home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cs.utexas.edu!zaphod.mps.ohio-state.edu!usc!elroy.jpl.nasa.gov!ames!agate!EXPO.LCS.MIT.EDU!gildea
- From: gildea@EXPO.LCS.MIT.EDU (Stephen Gildea)
- Newsgroups: comp.archives
- Subject: [comp.windows.x.announce] X11R5 public fix #13 available
- Followup-To: comp.windows.x.announce
- Date: 24 Jul 1992 10:18:10 GMT
- Organization: X Consortium, MIT Laboratory for Computer Science
- Lines: 80
- Approved: adam@soda.berkeley.edu
- Distribution: world
- Message-ID: <14old2INN8no@agate.berkeley.edu>
- References: <9207211733.AA23976@expo.lcs.mit.edu>
- NNTP-Posting-Host: soda.berkeley.edu
- X-Original-Newsgroups: comp.windows.x.announce
- X-Original-Date: 21 Jul 92 17:33:43 GMT
-
- Archive-name: auto/comp.windows.x.announce/X11R5-public-fix-13-available
-
- Patches for a variety of font support problems in R5 are now available
- via anonymous ftp on export.lcs.mit.edu and via the xstuff mail archive
- server on expo.lcs.mit.edu. Instructions for applying the fix are
- included in the patch file itself.
-
- Fixes are available via anonymous ftp to export.lcs.mit.edu (18.24.0.12),
- in the /pub/R5/fixes/ directory. The file "fix-13" contains this fix.
- Fixes usually propagate to other distribution sites as well, so it may
- pay to check at a nearer site first.
-
- For those without ftp access, individual fixes can be obtained by mail by
- sending a message to xstuff@expo.lcs.mit.edu (18.24.0.11). In the usual
- case, the message should have a subject line and no body, or a single-line
- body and no subject. Fix 13 from xstuff is multi-part, you need to ask for
- fixes "13a", "13b", and "13c" individually, e.g.,
- send fixes 13a
- to get part 13a. Concatenate all three parts together before applying.
- To get a summary of fixes, send the line:
- index fixes
- If you need help, send the line:
- help
- Some mailers produce mail headers that are unusable for extracting return
- addresses. If you use such a mailer, you won't get any response. If you
- happen to know an explicit path, you can include a line like
- path foo%bar.bitnet@mitvma.mit.edu
- or
- path zot!gzork!me@uunet.uu.net
- in the body of your message, and the daemon will use it.
-
-
- Fix 13 corrects some serious Font Service protocol errors in both the
- font server and font client libraries. Because these fixes change the
- implemented protocol in an incompatible way, the version number is
- changed to 2.0. The old, broken implementation is still supported as
- version 1; new clients can talk to old font servers, and vice versa.
-
- These protocol fixes and other bug fixes are detailed below:
-
- FS protocol: fontserver and fontlib have fsChar2B backwards
- FS protocol: ListFontsWithXInfo reply incorrectly formatted
- FS protocol: fs returns font metrics rather than ink metrics for some fonts
- FS protocol: ListFontsWithXInfo reply length incorrect
- FS protocol: ListFontsWithXInfo last reply length incorrect
- FS protocol: fs can return -1 authorization index
- FS protocol: auth_len is miscalculated in initial connection
- config: bdftopcf and mkfontdir not made if BuildFonts is NO
- fontlib: fs crashes if filename suggests scalable but XLFD name does not
- fontlib: bdf file with empty XLFD field crashes fs
- fontlib: bdf reader doesn't check for end of file
- fontlib: pcf reader frees uninitialized pointer on error
- fontlib: "byte" used as a variable name
- fontlib: fs_send_load_bitmap uses QueryXBitamp8 with 16-bit encoded character
- fontlib: attempt to access fonts through chained font servers fails
- fontlib: Speedo reader frees uninitialized pointer on error
- fontlib: Speedo generation fails at tiny sizes
- fontlib: ListFontsWithXInfo reply incorrectly parsed
- fontlib: libFS tries to read extension data twice
- fontlib: X server can't detect when font server has died
- fontlib: memory leaks
- fontserver: fs dies if garbage characters fed to its socket
- fontserver: fs doesn't manage signals correctly
- fontserver: syntax errors in osdep.h
- fontserver: fs dispatch routines poorly named
- fontserver: parameters swapped on call to WriteSwappedDataToClient
- fontserver: some error states return the same value as FSSuccess
- fontserver: fs can dereference a bogus pointer when reseting
- fontserver: ListFonts can attempt to free nonexistent memory
- fontserver: fs can dump core when chaining fontservers
- fontserver: when no font path elements, ListFonts sends no reply
- fontserver: num_range not byte-swapped for QueryXExtents and QueryXBitmaps
- fontserver: QueryExtension request doesn't get byte swapped
- fontserver: memory leaks
- FS protocol document: changed to read version 2.0
- showfont: handles nonexistent characters incorrectly
- showfont: incorrectly generates 2-byte ranges
- showfont: bitmap format is not always passed to the font server correctly
- misc fonts: some Scandinavian characters wrong or poorly-formed
-
-