dblfloatfix bug
mattias...@googlemail.com
I've had a problem with the figure placement a paper I've written. I
had loaded the dblfloatfix package in order to allow for twocolumn
floats at the bottom of the page. The result was that ordinary single-
column floats had sometimes also treated as twocolumn floats, thus
producing large empty areas next to them. I took me quite some time to
hunt down the bug, so I thought I'd share the resolution. It turned
out the dblfloatfix.sty was missing a check for twocolumn floats which
resulted in single-line bottom floats to be treated as twocolumn
floats. A corrected version of dblfloatfix is reproduced below, I've
commented the added check.
Btw. Whom should I contact to merge this fix back into the original
dblfloatfix package?
Cheers,
Mattias
%% This is the LaTeX package dblfloatfix which solves two problems:
%% a)~floats in a twocolumn document come out in the right order and
%% b)~allowed float positions are now [tbp].
%%
%% It may be distributed and/or modified under the
%% conditions of the LaTeX Project Public License, either version 1.2
%% of this license or (at your option) any later version.
%% The latest version of this license is in
%% http://www.latex-project.org/lppl.txt
\ProvidesPackage{dblfloatfix}[2003/11/18 v1.0 (MH)]
%%
%% USAGE:
%%
%% This package redefines internals of the LaTeX kernel by merging
%% the code of the packages fixltx2e (the fix2col part) and stfloats.
%%
%% If either of the packages stfloats or fix2col are loaded you will
%% get an error message and your document will probably not be able
%% to compile. So don't load them!
%%
%% When the new output routine for LaTeX3 is done, this package will
%% be obsolete. The sooner the better...
%%
%% If you're using the subfigure package this package will provide a
%% fix for \end@dblfloat by Steven Cochran (the author of subfigure).
%%
%% Morten H\o gholm, <moho...@student.cbs.dk>
\RequirePackage{fixltx2e}
%% An error message for those who didn't read the instructions.
\newcommand*{\DBLFIXerrormessage}[1]{%
\PackageError{dblfloatfix}{%
Don't use the #1 package!^^J%
Didn't you read the instructions in dblfloatfix.sty?!^^J%
You're in big trouble now...^^J%
Quit immediately (type `X') and correct the mistake}%
{Type `X' dammit!}%
}
%% A check for those users who don't read manuals.
\AtBeginDocument{%
\@ifpackageloaded{stfloats}{\DBLFIXerrormessage{stfloats}}{}%
\@ifpackageloaded{fix2col}{\DBLFIXerrormessage{fix2col}}{}%
%% Here's a fix by Steven Douglas Cochran (the author of subfigure).
%% It is needed in order to make subfigure work with this package as
%% both packages redefine \end@dblfloat.
\@ifpackageloaded{subfigure}{%
\typeout{Redefining \string\end@dblfloat\ for use with subfigure}%
\def\end@dblfloat{%
\if@twocolumn
\@listsubcaptions{\@captype}%
\@endfloatbox
\ifnum\@floatpenalty <\z@
\@largefloatcheck
\global\dp\@currbox1sp %
\expandafter\@gobble\subfig@end@float
\fi
\else
\end@float
\fi}%
}{}%
}
%% Here we go...
\def\@dblbotlist{}
\newcount\@dblbotnum
\newcounter{dblbotnumber}
\setcounter{dblbotnumber}{2}
\def\@dbflt#1{\@ifnextchar[{\@xdblfloat{#1}}{\@xdblfloat{#1}[tbp]}}
\def \@dblfloatplacement {%
\global \@dbltopnum \c@dbltopnumber
\global \@dblbotnum \c@dblbotnumber
\global \@dbltoproom \dbltopfraction\@colht
\@textmin \@colht
\advance \@textmin -\@dbltoproom
\@fpmin \dblfloatpagefraction\textheight
\@fptop \@dblfptop
\@fpsep \@dblfpsep
\@fpbot \@dblfpbot
\def\f@depth{1sp}%
}
\def\@addtodblcol{%
\begingroup
\@insertfalse
\@setfloattypecounts
\@getfpsbit \tw@
\ifodd\@tempcnta
\@flsetnum \@dbltopnum
\ifnum \@dbltopnum>\z@
\@tempswafalse
\ifdim \@dbltoproom>\ht\@currbox
\@tempswatrue
\else
\ifnum \@fpstype<\sixt@@n
\advance \@dbltoproom \@textmin
\ifdim \@dbltoproom>\ht\@currbox
\@tempswatrue
\fi
\advance \@dbltoproom -\@textmin
\fi
\fi
\if@tempswa
\@bitor \@currtype \@deferlist
\@testwrongwidth\@currbox
\if@test
\else
\@bitor \@currtype \@dblbotlist
\if@test
\else
\@tempdima -\ht\@currbox
\advance\@tempdima
-\ifx \@dbltoplist\@empty \dbltextfloatsep
\else \dblfloatsep
\fi
\global \advance \@dbltoproom \@tempdima
\global \advance \@colht \@tempdima
\global \advance \@dbltopnum \m@ne
\@cons \@dbltoplist \@currbox
\@inserttrue
\fi
\fi
\fi
\fi
\fi
\if@insert
\else
\@getfpsbit 4\relax
\ifodd\@tempcnta
\@flsetnum \@dblbotnum
\ifnum \@dblbotnum>\z@
\@tempswafalse
\ifdim \@dbltoproom>\ht\@currbox
\@tempswatrue
\else
\ifnum \@fpstype<\sixt@@n
\advance \@dbltoproom \@textmin
\ifdim \@dbltoproom>\ht\@currbox
\@tempswatrue
\fi
\advance \@dbltoproom -\@textmin
\fi
\fi
\if@tempswa
\@bitor \@currtype \@deferlist
% Test for twocolumn floats (2008/08/21 Mattias Nissler)
\@testwrongwidth\@currbox
\if@test
\else
\@tempdima -\ht\@currbox
\advance\@tempdima
-\ifx \@dblbotlist\@empty \dbltextfloatsep
\else \dblfloatsep
\fi
\global \advance \@dbltoproom \@tempdima
\global \advance \@colht \@tempdima
\global \advance \@dblbotnum \m@ne
\@cons \@dblbotlist \@currbox
\@inserttrue
\fi
\fi
\fi
\fi
\fi
\if@insert
\else
\@cons\@deferlist\@currbox
\fi
\endgroup
}
\def\@combinedblfloats{%
\ifx \@dbltoplist\@empty \else \@cdblflt \fi
\ifx \@dblbotlist\@empty \else \@cdblflb \fi
}
\def\@cdblflt{%
\setbox\@tempboxa \vbox{}%
\let \@elt \@comdblflelt
\@dbltoplist
\let \@elt \relax
\xdef \@freelist {\@freelist\@dbltoplist}%
\global\let \@dbltoplist \@empty
\setbox\@outputbox \vbox% to\textheight
{%\boxmaxdepth\maxdepth %% probably not needed, CAR
\unvbox\@tempboxa\vskip-\dblfloatsep
\ifnum \@dbltopnum>\m@ne
\dblfigrule
\fi
\vskip \dbltextfloatsep
\unvbox \@outputbox
}%
}
\def\@cdblflb{%
\setbox\@tempboxa \vbox{}%
\let \@elt \@comdblflelt
\@dblbotlist
\let \@elt \relax
\xdef \@freelist {\@freelist\@dblbotlist}%
\global\let \@dblbotlist \@empty
\setbox\@outputbox \vbox% to\textheight
{%\boxmaxdepth\maxdepth %% probably not needed, CAR
\unvbox\@outputbox
\vskip \dbltextfloatsep
\ifnum \@dblbotnum>\m@ne
\dblfigrule
\fi
\unvbox\@tempboxa
\vskip -\dblfloatsep
}%
}
\def \@doclearpage {%
\ifvoid\footins
\setbox\@tempboxa\vsplit\@cclv to\z@ \unvbox\@tempboxa
\setbox\@tempboxa\box\@cclv
\xdef\@deferlist{\@toplist\@botlist\@deferlist}%
\global \let \@toplist \@empty
\global \let \@botlist \@empty
\global \@colroom \@colht
\ifx \@currlist\@empty
\else
\@latexerr{Float(s) lost}\@ehb
\global \let \@currlist \@empty
\fi
\@makefcolumn\@deferlist
\@whilesw\if@fcolmade \fi{\@opcol\@makefcolumn\@deferlist}%
\if@twocolumn
\if@firstcolumn
\xdef\@deferlist{\@dbltoplist\@dblbotlist\@deferlist}%
\global \let \@dbltoplist \@empty
\global \let \@dblbotlist \@empty
\global \@colht \textheight
\begingroup
\@dblfloatplacement
\@makefcolumn\@deferlist
\@whilesw\if@fcolmade \fi{\@outputpage
\@makefcolumn\@deferlist}%
\endgroup
\else
\vbox{}\clearpage
\fi
\fi
\ifx\@deferlist\@empty \else\clearpage \fi
\else
\setbox\@cclv\vbox{\box\@cclv\vfil}%
\@makecol\@opcol
\clearpage
\fi
}
\endinput
Joseph Wright
> Btw. Whom should I contact to merge this fix back into the original
> dblfloatfix package?
It has Morten H{\o}gholm's initials and e-mail address in it, so I would
try him: morten....@gmail.com
--
Joseph Wright
mattias...@googlemail.com
wrote:
> It has Morten H{\o}gholm's initials and e-mail address in it, so I would
Yes, I found it. I had somehow missed it when posting...
Mattias