Re: New CVS snapshot 20000213

From: Petter Reinholdtsen (pere@hungry.com)
Date: Sat Feb 19 2000 - 01:38:45 PST

  • Next message: John Vickers: "Re: Xsane-0.55: scan resolution selection (Epson 1200S, CVS Sane)"

    [Henning Meier-Geinitz]
    > glib version is 1.2.3, I think it was compiled with some sort of
    > gcc. /usr/local/include/glib.h looks like this (line numers
    > included by me)
    >
    > [...]
    > 1541 /* Bit tests
    > 1542 */
    > 1543 G_INLINE_FUNC gint g_bit_nth_lsf (guint32 mask,
    > 1544 gint nth_bit);
    > [...]
    >
    > Any idea? BTW I don't know how to check the version of the C
    > compiler (it's the one SGI provided with Irix 5.3). The obvious
    > options for cc didn't do it.

    I have two suggestions, both seem quite unlikely. One is that the
    wrong include file is is /usr/lcoa/lib/glib/, making gint, guint and
    all the other glib types unknown. The other is that 'inline' is a
    unknown keyword and that G_INLINE_FUNC should be defined to '' or
    __inline.

    Try compiling with '-DG_INLINE_FUNC=__inline' or without any content
    to see if the inline keyword is making it choke. Try to compile with
    -E to check the output after preprosessing, to check if the glib types
    are typedef'ed.

    This seem more like a problem with glib then with SANE, but I guess we
    should detect it to avoid using glib if it is broken.

    Why do saned use glib by the way? Anyone know?

    -- 
    ##>  Petter Reinholdtsen  <##  |  pere@td.org.uit.no
    

    -- Source code, list archive, and docs: http://www.mostang.com/sane/ To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com



    This archive was generated by hypermail 2b29 : Sat Feb 19 2000 - 01:36:52 PST