another doc merge

[Eric Blake]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Another round of doc merges.

head:
2006-11-16  Eric Blake  <address@hidden>

        * tests/builtins.at (esyscmd, syscmd): Redo, since spilled
        diversions are now closed when not active.
        (divert): Remove check now in the manual.
        * tests/others.at (stderr closed): Likewise.
        (stdout closed): Add a check.
        * doc/m4.texinfo (Include, Search Path, Diversions, Divert): Sync
        more docs with branch.

branch:
2006-11-16  Eric Blake  <address@hidden>

        * doc/m4.texinfo (Include, Search Path, Diversions, Divert):
        Minor tweaks noticed while porting to head.

- --
Life is short - so eat dessert first!

Eric Blake             address@hidden
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Cygwin)
Comment: Public key at home.comcast.net/~ericblake/eblake.gpg
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFXHgj84KuGfSFAYARAv79AKCpF2G33aq/b5nQDW59ew2rrHxjkACglLfE
QqztUtThhKVbDcURi6RBrKk=
=8Ae/
-----END PGP SIGNATURE-----

Index: doc/m4.texinfo
===================================================================
RCS file: /sources/m4/m4/doc/m4.texinfo,v
retrieving revision 1.79
diff -u -p -r1.79 m4.texinfo
--- doc/m4.texinfo      14 Nov 2006 05:58:01 -0000      1.79
+++ doc/m4.texinfo      16 Nov 2006 14:37:30 -0000
@@ -4236,36 +4236,45 @@ in which they were saved (LIFO---last in
 @node Include
 @section Including named files
 
address@hidden {Builtin (m4)} include (@var{filename})
address@hidden {Builtin (m4)} sinclude (@var{filename})
-These two builtin macros are for including files.  Both cause the file
-named @var{filename} to be read by @code{m4}.  When the end of the file
-is reached, input is resumed from the previous input file.
+There are two builtin macros in @code{m4} for including files:
+
address@hidden {Builtin (m4)} include (@var{file})
address@hidden {Builtin (m4)} sinclude (@var{file})
+Both macros cause the file named @var{file} to be read by
address@hidden  When the end of the file is reached, input is resumed from
+the previous input file.
 
 The expansion of @code{include} and @code{sinclude} is therefore the
-contents of @var{filename}.
+contents of @var{file}.
 
-It is an error for an @code{include}d file not to exist.  If you do
-not want error messages about non-existent files, @code{sinclude} can
-be used to include a file, if it exists, expanding to nothing if it
-does not.
+If @var{file} does not exist (or cannot be read), the expansion is void,
+and @code{include} will fail with an error while @code{sinclude} is
+silent.  The empty string counts as a file that does not exist.
 
-The builtin macros @code{include} and @code{sinclude} are recognized
-only when given arguments.
+The macros @code{include} and @code{sinclude} are recognized only with
+parameters.
 @end deffn
 
 @comment status: 1
 @example
 include(`n')
address@hidden
 @error{}m4:stdin:1: include: cannot open `n': No such file or directory
address@hidden
+include()
address@hidden:stdin:2: include: cannot open `': No such file or directory
address@hidden
 sinclude(`n')
 @result{}
+sinclude()
address@hidden
 @end example
 
-This section uses the file
address@hidden@value{VERSION}/@/examples/@/incl.m4} included in the
address@hidden M4 package:
+This section uses the @option{--include} command-line option (or
address@hidden, @pxref{Preprocessor features, , Invoking m4}) to grab
+files from the @address@hidden/@/examples}
+directory shipped as part of the @acronym{GNU} @code{m4} package.  The
+file @address@hidden/@/examples/@/incl.m4} in the distribution
+contains the lines:
 
 @comment ignore
 @example
@@ -4309,30 +4318,34 @@ This is `bar':  >>bar<<
 @end example
 
 This use of @code{include} is not trivial, though, as files can contain
-quotes, commas and parentheses, which can interfere with the way the
address@hidden parser works.
+quotes, commas, and parentheses, which can interfere with the way the
address@hidden parser works.  @acronym{GNU} @code{m4} seamlessly concatenates
+the file contents with the next character, even if the included file
+ended in the middle of a comment, string, or macro call.  These
+conditions are only treated as end of file errors if specified as input
+files on the command line.
 
-In GNU @code{m4}, an alternative method of reading files is using
address@hidden on a named file.
+In @acronym{GNU} @code{m4}, an alternative method of reading files is
+using @code{undivert} (@pxref{Undivert}) on a named file.
 
 @node Search Path
 @section Searching for include files
 
 @cindex search path for included files
 @cindex included files, search path for
address@hidden GNU extensions
-GNU @code{m4} allows included files to be found in other directories
address@hidden @acronym{GNU} extensions
address@hidden @code{m4} allows included files to be found in other directories
 than the current working directory.
 
-If the @option{--prepend-include} or @option{-B} option was provided
-(@pxref{Preprocessor features, , Invoking m4}), those directories are
-searched first, in reverse
-order that those options were listed on the command line.  Then
address@hidden looks in the current working directory.  Next comes the
-directories specified with the @option{--include} or @option{-I} option
-will be searched, in the order found on the command line.  Finally, if
-the @env{M4PATH} environment variable is set, it is expected to contain
-a colon-separated list of directories, which will be searched in order.
+If the @option{--prepend-include} or @option{-B} command-line option was
+provided (@pxref{Preprocessor features, , Invoking m4}), those
+directories are searched first, in reverse order that those options were
+listed on the command line.  Then @code{m4} looks in the current working
+directory.  Next comes the directories specified with the
address@hidden or @option{-I} option, in the order found on the
+command line.  Finally, if the @env{M4PATH} environment variable is set,
+it is expected to contain a colon-separated list of directories, which
+will be searched in order.
 
 If the automatic search for include-files causes trouble, the @samp{p}
 debug flag (@pxref{Debugmode}) can help isolate the problem.
@@ -4347,7 +4360,7 @@ time.
 
 Numbered diversions are counted from 0 upwards, diversion number 0
 being the normal output stream.  The number of simultaneous diversions
-is limited mainly by the memory used to describe them, because GNU
+is limited mainly by the memory used to describe them, because @acronym{GNU}
 @code{m4} tries to keep diversions in memory.  However, there is a
 limit to the overall memory usable by all diversions taken altogether
 (512K, currently).  When this maximum is about to be exceeded,
@@ -4355,11 +4368,16 @@ a temporary file is opened to receive th
 diversion still in memory, freeing this memory for other diversions.
 When creating the temporary file, @code{m4} honors the value of the
 environment variable @env{TMPDIR}, and falls back to @file{/tmp}.
-So, it is theoretically possible that the number of diversions be
-limited by the number of available file descriptors.
+So, it is theoretically possible that the number and aggregate size of
+diversions is limited only by available disk space.
 
address@hidden FIXME: need some explanation here why this is a useful feature, 
not
address@hidden just how you use it.
+Diversions make it possible to generate output in a different order than
+the input was read.  It is possible to implement topological sorting
+dependencies.  For example, @acronym{GNU} Autoconf makes use of
+diversions under the hood to ensure that the expansion of a prerequisite
+macro appears in the output prior to the expansion of a dependent macro,
+regardless of which order the two macros were invoked in the user's
+input file.
 
 @menu
 * Divert::                      Diverting output
@@ -4374,11 +4392,12 @@ limited by the number of available file 
 @cindex diverting output to files
 @cindex output, diverting to files
 @cindex files, diverting output to
address@hidden {Builtin (m4)} divert
address@hidden {Builtin (m4)} divert (@ovar{number})
-Output is diverted using @code{divert} where @var{number} is the
-diversion to be used.  If @var{number} is left out, it is assumed to be
-zero.
+Output is diverted using @code{divert}:
+
address@hidden {Builtin (m4)} divert (@dvar{number, 0})
+The current diversion is changed to @var{number}.  If @var{number} is left
+out or empty, it is assumed to be zero.  If @var{number} cannot be
+parsed, the diversion is unchanged.
 
 The expansion of @code{divert} is void.
 @end deffn
@@ -4387,7 +4406,7 @@ When all the @code{m4} input will have b
 diversions are automatically undiverted, in numerical order.
 
 @example
-divert(1)
+divert(`1')
 This text is diverted.
 divert
 @result{}
@@ -4399,22 +4418,67 @@ This text is not diverted.
 @end example
 
 Several calls of @code{divert} with the same argument do not overwrite
-the previous diverted text, but append to it.
+the previous diverted text, but append to it.  Diversions are printed
+after any wrapped text is expanded.
 
-If output is diverted to a non-existent diversion, it is simply
-discarded.  This can be used to suppress unwanted output.  A common
-example of unwanted output is the trailing newlines after macro
-definitions.  Here is how to avoid them.
address@hidden
+define(`text', `TEXT')
address@hidden
+divert(`1')`diverted text.'
+divert
address@hidden
+m4wrap(`Wrapped text preceeds ')
address@hidden
+^D
address@hidden TEXT preceeds diverted text.
address@hidden example
+
+If output is diverted to a negative diversion, it is simply discarded.
+This can be used to suppress unwanted output.  A common example of
+unwanted output is the trailing newlines after macro definitions.  Here
+is a common programming idiom in @code{m4} for avoiding them.
 
 @example
-divert(-1)
+divert(`-1')
 define(`foo', `Macro `foo'.')
 define(`bar', `Macro `bar'.')
 divert
 @result{}
 @end example
 
-This is a common programming idiom in @code{m4}.
address@hidden @acronym{GNU} extensions
+Traditional implementations only supported ten diversions.  But as a
address@hidden extension, diversion numbers can be as large as positive
+integers will allow, rather than treating a multi-digit diversion number
+as a request to discard text.
+
address@hidden
+divert(eval(`1<<28'))world
+divert(`2')hello
+^D
address@hidden
address@hidden
address@hidden example
+
+Note that @code{divert} is an English word, but also an active macro
+without arguments.  When processing plain text, the word might appear in
+normal text and be unintentionally swallowed as a macro invocation.  One
+way to avoid this is to use the @option{-P} option to rename all
+builtins (@pxref{Operation modes, , Invoking m4}).  Another is to write
+a wrapper that requires a parameter to be recognized.
+
address@hidden
+We decided to divert the stream for irrigation.
address@hidden decided to  the stream for irrigation.
+define(`divert', `ifelse(`$#', `0', ``$0'', `builtin(`$0', $@@)')')
address@hidden
+divert(`-1')
+Ignored text.
+divert(`0')
address@hidden
+We decided to divert the stream for irrigation.
address@hidden decided to divert the stream for irrigation.
address@hidden example
 
 @node Undivert
 @section Undiverting output
Index: tests/builtins.at
===================================================================
RCS file: /sources/m4/m4/tests/builtins.at,v
retrieving revision 1.34
diff -u -p -r1.34 builtins.at
--- tests/builtins.at   14 Nov 2006 05:58:01 -0000      1.34
+++ tests/builtins.at   16 Nov 2006 14:37:30 -0000
@@ -161,15 +161,6 @@ AT_CHECK_M4([divert.m4], 0,
 Text diverted a second time.
 ]])
 
-dnl Handle large diversions.  In 1.4.7, this caused core dumps; in 1.4.8,
-dnl it exhausted memory on 32-bit platforms.
-AT_DATA([[in]], [[divert(eval(`1<<28'))world
-divert(`2')hello
-]])
-AT_CHECK_M4([in], [0], [[hello
-world
-]])
-
 AT_CLEANUP
 
 
@@ -279,7 +270,7 @@ AT_CHECK([cat trace], [0], [[m4trace: -1
 dnl Ensure that esyscmd does not inherit any unnecessary fds from diversions.
 AT_DATA([in.m4], [M4_ONE_MEG_DEFN[divert(`1')f
 world
-divert`'esyscmd(`echo hi >&3')
+esyscmd(`echo hi >&3')divert
 hello
 ]])
 AT_CHECK_M4([3>&-], [0], [stdout], [experr], [in.m4])
@@ -865,7 +856,7 @@ AT_CHECK([cat trace], [0], [[m4trace: -1
 dnl Ensure that syscmd does not inherit any unnecessary fds from diversions.
 AT_DATA([in.m4], [M4_ONE_MEG_DEFN[divert(`1')f
 world
-divert`'syscmd(`echo hi >&3')
+syscmd(`echo hi >&3')divert
 hello
 ]])
 AT_CHECK_M4([3>&-], [0], [stdout], [experr], [in.m4])
Index: tests/others.at
===================================================================
RCS file: /sources/m4/m4/tests/others.at,v
retrieving revision 1.26
diff -u -p -r1.26 others.at
--- tests/others.at     19 Oct 2006 23:13:45 -0000      1.26
+++ tests/others.at     16 Nov 2006 14:37:30 -0000
@@ -362,7 +362,7 @@ AT_CHECK([cat trace], [0], [[m4trace: -1
 
 dnl spilled diversion file must not collide with closed stderr
 AT_DATA([in.m4], [M4_ONE_MEG_DEFN[divert(1)f
-and
+and`'dnl(not)
 divert
 hello`'dnl(world)
 undivert
@@ -510,10 +510,19 @@ AT_CHECK_M4([>&-], [0], [], [[hello
 ]], [in.m4])
 
 dnl syscmd inherits closed stdout
-AT_CHECK([cat >&- && { echo "skipping: can't detect closed stdout"; exit 77; 
}],
+AT_DATA([hi], [[hi
+]])
+AT_CHECK([cat hi >&- && { echo "skipping: can't detect closed stdout"; exit 
77; }],
 [1], [], [stderr])
 mv stderr experr
-AT_DATA([in.m4], [[syscmd(`cat')dnl
+AT_DATA([in.m4], [[syscmd(`cat hi')dnl
+]])
+AT_CHECK_M4([>&-], [0], [], [experr], [in.m4])
+
+dnl spilled diversion file must not collide with closed stdout
+AT_DATA([in.m4], [M4_ONE_MEG_DEFN[divert(1)f
+syscmd(`cat hi')
+divert(`-1')undivert
 ]])
 AT_CHECK_M4([>&-], [0], [], [experr], [in.m4])
 

Index: doc/m4.texinfo
===================================================================
RCS file: /sources/m4/m4/doc/m4.texinfo,v
retrieving revision 1.1.1.1.2.102
diff -u -p -r1.1.1.1.2.102 m4.texinfo
--- doc/m4.texinfo      13 Nov 2006 23:25:08 -0000      1.1.1.1.2.102
+++ doc/m4.texinfo      16 Nov 2006 14:39:12 -0000
@@ -3484,11 +3484,11 @@ parameters.
 
 @example
 include(`none')
address@hidden
 @error{}m4:stdin:1: cannot open `none': No such file or directory
-include()
 @result{}
+include()
 @error{}m4:stdin:2: cannot open `': No such file or directory
address@hidden
 sinclude(`none')
 @result{}
 sinclude()
@@ -3501,13 +3501,14 @@ pointing to the @address@hidden
 directory shipped as part of the @acronym{GNU} @code{m4} package.  The
 file @address@hidden/@/examples/@/incl.m4} in the distribution
 contains the lines:
+
 @comment ignore
 @example
 Include file start
 foo
 Include file end
 @end example
address@hidden
+
 Normally file inclusion is used to insert the contents of a file
 into the input stream.  The contents of the file will be read by
 @code{m4} and macro calls in the file will be expanded:
@@ -3540,11 +3541,13 @@ This is `bar':  >>bar<<
 This use of @code{include} is not trivial, though, as files can contain
 quotes, commas, and parentheses, which can interfere with the way the
 @code{m4} parser works.  @acronym{GNU} @code{m4} seamlessly concatenates
-the file
-contents with the next character, even if the included file ended in
-the middle of a comment, string, or macro call.  These conditions are
-only treated as end of file errors if specified as input files on the
-command line.
+the file contents with the next character, even if the included file
+ended in the middle of a comment, string, or macro call.  These
+conditions are only treated as end of file errors if specified as input
+files on the command line.
+
+In @acronym{GNU} @code{m4}, an alternative method of reading files is
+using @code{undivert} (@pxref{Undivert}) on a named file.
 
 @node Search Path
 @section Searching for include files
@@ -3555,13 +3558,15 @@ command line.
 @acronym{GNU} @code{m4} allows included files to be found in other directories
 than the current working directory.
 
-If a file is not found in the current working directory, and the file
-name is not absolute, the file will be looked for in a specified search
-path.  First, the directories specified with the @option{-I} option will
-be searched, in the order found on the command line (@pxref{Preprocessor
-features, , Invoking m4}).  Second, if the
address@hidden environment variable is set, it is expected to contain a
-colon-separated list of directories, which will be searched in order.
+If the @option{--prepend-include} or @option{-B} command-line option was
+provided (@pxref{Preprocessor features, , Invoking m4}), those
+directories are searched first, in reverse order that those options were
+listed on the command line.  Then @code{m4} looks in the current working
+directory.  Next comes the directories specified with the
address@hidden or @option{-I} option, in the order found on the
+command line.  Finally, if the @env{M4PATH} environment variable is set,
+it is expected to contain a colon-separated list of directories, which
+will be searched in order.
 
 If the automatic search for include-files causes trouble, the @samp{p}
 debug flag (@pxref{Debug Levels}) can help isolate the problem.
@@ -3654,8 +3659,13 @@ m4exit
 @end example
 @end ignore
 
address@hidden FIXME: need some explanation here why this is a useful feature, 
not
address@hidden just how you use it.
+Diversions make it possible to generate output in a different order than
+the input was read.  It is possible to implement topological sorting
+dependencies.  For example, @acronym{GNU} Autoconf makes use of
+diversions under the hood to ensure that the expansion of a prerequisite
+macro appears in the output prior to the expansion of a dependent macro,
+regardless of which order the two macros were invoked in the user's
+input file.
 
 @menu
 * Divert::                      Diverting output
@@ -3750,9 +3760,9 @@ We decided to divert the stream for irri
 @result{}We decided to  the stream for irrigation.
 define(`divert', `ifelse(`$#', `0', ``$0'', `builtin(`$0', $@@)')')
 @result{}
-divert(-1)
+divert(`-1')
 Ignored text.
-divert(0)
+divert(`0')
 @result{}
 We decided to divert the stream for irrigation.
 @result{}We decided to divert the stream for irrigation.