Mercurial > emacs
view configure.in @ 18212:3864648e9b4f
(bibtex-mode-map): Changed the binding of the C-TAB
key, such that XEmacs will understand it, too.
(bibtex-format-entry, bibtex-end-of-entry): Give specific
error message if not on valid BibTeX entry.
(bibtex-field-string-quoted): Small bug fix. Allow
backslash followed by newline.
(bibtex-reposition-window, bibtex-mark-entry): Two new
functions, bound to M-C-l and M-C-h, respectively.
(bibtex-reformat-previous-options, bibtex-reformat-previous-labels):
New internal variables used by bibtex-reformat.
(bibtex-clean-entry-hook): New hook to be called after
entry has been cleaned.
(bibtex-clean-entry): Run the hook bibtex-clean-entry-hook.
(bibtex-entry-field-alist): Small bug fix.
(bibtex-autokey-title-terminators): Made -- a terminator instead
of ---.
(bibtex-font-lock-keywords): Don't treat ALT prefixed entries as
comments.
(bibtex-entry): Fixed parameter list. This function is not
intended to be called with required and optional fields as
optional arguments anymore.
(bibtex-generate-autokey): Now split into various
small functions.
(bibtex-autokey-names-stretch, bibtex-autokey-additional-names):
New variables used by bibtex-generate-autokey.
(bibtex-autokey-get-namefield, bibtex-autokey-get-names)
(bibtex-autokey-demangle-name, bibtex-autokey-get-namelist)
(bibtex-autokey-get-yearfield, bibtex-autokey-get-titlestring)
(bibtex-autokey-get-titles, bibtex-autokey-get-titlelist):
New helper functions for bibtex-generate-autokey.
(bibtex-submit-bug-report): Report all variables.
(bibtex-contline-indentation): New user option.
(bibtex-entry-offset): Renamed from bibtex-entry-indentation.
(bibtex-entry-field-alist): Used different order for
some fields (as documented in btxdoc.tex). Changed one of the
comment strings.
(bibtex-mode-hook, bibtex-add-entry-hook): Add var doc.
(bibtex-autokey-before-presentation-hook): New variable to be
called before autokey presentation.
(bibtex-generate-autokey):
Call bibtex-autokey-before-presentation-hook. Doc fix.
(bibtex-reference-key): Reincluded parentheses.
Parentheses should be disallowed only in field constants.
(bibtex-autokey-transcriptions): Fixed bug (two
entries for `\o' while `\oe' entry was missing).
(bibtex-entry-indentation): New variable to determine
the indentation of all entries.
(bibtex-move-outside-of-entry): Use `skip-chars-forward' instead
of `re-search-forward'.
(bibtex-beginning-of-first-entry, bibtex-beginning-of-last-entry):
Renamed from beginning-of-first-bibtex-entry and
beginning-of-last-bibtex-entry. Go to beginning of line, return point.
(bibtex-do-auto-fill, bibtex-make-field, bibtex-entry)
(bibtex-String, bibtex-Preamble): Respect `bibtex-entry-indentation'.
(bibtex-beginning-of-entry, bibtex-end-of-entry): Make it work
with indented entries.
(bibtex-count-entries, bibtex-sort-buffer, bibtex-validate,
bibtex-kill-entry, bibtex-reformat): Use return value from
bibtex-beginning-of-first-entry.
(bibtex-clean-entry): Use `bibtex-reference-maybe-empty-head'
instead of a fixed string.
(bibtex-beginning-of-entry, bibtex-end-of-entry): Now
return point if called from a program.
(bibtex-enclosing-field, bibtex-format-entry,
bibtex-generate-autokey, bibtex-parse-keys, bibtex-mode,
bibtex-ispell-entry, bibtex-narrow-to-entry, bibtex-sort-buffer,
bibtex-find-entry-location, bibtex-validate, bibtex-clean-entry,
bibtex-fill-entry): Use new return values of these functions.
(bibtex-ispell-abstract): Bug fix (inherently by the change to
bibtex-end-of-entry).
(bibtex-field-history): New variable for history
buffer of field name reading.
(bibtex-make-field): Use completion. Object to completion are all
standard fields defined for the current entry. Bound to `C-c C-f'.
(bibtex-mode): Set `comment-start' and `comment-start-skip' to
"@Comment ", `comment-column' to 0.
(bibtex-autokey-transcriptions): New variable.
(bibtex-autokey-name-change-strings)
(bibtex-autokey-titleword-change-strings):
Use bibtex-autokey-transcriptions as default value.
(bibtex-find-entry-location): Handle empty buffer.
(bibtex-entry-delimiters): New variable to determine
if entries shall be delimited by braces or parentheses.
(bibtex-entry-left-delimiter, bibtex-entry-right-delimiter): New
helper functions.
(bibtex-entry, bibtex-String, bibtex-Preamble): Respect
`bibtex-entry-delimiters'.
(bibtex-entry-format): Doc fix.
(bibtex-reference-key, bibtex-field-const): Removed parentheses
from allowed characters.
(bibtex-end-of-entry): Better handling of incorrect preambles.
(bibtex-validate): Small change to avoid reparsing of
errors, if you enter `compile-goto-error' in compilation buffer.
(bibtex-progress-message): New function to show
progress of some long-lasting functions in BibTeX mode by
indicating percentage of done work.
(bibtex-progress-lastperc, bibtex-progress-lastmes)
(bibtex-progress-interval):
New internal variables for `bibtex-progress-message'.
(bibtex-parse-keys, bibtex-reformat, bibtex-validate): Use new
function `bibtex-progress-message'.
(current-line): New helper function to calculate current
linenumber. Something like this should really be defined somewhere
else in Emacs.
(bibtex-validate): Changed to show all errors in buffer in a
`compilation mode' buffer. If there are syntax errors, it aborts
after the syntax check, since higher-level check functions rely on
the syntactical correctness of buffer. If called from another lisp
function (as `bibtex-convert-alien') the return value shows
whether validating has been successful. Fixed bug which made
checking for absent required fields work incorrectly.
(bibtex-parse-keys-timeout): Set to a value reasonable higher
(twice as high) than `lazy-lock-stealth-time'.
(bibtex-member-of-regexp, assoc-of-regexp): Small cosmetic changes.
(bibtex-buffer-last-parsed-tick): Renamed from
bibtex-buffer-last-parsed-for-keys-tick and made it really
buffer-local (bug fix).
(bibtex-parse-keys): Make it use bibtex-buffer-last-parsed-tick.
(bibtex-parse-buffers-stealthily): New function which parses all
BibTeX buffers if emacs has been idle an efficient amount of time.
(bibtex-parse-idle-timer): New variable which stores whether idle
timer for parsing already is installed.
(bibtex-parse-keys): Make it callable verbosely. Returns now nil
if it has been aborted.
(bibtex-mode): Run the new function bibtex-parse-buffers-stealthily.
(bibtex-generate-autokey): Changed the name part
generation (bugfix). This function handles now correctly all three
forms of BibTeX names, "First von Last", "von Last, First", "von
Last, Jr, First". In every case the "Last" part is correctly
extracted. If the "Last" part consists of more than one token only
the first is used. Name fields spread over more than one line are
no problem anymore.
(bibtex-entry-format): Changed default value to
exclude 'page-dashes. Modified documentation.
(bibtex-autokey-name-change-strings)
(bibtex-autokey-titleword-abbrevs)
(bibtex-autokey-titleword-change-strings, bibtex-entry)
(bibtex-validate): Doc fixes.
(bibtex-mode-map): Bound `C-c $' to bibtex-ispell-abstract.
(bibtex-generate-autokey): Changed documentation. Small
modification in calculating title field.
(bibtex-mode): Included bibtex-ispell-entry into the list of
`interesting' functions.
(bibtex-kill-field): Bug fix (killing of first field in entry
yielded error).
(bibtex-string-file-path): New variable which defines
the path to search for bibtex-string-files. It defaults to
contents of environment variable BIBINPUTS.
(bibtex-mode): Use this variable.
(bibtex-next-field, bibtex-find-text): Will now work
with string entries as well.
(bibtex-mode-map): bibtex-complete-key wasn't bound
correctly.
(bibtex-complete): Fixed bug (used string entries defined in
buffer as object to completion).
(Menu): Use easymenu. More menu items for
`BibTeX-Edit' menu. Use nested menus.
(bibtex-field-kill-ring-max)
(bibtex-entry-kill-ring-max): Two new variables substituting
bibtex-kill-ring-max.
(bibtex-field-kill-ring, bibtex-entry-kill-ring): Two new
variables substituting bibtex-kill-ring.
(bibtex-field-kill-ring-yank-pointer)
(bibtex-entry-kill-ring-yank-pointer): Two new variables
substituting bibtex-kill-ring-yank-pointer.
(bibtex-last-kill-command): New variable keeping the type of the
last kill command.
(bibtex-insert-current-kill): Clean distinction between the two
BibTeX kill rings.
(bibtex-kill-field): Use new variable bibtex-field-kill-ring.
(bibtex-kill-entry): Use new variable bibtex-entry-kill-ring.
(bibtex-kill-ring, bibtex-kill-ring-yank-pointer): New
internal variables like kill-ring and kill-ring-yank-pointer, but
bibtex-kill-ring holds fields or complete reference entries
instead of raw strings.
(bibtex-kill-ring-max): New user option similar to kill-ring-max.
(bibtex-kill-field): Renamed from bibtex-delete-field again. It
now supports the new variable bibtex-kill-ring.
(bibtex-copy-field-as-kill, bibtex-kill-entry)
(bibtex-copy-entry-as-kill, bibtex-yank, bibtex-yank-pop): New
interactive functions, which work on the bibtex-kill-ring
variable.
(bibtex-insert-current-kill): New helper function to insert
contents of bibtex-kill-ring in an appropriate way.
(bibtex-make-field): New optional argument to tell the function if
it was called by bibtex-yank. It operates slightly different then.
(bibtex-find-text): New optional argument to tell the function if
it was called by bibtex-make-field. It moves in this case to the
end of the key if it was called on the reference head line. If it
is called interactively on the head line, it works in an
appropriate manner, too.
(bibtex-enclosing-field): New optional argument to tell
bibtex-enclosing-field not to print an error message if enclosing
field isn't found, but to return nil in this case (t is returned
otherwise). This is used by bibtex-find-text such that no error
message is given if bibtex-find-text is called interactively in
the key line of an entry.
(bibtex-autokey-year-use-crossref-entry): New variable
to determine if crossreferenced entry should be used for autokey
generation, if year field of current entry is absent.
(bibtex-generate-autokey): Use this new variable.
(bibtex-include-OPTannote): Deleted (is set in
bibtex-user-optional-fields).
(bibtex-entry, bibtex-print-help-message): Removed support for
bibtex-include-OPTannote.
(bibtex-entry-format): New constant
`inherit-booktitle' allowed.
(bibtex-mode): Set value for
font-lock-mark-block-function.
(bibtex-font-lock-keywords): Changed to distinguish
optional from ordinary fields.
(bibtex-format-entry, bibtex-print-help-message)
(bibtex-remove-OPT-or-ALT, bibtex-pop): Used simpler regexps.
(bibtex-delete-field): Changed from
bibtex-delete-optional-or-alternative-field. Deletes now mandatory
fields as well.
(bibtex-mode): Changed documentation.
(bibtex-entry-type-history, bibtex-key-history): New
variables to use own histories in BibTeX buffers.
(bibtex-entry, bibtex-clean-entry, bibtex-String): Use these new
variables.
(bibtex-entry, bibtex-make-field): A function can now
be used to generate a fields init string.
(bibtex-include-OPTkey, bibtex-include-OPTannote)
(bibtex-entry-field-alist): Changed documentation accordingly.
(bibtex-mode): bibtex-parse-keys on start of mode is
now abortable, too.
(bibtex-entry, bibtex-complete-key, bibtex-String): If bibtex-keys
isn't set correctly due to abortion of bibtex-parse-keys, this
function is called now non-abortable.
(bibtex-entry-field-alist): Small change in comments.
(bibtex-find-entry-location): Bug fix: Insertion into
completely empty buffer didn't work.
(bibtex-user-optional-fields): Renamed from
bibtex-mode-user-optional-fields.
(bibtex-submit-bug-report, bibtex-entry, bibtex-print-help-message):
Use bibtex-user-optional-fields.
(bibtex-remove-delimiters): Bug fix: Only remove
delimiting braces and not those inside fields.
(skip-to-valid-bibtex-entry, bibtex-parse-keys)
(bibtex-end-of-entry, bibtex-validate, bibtex-reformat): Calculate
complex regexps outside of loops.
(bibtex-mode): Changed documentation on how to convert third party
buffers.
(bibtex-convert-alien): New function to convert a
buffer not created by BibTeX mode to a format, whichs enables all
features of BibTeX mode.
(bibtex-mode): Small bug fix for call to bibtex-parse-keys.
(bibtex-mode): Bug fix for using bibtex-parse-keys.
(bibtex-mode): Used other policy to initiate the first
call of bibtex-parse-keys. This avoids unnecessary double call if
Font Lock mode is chosen for buffer at startup.
(bibtex-String, bibtex-Preamble): Renamed from
bibtex-string and bibtex-preamble.
(bibtex-String): If bibtex-maintain-sorted-entries and
bibtex-sort-ignore-string-entries are both non-nil, read string
key from minibuffer (with completion) and insert entry at correct
location (as for normal entries).
(bibtex-autokey-titleword-first-ignore)
(bibtex-autokey-titleword-abbrevs): Changed documentation: case of
regexps doesn't matter anymore.
(bibtex-field-const, bibtex-reference-key): Simplified to not
contain uppercase letters.
(member-of-regexp, assoc-of-regexp): Ignore case of regexp.
(map-bibtex-entries): Call function not for every syntactical correct
entry, but only for entries with known type.
(map-bibtex-entries, skip-to-valid-bibtex-entry)
(bibtex-flash-head, bibtex-enclosing-field)
(bibtex-enclosing-reference-maybe-empty-head, bibtex-format-entry)
(bibtex-autokey-change, bibtex-generate-autokey, bibtex-parse-keys)
(bibtex-ispell-abstract, bibtex-sort-buffer)
(bibtex-find-entry-location, bibtex-validate, bibtex-remove-delimiters)
(bibtex-delete-optional-or-alternative-field, bibtex-pop)
(bibtex-clean-entry, bibtex-print-help-message): Make them work
regardless of user's value of case-fold-search.
(bibtex-generate-autokey): Use bibtex-cfield for generating the
year field regexp.
(bibtex-parse-keys): Only gather keys of known (and not of
syntactical correct) entries.
(bibtex-end-of-entry): Only report an "unknown entry" message if
called interactively.
(bibtex-sort-ignore-string-entries): Renamed back from
bibtex-sort-ignore-string-and-preamble. Of course, preambles are
always ignored, since they have no key at all.
(bibtex-string): Slightly less complex regexp.
(skip-to-valid-bibtex-entry): New helper function to skip forward
(or backward) to beginning of next syntactical correct known
BibTeX entry, if not already there. Respects
bibtex-sort-ignore-string-entries.
(map-bibtex-entries): Bug fix: It wasn't called for string entries
even if bibtex-sort-ignore-string-entries was nil.
(beginning-of-last-bibtex-entry): New helper function to go to
last entry in buffer.
(bibtex-end-of-entry): Bug fix: Now works with string and preamble
entries as well.
(bibtex-sort-buffer): Renamed from bibtex-sort-entries. Simplified
by using new function skip-to-valid-bibtex-entry. Now only known
entries are checked.
(bibtex-find-entry-location): Simplified by using new functions
skip-to-valid-bibtex-entry and beginning-of-last-bibtex-entry.
Only known entries are used to determine location.
(bibtex-validate): Now checks string entries, too.
(bibtex-move-outside-of-entry): Don't use forward-paragraph, but
bibtex-end-of-entry.
(bibtex-end-of-entry): Don't use forward-sexp anymore,
since this fails on entries with non-escaped souble-quotes. Use
search-bibtex-reference instead (though it is slower, it is more
reliable).
(bibtex-ispell-abstract): Use normal regexps created by
bibtex-cfield instead of special ones.
(beginning-of-first-bibtex-entry): No warning on
empty buffer.
(bibtex-validate): Bug fix. Syntactical check didn't
work, since due to a bug all entries were simply skipped.
(bibtex-mode): Doc fix.
(bibtex-delete-optional-or-alternative-field): Renamed from
bibtex-kill-optional-or-alternative-field.
(bibtex-delete-optional-or-alternative-field, bibtex-empty-field):
Use delete-region, not kill-region.
(bibtex-clean-entry): New second argument called-by-reformat
indicates if bibtex-clean-entry was called by reformat. Don't try
to find correct buffer position for newly generated key and don't
call bibtex-parse-keys in this case.
(map-bibtex-entries): Bugfix (missed first entry in
buffer, died on entries with `@' in other than first column).
(beginning-of-first-bibtex-entry, bibtex-format-entry)
(bibtex-beginning-of-entry, bibtex-validate, bibtex-clean-entry):
Changed to allow BibTeX entries to start in a column different
from 1 (but still for speed reasons only whitespace is allowed
prior to the `@' on the same line.
(map-bibtex-entries): Call it for known BibTeX entries
only. This allows entries as @Comment{...} which follow no
specific structure without breaking the validation functions.
(bibtex-validate): Check syntactical structure for known entries
only.
(bibtex-autokey-abbrev): Sped up and changed to allow
a length of zero.
(bibtex-entry-format): Remove option `month-strings'.
(bibtex-validate): If given a prefix argument, check for
questionable month fields.
(bibtex-generate-autokey): Use normal regexps created
by bibtex-cfield instead of special ones.
(bibtex-hide-entry-bodies): Sped up by using subst-char-in-region
instead of using replace-regexp or replace-match.
(bibtex-find-entry-location): A bug had been introduced by using
search-bibtex-reference instead of re-search-forward (fixed).
(bibtex-field-delimiters): Renamed from
bibtex-field-delimiter.
(bibtex-entry-format): Constant empty-opts renamed to
empty-opts-or-alts.
(bibtex-remove-delimiters): Renamed from
bibtex-remove-double-quotes-or-braces.
(bibtex-reformat): New function.
(bibtex-fill-entry): New function to refill entry.
(bibtex-mode-map): Defined key for bibtex-fill-entry.
(bibtex-field-delimiter): Substitutes variables
bibtex-field-left-delimiter and bibtex-field-right-delimiter.
(bibtex-field-left-delimiter, bibtex-field-right-delimiter): New
helper functions.
(bibtex-make-field, bibtex-pop): Use new variable
bibtex-field-delimiter.
(bibtex-empty-field, bibtex-string): Use new functions
bibtex-field-left-delimiter and bibtex-field-right-delimiter.
(bibtex-predefined-month-strings): New variable.
(bibtex-predefined-strings): Use bibtex-predefined-month-strings.
(bibtex-submit-bug-report): Use new variables
bibtex-field-delimiter and bibtex-predefined-month-strings.
(bibtex-entry-format): Substitutes variable
bibtex-clean-entry-zap-empty-opts-or-alts. Various types of
formatting options are available (see variable documentation).
(bibtex-format-entry): New function doing the formatting of entries.
Taken from bibtex-clean-entry and enhanced to support new variable
bibtex-entry-format.
(delete-whitespace): New helper function used by bibtex-format-entry.
(bibtex-clean-entry): Call new function bibtex-format-entry.
(bibtex-submit-bug-report): Use bibtex-entry-format instead of
bibtex-clean-entry-zap-empty-opts-or-alts.
(bibtex-do-auto-fill): New function to perform auto-filling in
BibTeX mode. In fact, this is the old function
bibtex-auto-fill-function.
(bibtex-mode): Don't set fill-prefix anymore, but use new function
bibtex-do-auto-fill.
(bibtex-find-entry-location): Fixed bug (when
bibtex-maintain-sorted-entries was non-nil, an entry with a key
greater than all other keys wasn't inserted in the correct place).
(bibtex-mode): Don't use bibtex-auto-fill-function
anymore, but use directly variable fill-prefix.
(bibtex-find-entry-location): Fixed bug (on duplicate
keys, point must move to beginning of entry, so that bibtex-entry
works correctly).
(bibtex-complete): Fixed bug (parameter string-list
was mistakenly altered by the function itself).
(bibtex-mode-map): Bind bibtex-complete-key to C-TAB.
(bibtex-validate): Renamed from bibtex-validate-buffer
since it can acts on region if active. Use search-bibtex-reference.
(search-bibtex-reference): New function to be used in places where
prior a re-search-{forward|backward} for bibtex-reference or
something alike was used. This function is necessary due to the
new limit of failure stack size in 19.32 and above.
(bibtex-enclosing-reference-maybe-empty-head)
(bibtex-sort-entries, bibtex-find-entry-location, bibtex-validate):
Use new function search-bibtex-reference.
(bibtex-pop, bibtex-clean-entry): Small change due to change in
bibtex-enclosing-reference-maybe-empty-head.
(bibtex-reference-infix, bibtex-reference-postfix): New constants
necessary due to splitting bibtex-reference.
(bibtex-reference): Deleted.
(bibtex-type-in-reference, skip-whitespace-and-comments): Deleted.
(bibtex-mode): Don't turn auto-fill-mode on. Use new
variable normal-auto-fill-function.
(bibtex-field-string): Simplified.
(bibtex-mode-syntax-table): Changed syntax of
double-quote back to quote syntax.
(bibtex-complete): New generic function for interface
functions bibtex-complete-string and bibtex-complete-key.
(bibtex-complete-key): New function.
(bibtex-sort-ignore-string-and-preamble): Renamed from
bibtex-sort-ignore-string-entries.
(map-bibtex-entries): Use bibtex-sort-ignore-string-and-preamble
and ignore preamble entries as well.
(bibtex-mode, bibtex-submit-bug-report, bibtex-sort-entries):
Use bibtex-sort-ignore-string-and-preamble.
(bibtex-count-entries): New function to count entries in buffer or
region.
(bibtex-comma-after-last-field): New variable to
decide if comma should be inserted at end of last field.
(bibtex-entry): Support new variable bibtex-comma-after-last-field.
(bibtex-field-indentation, bibtex-text-indentation):
Two new user options (replacing the former constants
bibtex-field-alignment and bibtex-text-alignment).
(bibtex-align-at-equal-sign): New user option.
(bibtex-auto-fill-function, bibtex-make-field)
(bibtex-remove-OPT-or-ALT, bibtex-clean-entry): Support new variables.
(bibtex-validate-buffer): Now only checks region if
mark is active. With optional argument checks if required fields
are missing, too.
(bibtex-mode): Added support for imenu.
(bibtex-entry-field-alist)
(bibtex-mode-user-optional-fields): Modified syntax to allow
preinitialization of fields.
(bibtex-make-field, bibtex-make-optional-field):
Support preinitialization of fields.
(bibtex-autokey-prefix-string)
(bibtex-autokey-preserve-case): New variables to support
bibtex-generate-autokey.
(bibtex-generate-autokey): Use new variables.
(bibtex-field-const, bibtex-reference-type)
(bibtex-reference-key): Changed to match the (according to Oren
Patashnik) allowed characters.
(bibtex-clean-entry-zap-empty-opts-or-alts): Renamed
from bibtex-clean-entry-zap-empty-opts.
(bibtex-entry-field-alist): Slightly modified syntax to support
alternative fields needed for Book and InBook references.
(bibtex-font-lock-keywords, bibtex-print-help-message)
(bibtex-make-field, bibtex-pop, bibtex-clean-entry):
Support ALT prefixed entries.
(bibtex-mode): Documented new ALT prefixed fields.
(bibtex-make-optional-field): Modified to give only field name as
arg to bibtex-make-field.
(bibtex-remove-OPT-or-ALT, bibtex-kill-optional-or-alternative-field):
Renamed from bibtex-remove-OPT and bibtex-kill-optional-field,
respectively. Modified to support ALT prefixes.
(bibtex-enclosing-field, bibtex-print-help-message):
Speed up things by not using bibtex-enclosing-regexp anymore.
(bibtex-enclosing-regexp): Function deleted.
| author | Richard M. Stallman <rms@gnu.org> |
|---|---|
| date | Tue, 10 Jun 1997 17:58:43 +0000 |
| parents | 42b9cdc157b1 |
| children | 9963d76b1017 |
line wrap: on
line source
dnl Autoconf script for GNU Emacs dnl To rebuild the `configure' script from this, execute the command dnl autoconf dnl in the directory containing this script. dnl dnl Copyright (C) 1994, 1995, 1996 Free Software Foundation, Inc. dnl dnl This file is part of GNU Emacs. dnl dnl GNU Emacs is free software; you can redistribute it and/or modify dnl it under the terms of the GNU General Public License as published by dnl the Free Software Foundation; either version 2, or (at your option) dnl any later version. dnl dnl GNU Emacs is distributed in the hope that it will be useful, dnl but WITHOUT ANY WARRANTY; without even the implied warranty of dnl MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the dnl GNU General Public License for more details. dnl dnl You should have received a copy of the GNU General Public License dnl along with GNU Emacs; see the file COPYING. If not, write to the dnl Free Software Foundation, Inc., 59 Temple Place - Suite 330, dnl Boston, MA 02111-1307, USA. AC_PREREQ(2.8)dnl AC_INIT(src/lisp.h) AC_CONFIG_HEADER(src/config.h:src/config.in) lispdir='${datadir}/emacs/${version}/lisp' locallisppath='${datadir}/emacs/${version}/site-lisp:'\ '${datadir}/emacs/site-lisp:${datadir}/emacs/${version}/leim' lisppath='${locallisppath}:${lispdir}' etcdir='${datadir}/emacs/${version}/etc' lockdir='${sharedstatedir}/emacs/lock' archlibdir='${libexecdir}/emacs/${version}/${configuration}' docdir='${datadir}/emacs/${version}/etc' AC_ARG_WITH(gcc, [ --with-gcc use GCC to compile Emacs]) AC_ARG_WITH(pop, [ --with-pop support POP for mail retrieval], [AC_DEFINE(MAIL_USE_POP)]) AC_ARG_WITH(kerberos, [ --with-kerberos support Kerberos-authenticated POP], [AC_DEFINE(KERBEROS)]) AC_ARG_WITH(hesiod, [ --with-hesiod support Hesiod to get the POP server host], [AC_DEFINE(HESIOD)]) dnl This should be the last --with option, because --with-x is dnl added later on when we find the path of X, and it's best to dnl keep them together visually. AC_ARG_WITH(x-toolkit, [ --with-x-toolkit=KIT use an X toolkit (KIT = yes/lucid/athena/motif/no)], [ case "${withval}" in y | ye | yes ) val=athena ;; n | no ) val=no ;; l | lu | luc | luci | lucid ) val=lucid ;; a | at | ath | athe | athen | athena ) val=athena ;; m | mo | mot | moti | motif ) val=motif ;; dnl These don't currently work. dnl o | op | ope | open | open- | open-l | open-lo \ dnl | open-loo | open-look ) val=open-look ;; * ) dnl AC_MSG_ERROR([the \`--with-x-toolkit' option is supposed to have a value dnl which is \`yes', \`no', \`lucid', \`athena', \`motif' or \`open-look'.]) AC_MSG_ERROR([\`--with-x-toolkit=$withval' is invalid\; this option's value should be \`yes', \`no', \`lucid', \`athena', or \`motif'. Currently, \`yes', \`athena' and \`lucid' are synonyms.]) ;; esac with_x_toolkit=$val ]) #### Make srcdir absolute, if it isn't already. It's important to #### avoid running the path through pwd unnecessary, since pwd can #### give you automounter prefixes, which can go away. We do all this #### so Emacs can find its files when run uninstalled. case "${srcdir}" in /* ) ;; . ) ## We may be able to use the $PWD environment variable to make this ## absolute. But sometimes PWD is inaccurate. ## Make sure CDPATH doesn't affect cd (in case PWD is relative). CDPATH= if test "${PWD}" != "" && test "`(cd ${PWD} ; sh -c pwd)`" = "`pwd`" ; then srcdir="$PWD" else srcdir="`(cd ${srcdir}; pwd)`" fi ;; * ) srcdir="`(cd ${srcdir}; pwd)`" ;; esac #### Check if the source directory already has a configured system in it. if test `pwd` != `(cd ${srcdir} && pwd)` \ && test -f "${srcdir}/src/config.h" ; then AC_MSG_WARN([The directory tree \`${srcdir}' is being used as a build directory right now; it has been configured in its own right. To configure in another directory as well, you MUST use GNU make. If you do not have GNU make, then you must now do \`make distclean' in ${srcdir}, and then run $0 again.]) changequote(, )dnl extrasub='/^VPATH[ ]*=/c\ changequote([, ])dnl vpath %.c $(srcdir)\ vpath %.h $(srcdir)\ vpath %.y $(srcdir)\ vpath %.l $(srcdir)\ vpath %.s $(srcdir)\ vpath %.in $(srcdir)\ vpath %.texi $(srcdir)' fi #### Given the configuration name, set machfile and opsysfile to the #### names of the m/*.h and s/*.h files we should use. ### Canonicalize the configuration name. AC_CANONICAL_HOST canonical=$host configuration=$host_alias changequote(, )dnl ### If you add support for a new configuration, add code to this ### switch statement to recognize your configuration name and select ### the appropriate operating system and machine description files. ### You would hope that you could choose an m/*.h file pretty much ### based on the machine portion of the configuration name, and an s- ### file based on the operating system portion. However, it turns out ### that each m/*.h file is pretty manufacturer-specific - for ### example, apollo.h, hp9000s300.h, mega68k, news.h, and tad68k are ### all 68000 machines; mips.h, pmax.h, and news-risc are all MIPS ### machines. So we basically have to have a special case for each ### configuration name. ### ### As far as handling version numbers on operating systems is ### concerned, make sure things will fail in a fixable way. If ### /etc/MACHINES doesn't say anything about version numbers, be ### prepared to handle anything reasonably. If version numbers ### matter, be sure /etc/MACHINES says something about it. ### ### Eric Raymond says we should accept strings like "sysvr4" to mean ### "System V Release 4"; he writes, "The old convention encouraged ### confusion between `system' and `release' levels'." machine='' opsys='' unported=no case "${canonical}" in ## NetBSD ports *-*-netbsd* ) opsys=netbsd case "${canonical}" in alpha-*-netbsd*) machine=alpha ;; i[3456]86-*-netbsd*) machine=intel386 ;; m68k-*-netbsd*) # This is somewhat bogus. machine=hp9000s300 ;; mips-*-netbsd*) machine=pmax ;; ns32k-*-netbsd*) machine=ns32000 ;; sparc-*-netbsd*) machine=sparc ;; vax-*-netbsd*) machine=vax ;; esac ;; ## Acorn RISCiX: arm-acorn-riscix1.1* ) machine=acorn opsys=riscix1-1 ;; arm-acorn-riscix1.2* | arm-acorn-riscix ) ## This name is riscix12 instead of riscix1.2 ## to avoid a file name conflict on MSDOS. machine=acorn opsys=riscix12 ;; ## Alliant machines ## Strictly speaking, we need the version of the alliant operating ## system to choose the right machine file, but currently the ## configuration name doesn't tell us enough to choose the right ## one; we need to give alliants their own operating system name to ## do this right. When someone cares, they can help us. fx80-alliant-* ) machine=alliant4 opsys=bsd4-2 ;; i860-alliant-* ) machine=alliant-2800 opsys=bsd4-3 ;; ## Alpha (DEC) machines. alpha-dec-osf* ) machine=alpha opsys=osf1 # This is needed to find X11R6.1 libraries for certain tests. NON_GCC_LINK_TEST_OPTIONS=-Wl,-rpath,/usr/X11R6/lib GCC_LINK_TEST_OPTIONS=-Wl,-rpath,/usr/X11R6/lib ;; alpha-*-linux-gnu* ) machine=alpha opsys=gnu-linux ;; ## Altos 3068 m68*-altos-sysv* ) machine=altos opsys=usg5-2 ;; ## Amdahl UTS 580-amdahl-sysv* ) machine=amdahl opsys=usg5-2-2 ;; ## Apollo, Domain/OS m68*-apollo-* ) machine=apollo opsys=bsd4-3 ;; ## AT&T 3b2, 3b5, 3b15, 3b20 we32k-att-sysv* ) machine=att3b opsys=usg5-2-2 ;; ## AT&T 3b1 - The Mighty Unix PC! m68*-att-sysv* ) machine=7300 opsys=usg5-2-2 ;; ## Bull dpx20 rs6000-bull-bosx* ) machine=ibmrs6000 opsys=aix3-2 ;; ## Bull dpx2 m68*-bull-sysv3* ) machine=dpx2 opsys=usg5-3 ;; ## Bull sps7 m68*-bull-sysv2* ) machine=sps7 opsys=usg5-2 ;; ## CCI 5/32, 6/32 -- see "Tahoe". ## Celerity ## I don't know what configuration name to use for this; config.sub ## doesn't seem to know anything about it. Hey, Celerity users, get ## in touch with us! celerity-celerity-bsd* ) machine=celerity opsys=bsd4-2 ;; ## Clipper ## What operating systems does this chip run that Emacs has been ## tested on? clipper-* ) machine=clipper ## We'll use the catch-all code at the bottom to guess the ## operating system. ;; ## Convex *-convex-bsd* | *-convex-convexos* ) machine=convex opsys=bsd4-3 ## Prevents spurious white space in makefiles - d.m.cooke@larc.nasa.gov NON_GNU_CPP="cc -E -P" ;; ## Cubix QBx/386 i[3456]86-cubix-sysv* ) machine=intel386 opsys=usg5-3 ;; ## Cydra 5 cydra*-cydrome-sysv* ) machine=cydra5 opsys=usg5-3 ;; ## Data General AViiON Machines ## DG changed naming conventions with the release of 5.4.4.10, they ## dropped the initial 5.4 but left the intervening R. Because of the ## R this shouldn't conflict with older versions of the OS (which I ## think were named like dgux4.*). m88k-dg-dguxR4.* | m88k-dg-dgux4* ) machine=aviion opsys=dgux4 ;; m88k-dg-dgux5.4R3* | m88k-dg-dgux5.4.3* ) ## This name is dgux5-4-3 instead of dgux5-4r3 ## to avoid a file name conflict on MSDOS. machine=aviion opsys=dgux5-4-3 ;; m88k-dg-dgux5.4R2* | m88k-dg-dgux5.4.2* ) machine=aviion opsys=dgux5-4r2 ;; m88k-dg-dgux* ) machine=aviion opsys=dgux ;; ## DECstations mips-dec-ultrix[0-3].* | mips-dec-ultrix4.0* | mips-dec-bsd4.2* ) machine=pmax opsys=bsd4-2 ;; mips-dec-ultrix4.[12]* | mips-dec-bsd* ) machine=pmax opsys=bsd4-3 ;; mips-dec-ultrix* ) machine=pmax opsys=ultrix4-3 ;; mips-dec-osf* ) machine=pmax opsys=osf1 ;; mips-dec-mach_bsd4.3* ) machine=pmax opsys=mach-bsd4-3 ;; ## Motorola Delta machines m68k-motorola-sysv* | m68000-motorola-sysv* ) machine=delta opsys=usg5-3 if test -z "`type gnucc | grep 'not found'`" then if test -s /etc/167config then CC="gnucc -m68040" else CC="gnucc -m68881" fi else if test -z "`type gcc | grep 'not found'`" then CC=gcc else CC=cc fi fi ;; m88k-motorola-sysv4* ) # jbotte@bnr.ca says that UNIX_System_V <hostName> 4.0 R40V4.3 m88k mc88110 # needs POSIX_SIGNALS and therefore needs usg5-4-2. # I hope there are not other 4.0 versions for this machine # which really need usg5-4 instead. machine=delta88k opsys=usg5-4-2 ;; m88k-motorola-sysv* | m88k-motorola-m88kbcs* ) machine=delta88k opsys=usg5-3 ;; ## Dual machines m68*-dual-sysv* ) machine=dual opsys=usg5-2 ;; m68*-dual-uniplus* ) machine=dual opsys=unipl5-2 ;; ## Elxsi 6400 elxsi-elxsi-sysv* ) machine=elxsi opsys=usg5-2 ;; ## Encore machines ns16k-encore-bsd* ) machine=ns16000 opsys=umax ;; ## The GEC 93 - apparently, this port isn't really finished yet. ## Gould Power Node and NP1 pn-gould-bsd4.2* ) machine=gould opsys=bsd4-2 ;; pn-gould-bsd4.3* ) machine=gould opsys=bsd4-3 ;; np1-gould-bsd* ) machine=gould-np1 opsys=bsd4-3 ;; ## Harris Night Hawk machines running CX/UX (a 5000 looks just like a 4000 ## as far as Emacs is concerned). m88k-harris-cxux* ) # Build needs to be different on 7.0 and later releases case "`uname -r`" in [56].[0-9] ) machine=nh4000 opsys=cxux ;; [7].[0-9] ) machine=nh4000 opsys=cxux7 ;; esac NON_GNU_CPP="/lib/cpp" ;; ## Harris ecx or gcx running CX/UX (Series 1200, Series 3000) m68k-harris-cxux* ) machine=nh3000 opsys=cxux ;; ## Harris power pc NightHawk running Power UNIX (Series 6000) powerpc-harris-powerunix ) machine=nh6000 opsys=powerunix NON_GNU_CPP="cc -Xo -E -P" ;; ## SR2001/SR2201 running HI-UX/MPP hppa1.1-hitachi-hiuxmpp* ) machine=sr2k opsys=hiuxmpp ;; ## Honeywell XPS100 xps*-honeywell-sysv* ) machine=xps100 opsys=usg5-2 ;; ## HP 9000 series 200 or 300 m68*-hp-bsd* ) machine=hp9000s300 opsys=bsd4-3 ;; ## HP/UX 7, 8, 9, and 10 are supported on these machines. m68*-hp-hpux* ) case "`uname -r`" in ## Someone's system reports A.B8.05 for this. ## I wonder what other possibilities there are. *.B8.* ) machine=hp9000s300 opsys=hpux8 ;; *.08.* ) machine=hp9000s300 opsys=hpux8 ;; *.09.* ) machine=hp9000s300 opsys=hpux9 ;; *.10.* ) machine=hp9000s300 opsys=hpux9shr ;; *) machine=hp9000s300 opsys=hpux ;; esac ;; ## HP 9000 series 700 and 800, running HP/UX hppa*-hp-hpux7* ) machine=hp800 opsys=hpux ;; hppa*-hp-hpux8* ) machine=hp800 opsys=hpux8 ;; hppa*-hp-hpux9shr* ) machine=hp800 opsys=hpux9shr ;; hppa*-hp-hpux9* ) machine=hp800 opsys=hpux9 ;; hppa*-hp-hpux10* ) machine=hp800 opsys=hpux10 ;; ## HP 9000 series 700 and 800, running HP/UX hppa*-hp-hpux* ) ## Cross-compilation? Nah! case "`uname -r`" in ## Someone's system reports A.B8.05 for this. ## I wonder what other possibilities there are. *.B8.* ) machine=hp800 opsys=hpux8 ;; *.08.* ) machine=hp800 opsys=hpux8 ;; *.09.* ) machine=hp800 opsys=hpux9 ;; *) machine=hp800 opsys=hpux ;; esac ;; hppa*-*-nextstep* ) machine=hp800 opsys=nextstep ;; ## Orion machines orion-orion-bsd* ) machine=orion opsys=bsd4-2 ;; clipper-orion-bsd* ) machine=orion105 opsys=bsd4-2 ;; ## IBM machines i[3456]86-ibm-aix1.1* ) machine=ibmps2-aix opsys=usg5-2-2 ;; i[3456]86-ibm-aix1.[23]* | i[3456]86-ibm-aix* ) machine=ibmps2-aix opsys=usg5-3 ;; i370-ibm-aix*) machine=ibm370aix opsys=usg5-3 ;; rs6000-ibm-aix3.1* | powerpc-ibm-aix3.1* ) machine=ibmrs6000 opsys=aix3-1 ;; rs6000-ibm-aix3.2.5 | powerpc-ibm-aix3.2.5 ) machine=ibmrs6000 opsys=aix3-2-5 ;; rs6000-ibm-aix4.1* | powerpc-ibm-aix4.1* ) machine=ibmrs6000 opsys=aix4-1 ;; rs6000-ibm-aix4.2* | powerpc-ibm-aix4.2* ) machine=ibmrs6000 opsys=aix4-2 ;; rs6000-ibm-aix4.0* | powerpc-ibm-aix4.0* ) machine=ibmrs6000 opsys=aix4 ;; rs6000-ibm-aix4* | powerpc-ibm-aix4* ) machine=ibmrs6000 opsys=aix4-1 ;; rs6000-ibm-aix* | powerpc-ibm-aix* ) machine=ibmrs6000 opsys=aix3-2 ;; romp-ibm-bsd4.3* ) machine=ibmrt opsys=bsd4-3 ;; romp-ibm-bsd4.2* ) machine=ibmrt opsys=bsd4-2 ;; romp-ibm-aos4.3* ) machine=ibmrt opsys=bsd4-3 ;; romp-ibm-aos4.2* ) machine=ibmrt opsys=bsd4-2 ;; romp-ibm-aos* ) machine=ibmrt opsys=bsd4-3 ;; romp-ibm-bsd* ) machine=ibmrt opsys=bsd4-3 ;; romp-ibm-aix* ) machine=ibmrt-aix opsys=usg5-2-2 ;; ## Integrated Solutions `Optimum V' m68*-isi-bsd4.2* ) machine=isi-ov opsys=bsd4-2 ;; m68*-isi-bsd4.3* ) machine=isi-ov opsys=bsd4-3 ;; ## Intel 386 machines where we do care about the manufacturer i[3456]86-intsys-sysv* ) machine=is386 opsys=usg5-2-2 ;; ## Prime EXL i[3456]86-prime-sysv* ) machine=i386 opsys=usg5-3 ;; ## Sequent Symmetry running Dynix i[3456]86-sequent-bsd* ) machine=symmetry opsys=bsd4-3 ;; ## Sequent Symmetry running ptx 4, which is a modified SVR4. i[3456]86-sequent-ptx4* | i[3456]86-sequent-sysv4* ) machine=sequent-ptx opsys=ptx4 NON_GNU_CPP=/lib/cpp ;; ## Sequent Symmetry running DYNIX/ptx ## Use the old cpp rather than the newer ANSI one. i[3456]86-sequent-ptx* ) machine=sequent-ptx opsys=ptx NON_GNU_CPP="/lib/cpp" ;; ## ncr machine running svr4.3. i[3456]86-ncr-sysv4.3 ) machine=ncr386 opsys=usg5-4-3 ;; ## Unspecified sysv on an ncr machine defaults to svr4.2. ## (Plain usg5-4 doesn't turn on POSIX signals, which we need.) i[3456]86-ncr-sysv* ) machine=ncr386 opsys=usg5-4-2 ;; ## Intel Paragon OSF/1 i860-intel-osf1* ) machine=paragon opsys=osf1 NON_GNU_CPP=/usr/mach/lib/cpp ;; ## Intel 860 i860-*-sysv4* ) machine=i860 opsys=usg5-4 NON_GNU_CC="/bin/cc" # Ie, not the one in /usr/ucb/cc. NON_GNU_CPP="/usr/ccs/lib/cpp" # cc -E tokenizes macro expansion. ;; ## Masscomp machines m68*-masscomp-rtu* ) machine=masscomp opsys=rtu ;; ## Megatest machines m68*-megatest-bsd* ) machine=mega68 opsys=bsd4-2 ;; ## Workstations sold by MIPS ## This is not necessarily all workstations using the MIPS processor - ## Irises are produced by SGI, and DECstations by DEC. ## etc/MACHINES lists mips.h and mips4.h as possible machine files, ## and usg5-2-2 and bsd4-3 as possible OS files. The only guidance ## it gives for choosing between the alternatives seems to be "Use ## -machine=mips4 for RISCOS version 4; use -opsystem=bsd4-3 with ## the BSD world." I'll assume that these are instructions for ## handling two odd situations, and that every other situation ## should use mips.h and usg5-2-2, they being listed first. mips-mips-usg* ) machine=mips4 ## Fall through to the general code at the bottom to decide on the OS. ;; mips-mips-riscos4* ) machine=mips4 opsys=bsd4-3 NON_GNU_CC="cc -systype bsd43" NON_GNU_CPP="cc -systype bsd43 -E" ;; mips-mips-riscos5* ) machine=mips4 opsys=riscos5 NON_GNU_CC="cc -systype bsd43" NON_GNU_CPP="cc -systype bsd43 -E" ;; mips-mips-bsd* ) machine=mips opsys=bsd4-3 ;; mips-mips-* ) machine=mips opsys=usg5-2-2 ;; ## NeXT m68*-next-* | m68k-*-nextstep* ) machine=m68k opsys=nextstep ;; ## The complete machine from National Semiconductor ns32k-ns-genix* ) machine=ns32000 opsys=usg5-2 ;; ## NCR machines m68*-ncr-sysv2* | m68*-ncr-sysvr2* ) machine=tower32 opsys=usg5-2-2 ;; m68*-ncr-sysv3* | m68*-ncr-sysvr3* ) machine=tower32v3 opsys=usg5-3 ;; ## Nixdorf Targon 31 m68*-nixdorf-sysv* ) machine=targon31 opsys=usg5-2-2 ;; ## Nu (TI or LMI) m68*-nu-sysv* ) machine=nu opsys=usg5-2 ;; ## Plexus m68*-plexus-sysv* ) machine=plexus opsys=usg5-2 ;; ## PowerPC reference platform powerpcle-*-solaris2* ) machine=prep opsys=sol2-5 ;; ## Pyramid machines ## I don't really have any idea what sort of processor the Pyramid has, ## so I'm assuming it is its own architecture. pyramid-pyramid-bsd* ) machine=pyramid opsys=bsd4-2 ;; ## Sequent Balance ns32k-sequent-bsd4.2* ) machine=sequent opsys=bsd4-2 ;; ns32k-sequent-bsd4.3* ) machine=sequent opsys=bsd4-3 ;; ## Siemens Nixdorf mips-siemens-sysv* | mips-sni-sysv*) machine=mips-siemens opsys=usg5-4 NON_GNU_CC=/usr/ccs/bin/cc NON_GNU_CPP=/usr/ccs/lib/cpp ;; ## Silicon Graphics machines ## Iris 2500 and Iris 2500 Turbo (aka the Iris 3030) m68*-sgi-iris3.5* ) machine=irist opsys=iris3-5 ;; m68*-sgi-iris3.6* | m68*-sgi-iris*) machine=irist opsys=iris3-6 ;; ## Iris 4D mips-sgi-irix3* ) machine=iris4d opsys=irix3-3 ;; mips-sgi-irix4* ) machine=iris4d opsys=irix4-0 ;; mips-sgi-irix6* ) machine=iris4d opsys=irix6-0 NON_GNU_CPP=/lib/cpp NON_GCC_TEST_OPTIONS=-32 ;; mips-sgi-irix5.[01]* ) machine=iris4d opsys=irix5-0 ;; mips-sgi-irix5* | mips-sgi-irix* ) machine=iris4d opsys=irix5-2 ;; ## SONY machines m68*-sony-bsd4.2* ) machine=news opsys=bsd4-2 ;; m68*-sony-bsd4.3* ) machine=news opsys=bsd4-3 ;; m68*-sony-newsos3* | m68*-sony-news3*) machine=news opsys=bsd4-3 ;; mips-sony-bsd* | mips-sony-newsos4* | mips-sony-news4*) machine=news-risc opsys=bsd4-3 ;; mips-sony-news* ) machine=news-risc opsys=newsos5 ;; ## Stride m68*-stride-sysv* ) machine=stride opsys=usg5-2 ;; ## Suns sparc-*-linux-gnu* ) machine=sparc opsys=gnu-linux ;; *-sun-sunos* | *-sun-bsd* | *-sun-solaris* \ | i[3456]86-*-solaris2* | i[3456]86-*-sunos5* | powerpc*-*-solaris2* \ | rs6000-*-solaris2*) case "${canonical}" in m68*-sunos1* ) machine=sun1 ;; m68*-sunos2* ) machine=sun2 ;; m68* ) machine=sun3 ;; i[3456]86-sun-sunos[34]* ) machine=sun386 ;; i[3456]86-*-* ) machine=intel386 ;; powerpcle* ) machine=powerpcle ;; powerpc* | rs6000* ) machine=ibmrs6000 ;; sparc* ) machine=sparc ;; * ) unported=yes ;; esac case "${canonical}" in ## The Sun386 didn't get past 4.0. i[3456]86-*-sunos4 ) opsys=sunos4-0 ;; *-sunos4.0* ) opsys=sunos4-0 ;; *-sunos4.1.[3-9]*noshare ) ## This name is sunos413 instead of sunos4-1-3 ## to avoid a file name conflict on MSDOS. opsys=sunos413 NON_GNU_CPP=/usr/lib/cpp NON_GCC_TEST_OPTIONS=-Bstatic GCC_TEST_OPTIONS=-static ;; *-sunos4.1.[3-9]* | *-sunos4shr*) opsys=sunos4shr NON_GNU_CPP=/usr/lib/cpp ;; *-sunos4* | *-sunos ) opsys=sunos4-1 NON_GCC_TEST_OPTIONS=-Bstatic GCC_TEST_OPTIONS=-static ;; *-sunos5.3* | *-solaris2.3* ) opsys=sol2-3 NON_GNU_CPP=/usr/ccs/lib/cpp ;; *-sunos5.4* | *-solaris2.4* ) opsys=sol2-4 NON_GNU_CPP=/usr/ccs/lib/cpp RANLIB="ar -ts" ;; *-sunos5.5* | *-solaris2.5* ) opsys=sol2-5 NON_GNU_CPP=/usr/ccs/lib/cpp RANLIB="ar -ts" ;; *-sunos5* | *-solaris* ) opsys=sol2-4 NON_GNU_CPP=/usr/ccs/lib/cpp ;; * ) opsys=bsd4-2 ;; esac ## Watch out for a compiler that we know will not work. case "${canonical}" in *-solaris* | *-sunos5* ) if [ "x$CC" = x/usr/ucb/cc ]; then ## /usr/ucb/cc doesn't work; ## we should find some other compiler that does work. unset CC fi ;; *) ;; esac ;; sparc-*-nextstep* ) machine=sparc opsys=nextstep ;; ## Tadpole 68k m68*-tadpole-sysv* ) machine=tad68k opsys=usg5-3 ;; ## Tahoe machines tahoe-tahoe-bsd4.2* ) machine=tahoe opsys=bsd4-2 ;; tahoe-tahoe-bsd4.3* ) machine=tahoe opsys=bsd4-3 ;; ## Tandem Integrity S2 mips-tandem-sysv* ) machine=tandem-s2 opsys=usg5-3 ;; ## Tektronix XD88 m88k-tektronix-sysv3* ) machine=tekxd88 opsys=usg5-3 ;; ## Tektronix 16000 box (6130?) ns16k-tektronix-bsd* ) machine=ns16000 opsys=bsd4-2 ;; ## Tektronix 4300 ## src/m/tek4300.h hints that this is a m68k machine. m68*-tektronix-bsd* ) machine=tek4300 opsys=bsd4-3 ;; ## Titan P2 or P3 ## We seem to have lost the machine-description file titan.h! titan-titan-sysv* ) machine=titan opsys=usg5-3 ;; ## Ustation E30 (SS5E) m68*-unisys-uniplus* ) machine=ustation opsystem=unipl5-2 ;; ## Vaxen. vax-dec-* ) machine=vax case "${canonical}" in *-bsd4.1* ) opsys=bsd4-1 ;; *-bsd4.2* | *-ultrix[0-3].* | *-ultrix4.0* ) opsys=bsd4-2 ;; *-bsd4.3* | *-ultrix* ) opsys=bsd4-3 ;; *-sysv[01]* | *-sysvr[01]* ) opsys=usg5-0 ;; *-sysv2* | *-sysvr2* ) opsys=usg5-2 ;; *-vms* ) opsys=vms ;; * ) unported=yes esac ;; ## Whitechapel MG1 ns16k-whitechapel-* ) machine=mg1 ## We don't know what sort of OS runs on these; we'll let the ## operating system guessing code below try. ;; ## Wicat m68*-wicat-sysv* ) machine=wicat opsys=usg5-2 ;; ## Intel 386 machines where we don't care about the manufacturer i[3456]86-*-* ) machine=intel386 case "${canonical}" in *-isc1.* | *-isc2.[01]* ) opsys=386-ix ;; *-isc2.2* ) opsys=isc2-2 ;; *-isc4.0* ) opsys=isc4-0 ;; *-isc4.* ) opsys=isc4-1 GCC_TEST_OPTIONS=-posix NON_GCC_TEST_OPTIONS=-Xp ;; *-isc* ) opsys=isc3-0 ;; *-esix5* ) opsys=esix5r4; NON_GNU_CPP=/usr/lib/cpp ;; *-esix* ) opsys=esix ;; *-xenix* ) opsys=xenix ;; *-linux-gnu* ) opsys=gnu-linux ;; *-sco3.2v4* ) opsys=sco4 ; NON_GNU_CPP=/lib/cpp ;; *-sco3.2v5* ) opsys=sco5 NON_GNU_CPP=/lib/cpp # Prevent -belf from being passed to $CPP. # /lib/cpp does not accept it. OVERRIDE_CPPFLAGS=" " ;; *-bsd386* | *-bsdi1* ) opsys=bsd386 ;; *-bsdi2.0* ) opsys=bsdos2 ;; *-bsdi2* ) opsys=bsdos2-1 ;; *-386bsd* ) opsys=386bsd ;; *-freebsd* ) opsys=freebsd ;; *-nextstep* ) opsys=nextstep ;; ## Otherwise, we'll fall through to the generic opsys code at the bottom. esac ;; ## Linux/68k-based GNU system m68k-*-linux-gnu* ) machine=m68k opsys=gnu-linux ;; * ) unported=yes ;; esac ### If the code above didn't choose an operating system, just choose ### an operating system based on the configuration name. You really ### only want to use this when you have no idea what the right ### operating system is; if you know what operating systems a machine ### runs, it's cleaner to make it explicit in the case statement ### above. if test x"${opsys}" = x; then case "${canonical}" in *-gnu* ) opsys=gnu ;; *-bsd4.[01] ) opsys=bsd4-1 ;; *-bsd4.2 ) opsys=bsd4-2 ;; *-bsd4.3 ) opsys=bsd4-3 ;; *-sysv0 | *-sysvr0 ) opsys=usg5-0 ;; *-sysv2 | *-sysvr2 ) opsys=usg5-2 ;; *-sysv2.2 | *-sysvr2.2 ) opsys=usg5-2-2 ;; *-sysv3* | *-sysvr3* ) opsys=usg5-3 ;; *-sysv4.1* | *-sysvr4.1* ) NON_GNU_CPP=/usr/lib/cpp opsys=usg5-4 ;; *-sysv4.[2-9]* | *-sysvr4.[2-9]* ) if [ x$NON_GNU_CPP = x ]; then if [ -f /usr/ccs/lib/cpp ]; then NON_GNU_CPP=/usr/ccs/lib/cpp else NON_GNU_CPP=/lib/cpp fi fi opsys=usg5-4-2 ;; *-sysv4* | *-sysvr4* ) opsys=usg5-4 ;; * ) unported=yes ;; esac fi if test "x$RANLIB" = x; then RANLIB=ranlib fi changequote([, ])dnl if test $unported = yes; then AC_MSG_ERROR([Emacs hasn't been ported to \`${canonical}' systems. Check \`etc/MACHINES' for recognized configuration names.]) fi machfile="m/${machine}.h" opsysfile="s/${opsys}.h" #### Choose a compiler. test -n "$CC" && cc_specified=yes # Save the value of CFLAGS that the user specified. SPECIFIED_CFLAGS="$CFLAGS" case ${with_gcc} in "yes" ) CC="gcc" GCC=yes ;; "no" ) : ${CC=cc} ;; * ) AC_PROG_CC esac # On Suns, sometimes $CPP names a directory. if test -n "$CPP" && test -d "$CPP"; then CPP= fi #### Some systems specify a CPP to use unless we are using GCC. #### Now that we know whether we are using GCC, we can decide whether #### to use that one. if test "x$NON_GNU_CPP" != x && test x$GCC != xyes && test "x$CPP" = x then CPP="$NON_GNU_CPP" fi #### Some systems specify a CC to use unless we are using GCC. #### Now that we know whether we are using GCC, we can decide whether #### to use that one. if test "x$NON_GNU_CC" != x && test x$GCC != xyes && test x$cc_specified != xyes then CC="$NON_GNU_CC" fi if test x$GCC = xyes && test "x$GCC_TEST_OPTIONS" != x then CC="$CC $GCC_TEST_OPTIONS" fi if test x$GCC = x && test "x$NON_GCC_TEST_OPTIONS" != x then CC="$CC $NON_GCC_TEST_OPTIONS" fi if test x$GCC = xyes && test "x$GCC_LINK_TEST_OPTIONS" != x then ac_link="$ac_link $GCC_LINK_TEST_OPTIONS" fi if test x$GCC = x && test "x$NON_GCC_LINK_TEST_OPTIONS" != x then ac_link="$ac_link $NON_GCC_LINK_TEST_OPTIONS" fi #### Some other nice autoconf tests. If you add a test here which #### should make an entry in src/config.h, don't forget to add an #### #undef clause to src/config.h.in for autoconf to modify. dnl checks for programs AC_PROG_LN_S AC_PROG_CPP AC_PROG_INSTALL AC_PROG_YACC dnl checks for Unix variants AC_AIX dnl checks for header files AC_CHECK_HEADERS(sys/select.h sys/timeb.h sys/time.h unistd.h utime.h linux/version.h sys/systeminfo.h termios.h limits.h) AC_HEADER_STDC AC_HEADER_TIME AC_DECL_SYS_SIGLIST dnl Some systems have utime.h but don't declare the struct anyplace. AC_MSG_CHECKING(for struct utimbuf) AC_TRY_COMPILE([#ifdef TIME_WITH_SYS_TIME #include <sys/time.h> #include <time.h> #else #ifdef HAVE_SYS_TIME_H #include <sys/time.h> #else #include <time.h> #endif #endif #ifdef HAVE_UTIME_H #include <utime.h> #endif], [static struct utimbuf x; x.actime = x.modtime;], [AC_MSG_RESULT(yes) AC_DEFINE(HAVE_STRUCT_UTIMBUF)], AC_MSG_RESULT(no)) dnl checks for typedefs AC_TYPE_SIGNAL AC_MSG_CHECKING(for struct timeval) AC_TRY_COMPILE([#ifdef TIME_WITH_SYS_TIME #include <sys/time.h> #include <time.h> #else #ifdef HAVE_SYS_TIME_H #include <sys/time.h> #else #include <time.h> #endif #endif], [static struct timeval x; x.tv_sec = x.tv_usec;], [AC_MSG_RESULT(yes) HAVE_TIMEVAL=yes AC_DEFINE(HAVE_TIMEVAL)], [AC_MSG_RESULT(no) HAVE_TIMEVAL=no]) dnl checks for structure members AC_STRUCT_TM AC_STRUCT_TIMEZONE dnl checks for compiler characteristics AC_C_CONST dnl check for Make feature AC_PROG_MAKE_SET dnl checks for operating system services AC_SYS_LONG_FILE_NAMES #### Choose a window system. AC_PATH_X if test "$no_x" = yes; then window_system=none else window_system=x11 fi if test "${x_libraries}" != NONE && test -n "${x_libraries}"; then LD_SWITCH_X_SITE=-L`echo ${x_libraries} | sed -e "s/:/ -L/g"` LD_SWITCH_X_SITE_AUX=-R`echo ${x_libraries} | sed -e "s/:/ -R/g"` fi if test "${x_includes}" != NONE && test -n "${x_includes}"; then C_SWITCH_X_SITE=-I`echo ${x_includes} | sed -e "s/:/ -I/g"` fi if test x"${x_includes}" = x; then bitmapdir=/usr/include/X11/bitmaps else # accumulate include directories that have X11 bitmap subdirectories bmd_acc="dummyval" for bmd in `echo ${x_includes} | sed -e "s/:/ /g"`; do if test -d "${bmd}/X11/bitmaps"; then bmd_acc="${bmd_acc}:${bmd}/X11/bitmaps" elif test -d "${bmd}/bitmaps"; then bmd_acc="${bmd_acc}:${bmd}/bitmaps" fi done if test ${bmd_acc} != "dummyval"; then bitmapdir=`echo ${bmd_acc} | sed -e "s/^dummyval://"` fi fi case "${window_system}" in x11 ) HAVE_X_WINDOWS=yes HAVE_X11=yes case "${with_x_toolkit}" in athena | lucid ) USE_X_TOOLKIT=LUCID ;; motif ) USE_X_TOOLKIT=MOTIF ;; dnl open-look ) USE_X_TOOLKIT=OPEN_LOOK ;; no ) USE_X_TOOLKIT=none ;; dnl If user did not say whether to use a toolkit, dnl make this decision later: use the toolkit if we have X11R5 or newer. * ) USE_X_TOOLKIT=maybe ;; esac ;; none ) HAVE_X_WINDOWS=no HAVE_X11=no USE_X_TOOLKIT=none ;; esac ### If we're using X11, we should use the X menu package. HAVE_MENUS=no case ${HAVE_X11} in yes ) HAVE_MENUS=yes ;; esac if test "${opsys}" = "hpux9"; then case "${x_libraries}" in *X11R4* ) opsysfile="s/hpux9-x11r4.h" ;; esac fi if test "${opsys}" = "hpux9shr"; then case "${x_libraries}" in *X11R4* ) opsysfile="s/hpux9shxr4.h" ;; esac fi #### Extract some information from the operating system and machine files. AC_CHECKING([the machine- and system-dependent files to find out - which libraries the lib-src programs will want, and - whether the GNU malloc routines are usable]) ### First figure out CFLAGS (which we use for running the compiler here) ### and REAL_CFLAGS (which we use for real compilation). ### The two are the same except on a few systems, where they are made ### different to work around various lossages. For example, ### GCC 2.5 on GNU/Linux needs them to be different because it treats -g ### as implying static linking. ### If the CFLAGS env var is specified, we use that value ### instead of the default. ### It's not important that this name contain the PID; you can't run ### two configures in the same directory and have anything work ### anyway. tempcname="conftest.c" echo ' #include "'${srcdir}'/src/'${opsysfile}'" #include "'${srcdir}'/src/'${machfile}'" #ifndef LIBS_MACHINE #define LIBS_MACHINE #endif #ifndef LIBS_SYSTEM #define LIBS_SYSTEM #endif #ifndef C_SWITCH_SYSTEM #define C_SWITCH_SYSTEM #endif #ifndef C_SWITCH_MACHINE #define C_SWITCH_MACHINE #endif configure___ libsrc_libs=LIBS_MACHINE LIBS_SYSTEM configure___ c_switch_system=C_SWITCH_SYSTEM configure___ c_switch_machine=C_SWITCH_MACHINE #ifndef LIB_X11_LIB #define LIB_X11_LIB -lX11 #endif #ifndef LIBX11_MACHINE #define LIBX11_MACHINE #endif #ifndef LIBX11_SYSTEM #define LIBX11_SYSTEM #endif configure___ LIBX=LIB_X11_LIB LIBX11_MACHINE LIBX11_SYSTEM #ifdef UNEXEC configure___ unexec=UNEXEC #else configure___ unexec=unexec.o #endif #ifdef SYSTEM_MALLOC configure___ system_malloc=yes #else configure___ system_malloc=no #endif #ifndef C_DEBUG_SWITCH #define C_DEBUG_SWITCH -g #endif #ifndef C_OPTIMIZE_SWITCH #define C_OPTIMIZE_SWITCH -O #endif #ifndef LD_SWITCH_MACHINE #define LD_SWITCH_MACHINE #endif #ifndef LD_SWITCH_SYSTEM #define LD_SWITCH_SYSTEM #endif #ifndef LD_SWITCH_X_SITE_AUX #define LD_SWITCH_X_SITE_AUX #endif configure___ ld_switch_system=LD_SWITCH_SYSTEM configure___ ld_switch_machine=LD_SWITCH_MACHINE #ifdef THIS_IS_CONFIGURE /* Get the CFLAGS for tests in configure. */ #ifdef __GNUC__ configure___ CFLAGS=C_DEBUG_SWITCH C_OPTIMIZE_SWITCH '${SPECIFIED_CFLAGS}' #else configure___ CFLAGS=C_DEBUG_SWITCH '${SPECIFIED_CFLAGS}' #endif #else /* not THIS_IS_CONFIGURE */ /* Get the CFLAGS for real compilation. */ #ifdef __GNUC__ configure___ REAL_CFLAGS=C_DEBUG_SWITCH C_OPTIMIZE_SWITCH '${SPECIFIED_CFLAGS}' #else configure___ REAL_CFLAGS=C_DEBUG_SWITCH '${SPECIFIED_CFLAGS}' #endif #endif /* not THIS_IS_CONFIGURE */ ' > ${tempcname} # The value of CPP is a quoted variable reference, so we need to do this # to get its actual value... CPP=`eval "echo $CPP"` changequote(, )dnl eval `${CPP} -Isrc ${tempcname} \ | sed -n -e 's/^configure___ \([^=]*=\)\(.*\)$/\1"\2"/p'` if test "x$SPECIFIED_CFLAGS" = x; then eval `${CPP} -Isrc -DTHIS_IS_CONFIGURE ${tempcname} \ | sed -n -e 's/^configure___ \([^=]*=\)\(.*\)$/\1"\2"/p'` else REAL_CFLAGS="$CFLAGS" fi changequote([, ])dnl rm ${tempcname} ac_link="$ac_link $ld_switch_machine $ld_switch_system" ### Compute the unexec source name from the object name. UNEXEC_SRC="`echo ${unexec} | sed 's/\.o/.c/'`" # Do the opsystem or machine files prohibit the use of the GNU malloc? # Assume not, until told otherwise. GNU_MALLOC=yes doug_lea_malloc=yes AC_CHECK_FUNC(malloc_get_state, ,doug_lea_malloc=no) AC_CHECK_FUNC(malloc_set_state, ,doug_lea_malloc=no) if test "${system_malloc}" = "yes"; then GNU_MALLOC=no GNU_MALLOC_reason=" (The GNU allocators don't work with this system configuration.)" fi if test "$doug_lea_malloc" = "yes" ; then if test "$GNU_MALLOC" = yes ; then GNU_MALLOC_reason=" (Using new Doug Lea's new malloc from the GNU C Library.)" fi AC_DEFINE(DOUG_LEA_MALLOC) fi if test x"${REL_ALLOC}" = x; then REL_ALLOC=${GNU_MALLOC} fi LISP_FLOAT_TYPE=yes #### Add the libraries to LIBS and check for some functions. if test x"${OVERRIDE_CPPFLAGS}" != x; then CPPFLAGS="${OVERRIDE_CPPFLAGS}" else CPPFLAGS="$c_switch_system $c_switch_machine $CPPFLAGS" fi LIBS="$libsrc_libs $LIBS" dnl If found, this defines HAVE_LIBDNET, which m/pmax.h checks, dnl and also adds -ldnet to LIBS, which Autoconf uses for checks. AC_CHECK_LIB(dnet, dnet_ntoa) dnl This causes -lresolv to get used in subsequent tests, dnl which causes failures on some systems such as HPUX 9. dnl AC_CHECK_LIB(resolv, gethostbyname) dnl FIXME replace main with a function we actually want from this library. AC_CHECK_LIB(Xbsd, main, LD_SWITCH_X_SITE="$LD_SWITCH_X_SITE -lXbsd") AC_CHECK_LIB(pthreads, cma_open) AC_MSG_CHECKING(for XFree86) if test -d /usr/X386/include; then HAVE_XFREE386=yes : ${C_SWITCH_X_SITE="-I/usr/X386/include"} else HAVE_XFREE386=no fi AC_MSG_RESULT($HAVE_XFREE386) # Change CFLAGS temporarily so that C_SWITCH_X_SITE gets used # for the tests that follow. We set it back to REAL_CFLAGS later on. if test "${HAVE_X11}" = "yes"; then DEFS="$C_SWITCH_X_SITE $DEFS" LDFLAGS="$LDFLAGS $LD_SWITCH_X_SITE" LIBS="$LIBX $LIBS" CFLAGS="$C_SWITCH_X_SITE $CFLAGS" # On Solaris, arrange for LD_RUN_PATH to point to the X libraries for tests. # This is handled by LD_SWITCH_X_SITE_AUX during the real build, # but it's more convenient here to set LD_RUN_PATH # since this also works on hosts that don't understand LD_SWITCH_X_SITE_AUX. if test "${x_libraries}" != NONE && test -n "${x_libraries}"; then LD_RUN_PATH=$x_libraries${LD_RUN_PATH+:}$LD_RUN_PATH export LD_RUN_PATH fi if test "${opsys}" = "gnu-linux"; then AC_MSG_CHECKING(whether X on GNU/Linux needs -b to link) AC_TRY_LINK([], [XOpenDisplay ("foo");], [xlinux_first_failure=no], [xlinux_first_failure=yes]) if test "${xlinux_first_failure}" = "yes"; then OLD_LD_SWITCH_X_SITE="$LD_SWITCH_X_SITE" OLD_C_SWITCH_X_SITE="$C_SWITCH_X_SITE" OLD_CPPFLAGS="$CPPFLAGS" OLD_LIBS="$LIBS" LD_SWITCH_X_SITE="$LD_SWITCH_X_SITE -b i486-linuxaout" C_SWITCH_X_SITE="$C_SWITCH_X_SITE -b i486-linuxaout" CPPFLAGS="$CPPFLAGS -b i486-linuxaout" LIBS="$LIBS -b i486-linuxaout" AC_TRY_LINK([], [XOpenDisplay ("foo");], [xlinux_second_failure=no], [xlinux_second_failure=yes]) if test "${xlinux_second_failure}" = "yes"; then # If we get the same failure with -b, there is no use adding -b. # So take it out. This plays safe. LD_SWITCH_X_SITE="$OLD_LD_SWITCH_X_SITE" C_SWITCH_X_SITE="$OLD_C_SWITCH_X_SITE" CPPFLAGS="$OLD_CPPFLAGS" LIBS="$OLD_LIBS" AC_MSG_RESULT(no) else AC_MSG_RESULT(yes) fi else AC_MSG_RESULT(no) fi fi AC_CHECK_FUNCS(XrmSetDatabase XScreenResourceString \ XScreenNumberOfScreen XSetWMProtocols) fi if test "${window_system}" = "x11"; then AC_MSG_CHECKING(X11 version 6) AC_TRY_LINK([#include <X11/Xlib.h>], [#if XlibSpecificationRelease < 6 fail; #endif ], [AC_MSG_RESULT(6 or newer) AC_DEFINE(HAVE_X11R6)], [AC_MSG_RESULT(before 6)]) fi if test "${window_system}" = "x11"; then AC_MSG_CHECKING(X11 version 5) AC_TRY_LINK([#include <X11/Xlib.h>], [#if XlibSpecificationRelease < 5 fail; #endif ], [AC_MSG_RESULT(5 or newer) HAVE_X11R5=yes AC_DEFINE(HAVE_X11R5)], [ HAVE_X11R5=no AC_MSG_RESULT(before 5)]) fi dnl Do not put whitespace before the #include statements below. dnl Older compilers (eg sunos4 cc) choke on it. if test x"${USE_X_TOOLKIT}" = xmaybe; then if test x"${HAVE_X11R5}" = xyes; then AC_MSG_CHECKING(X11 version 5 with Xaw) AC_TRY_LINK([ #include <X11/Intrinsic.h> #include <X11/Xaw/Simple.h>], [], [AC_MSG_RESULT(5 or newer, with Xaw; use toolkit by default) USE_X_TOOLKIT=LUCID], [AC_MSG_RESULT(before 5 or no Xaw; do not use toolkit by default) USE_X_TOOLKIT=none]) else USE_X_TOOLKIT=none fi fi X_TOOLKIT_TYPE=$USE_X_TOOLKIT if test "${USE_X_TOOLKIT}" != "none"; then AC_MSG_CHECKING(X11 toolkit version) AC_TRY_LINK([#include <X11/Intrinsic.h>], [#if XtSpecificationRelease < 6 fail; #endif ], [AC_MSG_RESULT(6 or newer) HAVE_X11XTR6=yes AC_DEFINE(HAVE_X11XTR6)], [AC_MSG_RESULT(before 6) HAVE_X11XTR6=no]) dnl If using toolkit, check whether libXmu.a exists. dnl tranle@intellicorp.com says libXmu.a can need XtMalloc in libXt.a to link. OLDLIBS="$LIBS" if test x$HAVE_X11XTR6 = xyes; then LIBS="-lXt -lSM -lICE $LIBS" else LIBS="-lXt $LIBS" fi AC_CHECK_LIB(Xmu, XmuConvertStandardSelection) LIBS="$OLDLIBS" fi # If netdb.h doesn't declare h_errno, we must declare it by hand. AC_MSG_CHECKING(whether netdb declares h_errno) AC_TRY_LINK([#include <netdb.h>], [return h_errno;], [AC_MSG_RESULT(yes) AC_DEFINE(HAVE_H_ERRNO)], [AC_MSG_RESULT(no)]) AC_FUNC_ALLOCA # fmod, logb, and frexp are found in -lm on most systems. # On HPUX 9.01, -lm does not contain logb, so check for sqrt. AC_CHECK_LIB(m, sqrt) # Check for mail-locking functions in a "mail" library AC_CHECK_LIB(mail, maillock, AC_DEFINE(HAVE_LIBMAIL) AC_CHECK_FUNCS(touchlock) AC_CHECK_HEADERS(maillock.h)) AC_CHECK_FUNCS(gettimeofday gethostname getdomainname dup2 \ rename closedir mkdir rmdir sysinfo \ random lrand48 bcopy bcmp logb frexp fmod rint cbrt ftime res_init setsid \ strerror fpathconf select mktime euidaccess getpagesize tzset setlocale \ utimes setrlimit setpgid getcwd) # Check this now, so that we will NOT find the above functions in ncurses. # That is because we have not set up to link ncurses in lib-src. # It's better to believe a function is not available # than to expect to find it in ncurses. AC_CHECK_LIB(ncurses, tparm) # These tell us which Kerberos-related libraries to use. if test "${with_kerberos+set}" = set; then AC_CHECK_LIB(krb, krb_get_cred) AC_CHECK_LIB(des, des_cbc_encrypt) AC_CHECK_LIB(com_err, com_err) fi AC_MSG_CHECKING(whether localtime caches TZ) AC_CACHE_VAL(emacs_cv_localtime_cache, [if test x$ac_cv_func_tzset = xyes; then AC_TRY_RUN([#include <time.h> #if STDC_HEADERS # include <stdlib.h> #endif extern char **environ; unset_TZ () { char **from, **to; for (to = from = environ; (*to = *from); from++) if (! (to[0][0] == 'T' && to[0][1] == 'Z' && to[0][2] == '=')) to++; } char TZ_GMT0[] = "TZ=GMT0"; char TZ_PST8[] = "TZ=PST8"; main() { time_t now = time ((time_t *) 0); int hour_GMT0, hour_unset; if (putenv (TZ_GMT0) != 0) exit (1); hour_GMT0 = localtime (&now)->tm_hour; unset_TZ (); hour_unset = localtime (&now)->tm_hour; if (putenv (TZ_PST8) != 0) exit (1); if (localtime (&now)->tm_hour == hour_GMT0) exit (1); unset_TZ (); if (localtime (&now)->tm_hour != hour_unset) exit (1); exit (0); }], emacs_cv_localtime_cache=no, emacs_cv_localtime_cache=yes, [# If we have tzset, assume the worst when cross-compiling. emacs_cv_localtime_cache=yes]) else # If we lack tzset, report that localtime does not cache TZ, # since we can't invalidate the cache if we don't have tzset. emacs_cv_localtime_cache=no fi])dnl AC_MSG_RESULT($emacs_cv_localtime_cache) if test $emacs_cv_localtime_cache = yes; then AC_DEFINE(LOCALTIME_CACHE) fi if test "x$HAVE_TIMEVAL" = xyes; then AC_MSG_CHECKING(whether gettimeofday can't accept two arguments) AC_TRY_LINK([ #ifdef TIME_WITH_SYS_TIME #include <sys/time.h> #include <time.h> #else #ifdef HAVE_SYS_TIME_H #include <sys/time.h> #else #include <time.h> #endif #endif ], [ struct timeval time; struct timezone dummy; gettimeofday (&time, &dummy); ], [AC_MSG_RESULT(no)], [AC_MSG_RESULT(yes) AC_DEFINE(GETTIMEOFDAY_ONE_ARGUMENT)]) fi ok_so_far=yes AC_CHECK_FUNC(socket, , ok_so_far=no) if test $ok_so_far = yes; then AC_CHECK_HEADER(netinet/in.h, , ok_so_far=no) fi if test $ok_so_far = yes; then AC_CHECK_HEADER(arpa/inet.h, , ok_so_far=no) fi if test $ok_so_far = yes; then AC_DEFINE(HAVE_INET_SOCKETS) fi if test -f /usr/lpp/X11/bin/smt.exp; then AC_DEFINE(HAVE_AIX_SMT_EXP) fi # Set up the CFLAGS for real compilation, so we can substitute it. CFLAGS="$REAL_CFLAGS" changequote(, )dnl #### Find out which version of Emacs this is. version=`grep 'defconst[ ]*emacs-version' ${srcdir}/lisp/version.el \ | sed -e 's/^[^"]*"\([^"]*\)".*$/\1/'` changequote([, ])dnl if test x"${version}" = x; then AC_MSG_ERROR(can't find current emacs version in \`${srcdir}/lisp/version.el'.) fi ### Specify what sort of things we'll be editing into Makefile and config.h. ### Use configuration here uncanonicalized to avoid exceeding size limits. AC_SUBST(version) AC_SUBST(configuration) AC_SUBST(canonical) AC_SUBST(srcdir) AC_SUBST(prefix) AC_SUBST(exec_prefix) AC_SUBST(bindir) AC_SUBST(datadir) AC_SUBST(sharedstatedir) AC_SUBST(libexecdir) AC_SUBST(mandir) AC_SUBST(infodir) AC_SUBST(lispdir) AC_SUBST(locallisppath) AC_SUBST(lisppath) AC_SUBST(etcdir) AC_SUBST(lockdir) AC_SUBST(archlibdir) AC_SUBST(docdir) AC_SUBST(bitmapdir) AC_SUBST(c_switch_system) AC_SUBST(c_switch_machine) AC_SUBST(LD_SWITCH_X_SITE) AC_SUBST(LD_SWITCH_X_SITE_AUX) AC_SUBST(C_SWITCH_X_SITE) AC_SUBST(CFLAGS) AC_SUBST(X_TOOLKIT_TYPE) AC_SUBST(machfile) AC_SUBST(opsysfile) AC_SUBST(RANLIB) AC_DEFINE_UNQUOTED(EMACS_CONFIGURATION, "${canonical}") AC_DEFINE_UNQUOTED(EMACS_CONFIG_OPTIONS, "${ac_configure_args}") AC_DEFINE_UNQUOTED(config_machfile, "${machfile}") AC_DEFINE_UNQUOTED(config_opsysfile, "${opsysfile}") AC_DEFINE_UNQUOTED(LD_SWITCH_X_SITE, ${LD_SWITCH_X_SITE}) AC_DEFINE_UNQUOTED(LD_SWITCH_X_SITE_AUX, ${LD_SWITCH_X_SITE_AUX}) AC_DEFINE_UNQUOTED(C_SWITCH_X_SITE, ${C_SWITCH_X_SITE}) AC_DEFINE_UNQUOTED(UNEXEC_SRC, ${UNEXEC_SRC}) if test "${HAVE_X_WINDOWS}" = "yes" ; then AC_DEFINE(HAVE_X_WINDOWS) fi if test "${USE_X_TOOLKIT}" != "none" ; then AC_DEFINE(USE_X_TOOLKIT) fi if test "${HAVE_X11}" = "yes" ; then AC_DEFINE(HAVE_X11) fi if test "${HAVE_XFREE386}" = "yes" ; then AC_DEFINE(HAVE_XFREE386) fi if test "${HAVE_MENUS}" = "yes" ; then AC_DEFINE(HAVE_MENUS) fi if test "${GNU_MALLOC}" = "yes" ; then AC_DEFINE(GNU_MALLOC) fi if test "${REL_ALLOC}" = "yes" ; then AC_DEFINE(REL_ALLOC) fi if test "${LISP_FLOAT_TYPE}" = "yes" ; then AC_DEFINE(LISP_FLOAT_TYPE) fi #### Report on what we decided to do. echo " Configured for \`${canonical}'. Where should the build process find the source code? ${srcdir} What operating system and machine description files should Emacs use? \`${opsysfile}' and \`${machfile}' What compiler should emacs be built with? ${CC} ${CFLAGS} Should Emacs use the GNU version of malloc? ${GNU_MALLOC}${GNU_MALLOC_reason} Should Emacs use the relocating allocator for buffers? ${REL_ALLOC} What window system should Emacs use? ${window_system} What toolkit should Emacs use? ${USE_X_TOOLKIT}" if test -n "${x_includes}"; then echo " Where do we find X Windows header files? ${x_includes}" else echo " Where do we find X Windows header files? Standard dirs" fi if test -n "${x_libraries}"; then echo " Where do we find X Windows libraries? ${x_libraries}" else echo " Where do we find X Windows libraries? Standard dirs" fi echo # Remove any trailing slashes in these variables. changequote(, )dnl test "${prefix}" != NONE && prefix=`echo "${prefix}" | sed 's,\([^/]\)/*$,\1,'` test "${exec_prefix}" != NONE && exec_prefix=`echo "${exec_prefix}" | sed 's,\([^/]\)/*$,\1,'` changequote([, ])dnl AC_OUTPUT(Makefile lib-src/Makefile.c:lib-src/Makefile.in oldXMenu/Makefile \ man/Makefile lwlib/Makefile src/Makefile.c:src/Makefile.in, [ ### Make the necessary directories, if they don't exist. for dir in cpp etc lisp ; do test -d ${dir} || mkdir ${dir} done # Build src/Makefile from ${srcdir}/src/Makefile.c # and lib-src/Makefile from ${srcdir}/lib-src/Makefile.c # This must be done after src/config.h is built, since we rely on that file. changequote(, )dnl The horror, the horror. # Now get this: Some word that is part of the ${srcdir} directory name # or the ${configuration} value might, just might, happen to be an # identifier like `sun4' or `i386' or something, and be predefined by # the C preprocessor to some helpful value like 1, or maybe the empty # string. Needless to say consequent macro substitutions are less # than conducive to the makefile finding the correct directory. undefs="`echo $top_srcdir $configuration $canonical | sed -e 's/[^a-zA-Z0-9_]/ /g' -e 's/^/ /' -e 's/ *$//' \ -e 's/ */ -U/g' -e 's/-U[0-9][^ ]*//g' \ `" changequote([, ])dnl echo creating src/paths.h make paths-force echo creating lib-src/Makefile ( cd lib-src rm -f junk.c junk1.c junk2.c sed -e '/start of cpp stuff/q' \ < Makefile.c > junk1.c sed -e '1,/start of cpp stuff/d'\ -e 's@/\*\*/#\(.*\)$@/* \1 */@' \ < Makefile.c > junk.c $CPP $undefs -I. -I$top_srcdir/src $CPPFLAGS junk.c | \ sed -e 's/^ / /' -e '/^#/d' -e '/^[ ]*$/d' > junk2.c cat junk1.c junk2.c > Makefile.new rm -f junk.c junk1.c junk2.c chmod 444 Makefile.new mv -f Makefile.new Makefile ) echo creating src/Makefile ( cd src rm -f junk.c junk1.c junk2.c sed -e '/start of cpp stuff/q' \ < Makefile.c > junk1.c sed -e '1,/start of cpp stuff/d'\ -e 's@/\*\*/#\(.*\)$@/* \1 */@' \ < Makefile.c > junk.c $CPP $undefs -I. -I$top_srcdir/src $CPPFLAGS junk.c | \ sed -e 's/^ / /' -e '/^#/d' -e '/^[ ]*$/d' > junk2.c cat junk1.c junk2.c > Makefile.new rm -f junk.c junk1.c junk2.c chmod 444 Makefile.new mv -f Makefile.new Makefile ) if test ! -f src/.gdbinit && test -f $top_srcdir/src/.gdbinit; then echo creating src/.gdbinit echo source $top_srcdir/src/.gdbinit > src/.gdbinit fi ], [CPP="$CPP" CPPFLAGS="$CPPFLAGS"])