. This is used for
-any global C variables that hold Lisp objects. A call to
-@code{staticpro()} happens implicitly as a result of any symbols
-declared with @code{defsymbol()} and any variables declared with
-@code{DEFVAR_FOO()}. You need to explicitly call @code{staticpro()}
-(in the @code{vars_of_foo()} method of a module) for other global
-C variables holding Lisp objects. (This typically includes
-internal lists and such things.)
+All objects that have been @code{staticpro()}d or
+@code{staticpro_nodump()}ed. This is used for any global C variables
+that hold Lisp objects. A call to @code{staticpro()} happens implicitly
+as a result of any symbols declared with @code{defsymbol()} and any
+variables declared with @code{DEFVAR_FOO()}. You need to explicitly
+call @code{staticpro()} (in the @code{vars_of_foo()} method of a module)
+for other global C variables holding Lisp objects. (This typically
+includes internal lists and such things.). Use
+@code{staticpro_nodump()} only in the rare cases when you do not want
+the pointed variable to be saved at dump time but rather recompute it at
+startup.
Note that @code{obarray} is one of the @code{staticpro()}d things.
Therefore, all functions and variables get marked through this.
in the next enclosing stack frame. Each @code{GCPRO}ed thing is an
lvalue, and the @code{struct gcpro} local variable contains a pointer to
this lvalue. This is why things will mess up badly if you don't pair up
-the @code{GCPRO}s and @code{UNGCPRO}s -- you will end up with
+the @code{GCPRO}s and @code{UNGCPRO}s---you will end up with
@code{gcprolist}s containing pointers to @code{struct gcpro}s or local
@code{Lisp_Object} variables in no-longer-active stack frames.
different section of code.
@end enumerate
+If you don't understand whether to @code{GCPRO} in a particular
+instance, ask on the mailing lists. A general hint is that @code{prog1}
+is the canonical example.
+
@cindex garbage collection, conservative
@cindex conservative garbage collection
Given the extremely error-prone nature of the @code{GCPRO} scheme, and
@node Garbage Collection - Step by Step
@section Garbage Collection - Step by Step
-@cindex garbage collection step by step
+@cindex garbage collection - step by step
@menu
* Invocation::
@cindex garbage collection, invocation
The first thing that anyone should know about garbage collection is:
-when and how the garbage collector is invoked. One might think that this
+when and how the garbage collector is invoked. One might think that this
could happen every time new memory is allocated, e.g. new objects are
created, but this is @emph{not} the case. Instead, we have the following
situation:
of the function @code{garbage_collect_1} in file @code{alloc.c}. The
invocation can occur @emph{explicitly} by calling the function
@code{Fgarbage_collect} (in addition this function provides information
-about the freed memory), or can occur @emph{implicitly} in four different
+about the freed memory), or can occur @emph{implicitly} in four different
situations:
@enumerate
@item
@item
In function @code{disksave_object_finalization} in file
@code{alloc.c}. The only purpose of this function is to clear the
-objects from memory which need not be stored with xemacs when we dump out
+objects from memory which need not be stored with xemacs when we dump out
an executable. This is only done by @code{Fdump_emacs} or by
@code{Fdump_emacs_data} respectively (both in @code{emacs.c}). The
actual clearing is accomplished by making these objects unreachable and
The upshot is that garbage collection can basically occur everywhere
@code{Feval}, respectively @code{Ffuncall}, is used - either directly or
-through another function. Since calls to these two functions are
-hidden in various other functions, many calls to
-@code{garabge_collect_1} are not obviously foreseeable, and therefore
-unexpected. Instances where they are used that are worth remembering are
-various elisp commands, as for example @code{or},
-@code{and}, @code{if}, @code{cond}, @code{while}, @code{setq}, etc.,
-miscellaneous @code{gui_item_...} functions, everything related to
-@code{eval} (@code{Feval_buffer}, @code{call0}, ...) and inside
-@code{Fsignal}. The latter is used to handle signals, as for example the
-ones raised by every @code{QUITE}-macro triggered after pressing Ctrl-g.
+through another function. Since calls to these two functions are hidden
+in various other functions, many calls to @code{garbage_collect_1} are
+not obviously foreseeable, and therefore unexpected. Instances where
+they are used that are worth remembering are various elisp commands, as
+for example @code{or}, @code{and}, @code{if}, @code{cond}, @code{while},
+@code{setq}, etc., miscellaneous @code{gui_item_...} functions,
+everything related to @code{eval} (@code{Feval_buffer}, @code{call0},
+...) and inside @code{Fsignal}. The latter is used to handle signals, as
+for example the ones raised by every @code{QUIT}-macro triggered after
+pressing Ctrl-g.
@node garbage_collect_1
@subsection @code{garbage_collect_1}
place.
@enumerate
@item
-There are several cases in which the garbage collector is left immediately:
+There are several cases in which the garbage collector is left immediately:
when we are already garbage collecting (@code{gc_in_progress}), when
the garbage collection is somehow forbidden
(@code{gc_currently_forbidden}), when we are currently displaying something
all the output occurring during garbage collecting is determined. In
order to be able to restore the old display's state after displaying the
message, some data about the current cursor position has to be
-saved. The variables @code{pre_gc_curser} and @code{cursor_changed} take
+saved. The variables @code{pre_gc_cursor} and @code{cursor_changed} take
care of that.
@item
The state of @code{gc_currently_forbidden} must be restored after
the garbage collection, no matter what happens during the process. We
accomplish this by @code{record_unwind_protect}ing the suitable function
@code{restore_gc_inhibit} together with the current value of
-@code{gc_currently_forbidden}.
+@code{gc_currently_forbidden}.
@item
If we are concurrently running an interactive xemacs session, the next step
is simply to show the garbage collector's cursor/message.
Before actually starting to go over all live objects, references to
objects that are no longer used are pruned. We only have to do this for events
(@code{clear_event_resource}) and for specifiers
-(@code{cleanup_specifiers}).
+(@code{cleanup_specifiers}).
@item
Now the mark phase begins and marks all accessible elements. In order to
start from
@itemize @bullet
@item
all constant symbols and static variables that are registered via
-@code{staticpro}@ in the array @code{staticvec}.
-@xref{Adding Global Lisp Variables}.
+@code{staticpro}@ in the dynarr @code{staticpros}.
+@xref{Adding Global Lisp Variables}.
@item
all Lisp objects that are created in C functions and that must be
protected from freeing them. They are registered in the global
list @code{gcprolist}.
@xref{GCPROing}.
-@item
+@item
all local variables (i.e. their name fields @code{symbol} and old
values @code{old_values}) that are bound during the evaluation by the Lisp
engine. They are stored in @code{specbinding} structs pushed on a stack
are freshly created objects and therefore have to be marked.
@xref{Catch and Throw}.
@item
-every function application pushes new structs @code{backtrace}
-on the call stack of the Lisp engine (@code{backtrace_list}). The unique
+every function application pushes new structs @code{backtrace}
+on the call stack of the Lisp engine (@code{backtrace_list}). The unique
parts that have to be marked are the fields for each function
(@code{function}) and all their arguments (@code{args}).
@xref{Evaluation}.
@item
-all objects that are used by the redisplay engine that must not be freed
+all objects that are used by the redisplay engine that must not be freed
are marked by a special function called @code{mark_redisplay} (in
@code{redisplay.c}).
@item
manually. That is done by the function @code{mark_profiling_info}
@end itemize
@item
-Hash tables in Xemacs belong to a kind of special objects that
+Hash tables in XEmacs belong to a kind of special objects that
make use of a concept often called 'weak pointers'.
To make a long story short, these kind of pointers are not followed
during the estimation of the live objects during garbage collection.
anyway, and the reference to it is cleared. In hash tables there are
different usage patterns of them, manifesting in different types of hash
tables, namely 'non-weak', 'weak', 'key-weak' and 'value-weak'
-(internally also 'key-car-weak' and 'value-car-weak') hash tables, each
-clearing entries depending on different conditions. More information can
+(internally also 'key-car-weak' and 'value-car-weak') hash tables, each
+clearing entries depending on different conditions. More information can
be found in the documentation to the function @code{make-hash-table}.
Because there are complicated dependency rules about when and what to
function @code{finish_marking_weak_hash_tables}. This function iterates
over each hash table entry @code{hentries} for each weak hash table in
@code{Vall_weak_hash_tables}. Depending on the type of a table, the
-appropriate action is performed.
+appropriate action is performed.
If a table is acting as @code{HASH_TABLE_KEY_WEAK}, and a key already marked,
-everything reachable from the @code{value} component is marked. If it is
+everything reachable from the @code{value} component is marked. If it is
acting as a @code{HASH_TABLE_VALUE_WEAK} and the value component is
-already marked, the marking starts beginning only from the
+already marked, the marking starts beginning only from the
@code{key} component.
-If it is a @code{HASH_TABLE_KEY_CAR_WEAK} and the car
+If it is a @code{HASH_TABLE_KEY_CAR_WEAK} and the car
of the key entry is already marked, we mark both the @code{key} and
@code{value} components.
Finally, if the table is of the type @code{HASH_TABLE_VALUE_CAR_WEAK}
@code{simple}, @code{assoc}, @code{key-assoc} and
@code{value-assoc}. You can find further details about them in the
description to the function @code{make-weak-list}. The scheme of their
-marking is similar: all weak lists are listed in @code{Qall_weak_lists},
+marking is similar: all weak lists are listed in @code{Qall_weak_lists},
therefore we iterate over them. The marking is advanced until we hit an
-already marked pair. Then we know that during a former run all
+already marked pair. Then we know that during a former run all
the rest has been marked completely. Again, depending on the special
type of the weak list, our jobs differ. If it is a @code{WEAK_LIST_SIMPLE}
and the elem is marked, we mark the @code{cons} part. If it is a
Since, by marking objects in reach from weak hash tables and weak lists,
other objects could get marked, this perhaps implies further marking of
-other weak objects, both finishing functions are redone as long as
+other weak objects, both finishing functions are redone as long as
yet unmarked objects get freshly marked.
@item
The function @code{prune_weak_hash_tables} does the job for weak hash
tables. Totally unmarked hash tables are removed from the list
@code{Vall_weak_hash_tables}. The other ones are treated more carefully
-by scanning over all entries and removing one as soon as one of
+by scanning over all entries and removing one as soon as one of
the components @code{key} and @code{value} is unmarked.
The same idea applies to the weak lists. It is accomplished by
@item
The function @code{prune_specifiers} checks all listed specifiers held
-in @code{Vall_speficiers} and removes the ones from the lists that are
+in @code{Vall_specifiers} and removes the ones from the lists that are
unmarked.
@item
All syntax tables are stored in a list called
-@code{Vall_syntax_tables}. The function @code{prune_syntax_tables} walks
+@code{Vall_syntax_tables}. The function @code{prune_syntax_tables} walks
through it and unlinks the tables that are unmarked.
@item
@code{gc_sweep} which holds the predominance.
@item
First, all the variables with respect to garbage collection are
-reset. @code{consing_since_gc} - the counter of the created cells since
+reset. @code{consing_since_gc} - the counter of the created cells since
the last garbage collection - is set back to 0, and
@code{gc_in_progress} is not @code{true} anymore.
@item
-In case the session is interactive, the displayed cursor and message are
+In case the session is interactive, the displayed cursor and message are
removed again.
@item
The state of @code{gc_inhibit} is restored to the former value by
@item
A small memory reserve is always held back that can be reached by
@code{breathing_space}. If nothing more is left, we create a new reserve
-and exit.
+and exit.
@end enumerate
@node mark_object
object is a real Lisp object @code{Lisp_Type_Record} or just an integer
or a character. Integers and characters are the only two types that are
stored directly - without another level of indirection, and therefore they
-don´t have to be marked and collected.
+don't have to be marked and collected.
@xref{How Lisp Objects Are Represented in C}.
The second case is the one we have to handle. It is the one when we are
already marked, and need not be marked for the second time (checked by
@code{MARKED_RECORD_HEADER_P}). If it is a special, unmarkable object
(@code{UNMARKABLE_RECORD_HEADER_P}, apparently, these are objects that
-sit in some CONST space, and can therefore not be marked, see
+sit in some const space, and can therefore not be marked, see
@code{this_one_is_unmarkable} in @code{alloc.c}).
Now, the actual marking is feasible. We do so by once using the macro
@code{MARK_RECORD_HEADER} to mark the object itself (actually the
special flag in the lrecord header), and calling its special marker
"method" @code{marker} if available. The marker method marks every
-other object that is in reach from our current object. Note, that these
+other object that is in reach from our current object. Note, that these
marker methods should not call @code{mark_object} recursively, but
instead should return the next object from where further marking has to
be performed.
@subsection @code{gc_sweep}
@cindex @code{gc_sweep}
-The job of this function is to free all unmarked records from memory. As
+The job of this function is to free all unmarked records from memory. As
we know, there are different types of objects implemented and managed, and
consequently different ways to free them from memory.
@xref{Introduction to Allocation}.
bulkier objects are allocated and handled using that scheme of
@code{lcrecords}. Each object is @code{malloc}ed separately
instead of placing it in one of the contiguous frob blocks. All types
-that are currently stored
-using @code{lcrecords}´s @code{alloc_lcrecord} and
+that are currently stored
+using @code{lcrecords}'s @code{alloc_lcrecord} and
@code{make_lcrecord_list} are the types: vectors, buffers,
char-table, char-table-entry, console, weak-list, database, device,
ldap, hash-table, command-builder, extent-auxiliary, extent-info, face,
Our next candidates are the other objects that behave quite differently
than everything else: the strings. They consists of two parts, a
-fixed-size portion (@code{struct Lisp_string}) holding the string's
+fixed-size portion (@code{struct Lisp_String}) holding the string's
length, its property list and a pointer to the second part, and the
actual string data, which is stored in string-chars blocks comparable to
frob blocks. In this block, the data is not only freed, but also a
compiled-functions, symbol, marker, extent, and event stored in
so-called "frob blocks", and therefore we can basically do the same on
every type objects, using the same macros, especially defined only to
-handle everything with respect to fixed-size blocks. The only fixed-size
+handle everything with respect to fixed-size blocks. The only fixed-size
type that is not handled here are the fixed-size portion of strings,
because we took special care of them earlier.
The only big exceptions are bit vectors stored differently and
-therefore treated differently by the function @code{sweep_bit_vectors_1}
+therefore treated differently by the function @code{sweep_bit_vectors_1}
described later.
At first, we need some brief information about how
stored in big blocks of memory - allocated at once - that can hold a
certain amount of objects of one type. The macro
@code{DECLARE_FIXED_TYPE_ALLOC} creates the suitable structures for
-every type. More precisely, we have the block struct
+every type. More precisely, we have the block struct
(holding a pointer to the previous block @code{prev} and the
objects in @code{block[]}), a pointer to current block
(@code{current_..._block)}) and its last index
macro @code{ADDITIONAL_FREE_...} that defines additional work that has
to be done when converting an object from in use to not in use (so far,
only markers use it in order to unchain them). Then, they all call
-the macro @code{SWEEP_FIXED_TYPE_BLOCK} instantiated with their type name
+the macro @code{SWEEP_FIXED_TYPE_BLOCK} instantiated with their type name
and their struct name.
This call in particular does the following: we go over all blocks
starting with the current moving towards the oldest.
For each block, we look at every object in it. If the object already
freed (checked with @code{FREE_STRUCT_P} using the first pointer of the
-object), or if it is
+object), or if it is
set to read only (@code{C_READONLY_RECORD_HEADER_P}, nothing must be
done. If it is unmarked (checked with @code{MARKED_RECORD_HEADER_P}), it
is put in the free list and set free (using the macro
-@code{FREE_FIXED_TYPE}, otherwise it stays in the block, but is unmarked
+@code{FREE_FIXED_TYPE}, otherwise it stays in the block, but is unmarked
(by @code{UNMARK_...}). While going through one block, we note if the
whole block is empty. If so, the whole block is freed (using
@code{xfree}) and the free list state is set to the state it had before
@cindex @code{sweep_lcrecords_1}
After nullifying the complete lcrecord statistics, we go over all
-lcrecords two separate times. They are all chained together in a list with
-a head called @code{all_lcrecords}.
+lcrecords two separate times. They are all chained together in a list with
+a head called @code{all_lcrecords}.
-The first loop calls for each object its @code{finalizer} method, but only
+The first loop calls for each object its @code{finalizer} method, but only
in the case that it is not read only
(@code{C_READONLY_RECORD_HEADER_P)}, it is not already marked
(@code{MARKED_RECORD_HEADER_P}), it is not already in a free list (list of
freed objects, field @code{free}) and finally it owns a finalizer
method.
-
-The second loop actually frees the appropriate objects again by iterating
-through the whole list. In case an object is read only or marked, it
+
+The second loop actually frees the appropriate objects again by iterating
+through the whole list. In case an object is read only or marked, it
has to persist, otherwise it is manually freed by calling
@code{xfree}. During this loop, the lcrecord statistics are kept up to
-date by calling @code{tick_lcrecord_stats} with the right arguments,
+date by calling @code{tick_lcrecord_stats} with the right arguments,
@node compact_string_chars
@subsection @code{compact_string_chars}
positions in the @code{string_chars_block}s using two pointer/integer
pairs, namely @code{from_sb}/@code{from_pos} and
@code{to_sb}/@code{to_pos}. They stand for the actual positions, from
-where to where, to copy the actually handled string.
+where to where, to copy the actually handled string.
While going over all chained @code{string_char_block}s and their held
strings, staring at @code{first_string_chars_block}, both pointers
@itemize @bullet
@item
The string at @code{from_sb}'s position could be marked as free, which
-is indicated by an invalid pointer to the pointer that should point back
+is indicated by an invalid pointer to the pointer that should point back
to the fixed size string object, and which is checked by
@code{FREE_STRUCT_P}. In this case, the @code{from_sb}/@code{from_pos}
is advanced to the next string, and nothing has to be copied.
copied. We likewise advance the @code{from_sb}/@code{from_pos}
pair as described above.
@item
-In all other cases, we have a marked string at hand. The string data
+In all other cases, we have a marked string at hand. The string data
must be moved from the from-position to the to-position. In case
there is not enough space in the actual @code{to_sb}-block, we advance
this pointer to the beginning of the next block before copying. In case the
definitions are a little bit special compared to the ones used
for the other fixed size types.
-@code{UNMARK_string} is defined the same way except some additional code
+@code{UNMARK_string} is defined the same way except some additional code
used for updating the bookkeeping information.
For strings, @code{ADDITIONAL_FREE_string} has to do something in
bit vectors must be freed by hand. This is done, as one might imagine,
the expected way: since they are all registered in a list called
@code{all_bit_vectors}, all elements of that list are traversed,
-all unmarked bit vectors are unlinked by calling @code{xfree} and all of
+all unmarked bit vectors are unlinked by calling @code{xfree} and all of
them become unmarked.
-In addition, the bookkeeping information used for garbage
+In addition, the bookkeeping information used for garbage
collector's output purposes is updated.
@node Integers and Characters
@section Integers and Characters
+@cindex integers and characters
+@cindex characters, integers and
Integer and character Lisp objects are created from integers using the
macros @code{XSETINT()} and @code{XSETCHAR()} or the equivalent
@node Allocation from Frob Blocks
@section Allocation from Frob Blocks
+@cindex allocation from frob blocks
+@cindex frob blocks, allocation from
The uninitialized memory required by a @code{Lisp_Object} of a particular type
is allocated using
@node lrecords
@section lrecords
+@cindex lrecords
[see @file{lrecord.h}]
All lrecords have at the beginning of their structure a @code{struct
-lrecord_header}. This just contains a pointer to a @code{struct
+lrecord_header}. This just contains a type number and some flags,
+including the mark bit. All builtin type numbers are defined as
+constants in @code{enum lrecord_type}, to allow the compiler to generate
+more efficient code for @code{@var{type}P}. The type number, thru the
+@code{lrecord_implementation_table}, gives access to a @code{struct
lrecord_implementation}, which is a structure containing method pointers
and such. There is one of these for each type, and it is a global,
constant, statically-declared structure that is declared in the
-@code{DEFINE_LRECORD_IMPLEMENTATION()} macro. (This macro actually
-declares an array of two @code{struct lrecord_implementation}
-structures. The first one contains all the standard method pointers,
-and is used in all normal circumstances. During garbage collection,
-however, the lrecord is @dfn{marked} by bumping its implementation
-pointer by one, so that it points to the second structure in the array.
-This structure contains a special indication in it that it's a
-@dfn{marked-object} structure: the finalize method is the special
-function @code{this_marks_a_marked_record()}, and all other methods are
-null pointers. At the end of garbage collection, all lrecords will
-either be reclaimed or unmarked by decrementing their implementation
-pointers, so this second structure pointer will never remain past
-garbage collection.
-
- Simple lrecords (of type (c) above) just have a @code{struct
+@code{DEFINE_LRECORD_IMPLEMENTATION()} macro.
+
+ Simple lrecords (of type (b) above) just have a @code{struct
lrecord_header} at their beginning. lcrecords, however, actually have a
@code{struct lcrecord_header}. This, in turn, has a @code{struct
lrecord_header} at its beginning, so sanity is preserved; but it also
Whenever you create an lrecord, you need to call either
@code{DEFINE_LRECORD_IMPLEMENTATION()} or
@code{DEFINE_LRECORD_SEQUENCE_IMPLEMENTATION()}. This needs to be
-specified in a C file, at the top level. What this actually does is
-define and initialize the implementation structure for the lrecord. (And
-possibly declares a function @code{error_check_foo()} that implements
-the @code{XFOO()} macro when error-checking is enabled.) The arguments
-to the macros are the actual type name (this is used to construct the C
-variable name of the lrecord implementation structure and related
-structures using the @samp{##} macro concatenation operator), a string
-that names the type on the Lisp level (this may not be the same as the C
-type name; typically, the C type name has underscores, while the Lisp
-string has dashes), various method pointers, and the name of the C
-structure that contains the object. The methods are used to encapsulate
-type-specific information about the object, such as how to print it or
-mark it for garbage collection, so that it's easy to add new object
-types without having to add a specific case for each new type in a bunch
-of different places.
+specified in a @file{.c} file, at the top level. What this actually
+does is define and initialize the implementation structure for the
+lrecord. (And possibly declares a function @code{error_check_foo()} that
+implements the @code{XFOO()} macro when error-checking is enabled.) The
+arguments to the macros are the actual type name (this is used to
+construct the C variable name of the lrecord implementation structure
+and related structures using the @samp{##} macro concatenation
+operator), a string that names the type on the Lisp level (this may not
+be the same as the C type name; typically, the C type name has
+underscores, while the Lisp string has dashes), various method pointers,
+and the name of the C structure that contains the object. The methods
+are used to encapsulate type-specific information about the object, such
+as how to print it or mark it for garbage collection, so that it's easy
+to add new object types without having to add a specific case for each
+new type in a bunch of different places.
The difference between @code{DEFINE_LRECORD_IMPLEMENTATION()} and
@code{DEFINE_LRECORD_SEQUENCE_IMPLEMENTATION()} is that the former is
For the purpose of keeping allocation statistics, the allocation
engine keeps a list of all the different types that exist. Note that,
since @code{DEFINE_LRECORD_IMPLEMENTATION()} is a macro that is
-specified at top-level, there is no way for it to add to the list of all
-existing types. What happens instead is that each implementation
-structure contains in it a dynamically assigned number that is
-particular to that type. (Or rather, it contains a pointer to another
-structure that contains this number. This evasiveness is done so that
-the implementation structure can be declared const.) In the sweep stage
-of garbage collection, each lrecord is examined to see if its
-implementation structure has its dynamically-assigned number set. If
-not, it must be a new type, and it is added to the list of known types
-and a new number assigned. The number is used to index into an array
-holding the number of objects of each type and the total memory
-allocated for objects of that type. The statistics in this array are
-also computed during the sweep stage. These statistics are returned by
-the call to @code{garbage-collect} and are printed out at the end of the
-loadup phase.
+specified at top-level, there is no way for it to initialize the global
+data structures containing type information, like
+@code{lrecord_implementations_table}. For this reason a call to
+@code{INIT_LRECORD_IMPLEMENTATION} must be added to the same source file
+containing @code{DEFINE_LRECORD_IMPLEMENTATION}, but instead of to the
+top level, to one of the init functions, typically
+@code{syms_of_@var{foo}.c}. @code{INIT_LRECORD_IMPLEMENTATION} must be
+called before an object of this type is used.
+
+The type number is also used to index into an array holding the number
+of objects of each type and the total memory allocated for objects of
+that type. The statistics in this array are computed during the sweep
+stage. These statistics are returned by the call to
+@code{garbage-collect}.
Note that for every type defined with a @code{DEFINE_LRECORD_*()}
macro, there needs to be a @code{DECLARE_LRECORD_IMPLEMENTATION()}
file. To create one of these, copy an existing model and modify as
necessary.
+ @strong{Please note:} If you define an lrecord in an external
+dynamically-loaded module, you must use @code{DECLARE_EXTERNAL_LRECORD},
+@code{DEFINE_EXTERNAL_LRECORD_IMPLEMENTATION}, and
+@code{DEFINE_EXTERNAL_LRECORD_SEQUENCE_IMPLEMENTATION} instead of the
+non-EXTERNAL forms. These macros will dynamically add new type numbers
+to the global enum that records them, whereas the non-EXTERNAL forms
+assume that the programmer has already inserted the correct type numbers
+into the enum's code at compile-time.
+
The various methods in the lrecord implementation structure are:
@enumerate
a function pointer (usually the @code{mark_object()} function), which is
used to mark an object. All Lisp objects that are contained within the
object need to be marked by applying this function to them. The mark
-method should also return a Lisp object, which should be either nil or
+method should also return a Lisp object, which should be either @code{nil} or
an object to mark. (This can be used in lieu of calling
@code{mark_object()} on the object, to reduce the recursion depth, and
consequently should be the most heavily nested sub-object, such as a
@node Low-level allocation
@section Low-level allocation
+@cindex low-level allocation
+@cindex allocation, low-level
Memory that you want to allocate directly should be allocated using
@code{xmalloc()} rather than @code{malloc()}. This implements
(On some systems, the memory warnings are not functional.)
Allocated memory that is going to be used to make a Lisp object
-is created using @code{allocate_lisp_storage()}. This calls @code{xmalloc()}
-but also verifies that the pointer to the memory can fit into
-a Lisp word (remember that some bits are taken away for a type
-tag and a mark bit). If not, an error is issued through @code{memory_full()}.
-@code{allocate_lisp_storage()} is called by @code{alloc_lcrecord()},
-@code{ALLOCATE_FIXED_TYPE()}, and the vector and bit-vector creation
-routines. These routines also call @code{INCREMENT_CONS_COUNTER()} at the
-appropriate times; this keeps statistics on how much memory is
-allocated, so that garbage-collection can be invoked when the
-threshold is reached.
-
-@node Pure Space
-@section Pure Space
-
- Not yet documented.
+is created using @code{allocate_lisp_storage()}. This just calls
+@code{xmalloc()}. It used to verify that the pointer to the memory can
+fit into a Lisp word, before the current Lisp object representation was
+introduced. @code{allocate_lisp_storage()} is called by
+@code{alloc_lcrecord()}, @code{ALLOCATE_FIXED_TYPE()}, and the vector
+and bit-vector creation routines. These routines also call
+@code{INCREMENT_CONS_COUNTER()} at the appropriate times; this keeps
+statistics on how much memory is allocated, so that garbage-collection
+can be invoked when the threshold is reached.
@node Cons
@section Cons
+@cindex cons
Conses are allocated in standard frob blocks. The only thing to
note is that conses can be explicitly freed using @code{free_cons()}
@node Vector
@section Vector
+@cindex vector
As mentioned above, each vector is @code{malloc()}ed individually, and
all are threaded through the variable @code{all_vectors}. Vectors are
@node Bit Vector
@section Bit Vector
+@cindex bit vector
+@cindex vector, bit
Bit vectors work exactly like vectors, except for more complicated
code to access an individual bit, and except for the fact that bit
@node Symbol
@section Symbol
+@cindex symbol
- Symbols are also allocated in frob blocks. Note that the code
-exists for symbols to be either lrecords (category (c) above)
-or simple types (category (b) above), and are lrecords by
-default (I think), although there is no good reason for this.
-
- Note that symbols in the awful horrible obarray structure are
-chained through their @code{next} field.
+ Symbols are also allocated in frob blocks. Symbols in the awful
+horrible obarray structure are chained through their @code{next} field.
Remember that @code{intern} looks up a symbol in an obarray, creating
one if necessary.
@node Marker
@section Marker
+@cindex marker
Markers are allocated in frob blocks, as usual. They are kept
in a buffer unordered, but in a doubly-linked list so that they
@node String
@section String
+@cindex string
As mentioned above, strings are a special case. A string is logically
two parts, a fixed-size object (containing the length, property list,
strings}, are all @code{malloc()}ed as their own block. (#### Although it
would make more sense for the threshold for big strings to be somewhat
lower, e.g. 1/2 or 1/4 the size of a string-chars block. It seems that
-this was indeed the case formerly -- indeed, the threshold was set at
-1/8 -- but Mly forgot about this when rewriting things for 19.8.)
+this was indeed the case formerly---indeed, the threshold was set at
+1/8---but Mly forgot about this when rewriting things for 19.8.)
Note also that the string data in string-chars blocks is padded as
necessary so that proper alignment constraints on the @code{struct
@node Compiled Function
@section Compiled Function
+@cindex compiled function
+@cindex function, compiled
Not yet documented.
-@node Events and the Event Loop, Evaluation; Stack Frames; Bindings, Allocation of Objects in XEmacs Lisp, Top
+
+@node Dumping, Events and the Event Loop, Allocation of Objects in XEmacs Lisp, Top
+@chapter Dumping
+@cindex dumping
+
+@section What is dumping and its justification
+@cindex dumping and its justification, what is
+
+The C code of XEmacs is just a Lisp engine with a lot of built-in
+primitives useful for writing an editor. The editor itself is written
+mostly in Lisp, and represents around 100K lines of code. Loading and
+executing the initialization of all this code takes a bit a time (five
+to ten times the usual startup time of current xemacs) and requires
+having all the lisp source files around. Having to reload them each
+time the editor is started would not be acceptable.
+
+The traditional solution to this problem is called dumping: the build
+process first creates the lisp engine under the name @file{temacs}, then
+runs it until it has finished loading and initializing all the lisp
+code, and eventually creates a new executable called @file{xemacs}
+including both the object code in @file{temacs} and all the contents of
+the memory after the initialization.
+
+This solution, while working, has a huge problem: the creation of the
+new executable from the actual contents of memory is an extremely
+system-specific process, quite error-prone, and which interferes with a
+lot of system libraries (like malloc). It is even getting worse
+nowadays with libraries using constructors which are automatically
+called when the program is started (even before main()) which tend to
+crash when they are called multiple times, once before dumping and once
+after (IRIX 6.x libz.so pulls in some C++ image libraries thru
+dependencies which have this problem). Writing the dumper is also one
+of the most difficult parts of porting XEmacs to a new operating system.
+Basically, `dumping' is an operation that is just not officially
+supported on many operating systems.
+
+The aim of the portable dumper is to solve the same problem as the
+system-specific dumper, that is to be able to reload quickly, using only
+a small number of files, the fully initialized lisp part of the editor,
+without any system-specific hacks.
+
+@menu
+* Overview::
+* Data descriptions::
+* Dumping phase::
+* Reloading phase::
+* Remaining issues::
+@end menu
+
+@node Overview
+@section Overview
+@cindex dumping overview
+
+The portable dumping system has to:
+
+@enumerate
+@item
+At dump time, write all initialized, non-quickly-rebuildable data to a
+file [Note: currently named @file{xemacs.dmp}, but the name will
+change], along with all informations needed for the reloading.
+
+@item
+When starting xemacs, reload the dump file, relocate it to its new
+starting address if needed, and reinitialize all pointers to this
+data. Also, rebuild all the quickly rebuildable data.
+@end enumerate
+
+@node Data descriptions
+@section Data descriptions
+@cindex dumping data descriptions
+
+The more complex task of the dumper is to be able to write lisp objects
+(lrecords) and C structs to disk and reload them at a different address,
+updating all the pointers they include in the process. This is done by
+using external data descriptions that give information about the layout
+of the structures in memory.
+
+The specification of these descriptions is in lrecord.h. A description
+of an lrecord is an array of struct lrecord_description. Each of these
+structs include a type, an offset in the structure and some optional
+parameters depending on the type. For instance, here is the string
+description:
+
+@example
+static const struct lrecord_description string_description[] = @{
+ @{ XD_BYTECOUNT, offsetof (Lisp_String, size) @},
+ @{ XD_OPAQUE_DATA_PTR, offsetof (Lisp_String, data), XD_INDIRECT(0, 1) @},
+ @{ XD_LISP_OBJECT, offsetof (Lisp_String, plist) @},
+ @{ XD_END @}
+@};
+@end example
+
+The first line indicates a member of type Bytecount, which is used by
+the next, indirect directive. The second means "there is a pointer to
+some opaque data in the field @code{data}". The length of said data is
+given by the expression @code{XD_INDIRECT(0, 1)}, which means "the value
+in the 0th line of the description (welcome to C) plus one". The third
+line means "there is a Lisp_Object member @code{plist} in the Lisp_String
+structure". @code{XD_END} then ends the description.
+
+This gives us all the information we need to move around what is pointed
+to by a structure (C or lrecord) and, by transitivity, everything that
+it points to. The only missing information for dumping is the size of
+the structure. For lrecords, this is part of the
+lrecord_implementation, so we don't need to duplicate it. For C
+structures we use a struct struct_description, which includes a size
+field and a pointer to an associated array of lrecord_description.
+
+@node Dumping phase
+@section Dumping phase
+@cindex dumping phase
+
+Dumping is done by calling the function pdump() (in dumper.c) which is
+invoked from Fdump_emacs (in emacs.c). This function performs a number
+of tasks.
+
+@menu
+* Object inventory::
+* Address allocation::
+* The header::
+* Data dumping::
+* Pointers dumping::
+@end menu
+
+@node Object inventory
+@subsection Object inventory
+@cindex dumping object inventory
+
+The first task is to build the list of the objects to dump. This
+includes:
+
+@itemize @bullet
+@item lisp objects
+@item C structures
+@end itemize
+
+We end up with one @code{pdump_entry_list_elmt} per object group (arrays
+of C structs are kept together) which includes a pointer to the first
+object of the group, the per-object size and the count of objects in the
+group, along with some other information which is initialized later.
+
+These entries are linked together in @code{pdump_entry_list} structures
+and can be enumerated thru either:
+
+@enumerate
+@item
+the @code{pdump_object_table}, an array of @code{pdump_entry_list}, one
+per lrecord type, indexed by type number.
+
+@item
+the @code{pdump_opaque_data_list}, used for the opaque data which does
+not include pointers, and hence does not need descriptions.
+
+@item
+the @code{pdump_struct_table}, which is a vector of
+@code{struct_description}/@code{pdump_entry_list} pairs, used for
+non-opaque C structures.
+@end enumerate
+
+This uses a marking strategy similar to the garbage collector. Some
+differences though:
+
+@enumerate
+@item
+We do not use the mark bit (which does not exist for C structures
+anyway); we use a big hash table instead.
+
+@item
+We do not use the mark function of lrecords but instead rely on the
+external descriptions. This happens essentially because we need to
+follow pointers to C structures and opaque data in addition to
+Lisp_Object members.
+@end enumerate
+
+This is done by @code{pdump_register_object()}, which handles Lisp_Object
+variables, and @code{pdump_register_struct()} which handles C structures,
+which both delegate the description management to @code{pdump_register_sub()}.
+
+The hash table doubles as a map object to pdump_entry_list_elmt (i.e.
+allows us to look up a pdump_entry_list_elmt with the object it points
+to). Entries are added with @code{pdump_add_entry()} and looked up with
+@code{pdump_get_entry()}. There is no need for entry removal. The hash
+value is computed quite simply from the object pointer by
+@code{pdump_make_hash()}.
+
+The roots for the marking are:
+
+@enumerate
+@item
+the @code{staticpro}'ed variables (there is a special @code{staticpro_nodump()}
+call for protected variables we do not want to dump).
+
+@item
+the variables registered via @code{dump_add_root_object}
+(@code{staticpro()} is equivalent to @code{staticpro_nodump()} +
+@code{dump_add_root_object()}).
+
+@item
+the variables registered via @code{dump_add_root_struct_ptr}, each of
+which points to a C structure.
+@end enumerate
+
+This does not include the GCPRO'ed variables, the specbinds, the
+catchtags, the backlist, the redisplay or the profiling info, since we
+do not want to rebuild the actual chain of lisp calls which end up to
+the dump-emacs call, only the global variables.
+
+Weak lists and weak hash tables are dumped as if they were their
+non-weak equivalent (without changing their type, of course). This has
+not yet been a problem.
+
+@node Address allocation
+@subsection Address allocation
+@cindex dumping address allocation
+
+
+The next step is to allocate the offsets of each of the objects in the
+final dump file. This is done by @code{pdump_allocate_offset()} which
+is called indirectly by @code{pdump_scan_by_alignment()}.
+
+The strategy to deal with alignment problems uses these facts:
+
+@enumerate
+@item
+real world alignment requirements are powers of two.
+
+@item
+the C compiler is required to adjust the size of a struct so that you
+can have an array of them next to each other. This means you can have an
+upper bound of the alignment requirements of a given structure by
+looking at which power of two its size is a multiple.
+
+@item
+the non-variant part of variable size lrecords has an alignment
+requirement of 4.
+@end enumerate
+
+Hence, for each lrecord type, C struct type or opaque data block the
+alignment requirement is computed as a power of two, with a minimum of
+2^2 for lrecords. @code{pdump_scan_by_alignment()} then scans all the
+@code{pdump_entry_list_elmt}'s, the ones with the highest requirements
+first. This ensures the best packing.
+
+The maximum alignment requirement we take into account is 2^8.
+
+@code{pdump_allocate_offset()} only has to do a linear allocation,
+starting at offset 256 (this leaves room for the header and keeps the
+alignments happy).
+
+@node The header
+@subsection The header
+@cindex dumping, the header
+
+The next step creates the file and writes a header with a signature and
+some random information in it. The @code{reloc_address} field, which
+indicates at which address the file should be loaded if we want to avoid
+post-reload relocation, is set to 0. It then seeks to offset 256 (base
+offset for the objects).
+
+@node Data dumping
+@subsection Data dumping
+@cindex data dumping
+@cindex dumping, data
+
+The data is dumped in the same order as the addresses were allocated by
+@code{pdump_dump_data()}, called from @code{pdump_scan_by_alignment()}.
+This function copies the data to a temporary buffer, relocates all
+pointers in the object to the addresses allocated in step Address
+Allocation, and writes it to the file. Using the same order means that,
+if we are careful with lrecords whose size is not a multiple of 4, we
+are ensured that the object is always written at the offset in the file
+allocated in step Address Allocation.
+
+@node Pointers dumping
+@subsection Pointers dumping
+@cindex pointers dumping
+@cindex dumping, pointers
+
+A bunch of tables needed to reassign properly the global pointers are
+then written. They are:
+
+@enumerate
+@item
+the pdump_root_struct_ptrs dynarr
+@item
+the pdump_opaques dynarr
+@item
+a vector of all the offsets to the objects in the file that include a
+description (for faster relocation at reload time)
+@item
+the pdump_root_objects and pdump_weak_object_chains dynarrs.
+@end enumerate
+
+For each of the dynarrs we write both the pointer to the variables and
+the relocated offset of the object they point to. Since these variables
+are global, the pointers are still valid when restarting the program and
+are used to regenerate the global pointers.
+
+The @code{pdump_weak_object_chains} dynarr is a special case. The
+variables it points to are the head of weak linked lists of lisp objects
+of the same type. Not all objects of this list are dumped so the
+relocated pointer we associate with them points to the first dumped
+object of the list, or Qnil if none is available. This is also the
+reason why they are not used as roots for the purpose of object
+enumeration.
+
+Some very important information like the @code{staticpros} and
+@code{lrecord_implementations_table} are handled indirectly using
+@code{dump_add_opaque} or @code{dump_add_root_struct_ptr}.
+
+This is the end of the dumping part.
+
+@node Reloading phase
+@section Reloading phase
+@cindex reloading phase
+@cindex dumping, reloading phase
+
+@subsection File loading
+@cindex dumping, file loading
+
+The file is mmap'ed in memory (which ensures a PAGESIZE alignment, at
+least 4096), or if mmap is unavailable or fails, a 256-bytes aligned
+malloc is done and the file is loaded.
+
+Some variables are reinitialized from the values found in the header.
+
+The difference between the actual loading address and the reloc_address
+is computed and will be used for all the relocations.
+
+
+@subsection Putting back the pdump_opaques
+@cindex dumping, putting back the pdump_opaques
+
+The memory contents are restored in the obvious and trivial way.
+
+
+@subsection Putting back the pdump_root_struct_ptrs
+@cindex dumping, putting back the pdump_root_struct_ptrs
+
+The variables pointed to by pdump_root_struct_ptrs in the dump phase are
+reset to the right relocated object addresses.
+
+
+@subsection Object relocation
+@cindex dumping, object relocation
+
+All the objects are relocated using their description and their offset
+by @code{pdump_reloc_one}. This step is unnecessary if the
+reloc_address is equal to the file loading address.
+
+
+@subsection Putting back the pdump_root_objects and pdump_weak_object_chains
+@cindex dumping, putting back the pdump_root_objects and pdump_weak_object_chains
+
+Same as Putting back the pdump_root_struct_ptrs.
+
+
+@subsection Reorganize the hash tables
+@cindex dumping, reorganize the hash tables
+
+Since some of the hash values in the lisp hash tables are
+address-dependent, their layout is now wrong. So we go through each of
+them and have them resorted by calling @code{pdump_reorganize_hash_table}.
+
+@node Remaining issues
+@section Remaining issues
+@cindex dumping, remaining issues
+
+The build process will have to start a post-dump xemacs, ask it the
+loading address (which will, hopefully, be always the same between
+different xemacs invocations) and relocate the file to the new address.
+This way the object relocation phase will not have to be done, which
+means no writes in the objects and that, because of the use of mmap, the
+dumped data will be shared between all the xemacs running on the
+computer.
+
+Some executable signature will be necessary to ensure that a given dump
+file is really associated with a given executable, or random crashes
+will occur. Maybe a random number set at compile or configure time thru
+a define. This will also allow for having differently-compiled xemacsen
+on the same system (mule and no-mule comes to mind).
+
+The DOC file contents should probably end up in the dump file.
+
+
+@node Events and the Event Loop, Evaluation; Stack Frames; Bindings, Dumping, Top
@chapter Events and the Event Loop
+@cindex events and the event loop
+@cindex event loop, events and the
@menu
* Introduction to Events::
@node Introduction to Events
@section Introduction to Events
+@cindex events, introduction to
An event is an object that encapsulates information about an
interesting occurrence in the operating system. Events are
nature of the most basic events that are received. Part of the
complex nature of the XEmacs event collection process involves
converting from the operating-system events into the proper
-Emacs events -- there may not be a one-to-one correspondence.
+Emacs events---there may not be a one-to-one correspondence.
Emacs events are documented in @file{events.h}; I'll discuss them
later.
@node Main Loop
@section Main Loop
+@cindex main loop
+@cindex events, main loop
The @dfn{command loop} is the top-level loop that the editor is always
running. It loops endlessly, calling @code{next-event} to retrieve an
This is documented elsewhere.
The guts of the command loop are in @code{command_loop_1()}. This
-function doesn't catch errors, though -- that's the job of
+function doesn't catch errors, though---that's the job of
@code{command_loop_2()}, which is a condition-case (i.e. error-trapping)
wrapper around @code{command_loop_1()}. @code{command_loop_1()} never
returns, but may get thrown out of.
@node Specifics of the Event Gathering Mechanism
@section Specifics of the Event Gathering Mechanism
+@cindex event gathering mechanism, specifics of the
Here is an approximate diagram of the collection processes
at work in XEmacs, under TTY's (TTY's are simpler than X
@node Specifics About the Emacs Event
@section Specifics About the Emacs Event
+@cindex event, specifics about the Lisp object
@node The Event Stream Callback Routines
@section The Event Stream Callback Routines
+@cindex event stream callback routines, the
+@cindex callback routines, the event stream
@node Other Event Loop Functions
@section Other Event Loop Functions
+@cindex event loop functions, other
@code{detect_input_pending()} and @code{input-pending-p} look for
input by calling @code{event_stream->event_pending_p} and looking in
@node Converting Events
@section Converting Events
+@cindex converting events
+@cindex events, converting
@code{character_to_event()}, @code{event_to_character()},
@code{event-to-character}, and @code{character-to-event} convert between
@node Dispatching Events; The Command Builder
@section Dispatching Events; The Command Builder
+@cindex dispatching events; the command builder
+@cindex events; the command builder, dispatching
+@cindex command builder, dispatching events; the
Not yet documented.
@node Evaluation; Stack Frames; Bindings, Symbols and Variables, Events and the Event Loop, Top
@chapter Evaluation; Stack Frames; Bindings
+@cindex evaluation; stack frames; bindings
+@cindex stack frames; bindings, evaluation;
+@cindex bindings, evaluation; stack frames;
@menu
* Evaluation::
@node Evaluation
@section Evaluation
+@cindex evaluation
@code{Feval()} evaluates the form (a Lisp object) that is passed to
it. Note that evaluation is only non-trivial for two types of objects:
are converted into an internal form for faster execution.
When a compiled function is executed for the first time by
-@code{funcall_compiled_function()}, or when it is @code{Fpurecopy()}ed
-during the dump phase of building XEmacs, the byte-code instructions are
-converted from a @code{Lisp_String} (which is inefficient to access,
-especially in the presence of MULE) into a @code{Lisp_Opaque} object
-containing an array of unsigned char, which can be directly executed by
-the byte-code interpreter. At this time the byte code is also analyzed
-for validity and transformed into a more optimized form, so that
+@code{funcall_compiled_function()}, or during the dump phase of building
+XEmacs, the byte-code instructions are converted from a
+@code{Lisp_String} (which is inefficient to access, especially in the
+presence of MULE) into a @code{Lisp_Opaque} object containing an array
+of unsigned char, which can be directly executed by the byte-code
+interpreter. At this time the byte code is also analyzed for validity
+and transformed into a more optimized form, so that
@code{execute_optimized_program()} can really fly.
Here are some of the optimizations performed by the internal byte-code
@code{nil}, or @code{keywordp}) symbols, so that the byte interpreter
doesn't have to.
@item
-The maxiumum number of variable bindings in the byte-code is
+The maximum number of variable bindings in the byte-code is
pre-computed, so that space on the @code{specpdl} stack can be
pre-reserved once for the whole function execution.
@item
@node Dynamic Binding; The specbinding Stack; Unwind-Protects
@section Dynamic Binding; The specbinding Stack; Unwind-Protects
+@cindex dynamic binding; the specbinding stack; unwind-protects
+@cindex binding; the specbinding stack; unwind-protects, dynamic
+@cindex specbinding stack; unwind-protects, dynamic binding; the
+@cindex unwind-protects, dynamic binding; the specbinding stack;
@example
struct specbinding
@node Simple Special Forms
@section Simple Special Forms
+@cindex special forms, simple
@code{or}, @code{and}, @code{if}, @code{cond}, @code{progn},
@code{prog1}, @code{prog2}, @code{setq}, @code{quote}, @code{function},
@code{let} and @code{let*}) using @code{specbind()} to create bindings
and @code{unbind_to()} to undo the bindings when finished.
-Note that, with the exeption of @code{Fprogn}, these functions are
+Note that, with the exception of @code{Fprogn}, these functions are
typically called in real life only in interpreted code, since the byte
compiler knows how to convert calls to these functions directly into
byte code.
@node Catch and Throw
@section Catch and Throw
+@cindex catch and throw
+@cindex throw, catch and
@example
struct catchtag
@node Symbols and Variables, Buffers and Textual Representation, Evaluation; Stack Frames; Bindings, Top
@chapter Symbols and Variables
+@cindex symbols and variables
+@cindex variables, symbols and
@menu
* Introduction to Symbols::
@node Introduction to Symbols
@section Introduction to Symbols
+@cindex symbols, introduction to
A symbol is basically just an object with four fields: a name (a
string), a value (some Lisp object), a function (some Lisp object), and
@node Obarrays
@section Obarrays
+@cindex obarrays
The identity of symbols with their names is accomplished through a
structure called an obarray, which is just a poorly-implemented hash
@node Symbol Values
@section Symbol Values
+@cindex symbol values
+@cindex values, symbol
The value field of a symbol normally contains a Lisp object. However,
a symbol can be @dfn{unbound}, meaning that it logically has no value.
@node Buffers and Textual Representation, MULE Character Sets and Encodings, Symbols and Variables, Top
@chapter Buffers and Textual Representation
+@cindex buffers and textual representation
+@cindex textual representation, buffers and
@menu
* Introduction to Buffers:: A buffer holds a block of text such as a file.
@node Introduction to Buffers
@section Introduction to Buffers
+@cindex buffers, introduction to
A buffer is logically just a Lisp object that holds some text.
In this, it is like a string, but a buffer is optimized for
@node The Text in a Buffer
@section The Text in a Buffer
+@cindex text in a buffer, the
+@cindex buffer, the text in a
The text in a buffer consists of a sequence of zero or more
characters. A @dfn{character} is an integer that logically represents
@node Buffer Lists
@section Buffer Lists
+@cindex buffer lists
Recall earlier that buffers are @dfn{permanent} objects, i.e. that
they remain around until explicitly deleted. This entails that there is
@node Markers and Extents
@section Markers and Extents
+@cindex markers and extents
+@cindex extents, markers and
Among the things associated with a buffer are things that are
logically attached to certain buffer positions. This can be used to
buffer positions in them as integers, and every time text is inserted or
deleted, these positions must be updated. In order to minimize the
amount of shuffling that needs to be done, the positions in markers and
-extents (there's one per marker, two per extent) and stored in Meminds.
+extents (there's one per marker, two per extent) are stored in Meminds.
This means that they only need to be moved when the text is physically
moved in memory; since the gap structure tries to minimize this, it also
minimizes the number of marker and extent indices that need to be
@node Bufbytes and Emchars
@section Bufbytes and Emchars
+@cindex Bufbytes and Emchars
+@cindex Emchars, Bufbytes and
Not yet documented.
@node The Buffer Object
@section The Buffer Object
+@cindex buffer object, the
+@cindex object, the buffer
Buffers contain fields not directly accessible by the Lisp programmer.
We describe them here, naming them by the names used in the C code.
@table @code
@item name
The buffer name is a string that names the buffer. It is guaranteed to
-be unique. @xref{Buffer Names,,, lispref, XEmacs Lisp Programmer's
+be unique. @xref{Buffer Names,,, lispref, XEmacs Lisp Reference
Manual}.
@item save_modified
This field contains the time when the buffer was last saved, as an
-integer. @xref{Buffer Modification,,, lispref, XEmacs Lisp Programmer's
+integer. @xref{Buffer Modification,,, lispref, XEmacs Lisp Reference
Manual}.
@item modtime
This field contains the modification time of the visited file. It is
set when the file is written or read. Every time the buffer is written
to the file, this field is compared to the modification time of the
-file. @xref{Buffer Modification,,, lispref, XEmacs Lisp Programmer's
+file. @xref{Buffer Modification,,, lispref, XEmacs Lisp Reference
Manual}.
@item auto_save_modified
@item undo_list
This field points to the buffer's undo list. @xref{Undo,,, lispref,
-XEmacs Lisp Programmer's Manual}.
+XEmacs Lisp Reference Manual}.
@item syntax_table_v
This field contains the syntax table for the buffer. @xref{Syntax
-Tables,,, lispref, XEmacs Lisp Programmer's Manual}.
+Tables,,, lispref, XEmacs Lisp Reference Manual}.
@item downcase_table
This field contains the conversion table for converting text to lower
-case. @xref{Case Tables,,, lispref, XEmacs Lisp Programmer's Manual}.
+case. @xref{Case Tables,,, lispref, XEmacs Lisp Reference Manual}.
@item upcase_table
This field contains the conversion table for converting text to upper
-case. @xref{Case Tables,,, lispref, XEmacs Lisp Programmer's Manual}.
+case. @xref{Case Tables,,, lispref, XEmacs Lisp Reference Manual}.
@item case_canon_table
This field contains the conversion table for canonicalizing text for
case-folding search. @xref{Case Tables,,, lispref, XEmacs Lisp
-Programmer's Manual}.
+Reference Manual}.
@item case_eqv_table
This field contains the equivalence table for case-folding search.
-@xref{Case Tables,,, lispref, XEmacs Lisp Programmer's Manual}.
+@xref{Case Tables,,, lispref, XEmacs Lisp Reference Manual}.
@item display_table
This field contains the buffer's display table, or @code{nil} if it
doesn't have one. @xref{Display Tables,,, lispref, XEmacs Lisp
-Programmer's Manual}.
+Reference Manual}.
@item markers
This field contains the chain of all markers that currently point into
the buffer. Deletion of text in the buffer, and motion of the buffer's
gap, must check each of these markers and perhaps update it.
-@xref{Markers,,, lispref, XEmacs Lisp Programmer's Manual}.
+@xref{Markers,,, lispref, XEmacs Lisp Reference Manual}.
@item backed_up
This field is a flag that tells whether a backup file has been made for
@item mark
This field contains the mark for the buffer. The mark is a marker,
hence it is also included on the list @code{markers}. @xref{The Mark,,,
-lispref, XEmacs Lisp Programmer's Manual}.
+lispref, XEmacs Lisp Reference Manual}.
@item mark_active
This field is non-@code{nil} if the buffer's mark is active.
in this buffer, and their values, with the exception of local variables
that have special slots in the buffer object. (Those slots are omitted
from this table.) @xref{Buffer-Local Variables,,, lispref, XEmacs Lisp
-Programmer's Manual}.
+Reference Manual}.
@item modeline_format
This field contains a Lisp object which controls how to display the mode
line for this buffer. @xref{Modeline Format,,, lispref, XEmacs Lisp
-Programmer's Manual}.
+Reference Manual}.
@item base_buffer
This field holds the buffer's base buffer (if it is an indirect buffer),
@node MULE Character Sets and Encodings, The Lisp Reader and Compiler, Buffers and Textual Representation, Top
@chapter MULE Character Sets and Encodings
+@cindex Mule character sets and encodings
+@cindex character sets and encodings, Mule
+@cindex encodings, Mule character sets and
Recall that there are two primary ways that text is represented in
XEmacs. The @dfn{buffer} representation sees the text as a series of
@node Character Sets
@section Character Sets
+@cindex character sets
A character set (or @dfn{charset}) is an ordered set of characters. A
particular character in a charset is indexed using one or more
@node Encodings
@section Encodings
+@cindex encodings, Mule
+@cindex Mule encodings
An @dfn{encoding} is a way of numerically representing characters from
one or more character sets. If an encoding only encompasses one
@node Japanese EUC (Extended Unix Code)
@subsection Japanese EUC (Extended Unix Code)
+@cindex Japanese EUC (Extended Unix Code)
+@cindex EUC (Extended Unix Code), Japanese
+@cindex Extended Unix Code, Japanese EUC
This encompasses the character sets Printing-ASCII, Japanese-JISX0201,
and Japanese-JISX0208-Kana (half-width katakana, the right half of
@node JIS7
@subsection JIS7
+@cindex JIS7
This encompasses the character sets Printing-ASCII,
Japanese-JISX0201-Roman (the left half of JISX0201; this character set
@node Internal Mule Encodings
@section Internal Mule Encodings
+@cindex internal Mule encodings
+@cindex Mule encodings, internal
+@cindex encodings, internal Mule
In XEmacs/Mule, each character set is assigned a unique number, called a
@dfn{leading byte}. This is used in the encodings of a character.
@node Internal String Encoding
@subsection Internal String Encoding
+@cindex internal string encoding
+@cindex string encoding, internal
+@cindex encoding, internal string
ASCII characters are encoded using their position code directly. Other
characters are encoded using their leading byte followed by their
@node Internal Character Encoding
@subsection Internal Character Encoding
+@cindex internal character encoding
+@cindex character encoding, internal
+@cindex encoding, internal character
One 19-bit word represents a single character. The word is
separated into three fields:
@node CCL
@section CCL
+@cindex CCL
@example
CCL PROGRAM SYNTAX:
other encoded/decoded data has been written out. This is not used for
charset CCL programs.
-REGISTER: 0..7 -- refered by RRR or rrr
+REGISTER: 0..7 -- referred by RRR or rrr
OPERATOR BIT FIELD (27-bit): XXXXXXXXXXXXXXX RRR TTTTT
TTTTT (5-bit): operator type
@node The Lisp Reader and Compiler, Lstreams, MULE Character Sets and Encodings, Top
@chapter The Lisp Reader and Compiler
+@cindex Lisp reader and compiler, the
+@cindex reader and compiler, the Lisp
+@cindex compiler, the Lisp reader and
Not yet documented.
@node Lstreams, Consoles; Devices; Frames; Windows, The Lisp Reader and Compiler, Top
@chapter Lstreams
+@cindex lstreams
An @dfn{lstream} is an internal Lisp object that provides a generic
buffering stream implementation. Conceptually, you send data to the
@node Creating an Lstream
@section Creating an Lstream
+@cindex lstream, creating an
Lstreams come in different types, depending on what is being interfaced
to. Although the primitive for creating new lstreams is
@node Lstream Types
@section Lstream Types
+@cindex lstream types
+@cindex types, lstream
@table @asis
@item stdio
@node Lstream Functions
@section Lstream Functions
+@cindex lstream functions
-@deftypefun {Lstream *} Lstream_new (Lstream_implementation *@var{imp}, CONST char *@var{mode})
+@deftypefun {Lstream *} Lstream_new (Lstream_implementation *@var{imp}, const char *@var{mode})
Allocate and return a new Lstream. This function is not really meant to
be called directly; rather, each stream type should provide its own
stream creation function, which creates the stream and does any other
@deftypefn Macro void Lstream_ungetc (Lstream *@var{stream}, int @var{c})
Push one byte back onto the input queue. This will be the next byte
read from the stream. Any number of bytes can be pushed back and will
-be read in the reverse order they were pushed back -- most recent
-first. (This is necessary for consistency -- if there are a number of
+be read in the reverse order they were pushed back---most recent
+first. (This is necessary for consistency---if there are a number of
bytes that have been unread and I read and unread a byte, it needs to be
the first to be read again.) This is a macro and so it is very
efficient. The @var{c} argument is only evaluated once but the @var{stream}
@deftypefun void Lstream_reopen (Lstream *@var{stream})
Reopen a closed stream. This enables I/O on it again. This is not
meant to be called except from a wrapper routine that reinitializes
-variables and such -- the close routine may well have freed some
+variables and such---the close routine may well have freed some
necessary storage structures, for example.
@end deftypefun
@node Lstream Methods
@section Lstream Methods
+@cindex lstream methods
@deftypefn {Lstream Method} ssize_t reader (Lstream *@var{stream}, unsigned char *@var{data}, size_t @var{size})
Read some data from the stream's end and store it into @var{data}, which
This function can be @code{NULL} if the stream is output-only.
@end deftypefn
-@deftypefn {Lstream Method} ssize_t writer (Lstream *@var{stream}, CONST unsigned char *@var{data}, size_t @var{size})
+@deftypefn {Lstream Method} ssize_t writer (Lstream *@var{stream}, const unsigned char *@var{data}, size_t @var{size})
Send some data to the stream's end. Data to be sent is in @var{data}
and is @var{size} bytes. Return the number of bytes sent. This
function can send and return fewer bytes than is passed in; in that
@end deftypefn
@deftypefn {Lstream Method} int seekable_p (Lstream *@var{stream})
-Indicate whether this stream is seekable -- i.e. it can be rewound.
+Indicate whether this stream is seekable---i.e. it can be rewound.
This method is ignored if the stream does not have a rewind method. If
this method is not present, the result is determined by whether a rewind
method is present.
@node Consoles; Devices; Frames; Windows, The Redisplay Mechanism, Lstreams, Top
@chapter Consoles; Devices; Frames; Windows
+@cindex consoles; devices; frames; windows
+@cindex devices; frames; windows, consoles;
+@cindex frames; windows, consoles; devices;
+@cindex windows, consoles; devices; frames;
@menu
* Introduction to Consoles; Devices; Frames; Windows::
@node Introduction to Consoles; Devices; Frames; Windows
@section Introduction to Consoles; Devices; Frames; Windows
+@cindex consoles; devices; frames; windows, introduction to
+@cindex devices; frames; windows, introduction to consoles;
+@cindex frames; windows, introduction to consoles; devices;
+@cindex windows, introduction to consoles; devices; frames;
A window-system window that you see on the screen is called a
@dfn{frame} in Emacs terminology. Each frame is subdivided into one or
@dfn{selected display}, @dfn{selected frame}, and @dfn{selected window}.
Each of these objects is distinguished in various ways, such as being the
default object for various functions that act on objects of that type.
-Note that every containing object rememembers the ``selected'' object
+Note that every containing object remembers the ``selected'' object
among the objects that it contains: e.g. not only is there a selected
window, but every frame remembers the last window in it that was
selected, and changing the selected frame causes the remembered window
@node Point
@section Point
+@cindex point
Recall that every buffer has a current insertion position, called
@dfn{point}. Now, two or more windows may be displaying the same buffer,
these are @dfn{hchild} (a list of horizontally-arrayed children),
@dfn{vchild} (a list of vertically-arrayed children), and @dfn{buffer}
(the buffer contained in a leaf window). Exactly one of
-these will be non-nil. Remember that @dfn{horizontally-arrayed}
+these will be non-@code{nil}. Remember that @dfn{horizontally-arrayed}
means ``side-by-side'' and @dfn{vertically-arrayed} means
@dfn{one above the other}.
@item
Leaf windows also have markers in their @code{start} (the
first buffer position displayed in the window) and @code{pointm}
-(the window's stashed value of @code{point} -- see above) fields,
-while combination windows have nil in these fields.
+(the window's stashed value of @code{point}---see above) fields,
+while combination windows have @code{nil} in these fields.
@item
The list of children for a window is threaded through the
GC purposes.
@item
-Most frames actually have two top-level windows -- one for the
+Most frames actually have two top-level windows---one for the
minibuffer and one (the @dfn{root}) for everything else. The modeline
(if present) separates these two. The @code{next} field of the root
points to the minibuffer, and the @code{prev} field of the minibuffer
@node The Window Object
@section The Window Object
+@cindex window object, the
+@cindex object, the window
Windows have the following accessible fields:
@node The Redisplay Mechanism, Extents, Consoles; Devices; Frames; Windows, Top
@chapter The Redisplay Mechanism
+@cindex redisplay mechanism, the
The redisplay mechanism is one of the most complicated sections of
XEmacs, especially from a conceptual standpoint. This is doubly so
@node Critical Redisplay Sections
@section Critical Redisplay Sections
+@cindex redisplay sections, critical
@cindex critical redisplay sections
Within this section, we are defenseless and assume that the
is sufficient to always provide the needed information. The second
thing we can do is be smart about invalidating the cache.
- TODO -- Be smart about invalidating the cache. Potential places:
+ TODO---Be smart about invalidating the cache. Potential places:
@itemize @bullet
@item
@node Redisplay Piece by Piece
@section Redisplay Piece by Piece
-@cindex Redisplay Piece by Piece
+@cindex redisplay piece by piece
As you can begin to see redisplay is complex and also not well
documented. Chuck no longer works on XEmacs so this section is my take
@code{redisplay-x.c}, @code{redisplay-msw.c} and @code{redisplay-tty.c}
@end enumerate
-Steps 1 and 2 are device-independant and relatively complex. Step 3 is
+Steps 1 and 2 are device-independent and relatively complex. Step 3 is
mostly device-dependent.
Determining the desired display
dynarr's of @code{display_line}'s are held by each window representing
the current display and the desired display.
-The @code{display_line} structures are tighly tied to buffers which
+The @code{display_line} structures are tightly tied to buffers which
presents a problem for redisplay as this connection is bogus for the
modeline. Hence the @code{display_line} generation routines are
duplicated for generating the modeline. This means that the modeline
The guts of @code{display_line} generation are in
@code{create_text_block}, which creates a single display line for the
desired locale. This incrementally parses the characters on the current
-line and generates redisplay structures for each.
+line and generates redisplay structures for each.
Gutter redisplay is different. Because the data to display is stored in
a string we cannot use @code{create_text_block}. Instead we use
@node Extents, Faces, The Redisplay Mechanism, Top
@chapter Extents
+@cindex extents
@menu
* Introduction to Extents:: Extents are ranges over text, with properties.
* Extent Ordering:: How extents are ordered internally.
* Format of the Extent Info:: The extent information in a buffer or string.
* Zero-Length Extents:: A weird special case.
-* Mathematics of Extent Ordering:: A rigorous foundation.
+* Mathematics of Extent Ordering:: A rigorous foundation.
* Extent Fragments:: Cached information useful for redisplay.
@end menu
@node Introduction to Extents
@section Introduction to Extents
+@cindex extents, introduction to
Extents are regions over a buffer, with a start and an end position
denoting the region of the buffer included in the extent. In
@node Extent Ordering
@section Extent Ordering
+@cindex extent ordering
Extents are compared using memory indices. There are two orderings
for extents and both orders are kept current at all times. The normal
@node Format of the Extent Info
@section Format of the Extent Info
+@cindex extent info, format of the
An extent-info structure consists of a list of the buffer or string's
extents and a @dfn{stack of extents} that lists all of the extents over
a particular position. The stack-of-extents info is used for
-optimization purposes -- it basically caches some info that might
+optimization purposes---it basically caches some info that might
be expensive to compute. Certain otherwise hard computations are easy
given the stack of extents over a particular position, and if the
stack of extents over a nearby position is known (because it was
@node Zero-Length Extents
@section Zero-Length Extents
+@cindex zero-length extents
+@cindex extents, zero-length
Extents can be zero-length, and will end up that way if their endpoints
-are explicitly set that way or if their detachable property is nil
+are explicitly set that way or if their detachable property is @code{nil}
and all the text in the extent is deleted. (The exception is open-open
zero-length extents, which are barred from existing because there is
no sensible way to define their properties. Deletion of the text in
@node Mathematics of Extent Ordering
@section Mathematics of Extent Ordering
+@cindex mathematics of extent ordering
@cindex extent mathematics
-@cindex mathematics of extents
@cindex extent ordering
@cindex display order of extents
@node Extent Fragments
@section Extent Fragments
-@cindex extent fragment
+@cindex extent fragments
+@cindex fragments, extent
Imagine that the buffer is divided up into contiguous, non-overlapping
@dfn{runs} of text such that no extent starts or ends within a run
An extent fragment is a structure that holds data about the run that
contains a particular buffer position (if the buffer position is at the
-junction of two runs, the run after the position is used) -- the
+junction of two runs, the run after the position is used)---the
beginning and end of the run, a list of all of the extents in that run,
the @dfn{merged face} that results from merging all of the faces
corresponding to those extents, the begin and end glyphs at the
@node Faces, Glyphs, Extents, Top
@chapter Faces
+@cindex faces
Not yet documented.
@node Glyphs, Specifiers, Faces, Top
@chapter Glyphs
+@cindex glyphs
Glyphs are graphical elements that can be displayed in XEmacs buffers or
gutters. We use the term graphical element here in the broadest possible
-sense since glyphs can be as mundane as text to as arcane as a native
+sense since glyphs can be as mundane as text or as arcane as a native
tab widget.
In XEmacs, glyphs represent the uninstantiated state of graphical
elements, i.e. they hold all the information necessary to produce an
-image on-screen but the image does not exist at this stage.
+image on-screen but the image need not exist at this stage, and multiple
+screen images can be instantiated from a single glyph.
Glyphs are lazily instantiated by calling one of the glyph
functions. This usually occurs within redisplay when
@code{Fglyph_height} is called. Instantiation causes an image-instance
-to be created and cached. This cache is on a device basis for all glyphs
-except glyph-widgets, and on a window basis for glyph widgets. The
+to be created and cached. This cache is on a per-device basis for all glyphs
+except widget-glyphs, and on a per-window basis for widgets-glyphs. The
caching is done by @code{image_instantiate} and is necessary because it
is generally possible to display an image-instance in multiple
domains. For instance if we create a Pixmap, we can actually display
this on multiple windows - even though we only need a single Pixmap
instance to do this. If caching wasn't done then it would be necessary
-to create image-instances for every displayable occurrance of a glyph -
+to create image-instances for every displayable occurrence of a glyph -
and every usage - and this would be extremely memory and cpu intensive.
Widget-glyphs (a.k.a native widgets) are not cached in this way. This is
because widget-glyph image-instances on screen are toolkit windows, and
thus cannot be reused in multiple XEmacs domains. Thus widget-glyphs are
-cached on a window basis.
+cached on an XEmacs window basis.
Any action on a glyph first consults the cache before actually
instantiating a widget.
-@section Widget-Glyphs in the MS-WIndows Environment
+@section Glyph Instantiation
+@cindex glyph instantiation
+@cindex instantiation, glyph
+
+Glyph instantiation is a hairy topic and requires some explanation. The
+guts of glyph instantiation is contained within
+@code{image_instantiate}. A glyph contains an image which is a
+specifier. When a glyph function - for instance @code{Fglyph_height} -
+asks for a property of the glyph that can only be determined from its
+instantiated state, then the glyph image is instantiated and an image
+instance created. The instantiation process is governed by the specifier
+code and goes through a series of steps:
+
+@itemize @bullet
+@item
+Validation. Instantiation of image instances happens dynamically - often
+within the guts of redisplay. Thus it is often not feasible to catch
+instantiator errors at instantiation time. Instead the instantiator is
+validated at the time it is added to the image specifier. This function
+is defined by @code{image_validate} and at a simple level validates
+keyword value pairs.
+@item
+Duplication. The specifier code by default takes a copy of the
+instantiator. This is reasonable for most specifiers but in the case of
+widget-glyphs can be problematic, since some of the properties in the
+instantiator - for instance callbacks - could cause infinite recursion
+in the copying process. Thus the image code defines a function -
+@code{image_copy_instantiator} - which will selectively copy values.
+This is controlled by the way that a keyword is defined either using
+@code{IIFORMAT_VALID_KEYWORD} or
+@code{IIFORMAT_VALID_NONCOPY_KEYWORD}. Note that the image caching and
+redisplay code relies on instantiator copying to ensure that current and
+new instantiators are actually different rather than referring to the
+same thing.
+@item
+Normalization. Once the instantiator has been copied it must be
+converted into a form that is viable at instantiation time. This can
+involve no changes at all, but typically involves things like converting
+file names to the actual data. This function is defined by
+@code{image_going_to_add} and @code{normalize_image_instantiator}.
+@item
+Instantiation. When an image instance is actually required for display
+it is instantiated using @code{image_instantiate}. This involves calling
+instantiate methods that are specific to the type of image being
+instantiated.
+@end itemize
+
+The final instantiation phase also involves a number of steps. In order
+to understand these we need to describe a number of concepts.
+
+An image is instantiated in a @dfn{domain}, where a domain can be any
+one of a device, frame, window or image-instance. The domain gives the
+image-instance context and identity and properties that affect the
+appearance of the image-instance may be different for the same glyph
+instantiated in different domains. An example is the face used to
+display the image-instance.
+
+Although an image is instantiated in a particular domain the
+instantiation domain is not necessarily the domain in which the
+image-instance is cached. For example a pixmap can be instantiated in a
+window be actually be cached on a per-device basis. The domain in which
+the image-instance is actually cached is called the
+@dfn{governing-domain}. A governing-domain is currently either a device
+or a window. Widget-glyphs and text-glyphs have a window as a
+governing-domain, all other image-instances have a device as the
+governing-domain. The governing domain for an image-instance is
+determined using the governing_domain image-instance method.
+
+@section Widget-Glyphs
+@cindex widget-glyphs
+
+@section Widget-Glyphs in the MS-Windows Environment
+@cindex widget-glyphs in the MS-Windows environment
+@cindex MS-Windows environment, widget-glyphs in the
To Do
@section Widget-Glyphs in the X Environment
+@cindex widget-glyphs in the X environment
+@cindex X environment, widget-glyphs in the
-Widget-glyphs under X make heavy use of lwlib for manipulating the
-native toolkit objects. This is primarily so that different toolkits can
-be supported for widget-glyphs, just as they are supported for features
-such as menubars etc.
+Widget-glyphs under X make heavy use of lwlib (@pxref{Lucid Widget
+Library}) for manipulating the native toolkit objects. This is primarily
+so that different toolkits can be supported for widget-glyphs, just as
+they are supported for features such as menubars etc.
Lwlib is extremely poorly documented and quite hairy so here is my
understanding of what goes on.
tree recursively.
This has desirable properties such as lw_modify_all_widgets which is
-called from glyphs-x.c and updates all the properties of a widget
+called from @file{glyphs-x.c} and updates all the properties of a widget
without having to know what the widget is or what toolkit it is from.
-Unfortunately this also has hairy properrties such as making the lwlib
+Unfortunately this also has hairy properties such as making the lwlib
code quite complex. And of course lwlib has to know at some level what
the widget is and how to set its properties.
@node Specifiers, Menus, Glyphs, Top
@chapter Specifiers
+@cindex specifiers
Not yet documented.
@node Menus, Subprocesses, Specifiers, Top
@chapter Menus
+@cindex menus
A menu is set by setting the value of the variable
@code{current-menubar} (which may be buffer-local) and then calling
its argument, which is the callback function or form given in the menu's
description.
-@node Subprocesses, Interface to X Windows, Menus, Top
+@node Subprocesses, Interface to the X Window System, Menus, Top
@chapter Subprocesses
+@cindex subprocesses
The fields of a process are:
or @code{nil} if it is using pipes.
@end table
-@node Interface to X Windows, Index, Subprocesses, Top
-@chapter Interface to X Windows
+@node Interface to the X Window System, Index, Subprocesses, Top
+@chapter Interface to the X Window System
+@cindex X Window System, interface to the
-Not yet documented.
+Mostly undocumented.
+
+@menu
+* Lucid Widget Library:: An interface to various widget sets.
+@end menu
+
+@node Lucid Widget Library
+@section Lucid Widget Library
+@cindex Lucid Widget Library
+@cindex widget library, Lucid
+@cindex library, Lucid Widget
+
+Lwlib is extremely poorly documented and quite hairy. The author(s)
+blame that on X, Xt, and Motif, with some justice, but also sufficient
+hypocrisy to avoid drawing the obvious conclusion about their own work.
+
+The Lucid Widget Library is composed of two more or less independent
+pieces. The first, as the name suggests, is a set of widgets. These
+widgets are intended to resemble and improve on widgets provided in the
+Motif toolkit but not in the Athena widgets, including menubars and
+scrollbars. Recent additions by Andy Piper integrate some ``modern''
+widgets by Edward Falk, including checkboxes, radio buttons, progress
+gauges, and index tab controls (aka notebooks).
+
+The second piece of the Lucid widget library is a generic interface to
+several toolkits for X (including Xt, the Athena widget set, and Motif,
+as well as the Lucid widgets themselves) so that core XEmacs code need
+not know which widget set has been used to build the graphical user
+interface.
+
+@menu
+* Generic Widget Interface:: The lwlib generic widget interface.
+* Scrollbars::
+* Menubars::
+* Checkboxes and Radio Buttons::
+* Progress Bars::
+* Tab Controls::
+@end menu
+
+@node Generic Widget Interface
+@subsection Generic Widget Interface
+@cindex widget interface, generic
+
+In general in any toolkit a widget may be a composite object. In Xt,
+all widgets have an X window that they manage, but typically a complex
+widget will have widget children, each of which manages a subwindow of
+the parent widget's X window. These children may themselves be
+composite widgets. Thus a widget is actually a tree or hierarchy of
+widgets.
+
+For each toolkit widget, lwlib maintains a tree of @code{widget_values}
+which mirror the hierarchical state of Xt widgets (including Motif,
+Athena, 3D Athena, and Falk's widget sets). Each @code{widget_value}
+has @code{contents} member, which points to the head of a linked list of
+its children. The linked list of siblings is chained through the
+@code{next} member of @code{widget_value}.
+
+@example
+ +-----------+
+ | composite |
+ +-----------+
+ |
+ | contents
+ V
+ +-------+ next +-------+ next +-------+
+ | child |----->| child |----->| child |
+ +-------+ +-------+ +-------+
+ |
+ | contents
+ V
+ +-------------+ next +-------------+
+ | grand child |----->| grand child |
+ +-------------+ +-------------+
+
+The @code{widget_value} hierarchy of a composite widget with two simple
+children and one composite child.
+@end example
+
+The @code{widget_instance} structure maintains the inverse view of the
+tree. As for the @code{widget_value}, siblings are chained through the
+@code{next} member. However, rather than naming children, the
+@code{widget_instance} tree links to parents.
+
+@example
+ +-----------+
+ | composite |
+ +-----------+
+ A
+ | parent
+ |
+ +-------+ next +-------+ next +-------+
+ | child |----->| child |----->| child |
+ +-------+ +-------+ +-------+
+ A
+ | parent
+ |
+ +-------------+ next +-------------+
+ | grand child |----->| grand child |
+ +-------------+ +-------------+
+
+The @code{widget_value} hierarchy of a composite widget with two simple
+children and one composite child.
+@end example
+
+This permits widgets derived from different toolkits to be updated and
+manipulated generically by the lwlib library. For instance
+@code{update_one_widget_instance} can cope with multiple types of widget
+and multiple types of toolkit. Each element in the widget hierarchy is
+updated from its corresponding @code{widget_value} by walking the
+@code{widget_value} tree. This has desirable properties. For example,
+@code{lw_modify_all_widgets} is called from @file{glyphs-x.c} and
+updates all the properties of a widget without having to know what the
+widget is or what toolkit it is from. Unfortunately this also has its
+hairy properties; the lwlib code quite complex. And of course lwlib has
+to know at some level what the widget is and how to set its properties.
+
+The @code{widget_instance} structure also contains a pointer to the root
+of its tree. Widget instances are further confi
+
+
+@node Scrollbars
+@subsection Scrollbars
+@cindex scrollbars
+
+@node Menubars
+@subsection Menubars
+@cindex menubars
+
+@node Checkboxes and Radio Buttons
+@subsection Checkboxes and Radio Buttons
+@cindex checkboxes and radio buttons
+@cindex radio buttons, checkboxes and
+@cindex buttons, checkboxes and radio
+
+@node Progress Bars
+@subsection Progress Bars
+@cindex progress bars
+@cindex bars, progress
+
+@node Tab Controls
+@subsection Tab Controls
+@cindex tab controls
@include index.texi
@c That's all
@bye
-
projects / chise / xemacs-chise.git.1 / blobdiff