bug-grub
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Bugs in Grub documentation


From: Ralf Medow
Subject: Bugs in Grub documentation
Date: Sun, 03 Dec 2000 04:01:36 +0100

What I wanted to do:
--------------------
Print the grub manual because while testing a boot loader the online
manual is not accessible :-) However, TeX complained about the files.


Version information:
--------------------
Texi-files extracted from grub-0_5_96_1_tar.gz, texi-file-dates 16. Oct.
2000 nearly 5pm
This is emTeX, Version 3.0 [3a] (preloaded format=plain 92.6.8)
texinfo.tex Loading texinfo package [Version 2.185]


What happened and how (some of) the problems can be fixed:
----------------------------------------------------------

During processing of grub.texi:

        ! Undefined control sequence.
        l.84 @ifnottex

As a matter of fact, @ifnottex is not defined in texinfo.tex.
Removing the block bracketed by @ifnottex ... @end ifnottex helps for
getting the file thru TeX but is not a general fix.


During processing of user-ref.texi:

! Undefined control sequence.
l.136 ...either using the utility @command
                                          {grub-install}
! Undefined control sequence.
l.284 @option
             {--root-directory}. The typical usage is that you create...

As a matter of fact, neither @command nor @option are defined in
texinfo.tex.

Fix: Add the following 4 lines after the first line of grub.texi (\input
texinfo):
        @iftex
        @address@hidden@samp
        @address@hidden@samp
        @end iftex
This helps for TeX and shouldn't hurt if the file is processed by
anything other than TeX.
Rather than @samp anything else might be used, of course. 
As in newer versions of texinfo.tex @command and @option might be
defined, it would be nicer to @let @command and @option only if it's not
defined yet, but I don't know enough about TeX to figure out how that
could be done.


During processing of prog-ref.texi:

        ! Illegal unit of measure (pt inserted).
        <to be read again>
                           .
        @col1 ->..
                  address@hidden
        <template> address@hidden @colcount @endcsname
                                                  @ifnum @colcount =1 @else
@...
        <to be read again>
                           @global
        @next ...ndent address@hidden address@hidden @global
                                                  @everycr address@hidden
address@hidden
        l.434 ...le @columnfractions 0.15 0.25 0.6

It turns out that @columnfractions doesn't like leading 0's in its
arguments, thus changing 
        l.434 ...le @columnfractions 0.15 0.25 0.6
to      l.434 ...le @columnfractions .15 .25 .6
in prog-ref.texi helps. It not only has to be done in l.434 but in 18
places altogether (out of 27 ocurrences of @columnfractions).


With these changes grub.texi passes thru TeX without errors.


Looking at the result I noticed three points worth mentioning:
--------------------------------------------------------------

- There are strange pagebreaks in tables (prog-ref section); it appears 
  that TeX avoids pagebreaks at the end of @multitable and rather 
  inserts a pagebreak before the last line of such a table, even if that 
  means that there are 3 rows of a table at the top of a page, the rest 
  of the page left blank, and the last line of the table placed on the 
  next page just above the header of the next chapter. This is probably 
  a problem with the definition of @multitable in texinfo.tex.

- To two entries in the list of partition type codes (chapter 27.2) 
  footnotes are attached, the second one to 83 Linux native stating 
  "This is not true. Use 83 for ext2fs even if the owner OS is 
  GNU/Hurd." This text seems more appropiate to the first footnote 
  attached to 63 GNU Hurd. 

- According to chapter 13.2 there are commands part_new and part_type. 
  According to the online-help (and what is really functioning) the 
  commands are partnew and parttype (without underscore).


Best regards
Ralf Medow




reply via email to

[Prev in Thread] Current Thread [Next in Thread]