May 2006 - XEmacs-Patches - XEmacs Mailing List Archives

[AC21.5] Improve beta.texi 18 years, 7 months

Stephen J. Turnbull

APPROVE COMMIT 21.5 Not for 21.4. This patch (1) Makes the Problem Reporting node more prominent by making it top-level, with other minor improvements. (2) Changes "Compiling Beta XEmacs" to "Building Beta XEmacs", and moves the "Getting the Sources" node here from "Introduction". This is more consistent with the way other projects document these activities. (3) Adds vertical space before ＠node commands so that node breaks are more visible. diff --git man/ChangeLog man/ChangeLog index fe5ad3d..a59970a 100644 --- man/ChangeLog +++ man/ChangeLog ＠＠ -0,0 +1,8 ＠＠ +2006-05-09 Stephen J. Turnbull <stephen(a)xemacs.org> + + * beta.texi (Building Beta XEmacs): Rename 'Compiling' -> 'Building'. + (Reporting Problems): Improve it, make it a top-level node, add + some vertical spacing to make node breaks more visible. + (Getting the Source): Flesh it out a bit and move it to the + Building Beta XEmacs node. + diff --git man/beta.texi man/beta.texi index 5da713f..2a73ee3 100644 --- man/beta.texi +++ man/beta.texi ＠＠ -57,6 +57,7 ＠＠ Boston, MA 02111-1307, USA. ＠vskip 0pt plus 1fill ＠noindent +Copyright ＠copyright{} 2006 Free Software Foundation. ＠* Copyright ＠copyright{} 2005 Ben Wing. ＠* This file is part of XEmacs. ＠＠ -83,10 +84,11 ＠＠ Boston, MA 02111-1307, USA. This Info file describes info relevant to beta versions of XEmacs. ＠menu * Introduction:: -* Compiling Beta XEmacs:: +* Building Beta XEmacs:: * Packages:: +* Reporting Problems:: * Improving XEmacs:: -* Index:: +* Index:: ＠detailmenu --- The Detailed Node Listing --- ＠＠ -95,8 +97,6 ＠＠ Introduction * Mailing Lists:: * Beta Release Schedule:: -* Reporting Problems:: -* Getting the Source:: Mailing Lists ＠＠ -108,8 +108,9 ＠＠ Mailing Lists * Subscribing by e-mail:: * Unsubscribing by e-mail:: -Compiling Beta XEmacs +Building Beta XEmacs +* Getting the Source:: * Building an XEmacs from patches:: * Building XEmacs from a full distribution:: ＠＠ -139,7 +140,8 ＠＠ Large contributions ＠end ifinfo -＠node Introduction, Compiling Beta XEmacs, Top, Top + +＠node Introduction, Building Beta XEmacs, Top, Top ＠chapter Introduction You are running a potentially unstable version of XEmacs. Please do ＠＠ -150,10 +152,9 ＠＠ them to ＠uref{mailto:xemacs-beta＠＠xemacs ＠menu * Mailing Lists:: * Beta Release Schedule:: -* Reporting Problems:: -* Getting the Source:: ＠end menu + ＠node Mailing Lists, Beta Release Schedule, Introduction, Introduction ＠section Mailing Lists ＠＠ -167,6 +168,7 ＠＠ them to ＠uref{mailto:xemacs-beta＠＠xemacs * Unsubscribing by e-mail:: ＠end menu + ＠node XEmacs Beta Mailing List, XEmacs Patches Mailing List, Mailing Lists, Mailing Lists ＠subsection XEmacs Beta Mailing List ＠＠ -176,6 +178,7 ＠＠ and coding discussion, takes place on th Only patches and administrative actions regarding patches are sent elsewhere (to the XEmacs Patches list). + ＠node XEmacs Patches Mailing List, XEmacs Design Mailing List, XEmacs Beta Mailing List, Mailing Lists ＠subsection XEmacs Patches Mailing List ＠＠ -189,6 +192,7 ＠＠ Besides patches, only actions by members should be posted to this list. All discussion should be redirected to XEmacs Beta or XEmacs Design. + ＠node XEmacs Design Mailing List, List Administrivia, XEmacs Patches Mailing List, Mailing Lists ＠subsection XEmacs Design Mailing List ＠＠ -217,6 +221,7 ＠＠ searchable archives. The URL is Note that the xemacs-LIST-admin address is used internally by the Mailman software; it is NOT a synonym for xemacs-LIST-request. + ＠node Managing your subscription via the Web, Subscribing by e-mail, List Administrivia, Mailing Lists ＠subsection Managing your subscription via the Web ＠＠ -224,19 +229,22 ＠＠ Subscription, unsubscription, and option temporarily suspending delivery) can be accomplished via the web interface at ＠uref{http://www.xemacs.org/Lists/#xemacs-LIST}. + ＠node Subscribing by e-mail, Unsubscribing by e-mail, Managing your subscription via the Web, Mailing Lists ＠subsection Subscribing by e-mail Send an email message to ＠uref{mailto:xemacs-LIST-request＠＠xemacs.org} with ＠samp{subscribe} (without the quotes) as the BODY of the message. + ＠node Unsubscribing by e-mail, , Subscribing by e-mail, Mailing Lists ＠subsection Unsubscribing by e-mail Send an email message to ＠uref{mailto:xemacs-LIST-request＠＠xemacs.org} with ＠samp{unsubscribe} (without the quotes) as the BODY of the message. -＠node Beta Release Schedule, Reporting Problems, Mailing Lists, Introduction + +＠node Beta Release Schedule, , Mailing Lists, Introduction ＠section Beta Release Schedule We would like to achieve a weekly or fortnightly release cycle (you ＠＠ -255,75 +263,46 ＠＠ can't use CVS for some reason and must u it will make it more likely that we release betas more often. -＠node Reporting Problems, Getting the Source, Beta Release Schedule, Introduction -＠section Reporting Problems +＠node Building Beta XEmacs, Packages, Introduction, Top +＠chapter Building Beta XEmacs -The best way to get problems fixed in XEmacs is to submit good problem -reports, ＠kbd{M-x report-xemacs-bug RET} will help you do this (assuming -you have a usable XEmacs). Since this is beta software, problems are -certain to exist. Please read through all of part II of the XEmacs -FAQ for an overview of problem reporting. Other items which are most -important are: - -＠enumerate -＠item -Do not submit C stack backtraces without line numbers. Since it -is possible to compile optimized with debug information with GCC -it is never a good idea to compile XEmacs without the -g flag. -XEmacs runs on a variety of platforms, and often it is not -possible to recreate problems which afflict a specific platform. -The line numbers in the C stack backtrace help isolate where the -problem is actually occurring. - -＠item -Attempt to recreate the problem starting with an invocation of -XEmacs with ＠code{xemacs -no-autoloads}. Quite often, problems are -due to package interdependencies, and the like. An actual bug -in XEmacs should be reproducible in a default configuration -without loading any special packages (or the one or two specific -packages that cause the bug to appear). If you have trouble -getting anything to work at all with the above invocation, use -＠code{xemacs -vanilla} instead. If you need to load your user init -file or the site file to get the problem to occur, then it has -something to do with them, and you should try to isolate the -issue in those files. +＠menu +* Getting the Source:: +* Building an XEmacs from patches:: +* Building XEmacs from a full distribution:: +＠end menu -＠item -A picture can be worth a thousand words. When reporting an -unusual display, it is generally best to capture the problem in a -screen dump and include that with the problem report. The easiest -way to get a screen dump is to use the xv program and its grab -function. Save the image as a GIF to keep bandwidth requirements -down without loss of information. MIME is the preferred method -for making the image attachments. -＠end enumerate -＠node Getting the Source, , Reporting Problems, Introduction +＠node Getting the Source, Building an XEmacs from patches, Building Beta XEmacs, Building Beta XEmacs ＠section Getting the Source +We provide the traditional tarballs and release-to-release patchkits for +each beta release. ＠xref{Beta Release Schedule}. These are available +at + + ＠uref{ftp://ftp.xemacs.org/pub/xemacs/beta/} + In addition to the normal tar distribution, XEmacs source is now available via CVS. Please see ＠uref{http://www.xemacs.org/Develop/cvsaccess.html} -＠node Compiling Beta XEmacs, Packages, Introduction, Top -＠chapter Compiling Beta XEmacs +You can also browse the repository via ViewCVS + + ＠uref{http://cvs.xemacs.org/} -＠menu -* Building an XEmacs from patches:: -* Building XEmacs from a full distribution:: -＠end menu -＠node Building an XEmacs from patches, Building XEmacs from a full distribution, Compiling Beta XEmacs, Compiling Beta XEmacs +＠node Building an XEmacs from patches, Building XEmacs from a full distribution, Getting the Source, Building Beta XEmacs ＠section Building an XEmacs from patches -All beta releases of XEmacs are included with patches from the previous -version in an attempt to keep bandwidth requirements down. Patches -should be applied with the GNU patch program in something like the -following. Let's say you're upgrading XEmacs 21.5-beta9 to XEmacs -21.5-beta10 and you have a full unmodified XEmacs 21.5-beta9 source -tree to work with. Change to the top level directory and issue the -shell command: +All beta releases of XEmacs provide patches from the previous version as +an alternative to keep bandwidth requirements down. These patches are +actually scripts generated by the ＠file{makepatch} program, and can be +run if you have the ＠file{applypatch} program. Patches may also be +applied with the GNU patch program in something like the following. +Let's say you're upgrading XEmacs 21.5-beta9 to XEmacs 21.5-beta10 and +you have a full unmodified XEmacs 21.5-beta9 source tree to work with. +Change to the top level directory and issue the shell command: ＠example $ gunzip -c /tmp/xemacs-21.5.9-21.5.10.patch.gz | patch -p1 ＠＠ -349,13 +328,15 ＠＠ $ make check > ./xemacs-make-check.err 2 Redirect the output from make to those files because you'll use them later when you send off a build report with ＠kbd{M-x build-report RET} -＠node Building XEmacs from a full distribution, , Building an XEmacs from patches, Compiling Beta XEmacs + +＠node Building XEmacs from a full distribution, , Building an XEmacs from patches, Building Beta XEmacs ＠section Building XEmacs from a full distribution ＠enumerate ＠item -Locate a convenient place where you have at least 100MB of free space -and issue the command +Locate a convenient place where you have at least 200MB of free space +(approximately 100MB for sources and compiled Lisp, and 100MB for +objects and other generated files) and issue the command ＠example $ gunzip -c /tmp/xemacs-21.5.10.tar.gz | tar xvf - ＠＠ -409,8 +390,8 ＠＠ If you have gcc, consider using --with-xemacs-compiler=g++ ＠end example -This will compile XEmacs using g++, which will turn on a lot of additional -error-checking. +This will compile XEmacs using g++, which is generally much stricter +about type-checking than C compilers like gcc. ＠item If your packages are not installed under /usr/local, you should add a ＠＠ -592,7 +573,8 ＠＠ of the developers. ＠end enumerate ＠end enumerate -＠node Packages, Improving XEmacs, Compiling Beta XEmacs, Top + +＠node Packages, Reporting Problems, Building Beta XEmacs, Top ＠chapter Packages [Note: these instructions have been partly updated, but not carefully ＠＠ -613,6 +595,7 ＠＠ by FTP at ＠uref{ftp://ftp.xemacs.org/pub * Building XEmacs and XEmacs packages from scratch:: ＠end menu + ＠node Binary package installation, Manual procedures for package management, Packages, Packages ＠section Binary package installation ＠＠ -623,6 +606,7 ＠＠ level of an XEmacs package hierarchy and in this directory, run the command ＠kbd{M-x package-admin-add-binary-package} and fill in appropriate values to the prompts. + ＠node Manual procedures for package management, Building XEmacs and XEmacs packages from scratch, Binary package installation, Packages ＠section Manual procedures for package management ＠＠ -656,7 +640,8 ＠＠ xemacs -vanilla -batch -f batch-byte-com Of course, being a beta tester, you'd be aware that it is much easier to manage your XEmacs packages with PUI. -＠node Building XEmacs and XEmacs packages from scratch, , Manual procedures for package management, Packages + +＠node Building XEmacs and XEmacs packages from scratch, , Manual procedures for package management, Packages ＠section Building XEmacs and XEmacs packages from scratch To build everything completely from scratch isn't hard, just time ＠＠ -705,7 +690,56 ＠＠ And then: $ make install ＠end example -＠node Improving XEmacs, Index, Packages, Top + +＠node Reporting Problems, Improving XEmacs, Packages, Top +＠section Reporting Problems + +The best way to get problems fixed in XEmacs is to submit good problem +reports, ＠kbd{M-x report-xemacs-bug RET} will help you do this (assuming +you have a usable XEmacs). Since this is beta software, problems are +certain to exist. Please read through all of part II of the XEmacs +FAQ for an overview of problem reporting. +＠xref{Installation, Installation and Troubleshooting, , xemacs-faq}. +The most relevant parts are in section 2.4, General Troubleshooting. +＠c #### Why doesn't this link work? +＠c ＠ref{Q2.4.1, General Troubleshooting, , xemacs-faq}. +Other items which are most important are: + +＠enumerate +＠item +Do not submit C stack backtraces without line numbers. Since it +is possible to compile optimized with debug information with GCC +it is never a good idea to compile XEmacs without the -g flag. +XEmacs runs on a variety of platforms, and often it is not +possible to recreate problems which afflict a specific platform. +The line numbers in the C stack backtrace help isolate where the +problem is actually occurring. + +＠item +Attempt to recreate the problem starting with an invocation of +XEmacs with ＠code{xemacs -no-autoloads}. Quite often, problems are +due to package interdependencies, and the like. An actual bug +in XEmacs should be reproducible in a default configuration +without loading any special packages (or the one or two specific +packages that cause the bug to appear). If you have trouble +getting anything to work at all with the above invocation, use +＠code{xemacs -vanilla} instead. If you need to load your user init +file or the site file to get the problem to occur, then it has +something to do with them, and you should try to isolate the +issue in those files. + +＠item +A picture can be worth a thousand words. When reporting an +unusual display, it is generally best to capture the problem in a +screen dump and include that with the problem report. The easiest +way to get a screen dump is to use the xv program and its grab +function. Save the image as a GIF to keep bandwidth requirements +down without loss of information. MIME is the preferred method +for making the image attachments. +＠end enumerate + + +＠node Improving XEmacs, Index, Reporting Problems, Top ＠chapter Improving XEmacs ＠menu ＠＠ -899,6 +933,7 ＠＠ Engineer are the most likely sources of ＠uref{http://www.xemacs.org/Develop/jobs.html#package,jobs.html} for current information about package release engineers. + ＠node Syncing with GNU Emacs, , New packages, Large contributions ＠subsection Syncing with GNU Emacs ＠＠ -956,6 +991,7 ＠＠ remains in the GNU Emacs core, as well t ＠c Print the tables of contents ＠contents ＠c That's all + ＠node Index, , Improving XEmacs, Top ＠unnumbered Index -- Graduate School of Systems and Information Engineering University of Tsukuba http://turnbull.sk.tsukuba.ac.jp/ Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Economics of Information Communication and Computation Systems Experimental Economics, Microeconomic Theory, Game Theory

1 participants
0 comments

[AC21.5] Xft customization for User's Manual 18 years, 7 months

Stephen J. Turnbull

APPROVE COMMIT 21.5 Not for 21.4. Thanks to Mats Lidell for the reminder! diff --git man/ChangeLog man/ChangeLog index a12b476..fe5ad3d 100644 --- man/ChangeLog +++ man/ChangeLog ＠＠ -0,0 +1,6 ＠＠ +2006-05-09 Stephen J. Turnbull <stephen(a)xemacs.org> + + * xemacs/custom.texi (Xft Font Customization): New node. + (Faces): Create menu, add new node. + * xemacs/xemacs.texi (Top): add new node to detail menu. + diff --git a/man/xemacs/custom.texi b/man/xemacs/custom.texi index 15a8e5b..4eaae64 100644 --- man/xemacs/custom.texi +++ man/xemacs/custom.texi ＠＠ -1,4 +1,6 ＠＠ +＠c FIXME -- we demand a detail menu! + ＠node Customization, Quitting, Emulation, Top ＠chapter Customization ＠cindex customization ＠＠ -2034,6 +2036,14 ＠＠ attributes when its buffer is viewed fro The display attributes of faces may be specified either in Lisp or through the X resource manager. +Basic Xft support has been merged into the mainline, and it looks pretty +good. However, customization UI and documentation still leaves a lot to +be desired. Here's a first cut, as a separate node. + +＠menu +* Xft Font Customization:: +＠end menu + ＠subsection Customizing Faces You can change the face of an extent with the functions in ＠＠ -2214,6 +2224,406 ＠＠ You can set underline the specified ＠var to make underlining an attribute of the face or not. If the optional ＠var{frame} argument is provided, the face is changed only in that frame; otherwise, it is changed in all frames. + + +＠node Xft Font Customization, , , Faces +＠section Xft Font Customization + +This section was written by ＠email{stephen＠(a)xemacs.org,Stephen Turnbull}, +and is very much a work in progress. I've tried to provide pointers to +as much of the relevant information as possible, but many of the APIs +and UIs are in flux, so it seemed more work than it was worth to +completely translate the tables of constants, and so on. Feel free to +ask for clarifications, ＠emph{etc.} Please Cc +＠email{xemacs-beta＠(a)xemacs.org,the XEmacs Beta Testers' mailing list}, +as that is the issue tracking channel of record, and there are a few +others who can answer the questions or improve documentation when I'm +not able to respond quickly. + +＠c Don't blame Ben (or Eric and Matthias, for that matter). Feel free to +＠c add, edit, and share the blame, everybody! + +＠c #### Make these ＠urlref's!! +As of mid-2005, we have added support for the +＠file{Xft} library, which provides a more robust ＠emph{font +configuration} mechanism via Keith Packard's ＠file{fontconfig} library; +improved glyph rendering, including antialiasing, via the +＠file{freetype} library; and client-side rendering (saving bandwidth and +server memory) via the ＠file{XRender extension}. + +＠c #### Describe Alexey Gladkov and Yury Konovalov's work. + +＠subheading Font configuration + +In XEmacs, font configuration is handled via ＠emph{faces}. Currently +XEmacs uses a special type of ＠emph{font specifier} to map XEmacs +locales to font names. Especially under X11, this can cause annoying +problems because of the unreliability of X servers' mappings from +＠samp{XLFD} names to X11 fonts, over which XEmacs has no influence +whatsoever. However, the ＠file{fontconfig} library which is used with +＠file{Xft} provides much more reliable mapping, along with a more +reliably parsable naming scheme similar to that used by TrueType fonts +on MS Windows and the Macintosh. + +＠subheading fontconfig + +Fontconfig is dramatically different from the X model in several ways. +In particular, when queried for a font ＠emph{fontconfig always returns a +font}, whereas X queries may return ``not found.'' However, the font +returned need not be anything like the desired font. This is not really +a problem in practice, because users generally have a pretty good idea +of what fonts are available on their display. However, users should be +aware that as of XEmacs 21.5.26 the font selection internals have not +been revised to account for this radically different model, so some +surprising results are possible. + +From the user's point of view, ＠file{fontconfig} provides a naming +convention which is ＠emph{precise}, ＠emph{accurate}, and +＠emph{convenient}. Precision means that all properties available in the +programming API can be individually specified. Accuracy means that the +truename of the font is exactly the list of all properties specified by +the font. Thus, the anomolies that occur with XLFDs on many servers +(including modern Linux distributions with XFree86 or X.org servers) +cannot occur. Convenience is subjective, of course. However, +＠file{fontconfig} provides a configuration system which (1) explicitly +specifies the defaults and substitutions that will be made in processing +user queries, (2) allows the user to specify search configuration, +abbreviations, substitutions, and defaults that override the system's, +in the same format as used by system files, and (3) allows flexible +aliases that can resolve to any of several actual fonts depending on +which fonts are installed. + +Further, a ＠dfn{standard minimal configuration} is defined that ensures +that at least the ＠emph{serif}, ＠emph{sans-serif}, and ＠emph{monospace} +font aliases are available on all ＠file{fontconfig} systems. + +＠subheading fontconfig font names + +＠file{fontconfig} font names are very regular, and constitute a precise +and extensible specification of a font's properties. + +The basic ＠file{fontconfig} font name has three components: the font +family name, the size, and a list of named attribute fields. All +attribute names and values are strings encoded in Unicode UTF-8, or +decimal numbers with optional decimal point and fraction. The +characters ＠samp{-}, ＠samp{:}, ＠samp{,}, ＠samp{=}, and ＠samp{\} are +syntactically significant to ＠file{fontconfig}. They may be used in +font name components by the familiar mechanism of backslash escaping. +This simply removes any special meaning from the following character. +It is always safe to use an escape even if it is not needed. + +Most string values are case-insensitive, but this is attribute-specific. + +The ＠dfn{font family name} is an arbitrary string, which may contain any +character, including spaces, hyphens, and commas. Don't forget to +escape hyphens, colons, commas, and backslashes! + +The ＠dfn{size} is a decimal number with optional decimal point and +fractional part. It is interpreted as point size. + +A ＠dfn{named attribute field} is a key-value pair separated by an equal +sign. Some attributes have predefined semantics. (These include such +familiar attributes as ＠samp{slant} and ＠samp{dpi} -- note that +＠file{fontconfig} does not distinguish between vertical and horizontal +resolution! + +The basic syntax of a font name is simple and regular. The ＠emph{font +family name} comes first, followed by a hyphen and the ＠emph{size}, +followed by a list of ＠emph{named attribute fields}, each introduced by +a colon: + +＠example +＠var{font family name}-＠var{size}:＠var{key1}=＠var{value1}:＠var{key2}=＠var{value2}... +＠end example + +There are four extensions to the basic syntax. First, all of the +fields are optional. If size is omitted, the hyphen should be omitted +as well. There should never be a trailing colon. Note that +＠file{fontconfig} does ＠emph{not} interpret an omitted field as +``default to current.'' ＠file{fontconfig} does not have access to that +context. Instead, omitted fields are filled in according to a set of +defaults specified in the system ＠file{fonts.conf} file and in the +user's ＠file{.fonts.cont} file. The user's configuration gets +precedence, of course. + +The second is that the ＠emph{font family name}, the ＠emph{size}, and the +＠emph{value} component of a ＠emph{named attribute field} may be a +comma-separated list of values. This is called a ＠dfn{pattern}. In +queries, ＠file{fontconfig} tries to match each entry in the list in +order. (I suspect that order of attributes is also significant. +(a)emph{I.e.}, font family always gets highest precedence, followed by +size, followed by the remaining named attributes. Testing and reports +to the ＠email{xemacs-beta＠(a)xemacs.org,XEmacs Beta Testers mailing list} +are very welcome!) + +The third is the definition of ＠dfn{constants} to represent numerical +values. For example, both ＠samp{slant} and ＠samp{weight} are defined as +integer attributes, but the magnitudes are arbitrary; only the ordering +and relative distances are useful. (``Relative distance'' is used to +implement the concept that although strictly speaking ＠emph{italic} and +＠emph{oblique} are different faces, most users don't know the +difference, and even professional typographers would agree that they are +much closer to each other than either is to ＠emph{roman}.) So +＠emph{constants} like ＠samp{roman} (0) and ＠samp{italic} (100) are +defined for ＠samp{slant}, and ＠samp{medium} (100) and ＠samp{bold} (200) +are defined for ＠samp{weight}. + +The fourth is that a ＠dfn{style} may be defined as an alias for an +instance of a ＠emph{named attribute field}, that is, specifying both key +and value. The ＠emph{styles} ＠samp{bold}, an alias for +＠samp{weight=200}, and ＠samp{italic}, an alias for ＠samp{slant=100}, are +commonly used. + +Styles and constants can be defined by the application. XEmacs +currently defines none, but suggestions are welcome if some convenient +alias is lacking from ＠file{fontconfig}. Note that we will not provide +additional aliases where standard ones exist, ＠emph{e.g.}, ＠samp{heavy} +as an additional alias for ＠samp{bold} would almost certainly be +rejected. These cause more confusion than they are worth, and would +decrease portability of user specifications to other applications. + +(Attributes can also be defined, but they must be implemented by the +fonts to be useful. Until XEmacs provides its own fonts with +non-standardized attributes, this is not useful.) + +Here are some examples of font names and query patterns: + +＠example +Times-12 12-point Times Roman +Times-12:bold 12-point Times Bold +Courier:italic Courier Italic in the default size +Monospace:matrix=1 .1 0 1 The user's preferred monospace font + with artificial obliquing +Mikachan\-PB-16 16-point Mikachan-PB +LucidaTypewriter,Courier-9 9-point LucidaTypewriter if available, + otherwise 9-point Courier +＠end example + +Note how ＠emph{styles} are used in the second and third examples, how +＠samp{matrix} has a complex value containing spaces, and the +escaped hyphen in the font family name ＠samp{Mikachan\-PB}. + +＠c #### FIXME here and also in fontconfig.texi (in general-docs package). +Here is a (somewhat outdated) list of current standard named attributes: + +＠example +Property CPP symbol Type Description + +family FC_FAMILY String Font family name +style FC_STYLE String Font style. Overrides weight and slant +slant FC_SLANT Int Italic, oblique or roman +weight FC_WEIGHT Int Light, medium, demibold, bold or black +size FC_SIZE Double Point size +aspect FC_ASPECT Double Stretch glyphs horizontally, then hint +pixelsize FC_PIXEL_SIZE Double Pixel size +spacing FC_SPACING Int Proportional, monospace or charcell +foundry FC_FOUNDRY String Font foundry name +antialias FC_ANTIALIAS Bool Should glyphs be antialiased? +hinting FC_HINTING Bool Should the rasterizer use hinting? +verticallayout FC_VERTICAL_LAYOUT Bool Use vertical layout +autohint FC_AUTOHINT Bool Use autohinter instead of normal hinter +globaladvance FC_GLOBAL_ADVANCE Bool Use font global advance data +file FC_FILE String The filename holding the font +index FC_INDEX Int The index of the font within the file +ftface FC_FT_FACE FT_Face Use the specified FreeType face object +rasterizer FC_RASTERIZER String Which rasterizer is in use +outline FC_OUTLINE Bool Whether the glyphs are outlines +scalable FC_SCALABLE Bool Whether glyphs can be scaled +scale FC_SCALE Double Point->pixel conversion scale factor +dpi FC_DPI Double Target dots per inch +rgba FC_RGBA Int unknown, rgb, bgr, vrgb, vbgr, none + - subpixel geometry +source FC_SOURCE String X11, freetype +minspace FC_MINSPACE Bool Eliminate leading from line spacing +charset FC_CHARSET CharSet Unicode chars encoded by the font +lang FC_LANG String List of RFC-3066-style languages + this font supports +fontversion FC_FONTVERSION Int From 'head' table +＠end example + +Here is a list of current standard constants: + +＠example +Constant Property CPP symbol + +light weight FC_WEIGHT_LIGHT +medium weight FC_WEIGHT_MEDIUM +demibold weight FC_WEIGHT_DEMIBOLD +bold weight FC_WEIGHT_BOLD +black weight FC_WEIGHT_BLACK +roman slant FC_SLANT_ROMAN +italic slant FC_SLANT_ITALIC +oblique slant FC_SLANT_OBLIQUE +proportional spacing FC_PROPORTIONAL +mono spacing FC_MONO +charcell spacing FC_CHARCELL +unknown rgba FC_RGBA_UNKNOWN +rgb rgba FC_RGBA_RGB +bgr rgba FC_RGBA_BGR +vrgb rgba FC_RGBA_VRGB +vbgr rgba FC_RGBA_VBGR +none rgba FC_RGBA_NONE +＠end example + +Note that this is the ＠file{fontconfig} API list; you can expect that +XEmacs will define corresponding keywords by substituting ＠samp{:} for +the leading ＠samp{FC_}, ＠samp{-} for ＠samp{_}, removing the key of the +attribute if present, and lowercasing the name. Thus +＠samp{FC_WEIGHT_BOLD} becomes ＠samp{:bold}. +＠kbd{M-x apropos RET fc-.*-mapping} will give a list of variables each +of which lists such keywords and their meanings. + +＠subheading Font menus + +The ＠samp{Options->Font} and ＠samp{Options->Font Sizes} menus are +broken, by design, not just by ＠file{Xft}. The problem is that many +fonts are unavailable because they don't match the current size---which +is very strange, since ＠file{Xft} fonts are of course scalable. But the +whole idea of requiring that the font match the size is strange. And +the ＠samp{Options->Font Weights} menu is just disabled, and has been for +eons. + +＠subheading X resources + +Currently there are ＠emph{four} treatments of font resources. There are +the ＠samp{XEmacs.(a)var{face}.attributeFont} resources used to set a +single global font specification. In the widgets, some (still) have a +＠samp{font} resource using the automatic ＠file{Xt} resource conversion +to X's ＠samp{FontStruct}, some have separate ＠samp{font} and ＠samp{xftFont} +resources with the former automatically converted to ＠samp{FontStruct} +by ＠file{Xt} and the latter left as a string, to be converted to a +fontconfig ＠samp{FcPattern} by +＠samp{FcParseName} later, and some have a single ＠samp{font} resource +which is converted to ＠samp{FontStruct} by ＠file{Xt} or the latter left +as a string, depending on whether ＠file{Xft} was enabled by +＠samp{configure} or not. + +Eventually these should be converted to use the face +approach, perhaps with some way to set specifications for individual +widgets, frames, or buffers. This will require some careful design work +to incorporate face support in the widgets. Stephen's current thinking +is that XEmacs should just accept any +or all of ＠samp{font}, ＠samp{fontSet}, and ＠samp{fontList} resources, +treat them all as lists of font names, either ＠samp{XLFD}- or +＠file{fontconfig}-style, parse them ourselves (ie, not use the ＠file{Xt} +resource manager), and add them to font specifiers as appropriate. But +this will require a bit of thought to obey POLA vis-a-vis usual ＠file{Xt} +conventions. + +＠subheading Specifiers, charsets, and languages + +Traditionally Mule uses a rather rigid and low-level abstraction, the +＠emph{charset}, to characterize font repertoires. Unfortunately, +support for a given charset is generally neither necessary nor +sufficient to support a language. Worse, in XEmacs's current +implementation Xft doesn't help much. Instead you need to add the fonts +for different charsets to the font specifier in the right order. + +There currently is no explicit way to specify that a particular font be +used only for a given language. However, since many fonts support only +a limited repertoire such as ISO 8859/1, you can use the precedence of +specifications for a given specifier locale to get something of this +effect for non-Latin character sets. This will normally work rather +poorly for multiple Latin character sets, however, because the +repertoires tend to have large amounts of overlap. Support for +specifying font by ＠emph{language} as well as by character set is +planned. + +Because fonts supporting other languages tend to support English as +well, if you want to use one font for English and another for the other +language, you must use the ＠code{append} method when adding font +specifications for the ＠emph{other} language. + +However, this simple method leaves you with a problem if you want to +change the other language's font: you have to remove the existing +specification so it won't shadow the new one when you append. + +In order to provide a convenient way to change ``other-language fonts'', +I use ＠code{remove-tag-set-append} and ＠code{define-specifier-tag} like +this: + +＠example +(define-specifier-tag 'lang-ja) +;; No, I don't try to do real work with this font! But it makes it +;; obvious that I got the requested font. :-) +(set-face-font 'default "AirCut-14") +(set-face-font 'default "Kochi Mincho-14" nil '(lang-ja) 'append) +;; Oops, too sober. Try something to match AirCut. +(set-face-font 'default "Mikachan-14" + nil '(lang-ja) 'remove-tag-set-append) +＠end example + +The only way to configure widget fonts at the present time is to use X +resources (or hack the source and rebuild). Currently supported widgets +are +＠itemize +＠item +menubars +＠item +tab controls +＠end itemize + +Here are the resources I use. ＠strong{Warning:} ＠emph{This interface +will change. Pay attention to beta announcements, and complain loudly +if changes aren't documented here!} The tab control and menubar have +separate Font and XftFont resources, and use the X resource manager to +instantiate a FontStruct from the Font resource. There is no converter +facility for XftFont yet, and creating one that handles both FontStruct +and XftFont depending on XEmacs's configuration and the font name seems +error-prone at best. Probably we will use a simple string +representation for this resource, and convert to a face in XEmacs rather +than a font in Xt/Xft. +＠example +XEmacs*Tabs.xftFont: Bitstream Vera Sans-16 +XEmacs*menubar*xftFont: Bitstream Vera Sans-16 +XEmacs.modeline.attributeFont: Bitstream Charter-16 +XEmacs.default.attributeFont: Bitstream Vera Sans Mono-16 +＠end example +I highly recommend use of a proportional font in the modeline because it +allows a lot more text to fit there. +＠c Previously the font sizes were quite varied, and there was a comment +＠c that this weirdness gave good balance. This isn't true on my main +＠c platform, Mac OS X, and needs to be rechecked on Linux, where it was +＠c observed. + +＠subheading Known Problems + +＠table ＠code +＠item Options->Font +＠itemx Options->Font Size +These menus don't work. All fonts are greyed out. All sizes are +available, but many (most?) faces don't change size, in particular, +＠samp{default} does not. + +＠item Antialiased text bleeding outside of reported extent +This is most obvious with the underscore character in that +font, and with cursors. The bottom of the underscore is antialiased, +and insertions or deletions in the same line before the underscore leave +a series of "phantom" underlines. + +I think this is probably an Xft bug, but I'm not sure. +＠end table + +＠subheading Variables Used with Xft and Fontconfig + +＠defvar xft-debug-level + +Level of debugging messages to issue to stderr for ＠var{Xft}. +A nonnegative integer. Set to 0 to suppress all warnings. +Default is 1 to ensure a minimum of debugging output at initialization. +Higher levels give more information. +＠end defvar + +＠defvar xft-version + +The major version number of the Xft library compiled with. +＠end defvar + +＠defvar xft-xlfd-font-regexp + +Regular expression matching XLFD font names. +＠end defvar + ＠node Frame Components ＠section Frame Components diff --git man/xemacs/xemacs.texi man/xemacs/xemacs.texi index b848210..eb7a396 100644 --- man/xemacs/xemacs.texi +++ man/xemacs/xemacs.texi ＠＠ -605,6 +605,10 ＠＠ The Init File * Init Examples:: How to do some things with an init file. * Terminal Init:: Each terminal type can have an init file. +Faces + +* Xft Font Customization:: Configuring the next generation of fonts. + Dealing with XEmacs Trouble * Stuck Recursive:: `[...]' in mode line around the parentheses. -- School of Systems and Information Engineering http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Ask not how you can "do" free software business; ask what your business can "do for" free software.

1 participants
0 comments

[PATCH] Autoload shell-prompt-pattern 18 years, 7 months

Tero Saarni

Hi, Function telnet in telnet.el is broken currently. Calling it results in error "Symbol's value as variable is void: shell-prompt-pattern". The variable has not been autoloaded after this change: http://cvs.xemacs.org/viewcvs.cgi/XEmacs/packages/xemacs-packages/xemacs-... I believe the trivial fix below will help. -- Tero --- xemacs-base-orig/shell.el 2004-11-23 06:10:54.000000000 +0200 +++ xemacs-base/shell.el 2006-03-10 18:44:32.000000000 +0200 ＠＠ -132,6 +132,7 ＠＠ :type 'regexp :group 'shell) +;;;###autoload (defcustom shell-prompt-pattern "^[^#$%>\n]*[#$%>] *" "Regexp to match prompts in the inferior shell. Defaults to \"^[^#$%>\\n]*[#$%>] *\", which works pretty well.

4 participants
3 comments

[COMMIT] Support builtin functions in find-function 18 years, 7 months

Aidan Kehoe

SUPERSEDES 17336.3580.904405.497258(a)parhasard.net Doesn’t work for ELL modules yet, but I’m not waiting for that to commit, because I have a feeling if I ever get to it, it’ll be a long time in the future, and the code is plenty useful as it is. NOTE: This patch has been committed. lib-src/ChangeLog addition: 2006-04-29 Aidan Kehoe <kehoea(a)parhasard.net> * make-docfile.c: * make-docfile.c (put_filename): * make-docfile.c (scan_c_file): * make-docfile.c (scan_lisp_file): Record file name information for built-in symbols. Based on the FSF's implementation of same. lisp/ChangeLog addition: 2006-04-29 Aidan Kehoe <kehoea(a)parhasard.net> * dumped-lisp.el (preloaded-file-list): Move loadhist earlier in the preloaded-file list. * help.el: * help.el (help-mode-map): Add bindings to find the source code of a function, notably when that function's in C. * help.el (describe-function-find-file, describe-symbol-find-file): Removed. Use symbol-file from loadhist.el instead. * help.el (frob-help-extents): * help.el (describe-function-1): Allow built-in function file names to be hyperlinks. * help.el (describe-variable): * help.el (help-find-source-or-scroll-up): New. * help.el (help-mouse-find-source-or-track): New. Make describe-function a bit more mouse-friendly, basically. * loadhist.el (symbol-file): Support looking up builtin symbols using built-in-symbol-file. src/ChangeLog addition: 2006-04-29 Aidan Kehoe <kehoea(a)parhasard.net> * doc.c: * doc.c (extract_object_file_name): * doc.c (get_object_file_name): * doc.c (Fbuilt_in_symbol_file): Support saving and recovering the source file name of a built-in symbol (that is, one created in C.) XEmacs Trunk source patch: Diff command: cvs -q diff -u Files affected: src/doc.c lisp/loadhist.el lisp/help.el lisp/dumped-lisp.el lib-src/make-docfile.c Index: lib-src/make-docfile.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lib-src/make-docfile.c,v retrieving revision 1.17 diff -u -u -r1.17 make-docfile.c --- lib-src/make-docfile.c 2005/02/22 08:05:58 1.17 +++ lib-src/make-docfile.c 2006/04/29 16:02:37 ＠＠ -44,6 +44,7 ＠＠ #include <config.h> #include <sysfile.h> +#include <assert.h> #include <stdio.h> #include <stdlib.h> #include <string.h> ＠＠ -58,6 +59,7 ＠＠ ('0' <= c && c <= '9') || \ (c == '_')) +static void put_filename (const char *filename); static int scan_file (const char *filename); static int read_c_string (FILE *, int, int); static void write_c_args (FILE *out, const char *func, char *buf, int minargs, ＠＠ -263,6 +265,30 ＠＠ return err_count > 0; } +/* Add a source file name boundary in the output file. */ +static void +put_filename (const char *filename) +{ + const char *tmp; + + /* Why are we cutting this off? */ + for (tmp = filename; *tmp; tmp++) + { + if (IS_DIRECTORY_SEP(*tmp)) + filename = tmp + 1; + } + + /* <= because sizeof includes the nul byte at the end. Not quite right, + because it should include the length of the symbol + "\037[VF]" instead + of simply 10. */ + assert(sizeof("\037S\n") + strlen(filename) + 10 + <= DOC_MAX_FILENAME_LENGTH); + + putc (037, outfile); + putc ('S', outfile); + fprintf (outfile, "%s\n", filename); +} + /* Read file FILENAME and output its doc strings to outfile. */ /* Return 1 if file is not found, 0 if it is found. */ ＠＠ -864,11 +890,14 ＠＠ if (defunflag || defvarflag || c == '"') { /* XEmacs change: the original code is in the "else" clause */ + /* XXX Must modify the documentation file name code to handle + ELLCCs */ if (ellcc) fprintf (outfile, " CDOC%s(\"%s\", \"\\\n", defvarflag ? "SYM" : "SUBR", globalbuf); else { + put_filename (filename); /* XEmacs addition */ putc (037, outfile); putc (defvarflag ? 'V' : 'F', outfile); fprintf (outfile, "%s\n", globalbuf); ＠＠ -963,6 +992,10 ＠＠ The NAME and DOCSTRING are output. NAME is preceded by `F' for a function or `V' for a variable. An entry is output only if DOCSTRING has \ newline just after the opening " + + Adds the filename a symbol or function was found in before its docstring; + there's no need for this with the load-history available, but we do it for + consistency with the C parsing code. */ static void ＠＠ -1356,6 +1389,7 ＠＠ In the latter case, the opening quote (and leading backslash-newline) have already been read. */ + put_filename (filename); /* XEmacs addition */ putc ('\n', outfile); /* XEmacs addition */ putc (037, outfile); putc (type, outfile); Index: lisp/dumped-lisp.el =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lisp/dumped-lisp.el,v retrieving revision 1.58 diff -u -u -r1.58 dumped-lisp.el --- lisp/dumped-lisp.el 2006/04/23 16:11:22 1.58 +++ lisp/dumped-lisp.el 2006/04/29 16:02:39 ＠＠ -92,6 +92,8 ＠＠ ; `emacs-user-extension-dir' "misc" ;; (pureload "profile") + "loadhist" ; Must be dumped before loaddefs is loaded + ; Used by help. "help" ;; (pureload "hyper-apropos") Soon... "files" ＠＠ -308,7 +310,6 ＠＠ ;; "sun-eos-debugger" ;; "sun-eos-debugger-extra" ;; "sun-eos-menubar")) - "loadhist" ; Must be dumped before loaddefs is loaded "loaddefs" ; <=== autoloads get loaded here )) Index: lisp/help.el =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lisp/help.el,v retrieving revision 1.47 diff -u -u -r1.47 help.el --- lisp/help.el 2005/11/16 12:13:02 1.47 +++ lisp/help.el 2006/04/29 16:02:39 ＠＠ -1,4 +1,4 ＠＠ -;;; help.el --- help commands for XEmacs. +;; help.el --- help commands for XEmacs. ;; Copyright (C) 1985, 1986, 1992-4, 1997 Free Software Foundation, Inc. ;; Copyright (C) 2001, 2002, 2003 Ben Wing. ＠＠ -41,6 +41,8 ＠＠ ;; or run interpreted, but not when the compiled code is loaded. (eval-when-compile (require 'help-macro)) +(require 'loadhist) ;; For symbol-file. + (defgroup help nil "Support for on-line help systems." :group 'emacs) ＠＠ -153,6 +155,8 ＠＠ (define-key help-mode-map "c" 'customize-variable) (define-key help-mode-map [tab] 'help-next-symbol) (define-key help-mode-map [(shift tab)] 'help-prev-symbol) +(define-key help-mode-map [return] 'help-find-source-or-scroll-up) +(define-key help-mode-map [button2] 'help-mouse-find-source-or-track) (define-key help-mode-map "n" 'help-next-section) (define-key help-mode-map "p" 'help-prev-section) ＠＠ -1091,14 +1095,14 ＠＠ :type 'boolean :group 'help-appearance) -(defun describe-symbol-find-file (symbol) - (loop for (file . load-data) in load-history - do (when (memq symbol load-data) - (return file)))) +(define-obsolete-function-alias + ;; Moved to using the version in loadhist.el + 'describe-function-find-symbol + 'symbol-file) (define-obsolete-function-alias 'describe-function-find-file - 'describe-symbol-find-file) + 'symbol-file) (defun describe-function (function) "Display the full documentation of FUNCTION (a symbol). ＠＠ -1340,6 +1344,7 ＠＠ (when (or var fun) (let ((ex (make-extent b e))) (require 'hyper-apropos) + (set-extent-property ex 'mouse-face 'highlight) (set-extent-property ex 'help-symbol sym) (set-extent-property ex 'face 'hyper-apropos-hyperlink) ＠＠ -1421,10 +1426,21 ＠＠ (if autoload-file (princ (format " -- autoloads from \"%s\"\n" autoload-file))) (or file-name - (setq file-name (describe-symbol-find-file function))) - (if file-name - (princ (format " -- loaded from \"%s\"\n" file-name))) -;; (terpri) + (setq file-name (symbol-file function))) + (when file-name + (princ " -- loaded from \"") + (if (not (bufferp standard-output)) + (princ file-name) + (let ((opoint (point standard-output)) + e) + (require 'hyper-apropos) + (princ file-name) + (setq e (make-extent opoint (point standard-output) + standard-output)) + (set-extent-property e 'face 'hyper-apropos-hyperlink) + (set-extent-property e 'mouse-face 'highlight) + (set-extent-property e 'find-function-symbol function))) + (princ "\"\n")) (if describe-function-show-arglist (let ((arglist (function-arglist function))) (when arglist ＠＠ -1469,7 +1485,6 ＠＠ (eq ?\n (aref doc (1- (length doc))))) (terpri))))))))) - ;;; [Obnoxious, whining people who complain very LOUDLY on Usenet ;;; are binding this to keys.] (defun describe-function-arglist (function) ＠＠ -1590,11 +1605,22 ＠＠ (princ (format "%s" aliases))) (princ (built-in-variable-doc variable)) (princ ".\n") - (let ((file-name (describe-symbol-find-file variable))) - (if file-name - (princ (format " -- loaded from \"%s\"\n" file-name)))) - (princ "\nValue: ") (require 'hyper-apropos) + (let ((file-name (symbol-file variable)) + opoint e) + (when file-name + (princ " -- loaded from \"") + (if (not (bufferp standard-output)) + (princ file-name) + (setq opoint (point standard-output)) + (princ file-name) + (setq e (make-extent opoint (point standard-output) + standard-output)) + (set-extent-property e 'face 'hyper-apropos-hyperlink) + (set-extent-property e 'mouse-face 'highlight) + (set-extent-property e 'find-variable-symbol variable)) + (princ"\"\n"))) + (princ "\nValue: ") (if (not (boundp variable)) (Help-princ-face "void\n" 'hyper-apropos-documentation) (Help-prin1-face (symbol-value variable) ＠＠ -1778,5 +1804,29 ＠＠ (if (stringp string) (with-displaying-help-buffer (insert string))))) + +(defun help-find-source-or-scroll-up (&optional pos) + "Follow any cross reference to source code; if none, scroll up. " + (interactive "d") + (let ((e (extent-at pos nil 'find-function-symbol))) + (if e + (find-function (extent-property e 'find-function-symbol)) + (setq e (extent-at pos nil 'find-variable-symbol)) + (if e + (find-variable (extent-property e 'find-variable-symbol)) + (view-scroll-lines-up 1))))) + +(defun help-mouse-find-source-or-track (event) + "Follow any cross reference to source code under the mouse; +if none, call mouse-track. " + (interactive "e") + (mouse-set-point event) + (let ((e (extent-at (point) nil 'find-function-symbol))) + (if e + (find-function (extent-property e 'find-function-symbol)) + (setq e (extent-at (point) nil 'find-variable-symbol)) + (if e + (find-variable (extent-property e 'find-variable-symbol)) + (mouse-track event))))) ;;; help.el ends here Index: lisp/loadhist.el =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lisp/loadhist.el,v retrieving revision 1.6 diff -u -u -r1.6 loadhist.el --- lisp/loadhist.el 2002/11/18 06:52:28 1.6 +++ lisp/loadhist.el 2006/04/29 16:02:39 ＠＠ -41,9 +41,15 ＠＠ "Return the input source from which SYM was loaded. This is a file name, or nil if the source was a buffer with no associated file." (interactive "SFind source file for symbol: ") ; XEmacs - (dolist (entry load-history) - (when (memq sym (cdr entry)) - (return (car entry))))) + (block look-up-symbol-file + (dolist (entry load-history) + (when (memq sym (cdr entry)) + (return-from look-up-symbol-file (car entry)))) + (when (or (and (boundp sym) (built-in-variable-type sym)) + (and (fboundp sym) (subrp (symbol-function sym)))) + (let ((built-in-file (built-in-symbol-file sym))) + (if built-in-file + (concat build-root "/src/" built-in-file)))))) (defun feature-symbols (feature) "Return the file and list of symbols associated with a given FEATURE." Index: src/doc.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/doc.c,v retrieving revision 1.35 diff -u -u -r1.35 doc.c --- src/doc.c 2005/10/25 11:16:22 1.35 +++ src/doc.c 2006/04/29 16:02:40 ＠＠ -39,11 +39,128 ＠＠ Lisp_Object QSsubstitute; -/* Read and return doc string or instructions from open file descriptor FD - at position POSITION. Does not close the file. Returns string; or if - error, returns a cons holding the error data to pass to Fsignal. - NAME_NONRELOC and NAME_RELOC are only used for the error messages. */ +/* Work out what source file a function or variable came from, taking the + information from the documentation file. */ +static Lisp_Object +extract_object_file_name (int fd, EMACS_INT doc_pos, + Ibyte *name_nonreloc, Lisp_Object name_reloc, + int standard_doc_file) +{ + Ibyte buf[DOC_MAX_FILENAME_LENGTH]; + Ibyte *buffer = buf; + int buffer_size = sizeof (buf), space_left; + Ibyte *from, *to; + REGISTER Ibyte *p = buffer; + Lisp_Object return_me; + Lisp_Object fdstream = Qnil, instream = Qnil; + struct gcpro gcpro1, gcpro2; + EMACS_INT position, seenS = 0; + + GCPRO2 (fdstream, instream); + + position = doc_pos > DOC_MAX_FILENAME_LENGTH ? + doc_pos - DOC_MAX_FILENAME_LENGTH : 0; + + if (0 > lseek (fd, position, 0)) + { + if (name_nonreloc) + name_reloc = build_intstring (name_nonreloc); + return_me = list3 (build_msg_string + ("Position out of range in doc string file"), + name_reloc, make_int (position)); + goto done; + } + + fdstream = make_filedesc_input_stream (fd, 0, -1, 0); + Lstream_set_buffering (XLSTREAM (fdstream), LSTREAM_UNBUFFERED, 0); + instream = + make_coding_input_stream + (XLSTREAM (fdstream), standard_doc_file ? Qescape_quoted : Qbinary, + CODING_DECODE, 0); + Lstream_set_buffering (XLSTREAM (instream), LSTREAM_UNBUFFERED, 0); + + space_left = buffer_size - (p - buffer); + while (space_left > 0) + { + int nread; + + nread = Lstream_read (XLSTREAM (instream), p, space_left); + if (nread < 0) + { + return_me = list1 (build_msg_string + ("Read error on documentation file")); + goto done; + } + + p[nread] = 0; + + if (!nread) + break; + + p += nread; + space_left = buffer_size - (p - buffer); + } + + /* First, search backward for the "\037S" that marks the beginning of the + file name, then search forward from that to the newline or to the end + of the buffer. */ + from = p; + + while (from > buf) + { + --from; + if (seenS) + { + if ('\037' == *from) + { + /* Got a file name; adjust `from' to point to it, break out of + the loop. */ + from += 2; + break; + } + } + /* Is *from 'S' ? */ + seenS = ('S' == *from); + } + + if (buf == from) + { + /* We've scanned back to the beginning of the buffer without hitting + the file name. Either the file name plus the symbol name is longer + than DOC_MAX_FILENAME_LENGTH--which shouldn't happen, because it'll + trigger an assertion failure in make-docfile, the DOC file is + corrupt, or it was produced by a version of make-docfile that + doesn't store the file name with the symbol name and docstring. */ + return_me = list1 (build_msg_string + ("Object file name not stored in doc file")); + goto done; + } + + to = from; + /* Search for the end of the file name. */ + while (++to < p) + { + if ('\n' == *to || '\037' == *to) + { + break; + } + } + + /* Don't require the file name to end in a newline. */ + return_me = make_string (from, to - from); + + done: + if (!NILP (instream)) + { + Lstream_delete (XLSTREAM (instream)); + Lstream_delete (XLSTREAM (fdstream)); + } + + UNGCPRO; + return return_me; +} + Lisp_Object unparesseuxify_doc_string (int fd, EMACS_INT position, Ibyte *name_nonreloc, Lisp_Object name_reloc, ＠＠ -287,6 +404,150 ＠＠ return Fread (string); } +static Lisp_Object +get_object_file_name (Lisp_Object filepos) +{ + REGISTER int fd; + REGISTER Ibyte *name_nonreloc = 0; + EMACS_INT position; + Lisp_Object file, tem; + Lisp_Object name_reloc = Qnil; + int standard_doc_file = 0; + + if (INTP (filepos)) + { + file = Vinternal_doc_file_name; + standard_doc_file = 1; + position = XINT (filepos); + } + else if (CONSP (filepos) && INTP (XCDR (filepos))) + { + file = XCAR (filepos); + position = XINT (XCDR (filepos)); + if (position < 0) + position = - position; + } + else + return Qnil; + + if (!STRINGP (file)) + return Qnil; + + /* Put the file name in NAME as a C string. + If it is relative, combine it with Vdoc_directory. */ + + tem = Ffile_name_absolute_p (file); + if (NILP (tem)) + { + Bytecount minsize; + /* XEmacs: Move this check here. OK if called during loadup to + load byte code instructions. */ + if (!STRINGP (Vdoc_directory)) + return Qnil; + + minsize = XSTRING_LENGTH (Vdoc_directory); + /* sizeof ("../lib-src/") == 12 */ + if (minsize < 12) + minsize = 12; + name_nonreloc = alloca_ibytes (minsize + XSTRING_LENGTH (file) + 8); + string_join (name_nonreloc, Vdoc_directory, file); + } + else + name_reloc = file; + + fd = qxe_open (name_nonreloc ? name_nonreloc : + XSTRING_DATA (name_reloc), O_RDONLY | OPEN_BINARY, 0); + if (fd < 0) + { + if (purify_flag) + { + /* sizeof ("../lib-src/") == 12 */ + name_nonreloc = alloca_ibytes (12 + XSTRING_LENGTH (file) + 8); + /* Preparing to dump; DOC file is probably not installed. + So check in ../lib-src. */ + qxestrcpy_ascii (name_nonreloc, "../lib-src/"); + qxestrcat (name_nonreloc, XSTRING_DATA (file)); + + fd = qxe_open (name_nonreloc, O_RDONLY | OPEN_BINARY, 0); + } + + if (fd < 0) + report_file_error ("Cannot open doc string file", + name_nonreloc ? build_intstring (name_nonreloc) : + name_reloc); + } + + tem = extract_object_file_name (fd, position, name_nonreloc, name_reloc, + standard_doc_file); + retry_close (fd); + + if (!STRINGP (tem)) + signal_error_1 (Qinvalid_byte_code, tem); + + return tem; +} + + +static void +weird_doc (Lisp_Object sym, const CIbyte *weirdness, const CIbyte *type, + int pos) +{ + if (!strcmp (weirdness, GETTEXT ("duplicate"))) return; + message ("Note: Strange doc (%s) for %s %s ＠ %d", + weirdness, type, XSTRING_DATA (XSYMBOL (sym)->name), pos); +} + +DEFUN ("built-in-symbol-file", Fbuilt_in_symbol_file, 1, 1, 0, /* +Return the C source file built-in symbol SYM comes from. +Don't use this. Use the more general `symbol-file' (q.v.) instead. +*/ + (symbol)) +{ + /* This function can GC */ + Lisp_Object fun; + Lisp_Object filename = Qnil; + + if (EQ(Ffboundp(symbol), Qt)) + { + fun = Findirect_function (symbol); + + if (SUBRP (fun)) + { + if (XSUBR (fun)->doc == 0) + return Qnil; + + if ((EMACS_INT) XSUBR (fun)->doc >= 0) + { + weird_doc (symbol, "No file info available for function", + GETTEXT("function"), 0); + return Qnil; + } + el + +#ifdef emacs #if defined (WIN32_NATIVE) #define PATHNAME_RESOLVE_LINKS(path, pathout) \ -- In the beginning God created the heavens and the earth. And God was a bug-eyed, hexagonal smurf with a head of electrified hair; and God said: “Si, mi chiamano Mimi...”

2 participants
10 comments

[PATCH] Avoid crashing DEBUG_XEMACS builds when debug-coding-detection is t 18 years, 7 months

Aidan Kehoe

(Yes, this could be done more elegantly.) src/ChangeLog addition: 2006-05-07 Aidan Kehoe <kehoea(a)parhasard.net> * file-coding.c (output_bytes_in_ascii_and_hex): Pass stderr_out internally-formatted data, don't use an external format. XEmacs Trunk source patch: Diff command: cvs -q diff -u Files affected: src/file-coding.c Index: src/file-coding.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/file-coding.c,v retrieving revision 1.52 diff -u -u -r1.52 file-coding.c --- src/file-coding.c 2006/03/26 13:05:17 1.52 +++ src/file-coding.c 2006/05/07 12:49:51 ＠＠ -3370,6 +3370,8 ＠＠ UExtbyte *ascii = alloca_array (UExtbyte, n + 1); UExtbyte *hex = alloca_array (UExtbyte, 3 * n + 1); int i; + DECLARE_EISTRING (eistr_ascii); + DECLARE_EISTRING (eistr_hex); for (i = 0; i < n; i++) { ＠＠ -3384,7 +3386,11 ＠＠ } ascii[i] = '\0'; hex[3 * i - 1] = '\0'; - stderr_out ("%s %s", ascii, hex); + + eicpy_ext(eistr_hex, hex, Qbinary); + eicpy_ext(eistr_ascii, ascii, Qbinary); + + stderr_out ("%s %s", eidata(eistr_ascii), eidata(eistr_hex)); } #endif /* DEBUG_XEMACS */ -- Aidan Kehoe, http://www.parhasard.net/

1 participants
0 comments

[AC (pkgs)] javascript-mode function regexp improvements 18 years, 7 months

Ville Skyttä

APPROVE COMMIT This one has been in my packages tree for a while, forgot to commit it. 2006-05-06 Ville Skyttä <scop(a)xemacs.org> * javascript-mode.el: Improve regexps for finding prototype and property style function declarations (thanks to Tsirkin Evgeny), miscellaneous other cleanups.

1 participants
0 comments

[AC21.5R21.4] Draft fix for metacity maximization bug 18 years, 7 months

Stephen J. Turnbull

APPROVE COMMIT 21.5 RECOMMEND 21.4 Vin, there's a bunch of "future work" cruft here, so I don't necessarily recommend this patch as-is. Also, I haven't actually checked if it applies, and the backwards-compatibility variable might be a little bogus in the stable line. Let me know what, if anything of this you want. Thanks to Glynn Clements, without whose discussion I never would have figured this out, to the good folks at Sun, HP, etc, who funded Real Documentation that Actually Helps, and to the various users who kept reporting the issue even after we thought we (mostly) had it licked. I provided an earlier patch, about which I wrote: sjt> It is possible to inhibit this, with no huge ill-effect on sjt> XEmacs functionality. I'm supplying a patch, with this sjt> action controlled from Lisp by the variable `wedge-metacity', sjt> defaulting to nil. This patch is *superseded* by the following patch, which may not apply to 21.4 (sorry), and if you applied the earlier patch you'll have to revert it (sorry again). If you can apply it and are willing to test it, it should "just work", no need to fiddle with anything. I will work up a 21.4 patch per Vin's comments (if any) within a day or so. And it's being committed to 21.5 immediately. If you run with the patch and have minor problems (but enough that you're willing to avoid maximizing under metacity et al), you can restore the old buggy behavior with (setq wedge-metacity t). I'm pretty sure this patch is "right", although I haven't been able to test with metacity. (1) The Xt conventions say "don't call geometry request functions from the resize callback". This patch moves the request from a resize callback into a function explicitly intended to change size. The theory is right. (2) With the old behavior, the resize callback called itself recursively, at least once. Now it doesn't. (3) The fluxbox window manager now gives "true maximization" where it used to flash to the true maximum, then drop back to the largest cell-oriented window. Index: src/ChangeLog =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/ChangeLog,v retrieving revision 1.954 diff -u -r1.954 ChangeLog --- src/ChangeLog 6 May 2006 08:46:38 -0000 1.954 +++ src/ChangeLog 6 May 2006 17:26:00 -0000 ＠＠ -0,0 +1,26 ＠＠ +2006-04-30 Stephen J. Turnbull <stephen(a)xemacs.org> + + Move geometry management from EmacsFrameResize to x_set_frame_size. + Should fix "metacity maximization" bug. + Provide (deprecated, temporary) backward compatibility option. + + * EmacsFrame.c (EmacsFrameResize): + * frame-x.c (x_set_frame_size): + Move call of ChangeEmacsManagerSize. + Don't bogusly notify WM about size changes the WM asked for. + + * console-x.c (wedge-metacity): New builtin Boolean Lisp variable. + * console-x-impl.h (wedge_metacity): Declare C variable. + * console-x.c (vars_of_console_x): New function to initialize it. + * symsinit.h (vars_of_console_x): declare it. + * emacs.c (main_1): Call vars_of_console_x. + + * EmacsFrameP.h (struct EmacsFrame): + * EmacsManager.c (ChangeEmacsManagerSize): + * EmacsShell-sub.c (SuperClassRootGeometryManager): + * console-x-impl.h (wedge_metacity): + Various comments, some improved documentation, mostly sad comments + on the state of the art of Xt programming. + + * frame-x.c (defi): #undef it. (Random code cleanliness.) + Index: src/EmacsFrame.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/EmacsFrame.c,v retrieving revision 1.28 diff -u -r1.28 EmacsFrame.c --- src/EmacsFrame.c 25 Oct 2005 11:16:20 -0000 1.28 +++ src/EmacsFrame.c 6 May 2006 17:26:00 -0000 ＠＠ -360,19 +360,32 ＠＠ pixel_to_char_size (f, ew->core.width, ew->core.height, &columns, &rows); change_frame_size (f, rows, columns, 0); - /* Now we tell the EmacsShell that we've changed the size of the non-fixed - portion of the frame. Note that, if we the resize occurred as a result - of EmacsFrameSetCharSize(), this information will be stored twice. - This is not a big deal, as storing this information doesn't actually - do anything until the next resize. */ - if (FRAME_X_TOP_LEVEL_FRAME_P (f)) - x_wm_set_variable_size (FRAME_X_SHELL_WIDGET (f), columns, rows); + /* The code below is just plain wrong. If the EmacsShell or EmacsManager + needs to know, they should just ask. If needed information is being + updated here, then we should set a dirty flag and have it updated on an + as-needed basis. + For now, conditionalize so people can get work done if this breaks + something. */ + if (wedge_metacity) /* cf. x_set_frame_size */ + { + /* Now we tell the EmacsShell that we've changed the size of the + non-fixed portion of the frame. Note that, if the resize occurred + as a result of EmacsFrameSetCharSize(), this information will be + stored twice. This is not a big deal, as storing this information + doesn't actually do anything until the next resize. */ + if (FRAME_X_TOP_LEVEL_FRAME_P (f)) + x_wm_set_variable_size (FRAME_X_SHELL_WIDGET (f), columns, rows); - /* Kick the manager so that it knows we've changed size. */ - req.request_mode = 0; - XtQueryGeometry (FRAME_X_CONTAINER_WIDGET (f), &req, &repl); - EmacsManagerChangeSize (FRAME_X_CONTAINER_WIDGET (f), repl.width, - repl.height); + /* Kick the manager so that it knows we've changed size. + #### No, no, no! If this does anything at all, it will involve + changing the manager's size. That's not something that a child + widget should initialize as part of a purely informational + method!! */ + req.request_mode = 0; + XtQueryGeometry (FRAME_X_CONTAINER_WIDGET (f), &req, &repl); + EmacsManagerChangeSize (FRAME_X_CONTAINER_WIDGET (f), + repl.width, repl.height); + } } static Boolean Index: src/EmacsFrameP.h =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/EmacsFrameP.h,v retrieving revision 1.7 diff -u -r1.7 EmacsFrameP.h --- src/EmacsFrameP.h 4 May 2003 02:34:35 -0000 1.7 +++ src/EmacsFrameP.h 6 May 2006 17:26:00 -0000 ＠＠ -65,13 +65,16 ＠＠ Boolean iconic; /* whether this frame is iconic */ /* The rest of this is crap and should be deleted. + #### Comments that start with + are fields that actually get referred + to somewhere aside from the init function. + I guess the "crap" has mostly moved to specifiers? */ Boolean minibuffer; /* 0: normal frames with minibuffers. * 1: frames without minibuffers * 2: minibuffer only. */ Boolean unsplittable; /* frame can only have one window */ - int internal_border_width; /* internal borders */ + int internal_border_width; /* + internal borders */ int scrollbar_width; /* width of frame vertical sb's */ int scrollbar_height; /* height of frame horizontal sb's */ int top_toolbar_height; /* height of top toolbar */ ＠＠ -82,9 +85,9 ＠＠ int bottom_toolbar_border_width; /* ... of bottom toolbar */ int left_toolbar_border_width; /* ... of left toolbar */ int right_toolbar_border_width; /* ... of right toolbar */ - Dimension toolbar_shadow_thickness; + Dimension toolbar_shadow_thickness; /* + of shadows */ unsigned char scrollbar_placement; - int interline; /* skips between lines */ + int interline; /* + skips between lines */ XFontStruct* font; /* font */ Pixel foreground_pixel; /* foreground */ ＠＠ -97,7 +100,7 ＠＠ int bell_volume; /* how loud is beep */ Boolean menubar_p; /* initially show a menubar? */ - Boolean initially_unmapped; /* inhibit initial window mapping */ + Boolean initially_unmapped; /* + inhibit initial window mapping */ Boolean use_backing_store; /* backing store for menubar & ew? */ Dimension preferred_width; /* if non-zero, preferred size for */ Index: src/EmacsManager.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/EmacsManager.c,v retrieving revision 1.7 diff -u -r1.7 EmacsManager.c --- src/EmacsManager.c 22 Dec 2004 10:59:09 -0000 1.7 +++ src/EmacsManager.c 6 May 2006 17:26:00 -0000 ＠＠ -241,6 +241,12 ＠＠ if (height == 0) height = w->core.height; + /* #### AFAICT this gets called in two places. One is in ChangeManaged(), + above. The other is in EmacsFrameResize(). Perhaps ChangeManaged() + should initiate resize requests, but EmacsFrameResize() should not. + Unfortunately, I've tried making this conditional on whether we're + called from EmacsFrameResize() or not, but that results in an infloop + via the callback to x_layout_widgets() in Resize(). Whee! */ /* do nothing if we're already that size */ if (w->core.width != width || w->core.height != height) { Index: src/EmacsShell-sub.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/EmacsShell-sub.c,v retrieving revision 1.7 diff -u -r1.7 EmacsShell-sub.c --- src/EmacsShell-sub.c 24 Jan 2005 23:33:46 -0000 1.7 +++ src/EmacsShell-sub.c 6 May 2006 17:26:00 -0000 ＠＠ -278,7 +278,8 ＠＠ GenericClassExtRec *gcer; /* find the shell extension record that specifies the - root geometry manager method */ + root geometry manager method + #### We could use XtGetClassExtension here. */ for (gcer = (GenericClassExtRec *) swc->shell_class.extension; gcer; gcer = (GenericClassExtRec *) gcer->next_extension) ＠＠ -287,6 +288,9 ＠＠ break; } + /* #### The R11.6.4 Xt specification says if we don't find NULLQUARK here, + we should assume root_geometry_manager = XtInheritRootGeometryManager. + Is that actually callable? */ if (!gcer) ABORT (); Index: src/console-x-impl.h =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/console-x-impl.h,v retrieving revision 1.5 diff -u -r1.5 console-x-impl.h --- src/console-x-impl.h 26 Nov 2005 11:46:07 -0000 1.5 +++ src/console-x-impl.h 6 May 2006 17:26:00 -0000 ＠＠ -40,6 +40,8 ＠＠ DECLARE_CONSOLE_TYPE (x); +extern int wedge_metacity; + struct x_device { #ifdef NEW_GC ＠＠ -234,7 +236,8 ＠＠ /* The maximum number of widgets that can be displayed above the text area at one time. Currently no more than 3 will ever actually be - displayed (menubar, psheet, debugger panel). */ + displayed (menubar, psheet, debugger panel). + #### Are "psheet" and "debugger panel" relevant any more? */ #define MAX_CONCURRENT_TOP_WIDGETS 8 struct x_frame ＠＠ -243,12 +246,17 ＠＠ struct lrecord_header header; #endif /* NEW_GC */ - /* The widget of this frame. This is an EmacsShell or an - ExternalShell. */ + /* The widget of this frame. + This is an EmacsShell or an ExternalShell. + It negotiates with the window manager or containing app on behalf of + the container widget. Should be (but isn't) invisible to Emacs. */ Widget widget; /* The parent of the EmacsFrame, the menubar, and the scrollbars. - This is an EmacsManager. */ + This is an EmacsManager. + It is responsible for managing the geometry of the frame. This is what + Emacs mostly talks to. Anything that affects its geometry will be + reflected in the Shell widget, and thus cause WM interaction. */ Widget container; /* The widget of the menubar, of whatever widget class it happens to be. */ Index: src/console-x.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/console-x.c,v retrieving revision 1.16 diff -u -r1.16 console-x.c --- src/console-x.c 17 Dec 2005 19:47:02 -0000 1.16 +++ src/console-x.c 6 May 2006 17:26:00 -0000 ＠＠ -41,6 +41,8 ＠＠ DEFINE_CONSOLE_TYPE (x); +int wedge_metacity; /* nonzero means update WM_HINTS always */ + extern void x_has_keysym (KeySym, Lisp_Object, int); static int ＠＠ -399,6 +401,22 ＠＠ CONSOLE_HAS_METHOD (x, perhaps_init_unseen_key_defaults); } + +void +vars_of_console_x (void) +{ + DEFVAR_BOOL ("wedge-metacity", &wedge_metacity /* +When non-nil, frame geometry management is backward-compatible. +This is known to create inflooping window jitter in metacity, et al. +It also does not conform to Xt conventions for geometry management. +Specifically, all frame resizes, XEmacs-initiated or not, update WM_HINTS. +Furthermore, geometry changes occur in the widget resize method. + +The default is nil. This probably gives correct behavior regardless of the +window manager used. +This variable is deprecated and will be removed. +*/ ); +} void reinit_console_type_create_x (void) Index: src/emacs.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/emacs.c,v retrieving revision 1.166 diff -u -r1.166 emacs.c --- src/emacs.c 25 Apr 2006 14:02:08 -0000 1.166 +++ src/emacs.c 6 May 2006 17:26:01 -0000 ＠＠ -2186,6 +2186,7 ＠＠ #ifdef HAVE_BALLOON_HELP vars_of_balloon_x (); #endif + vars_of_console_x (); vars_of_device_x (); #ifdef HAVE_X_DIALOGS vars_of_dialog_x (); Index: src/frame-x.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/frame-x.c,v retrieving revision 1.72 diff -u -r1.72 frame-x.c --- src/frame-x.c 26 Nov 2005 11:46:08 -0000 1.72 +++ src/frame-x.c 6 May 2006 17:26:01 -0000 ＠＠ -507,6 +507,7 ＠＠ defi(Qtop, XtNy); #undef def +#undef defi } static Lisp_Object ＠＠ -2307,6 +2308,17 ＠＠ x_set_frame_size (struct frame *f, int cols, int rows) { EmacsFrameSetCharSize (FRAME_X_TEXT_WIDGET (f), cols, rows); + + if (!wedge_metacity) /* cf. EmacsFrameResize */ + { + /* Kick the manager so that it knows we've changed size. */ + XtWidgetGeometry req, repl; + req.request_mode = 0; + XtQueryGeometry (FRAME_X_CONTAINER_WIDGET (f), &req, &repl); + EmacsManagerChangeSize (FRAME_X_CONTAINER_WIDGET (f), + repl.width, repl.height); + } + #if 0 /* this is not correct. x_set_frame_size() is called from Fset_frame_size(), which may or may not have been called Index: src/symsinit.h =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/symsinit.h,v retrieving revision 1.57 diff -u -r1.57 symsinit.h --- src/symsinit.h 25 Apr 2006 14:02:09 -0000 1.57 +++ src/symsinit.h 6 May 2006 17:26:01 -0000 ＠＠ -306,6 +306,7 ＠＠ void vars_of_console_stream (void); void vars_of_console_mswindows (void); void vars_of_console_tty (void); +void vars_of_console_x (void); void vars_of_data (void); void vars_of_database (void); void vars_of_debug (void); -- School of Systems and Information Engineering http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Ask not how you can "do" free software business; ask what your business can "do for" free software.

1 participants
1 comments

Re: [Failure] XEmacs 21.5-b26 "endive" (+CVS-20060430) i586-pc-win32 18 years, 7 months

Aidan Kehoe

Ar an chéad lá de mí Bealtaine, scríobh Adrian Aichner: > Doesn't help. > > I build on Windows XP using xemacs.mak, not configure && make. This does (should be applied from with src/, I don’t have patcher to hand on the XP machine). I intend committing a version of it tomorrow. diff -u sysfile.h~ sysfile.h --- sysfile.h~ Sat Apr 29 18:15:32 2006 +++ sysfile.h Fri May 05 16:36:07 2006 ＠＠ -467,6 +467,7 ＠＠ #define SEPCHAR ';' #define DEFAULT_DIRECTORY_SEP '\\' +#ifdef emacs DECLARE_INLINE_HEADER (Ibyte sysfile_get_directory_sep (void)) { if (!CHARP (Vdirectory_sep_char) ＠＠ -484,6 +485,12 ＠＠ } #define DIRECTORY_SEP sysfile_get_directory_sep() +#else /* emacs */ + +#define DIRECTORY_SEP DEFAULT_DIRECTORY_SEP + +#endif /* emacs */ + #else /* not WIN32_NATIVE */ #define SEPCHAR ':' ＠＠ -499,6 +506,8 ＠＠ #define IS_DEVICE_SEP(c) ((c) == DEVICE_SEP) +#ifdef emacs + DECLARE_INLINE_HEADER (int IS_DIRECTORY_SEP (Ichar c)) { return (c == '/' || c == '\\'); ＠＠ -508,6 +517,20 ＠＠ { return (c == '/' || c == '\\' || c == ':'); } + +#else /* emacs */ + +DECLARE_INLINE_HEADER (int IS_DIRECTORY_SEP (int c)) +{ + return (c == '/' || c == '\\'); +} + +DECLARE_INLINE_HEADER (int IS_ANY_SEP (int c)) +{ + return (c == '/' || c == '\\' || c == ':'); +} + +#endif #else /* not WIN32_ANY */ diff -u symbols.c~ symbols.c --- symbols.c~ Sat Apr 29 18:15:30 2006 +++ symbols.c Fri May 05 16:30:02 2006 ＠＠ -729,7 +729,7 ＠＠ if (!SUBRP (subr)) wrong_type_argument (Qsubrp, subr); name = XSUBR (subr)->name; - return make_string (name, strlen (name)); + return make_string ((const Ibyte *)name, strlen (name)); } DEFUN ("setplist", Fsetplist, 2, 2, 0, /* However, the build fails later with this: NMAKE: fatal error U1095: expanded command line 'echo link -nologo -debug -opt:noref -incremental:no -base:0x1000000 -stack:0x800000 -entry:mainCRTStartup -subsystem:windows -heap:0x00100000 -nodefaultlib setargv.obj -pdb:C:\xemacs-21.5\src\temacs.pdb -map:C:\xemacs-21.5\src\temacs.map -out:C:\xemacs-21.5\src\temacs.exe C:\xemacs-21.5\nt\obj\console-msw.obj C:\xemacs-21.5\nt\obj\device-msw.obj C:\xemacs-21.5\nt\obj\event-msw.obj C:\xemacs-21.5\nt\obj\frame-msw.obj C:\xemacs-21.5\nt\obj\glyphs-msw.obj C:\xemacs-21.5\nt\obj\gui-msw.obj C:\xemacs-21.5\nt\obj\objects-msw.obj C:\xemacs-21.5\nt\obj\redisplay-msw.obj C:\xemacs-21.5\nt\obj\select-msw.obj C:\xemacs-21.5\nt\obj\dired-msw.obj C:\xemacs-21.5\nt\obj\menubar.obj C:\xemacs-21.5\nt\obj\menubar-msw.obj C:\xemacs-21.5\nt\obj\scrollbar.obj C:\xemacs-21.5\nt\obj\scrollbar-msw.obj C:\xemacs-21.5\nt\obj\toolbar.obj C:\xemacs-21.5\nt\obj\toolbar-msw.obj C:\xemacs-21.5\nt\obj\dialog.obj C:\xemacs-21.5\nt\obj\dialog-m! sw.obj C:\xemacs-21.5\nt\obj\dgif_lib.obj C:\xemacs-21.5\nt\obj\gif_io.obj C:\xemacs-21.5\nt\obj\mule-ccl.obj C:\xemacs-21.5\nt\obj\mule-charset.obj C:\xemacs-21.5\nt\obj\mule-coding.obj C:\xemacs-21.5\nt\obj\debug.obj C:\xemacs-21.5\nt\obj\tests.obj C:\xemacs-21.5\nt\obj\dumper.obj C:\xemacs-21.5\nt\obj\abbrev.obj C:\xemacs-21.5\nt\obj\alloc.obj C:\xemacs-21.5\nt\obj\alloca.obj C:\xemacs-21.5\nt\obj\blocktype.obj C:\xemacs-21.5\nt\obj\buffer.obj C:\xemacs-21.5\nt\obj\bytecode.obj C:\xemacs-21.5\nt\obj\callint.obj C:\xemacs-21.5\nt\obj\casefiddle.obj C:\xemacs-21.5\nt\obj\casetab.obj C:\xemacs-21.5\nt\obj\chartab.obj C:\xemacs-21.5\nt\obj\cmdloop.obj C:\xemacs-21.5\nt\obj\cmds.obj C:\xemacs-21.5\nt\obj\console-stream.obj C:\xemacs-21.5\nt\obj\console.obj C:\xemacs-21.5\nt\obj\data.obj C:\xemacs-21.5\nt\obj\device.obj C:\xemacs-21.5\nt\obj\dired.obj C:\xemacs-21.5\nt\obj\doc.obj C:\xemacs-21.5\nt\obj\doprnt.obj C:\xemacs-21.5\nt\obj\dragdrop.obj C:\xema! cs-21.5\nt\obj\dynarr.obj C:\xemacs-21.5\nt\obj\editfns.obj ! C:\xemac s-21.5\nt\obj\elhash.obj C:\xemacs-21.5\nt\obj\emacs.obj C:\xemacs-21.5\nt\obj\emodules.obj C:\xemacs-21.5\nt\obj\eval.obj C:\xemacs-21.5\nt\obj\event-stream.obj C:\xemacs-21.5\nt\obj\events.obj C:\xemacs-21.5\nt\obj\extents.obj C:\xemacs-21.5\nt\obj\faces.obj C:\xemacs-21.5\nt\obj\file-coding.obj C:\xemacs-21.5\nt\obj\fileio.obj C:\xemacs-21.5\nt\obj\filemode.obj C:\xemacs-21.5\nt\obj\floatfns.obj C:\xemacs-21.5\nt\obj\fns.obj C:\xemacs-21.5\nt\obj\font-lock.obj C:\xemacs-21.5\nt\obj\frame.obj C:\xemacs-21.5\nt\obj\gc.obj C:\xemacs-21.5\nt\obj\general.obj C:\xemacs-21.5\nt\obj\getloadavg.obj C:\xemacs-21.5\nt\obj\glyphs.obj C:\xemacs-21.5\nt\obj\glyphs-eimage.obj C:\xemacs-21.5\nt\obj\glyphs-shared.obj C:\xemacs-21.5\nt\obj\glyphs-widget.obj C:\xemacs-21.5\nt\obj\gui.obj C:\xemacs-21.5\nt\obj\gutter.obj C:\xemacs-21.5\nt\obj\hash.obj C:\xemacs-21.5\nt\obj\indent.obj C:\xemacs-21.5\nt\obj\imgproc.obj C:\xemacs-21.5\nt\obj\insdel.obj C:\xemacs-21.5! \nt\obj\intl.obj C:\xemacs-21.5\nt\obj\intl-win32.obj C:\xemacs-21.5\nt\obj\intl-encap-win32.obj C:\xemacs-21.5\nt\obj\intl-auto-encap-win32.obj C:\xemacs-21.5\nt\obj\keymap.obj C:\xemacs-21.5\nt\obj\libinterface.obj C:\xemacs-21.5\nt\obj\line-number.obj C:\xemacs-21.5\nt\obj\lread.obj C:\xemacs-21.5\nt\obj\lstream.obj C:\xemacs-21.5\nt\obj\macros.obj C:\xemacs-21.5\nt\obj\marker.obj C:\xemacs-21.5\nt\obj\md5.obj C:\xemacs-21.5\nt\obj\minibuf.obj C:\xemacs-21.5\nt\obj\nt.obj C:\xemacs-21.5\nt\obj\ntplay.obj C:\xemacs-21.5\nt\obj\objects.obj C:\xemacs-21.5\nt\obj\opaque.obj C:\xemacs-21.5\nt\obj\print.obj C:\xemacs-21.5\nt\obj\process.obj C:\xemacs-21.5\nt\obj\process-nt.obj C:\xemacs-21.5\nt\obj\profile.obj C:\xemacs-21.5\nt\obj\rangetab.obj C:\xemacs-21.5\nt\obj\realpath.obj C:\xemacs-21.5\nt\obj\redisplay-output.obj C:\xemacs-21.5\nt\obj\redisplay.obj C:\xemacs-21.5\nt\obj\regex.obj C:\xemacs-21.5\nt\obj\search.obj C:\xemacs-21.5\nt\obj\select.! obj C:\xemacs-21.5\nt\obj\signal.obj C:\xemacs-21.5\nt\obj\s! ound.obj C:\xemacs-21.5\nt\obj\specifier.obj C:\xemacs-21.5\nt\obj\strftime.obj C:\xemacs-21.5\nt\obj\symbols.obj C:\xemacs-21.5\nt\obj\syntax.obj C:\xemacs-21.5\nt\obj\sysdep.obj C:\xemacs-21.5\nt\obj\text.obj C:\xemacs-21.5\nt\obj\tparam.obj C:\xemacs-21.5\nt\obj\undo.obj C:\xemacs-21.5\nt\obj\unicode.obj C:\xemacs-21.5\nt\obj\widget.obj C:\xemacs-21.5\nt\obj\window.obj C:\xemacs-21.5\nt\obj\win32.obj C:\xemacs-21.5\nt\obj\dump-id.obj comctl32.lib "c:\src\xpm-3.4k\lib\Xpm.lib" "c:\src\libpng-1.2.8\libpng.lib" "c:\src\zlib-1.2.3\zlib.lib" "c:\src\tiff-3.7.3\libtiff\libtiff.lib" "c:\src\jpeg-6b\libjpeg.lib""c:\src\compface-1.5.1\libcompface.lib" "c:\src\zlib-1.2.3\zlib.lib" oldnames.lib kernel32.lib user32.lib gdi32.lib comdlg32.lib advapi32.lib shell32.lib wsock32.lib netapi32.lib winmm.lib winspool.lib ole32.lib mpr.lib uuid.lib imm32.lib msvcrtd.lib' too long -- Aidan Kehoe, http://www.parhasard.net/

2 participants
2 comments

User Verification Request 18 years, 7 months

tetsuya＠pspinc.com

Dear xemacs-patches(a)xemacs.org The email address, tetsuya(a)pspinc.com, that you have sent is protected by white list service. Please clik below URL, and register your email address in my white list. http://whitelist.dreamersi.net/white_list/index.php?act=top&user_id=tetsu... Thank you for your time and cooperation. Tetsuya Sano Pacific Software Publishing, Inc.

1 participants
0 comments

[PATCH (pkgs)] Package compilation failure due to \U 18 years, 7 months

Jerry James

PATCH packages After the last patch, the package compilation went on for quite some time, then failed again with a non-hex digit in \U message. This patch fixes the problem for me. mule-packages/latin-unity/ChangeLog addition: 2006-05-02 Jerry James <james(a)xemacs.org> * latin-unity.el (latin-unity-list-coding-systems): Change malformed \U to \nU as it should have been. packages source patch: Diff command: cvs -q diff -uN Files affected: mule-packages/latin-unity/latin-unity.el Index: mule-packages/latin-unity/latin-unity.el =================================================================== RCS file: /pack/xemacscvs/XEmacs/packages/mule-packages/latin-unity/latin-unity.el,v retrieving revision 1.13 diff -d -u -r1.13 latin-unity.el --- mule-packages/latin-unity/latin-unity.el 2005/12/23 17:32:50 1.13 +++ mule-packages/latin-unity/latin-unity.el 2006/05/02 17:00:16 ＠＠ -337,7 +337,7 ＠＠ latin-unity-preferred-coding-system-list) (fill-region start (point)) - (insert "\Universal coding systems:\n ") + (insert "\nUniversal coding systems:\n ") (setq start (point)) (mapc (lambda (codesys) (insert (format " %s" codesys))) latin-unity-ucs-list) -- Jerry James, Assistant Professor james(a)xemacs.org Computer Science Department http://www.cs.usu.edu/~jerry/ Utah State University

1 participants
1 comments

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

XEmacs-Patches May 2006