latex2rtf-developers — Development and Usage of latex2rtf

Re: [Latex2rtf-developers] [Latex2rtf-users] release candidate for latex2rtf

[Tyler S. <tyl...@ma...>]

On Sat, Nov 10, 2007 at 05:56:44PM -0800, Scott Prahl wrote:
> 
> One issue that merits further discussion is that this release has a
> new restriction: any files used by the root latex document must be
> contained somewhere in the directory of the root document. This is
> similar to how TeX itself works. Absolute paths don't work anymore.
> There has been a bug report about both behaviors. Who thinks that
> this is an unnecessary restriction?

The latest build works fine for my multi-file documents, so the new
restriction won't impact my work at all.

Tyler

-- 
There is no theory of evolution. 
Just a list of animals Chuck Norris allows to live.

http://www.chucknorrisfacts.com/

Re: [Latex2rtf-developers] release candidate for latex2rtf

[<jue...@t-...>]

Scott Prahl wrote:
> Here is a release candidate of latex2rtf. =A0This release is driven by
> extensive testing done by Wilfried who uncovered a disconcerting number
> of bugs. =A0These are now fixed and the new candidate is at
>
> =A0 =A0http://omlc.ogi.edu/software/tex_convert/latex2rtf-1.9.19.tar.gz

Congratulations, this looks very promising.

One question wrt the new URL handling: I get now real hyperlinks in the RTF=
=20
when using \url (from url.sty). Can this be switched off (I usually don't=20
want clickable links in manuscripts for journals and proceedings)?

> One issue that merits further discussion is that this release has a new
> restriction: any files used by the root latex document must be contained
> somewhere in the directory of the root document. =A0This is similar to how
> TeX itself works. =A0Absolute paths don't work anymore. =A0There has been=
 =A0
> a bug
> report about both behaviors. =A0Who thinks that this is an unnecessary =A0
> restriction?

it might result in problems with LyX, where the files are copied into a tem=
p=20
directory before processing, but I'm not sure about that.

J=FCrgen

[Latex2rtf-developers] release candidate for latex2rtf

[Scott P. <pr...@bm...>]

Hey all,

Here is a release candidate of latex2rtf.  This release is driven by
extensive testing done by Wilfried who uncovered a disconcerting number
of bugs.  These are now fixed and the new candidate is at

   http://omlc.ogi.edu/software/tex_convert/latex2rtf-1.9.19.tar.gz

One issue that merits further discussion is that this release has a new
restriction: any files used by the root latex document must be contained
somewhere in the directory of the root document.  This is similar to how
TeX itself works.  Absolute paths don't work anymore.  There has been  
a bug
report about both behaviors.  Who thinks that this is an unnecessary  
restriction?

The 1.9.19 release also eliminates the rarely used '%latex2rtf:'  
sequence to
pass commands directly to latex2rtf (and hide them from latex).  Instead
use the much more powerful '\iflatextortf ... \else ... \fi' sequence.

New features (since 1.9.18) include

  * basic support for \usepackage{subfigure}
  * basic support for \usepackage{endfloat}
  * basic support for \documentclass{apa}
  * basic support for \usepackage{pstricks-add}
  * basic support for \usepackage{hyperref}
  * nearly complete support for \usepackage{apacite}

  * support for \symbol{\'23} and better support for \char'23
  * support for \overrightarrow
  * support for \namedref
  * support for \urlstyle and \UrlFont

Scott

********************  ChangeLog  ***********************************

r752 support \symbol{\'23} and \symbol{\"13}.  Improve \char'23 so that
      it changes if\tt is active

r753 support \char92 and \char`b and \symbol{92}

r754 rip out support for %latex2rtf: as a way of controlling parser.
      Replace uses in examples with \iflatex2rtf ... \else ... \fi.  Fix
      mistake in getSimpleCommand() that did not allow '\!' to be
      returned. Eliminate weird bug when a command follows an integral
      sign ... $\int \cos$. Improve debugging output at -d2 level.

r755 Do not try to ignore packages or package options.

r756 Fixes for the regressions in overstrike.tex

r757 WH: commented out unneeded 'show_buffer' function in preparse.c

r758 add \urlstyle{} ignore.cfg.  Cleaned up TryIgnoreVariable().  Made
      arguments to strdup_together be constant strings.  More of this
      should be done.

r759 silence \renewcommand{}{} warning.

r760 no functional change ... just different code comment

r762 Improve documentation for \iflatextortf and provide work-around for
      %latex2rtf:

r763 Fix for backslash-linefeed problem in .aux files.  Pretty obscure.

r764 Suppress multiple warnings for missing \bibcite{} entries.  Change
      Makefile in test directory to sup press all the output from latex.

r765 WH: modified cfg.c so that latex2rtf also tries to find the cfg dir
      in C:\%ProgramFiles%\latex2rtf\cfg (default if installed by  
windows
      installer)

r766 Fix typo in diagnotic message for \ensureeqn.  Fix Makefile for
      bib_super.tex. Revise my_fgetc() in the hopes of making it more
      portable.  Update ScanBbl() so that it uses the new routine.

r767 Fix regressions in bib_natbib1.tex and bib_natbib2.tex.  Don't know
      where this code went, but the use of plain old \cite{} in the
      natbib style was broken.

r768 Add tests for \documentclass{apa}

r769 Make section headers bold for \documentclass{apa}

r770 Fix the bug from the preamble of bib_apa.  Add 'const' to more
      functions. Eliminate all unneeded fprintf(stderr,...).  Rename a
      bunch of diagnostic(1,...) instances to diagnostic(WARNING, ...).

r771 Adding two more test files

r772 WH: Modified main.c so that -P option now works also under
      DOS/Windows. Updated documentation of -P option and -M option.
      Fixed bug in latex2png script.

r773 WH: Updated documentation of -P option and -M option (again).

r774 added diagnostic() line to my_fgets() to help sort out the problems
      on Windows.

r775 WH: xrefs.c: Open aux file in binary mode. Removed extra
      diagnostic() line.

r776 Use fprintf() in show_string() to produce better diagnostics

r777 Fix for bibentry_apalike.tex

r778 Make sure all fopen() calls use the mode "rb"

r779 Starting the process of rationalizing the debug levels

r780 remove a few instances of strcat

r781 More clean-up of debugging information.  -d2 now shows nothing but
      file manipulations.  Small tweaks to makefile in the test  
directory
      to create temporary files in a folder within the current  
directory.

r782 More diagnostics clean-up.  It is now possible to get relatively
      noise-free output using -d2.  This diagnostic level shows file
      handling and each section as it is handled.  The plan is to make
      -d3 echo system commands and each section as it is preparsed.  -d4
      will track each \command as it is used.

r783 More rationalization for debugging levels.  One day -d5 will be
      stuff that happens within CmdName() functions and -d6 will show
      nearly everything but character processing, and -d7 will show each
      character as it gets read and processed.

r784 Fix for fig_endfloat conversion.  It mostly works.  A few more
      diagnostic() tweaks.

r785 More diagnostics() tweaks.

r786 support \overrightarrow

r787 preliminary support for subfigure package

r788 test file for \subfigure[caption]{figure} command

r789 Add proper paragraph ending to subfigure pieces.  Fix numbering of
      subfigures.  Add tests.

r790 Initial code to support the pstricks-add package.  Three commands
      are recognized \psset{}, \begin{psgraph}, and \begin{pspicture}.
      One broken test file (at least I cannot latex it successfully) is
      also included.

r791 Emit information as latex2rtf is doing slow stuff (converting
      latex->PNG or pdf->PNG).  Eliminate first linefeed and emit a  
final
      linefeed when done.  This helps to group the output with the
      previous command.

r792 Clean up a few preamble problems in the test documents.

r793 revert to placing the DOCUMENT_MODE environment before PREAMBLE
      because there was no real reason to change and this might  
introduce
      new subtle bugs.

r794 add support for \href{url}{text}, \url{url}, \nolinkurl{url}.

r795 start of a test file for hyperref features

r796 refactoring the hyperlink code

r797 more \url{ref} clean-up.  add putRtfStrEscaped() function.

r798 eliminate some compiler warnings

r799 Minor change to getRawTexChar() to handle EOF condition better.
      Small change to CmdInclude() to properly terminate with long file
      names.

r800 Make is so that long file names fail properly when \input is used.
      Extend the possible file name length to 99.

r801 Fix bug in ScanAux() so that it handles \@input{file} properly.
      Enhance test/include.tex so that it can be properly processed by
      latex.  Add pagebreak before every instance of \include{file} as
      should be done.  Fix minor bug in subsubsection numbering when
      section but not subsection is used.

r802 preliminary code to support \newpsstyle{}{} ... unclear if it works
      yet.

r803 remove debugging commands since \newpsstyle{}{} seems to work.  Add
      more packages with 'limited' support.  Have a number of test files
      run latex first to generate .aux files.  Improve diagnostics for
      writelatexasbitmap().

r804 silently ignore \nobreak, \expandafter, \long.  Change warnings
      from 'ignored' to 'unknown'.  More general cleanup.

r805 WH: fixed a bug in main.c. Updated documentation.

r806 WH: fixed my bugfix in main.c (should have tested it before
      uploading!)

r807 Minor clean-up of -se# and -sf# scaling options.  We don't properly
      support long option names, currently '-se #' is an error.

Re: [Latex2rtf-developers] oowriter problems

[Tyler S. <tyl...@ma...>]

On Fri, Oct 19, 2007 at 11:51:34AM -0400, Tyler Smith wrote:
> 
> Unfortunately, I need to be able to send 'final' versions of my paper 
> to an editor who will expect it to appear perfectly formatted the 
> moment he opens it in Word. This really sucks, as it now appears that I 
> will have to transfer my l2r-translated documents to another machine, 
> running MS with Word, in order to complete the processing of my paper. 
> Latex2rtf keeps getting better, but if I have to do any processing in 
> MSWord at all there's no longer any point to using oowriter.

As a follow-up, I've just installed crossover-office and officeXP on
my Debian laptop, and it looks like it is more than serviceable. If
that turns out to be true, this will be a great relief. Well, except
for the fact that it will mean MSWord has followed me into the GNU/Linux
world, but at least I'll be able to do all my editing and
word-processing on the same machine.

Tyler


-- 
Research is what I'm doing when I don't know what I'm doing.
                                       --Wernher von Braun

Re: [Latex2rtf-developers] oowriter problems

[Tyler S. <tyl...@ma...>]

Quoting Wilfried Hennings <W.H...@fz...>:
> I would recommend that for converting to OO to use TeX4ht with the 
> oolatex script. This creates a native OO file (which is a set of xml 
> files and bitmaps packed in a zip archive) retaining all equations as 
> editable OO objects. From OO this can be saved as Word doc, with the 
> limitation that equations first look broken when opened in Word and 
> need to be double-clicked to fix them.
>

Thanks Wilfried.

Unfortunately, I need to be able to send 'final' versions of my paper 
to an editor who will expect it to appear perfectly formatted the 
moment he opens it in Word. This really sucks, as it now appears that I 
will have to transfer my l2r-translated documents to another machine, 
running MS with Word, in order to complete the processing of my paper. 
Latex2rtf keeps getting better, but if I have to do any processing in 
MSWord at all there's no longer any point to using oowriter.

The really frustrating thing is that the paper gets converted to pdf by 
the journal before it is sent out to reviewers anyways! After my paper 
is safely accepted I will try again to explain this to the editors 
involved, but they have been very firm in refusing to accept pdf files 
as submissions. Bizarre.

Sorry for ranting,

Tyler

> Wilfried
>
> -- 
> Dipl.-Ing.(=M.Eng.) Wilfried Hennings c./o.
> Research Center Juelich
> INB-MUT
> 52425 Juelich
> Germany
> Tel. (+49) 2461 61 5570 ; Fax (+49) 2461 61 8254
> email: W.H...@fz...
> PGP key available from key servers, e.g. from
>     http://pgpkeys.pca.dfn.de/
> work homepage: http://www.fz-juelich.de/mut/mut_home
> All opinions mentioned are strictly my own, not my employer's
>
>
>
> -----------------------------------------------------------------------------------------
> -----------------------------------------------------------------------------------------
> Forschungszentrum Juelich GmbH
> 52425 Juelich
>
> Sitz der Gesellschaft: Juelich
> Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
> Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
> Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft 
> (stellv. Vorsitzender)
> -----------------------------------------------------------------------------------------
> -----------------------------------------------------------------------------------------
>



-- 
Tyler Smith

PhD Candidate
Plant Science Department
Faculty of Agricultural and Environmental Sciences
McGill University
MacDonald Campus
21,111 Lakeshore
Ste. Anne de Bellevue, Quebec, Canada H9X 3V9

Tel.: 514-398-7851 ext. 8726
Fax: 514-398-7897
tsm...@po...

Re: [Latex2rtf-developers] oowriter problems

[Wilfried H. <W.H...@fz...>]

Hello,
At 19.10.2007 15:24, Tyler Smith wrote:
>I used the latest l2r from svn to convert a paper I'm working on. It
>looks great - the figures now appear much more sensibly, the tables
>look better, most of the inline special symbols are fine. Much
>improvement over the previous stable release. I checked this with
>oowriter, and it displays all my gammas and sqrt sybmbols, very nice.
>
>So then I add a single 'X' to the file and resave it, still in rtf.
>I follow this with a call to diff on the original file and the
>modified version, with the single additional character added. I get
>pages and pages and pages of differences! What's more, when I reopen
>the file all the gammas are gone, and random other bits get garbled.
>
>This has caused no end of grief as I painstakingly correct the
>latex2rtf output in oowriter, send it to my supervisor who opens the
>document in Word, and then complains to me that it looks like crap.
>
>What gives? Why does oowriter need to break files that it can read
>just fine?

I just confirmed with OO2.3 what I had tested with an earlier version:
Without any reference to latex2rtf, I created a document with 
OOwriter, containing a Unicode gamma in Arial (by "insert symbol"), a 
gamma in Symbol (type a "g" and choose font "Symbol"), and an OO equation.
Saving as odt and reopening works well (of course).
Saving as rtf, the gammas both are retained but the equation is gone. 
OO can not retain its own equations on opening in rtf, so there is no 
way to insert rtf code which would be interpreted by OOwriter as an equation.
But if you save as MS Word 97-2000-XP and have the "convert OO 
equations to MathType" checked, the equations are converted to 
editable EquationEditor objects. Although they look broken when first 
opened in Word, a double-click on each of the equations (in Word) 
makes it look fine.

Even worse, OO cannot interpret Word field functions (like the EQ 
fields inserted by latex2rtf) nor save them, this applies to both the 
doc and the rtf format.

If you open latex2rtf's output in word, double-click each EQ field to 
make it an EquationEditor object, and save the document as Word doc 
(not rtf), OOwriter converts the equations correctly (with the 
limitation that EquationEditor changes greek symbols into standard 
font whereas MathType handles them correctly; this is EE's fault, not OO's).

>Is this something we could fix at the l2r side?

As the rtf import to OO is limited if not broken, there is not much 
we can do about.
I would recommend that for converting to OO to use TeX4ht with the 
oolatex script. This creates a native OO file (which is a set of xml 
files and bitmaps packed in a zip archive) retaining all equations as 
editable OO objects. From OO this can be saved as Word doc, with the 
limitation that equations first look broken when opened in Word and 
need to be double-clicked to fix them.

Wilfried

-- 
Dipl.-Ing.(=M.Eng.) Wilfried Hennings c./o.
Research Center Juelich
INB-MUT
52425 Juelich
Germany
Tel. (+49) 2461 61 5570 ; Fax (+49) 2461 61 8254
email: W.H...@fz...
PGP key available from key servers, e.g. from
     http://pgpkeys.pca.dfn.de/
work homepage: http://www.fz-juelich.de/mut/mut_home
All opinions mentioned are strictly my own, not my employer's



-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich

Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv. 
Vorsitzender)
-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------

[Latex2rtf-developers] oowriter problems

[Tyler S. <tyl...@ma...>]

Hi gang,

I've just confirmed something that has been really pissing me off. I
used the latest l2r from svn to convert a paper I'm working on. It
looks great - the figures now appear much more sensibly, the tables
look better, most of the inline special symbols are fine. Much
improvement over the previous stable release. I checked this with
oowriter, and it displays all my gammas and sqrt sybmbols, very nice.

So then I add a single 'X' to the file and resave it, still in rtf.
I follow this with a call to diff on the original file and the
modified version, with the single additional character added. I get
pages and pages and pages of differences! What's more, when I reopen
the file all the gammas are gone, and random other bits get garbled.

This has caused no end of grief as I painstakingly correct the
latex2rtf output in oowriter, send it to my supervisor who opens the
document in Word, and then complains to me that it looks like crap.

What gives? Why does oowriter need to break files that it can read
just fine? Is this something we could fix at the l2r side? I'm going
to go download ABIWord and see if it isn't any better.

Tyler

-- 
Who is your computer working for?

http://www.defectivebydesign.org/faq

[Latex2rtf-developers] Release candidate 1.9.18 for latex2rtf

[Scott P. <pr...@bm...>]

Hey all,

Graphics and equation bitmap handling has been improved since the  
last release (especially on the windows platform).  Wilfried and I  
think that a new release would be nice.  The release candidate is at

    http://omlc.ogi.edu/software/tex_convert/latex2rtf-1.9.18.tar.gz

Noteworthy changes

* new script pdf2pnga to handle image conversion on windows machines
* -t option to include tables as bitmap images in the RTF document
* image scaling now works when just the width is specified
* support for the paralist package
* \appendix now affects numbering of chapters/sections that follow

Give it a spin.

Scott

Detailed changes between release 1.9.17 and 1.9.18

r673
first check-in for 1.9.18

r674
WH: update of the documentation (equation conversion)

r675
WH: update of the documentation (unicode input)

r676
Infrastructure changes to support stylesheets.  Modify
CmdStartParagraph() to take two parameters (paragraph style and
indenting).  Create style numbers in funct1.h for the different possible
paragraph styles and use these as appropriate. No changes to translation
or conversion have been made.

r677
Eliminate the numbers in favor of passing strings.  This will allow
InsertStyle() to be called directly when the time comes to hook up these
parameters.  I had forgotten about the existing style infrastructure in
style.c and style.h.

r678
WH: updated main.c to include "help" text for -M32

r679
Support for list environments from paralist package.  While implementing
this I noticed that some nested list environments were broken.  While
tracing the problems, I discovered that an old optimization was broken.
I have taken it out and everything still seems to work fine.

r680
Mention that there is some support for paralist

r681
Add compactitem and compactdesc environment handling. Like the rest of
paralist, there has been no attempt to reduce the line spacing.

r682
Add a few more test cases for paralist environments to list.tex

r683
Started to improve tabular widths.  Then realized that global variables
for the tabular environment were used oddly.  Further investigation
revealed that the global variable could be removed completely if the
active column was placed in the TabularT type.  So I did this, and then
needed to comment out some usage in funct1.c and convert.c.  Evidently,
that stuff was never used.  Maybe I will get farther next time.

r684
Modify function to extract the contents of a brace from tables.c and put
it in util.c because it is a handy little function.  Use in in xref.c to
simplify some code.

r685
Make getStringBraceParam() work properly with \{ and \}

r686
modify testfile so that it does not create unnecessary warnings.

r687
fix problem with multicolumn in the last line of a table

r688
Fix equation numbering in reports and books so that equations use (2.3)
format as appropriate.  Reset numbering at each chapter break.

r689
renaming files so that plurals are used more consistently

r690
More plurals for file names

r691
rename encode_tables.h to encoding_tables.h update dependencies in the
Makefile

r692
New option -t # that allows the user to have tabular environments
generated as bitmaps. Basically -t 1 is the default and generates RTF
only.  -t 2 will convert tabular environments to bitmaps, -t 3 will
generate both in the RTF document.

r693
WH: in counters.c corrected "util.h" -> "utils.h"

r694
WH: corrected latex2pn.bat to allow repeated call for reducing picture
size

r695
Do not scale tables that are inserted as bitmaps.  Modify tabular.tex so
that it typesets more nicely with latex.  Thanks Wilfried!

r696
Support tabbing environments as bitmaps.  Add a minimal documentation in
help message.

r697
Refactoring code.  Move WriteLatexAsBitmap() to graphics.c.  Create new
routines PrepareDisplayedBitmap() and FinishDisplayedBitmap() to
simplify to localize paragraph starting and stopping.

r698
Once again, try to use ghostscript for pdf->png conversion.  Last time
the command needed to be changed because it failed on Windows system.
Now, under unix, use ghostscript conversion.  Under everything else
continue to use ImageMagick.  Hopefully, this will make everyone happy.
Or at least happier.

r699
Quote filenames passed to system() in pdf_to_png

r700
Single quote all filenames passed to shell via system().  Adam Maxwell++

r701
Convert 'filename' to "filename" in system commands because Windows
treats single quotes differently than UNIX.  Also, fix up the lengths of
the command strings.  This needs revisiting.

r702
Collect most of the code for creating command-line graphics conversion
into a single routine SysGraphicsCommand().  This routine is both
simpler and more robust than the code used in many other places in the
code.

r703
hack to resolve doublespace environment problem

r704
fix for image scaling when the width is specified, but the height is
not.  Adam Maxwell++

r705
implement rounding properly ... what was I thinking?

r706
add yet another test image file.  Maxwell++

r707
WH: added pdf2pnga scripts, called in graphics.c

r708
supply rint() as a substitute for round() because windows does not have
round() defined in its math.h header.

r709
Collecting all the system dependent graphics conversion stuff into one
place.  This eliminates a lot of code duplication and makes it clearer
that all conversions are handled in a uniform fashion (i.e., -D -T and
-P command line flags).  Right now there are two basic sets of calls ...
one for UNIX and one for XP.  (There is one minor sub-branch in the UNIX
versions that uses native pdf handling on Apple systems.)

Scaling on apple systems looks messed up ... sigh

r710
WH: (DOS version) in graphics.c, removed quotes from calls to
latex2pn.bat and pdf2pnga.bat

r711
WH: (DOS version) in graphics.c, now calling bash instead of
command.com, so that quotes in calls to latex2png are supported

r712
WH: in graphics.c, calling pdf2pnga script instead of sips (Apple) or gs
(otherwise) Added Apple specific script pdf2pnga_apple (calling sips)

r713
Changes to pdf2pnga to make things work at least under OS X.  This
probably breaks the XP stuff, and is only intended to start the process
of coming up with script that will run properly cross-platform.

r714
refactoring the horrible horrible getSection().  In the process
misc4.tex and linux.tex no longer translate properly.  More fixes
coming.  This version will spew a whole bunch of diagnostic information.
Wait for a later version before downloading.

r715
Fixing remaining bugs from the refactoring of getSection().  Things
should make a bit more sense.  Files other than parser.c have ignorable
tweaks.

r716
Clean separation of unix and windows graphics conversion.  Filenames are
now single-quoted in unix as they should be.  Windows gets double
quotes.  Use sips by default on apple systems.  Windows uses the script
pdf2pnga for conversions.  I think that this should provide maximum
functionality to all platforms.

r717
Move getSection() out of parser.c and into preparse.c

r718
Actually add the files to svn this time.  Rename getSection() to
preParse()

r719
WH: latex2pnga script tests availability of Ghostscript or sips and
issues system dependent call; unneeded scripts deleted

r720
WH: Makefile: script latex2pn.bat removed, pdf2pnga added

r721
WH: equations.c: one comment changed graphics.c: comments about pdf2png
moved into function SysGraphicsConvert, command for debugging
(non-)insertion of bitmap into rtf file inserted (marked with "/* WH
*/")

r722
Fix problem with bitmapped equations not appearing when -M12 option is
used.

r723
Make equation conversion to bitmap less noisy.

r724
If SysGraphicsConvert() returns null, then make sure that the return
value is handled properly.  Catch a few cases when strings are not freed
properly.

r725
WH: main.c: help text corrected

r726
WH: parser.c: Added exit to catch extra blank in \begin{document}
Actually, it doesn't solve the problem that such extra blanks are
allowed but at least it exits the progrm with a helpful error message.

r727
fix for '\begin {document}' parse error.  This solution has not been
applied to any other \begin .. \end environments.

r728
remove debugging \end{document} from misc4.tex

r729
Fix numbering of chapters in the appendix ... not perfect, but the latex
default should work.

Re: [Latex2rtf-developers] \begin {document}

[Wilfried H. <W.H...@fz...>]

At 11.10.2007 09:26, I wrote:
>Hello,
>just for information:
>I posted a question in the German TeX newsgroup whether blanks
>between \begin and { are principally allowed or not.
>Although MikTeX accepts it with a blank in between, this is the first
>LaTeX input file I have seen with a blank in this place.

Answer (my translation from German):

David Kastrup wrote:
 > Markus Kohm writes:
 >
 > > Wilfried Hennings wrote:
 > >
 > >> Should latex2rtf accept blanks here?
 > >
 > > After commands, principally any number of blanks are allowed.
 > > Also between command and parameters.
 > > In LaTeX normally also before and after optional parameters.
 > > Also in the list of options \documentclass, \LoadClass, \usepackage etc.
 > > blanks are allowed after the commas.
 >
 > They are "allowed" anywhere (not restricted to after commas) and are
 > ignored. Some weeks ago I identified a
 >
 > \usepackage[Borderwhatever={0 0 0}]{hyperref}
 >
 > as responsible for a strange problem, as only "000" was inserted
 > in the PostScript
 >
 > > In vertical lists even any number of blanks are allowed:
 > >
 > > \documentclass   [
 > >         a4paper,   twoside     ]      {article}
 > > \begin             {document}
 >
 > Also
 >
 > \documentclass[a 4 paper, two
 >                           side]{article}
 >
 > should work. Sick (correctly translated?).
 > Documentation only sais that is is allowed after commas.

Now I don't think latex2rtf should handle all strange cases, but a 
better error message would be helpful, keeping in mind that LaTeX 
doesn't help finding the problem in this case.


>At 10.10.2007 00:06, Peter Lloyd wrote:
>  >I'm using latex2rtf-1.9.17a_dos. Thanks for creating a useful tool.
>  >However, I've encountered what I think is a bug.
>  >
>  >To reproduce, create a file eg min.tex, containing the following:
>  >
>  >\documentclass{article}
>  >\begin {document}
>  >This is some text
>  >\end{document}
>  >
>  >Now run 'l2r min.tex'
>  >
>  >I get the error:
>  >
>  >{
>  >(Not set):3 Mismatched '{' in RTF file, Conversion may cause problems.
>  >(Not set):3 This is often caused by having environments that span
>  >(Not set):3 \section{}s.  For example
>  >(Not set):3    \begin{small} ... \section{A} ... \section{B} ...
>  >\end{small}
>  >(Not set):3 will definitely fail.
>  >(Not set):3 Try translating with 'latex2rtf -Z2 min.tex'
>  >
>  >And there is nothing but header information in the resulting rtf file.
>  >(Using -Z2 doesn't make it work either)
>  >
>  >I've tracked it down to the space between \begin and {document} --
>  >removing this space makes everything work correctly.
>  >
>  >LaTeX (MiKTeX 2.5) parses the input file just fine.
>--
>Wilfried Hennings c./o.
>Forschungszentrum (Research Center) Juelich GmbH
>INB-MUT
>52425 Juelich, Germany
>Tel. (+49) 2461 61 5570 ; Fax (+49) 2461 61 8254
>email: W.H...@fz...
>PGP key available from key servers, e.g. from
>      http://pgpkeys.pca.dfn.de/
>work homepage: http://www.fz-juelich.de/mut/mut_home
>All opinions mentioned are strictly my own, not my employer's
>
>
>
>-----------------------------------------------------------------------------------------
>-----------------------------------------------------------------------------------------
>Forschungszentrum Juelich GmbH
>52425 Juelich
>
>Sitz der Gesellschaft: Juelich
>Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
>Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
>Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv.
>Vorsitzender)
>-----------------------------------------------------------------------------------------
>-----------------------------------------------------------------------------------------
>
>-------------------------------------------------------------------------
>This SF.net email is sponsored by: Splunk Inc.
>Still grepping through log files to find problems?  Stop.
>Now Search log events and configuration files using AJAX and a browser.
>Download your FREE copy of Splunk now >> http://get.splunk.com/
>_______________________________________________
>Latex2rtf-developers mailing list
>Lat...@li...
>https://lists.sourceforge.net/lists/listinfo/latex2rtf-developers

-- 
Dipl.-Ing.(=M.Eng.) Wilfried Hennings c./o.
Research Center Juelich
INB-MUT
52425 Juelich
Germany
Tel. (+49) 2461 61 5570 ; Fax (+49) 2461 61 8254
email: W.H...@fz...
PGP key available from key servers, e.g. from
     http://pgpkeys.pca.dfn.de/
work homepage: http://www.fz-juelich.de/mut/mut_home
All opinions mentioned are strictly my own, not my employer's



-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich

Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv. 
Vorsitzender)
-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------

Re: [Latex2rtf-developers] \begin {document}

[Wilfried H. <W.H...@fz...>]

Hello,
just for information:
I posted a question in the German TeX newsgroup whether blanks 
between \begin and { are principally allowed or not.
Although MikTeX accepts it with a blank in between, this is the first 
LaTeX input file I have seen with a blank in this place.

At 10.10.2007 00:06, Peter Lloyd wrote:
 >I'm using latex2rtf-1.9.17a_dos. Thanks for creating a useful tool.
 >However, I've encountered what I think is a bug.
 >
 >To reproduce, create a file eg min.tex, containing the following:
 >
 >\documentclass{article}
 >\begin {document}
 >This is some text
 >\end{document}
 >
 >Now run 'l2r min.tex'
 >
 >I get the error:
 >
 >{
 >(Not set):3 Mismatched '{' in RTF file, Conversion may cause problems.
 >(Not set):3 This is often caused by having environments that span
 >(Not set):3 \section{}s.  For example
 >(Not set):3    \begin{small} ... \section{A} ... \section{B} ...
 >\end{small}
 >(Not set):3 will definitely fail.
 >(Not set):3 Try translating with 'latex2rtf -Z2 min.tex'
 >
 >And there is nothing but header information in the resulting rtf file.
 >(Using -Z2 doesn't make it work either)
 >
 >I've tracked it down to the space between \begin and {document} --
 >removing this space makes everything work correctly.
 >
 >LaTeX (MiKTeX 2.5) parses the input file just fine.
-- 
Wilfried Hennings c./o.
Forschungszentrum (Research Center) Juelich GmbH
INB-MUT
52425 Juelich, Germany
Tel. (+49) 2461 61 5570 ; Fax (+49) 2461 61 8254
email: W.H...@fz...
PGP key available from key servers, e.g. from
     http://pgpkeys.pca.dfn.de/
work homepage: http://www.fz-juelich.de/mut/mut_home
All opinions mentioned are strictly my own, not my employer's



-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich

Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv. 
Vorsitzender)
-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------

Re: [Latex2rtf-developers] \begin {document}

[Wilfried H. <W.H...@fz...>]

Hello,
- Peter: You'd need to subscribe to the mailing list before your 
submitted mail goes through. I accepted it manually. -

At 10.10.2007 00:06, Peter Lloyd wrote:
>I'm using latex2rtf-1.9.17a_dos. Thanks for creating a useful tool.
>However, I've encountered what I think is a bug.
>
>To reproduce, create a file eg min.tex, containing the following:
>
>\documentclass{article}
>\begin {document}
>This is some text
>\end{document}
>
>Now run 'l2r min.tex'
>
>I get the error:
>
>{
>(Not set):3 Mismatched '{' in RTF file, Conversion may cause problems.
>(Not set):3 This is often caused by having environments that span
>(Not set):3 \section{}s.  For example
>(Not set):3    \begin{small} ... \section{A} ... \section{B} ...
>\end{small}
>(Not set):3 will definitely fail.
>(Not set):3 Try translating with 'latex2rtf -Z2 min.tex'
>
>And there is nothing but header information in the resulting rtf file.
>(Using -Z2 doesn't make it work either)
>
>I've tracked it down to the space between \begin and {document} --
>removing this space makes everything work correctly.
>
>LaTeX (MiKTeX 2.5) parses the input file just fine.
>
>Were you aware of this?

No (at least not me).
Thanks for pointing that out. (I can reproduce it.)


-- 
Wilfried Hennings c./o.
Forschungszentrum (Research Center) Juelich GmbH
INB-MUT
52425 Juelich, Germany
Tel. (+49) 2461 61 5570 ; Fax (+49) 2461 61 8254
email: W.H...@fz...
PGP key available from key servers, e.g. from
     http://pgpkeys.pca.dfn.de/
work homepage: http://www.fz-juelich.de/mut/mut_home
All opinions mentioned are strictly my own, not my employer's



-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich

Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv. 
Vorsitzender)
-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------

[Latex2rtf-developers] \begin {document}

[Peter L. <l-...@pg...>]

Hi,

I'm using latex2rtf-1.9.17a_dos. Thanks for creating a useful tool.
However, I've encountered what I think is a bug.

To reproduce, create a file eg min.tex, containing the following:

\documentclass{article}
\begin {document}
This is some text
\end{document}

Now run 'l2r min.tex'

I get the error:

{
(Not set):3 Mismatched '{' in RTF file, Conversion may cause problems.
(Not set):3 This is often caused by having environments that span
(Not set):3 \section{}s.  For example
(Not set):3    \begin{small} ... \section{A} ... \section{B} ... 
\end{small}
(Not set):3 will definitely fail.
(Not set):3 Try translating with 'latex2rtf -Z2 min.tex'

And there is nothing but header information in the resulting rtf file. 
(Using -Z2 doesn't make it work either)

I've tracked it down to the space between \begin and {document} -- 
removing this space makes everything work correctly.

LaTeX (MiKTeX 2.5) parses the input file just fine.

Were you aware of this?

Thanks,
Peter

Re: [Latex2rtf-developers] latex2rtf bugs

[Scott P. <pr...@bm...>]

Wilfried,

Hope your vacation was revitalizing!

On Oct 9, 2007, at 2:09 AM, whe...@us... wrote:

> The filenames must not be quoted. If quoted, the call to the batch  
> file is
>
>    latex2pn.bat -d 300 -o 6 "l2r_0001"
>
> which causes the batch file to try processing "l2r_0001".tex which
> obviously doesn't exist.
>
> - char format_xp[] = "command.com /e:2048 /c latex2pn -d %d -o %d  
> \"%s\"";
> + char format_xp[] = "command.com /e:2048 /c latex2pn -d %d -o %d %s";

If the problem is just the extension, i.e.,  if

    latex2pn.bat -d 300 -o 6 "l2r_0001.tex"

works, then the latex2rtf should append the extension before making  
the system call
and then we can retain the quotes.  If the above fails, then your fix  
is probably as
good as we can do on XP.

In general, I like your pdf2pnga ideas.  That stuff will go back in  
as soon as
I get a chance...as well as the problem with \sqrt{} in a field ...

Thanks for testing/fixing things on XP.  Welcome back!

Scott

[Latex2rtf-developers] latex2rtf bugs

[Wilfried H. <W.H...@fz...>]

Hello Scott,

First, there is a bug in the 1.9.17 release as well as in the current 
cvs version.
The conversion of the following equation has a bug:

\begin{equation}
L=\frac{L_0}{\sqrt{(1-\left(\frac{\upsilon}{c}\right)^2}}
\end{equation}

The produced field function content (with semicolon as delimiter) is

EQ L= \F(L\s\do5(0); \R(;(1- \b ( \F(u;c))\s\up14(2)))

while it should be

EQ L= \F(L\s\do5(0); \R(1- \b ( \F(u;c))\s\up14(2)))


Second, only in the current cvs version, conversion of pictures 
works, but conversion of equations to bitmaps fails under WinXP:

1.) The filenames must not be quoted. If quoted, the call to the batch file is
latex2pn.bat -d 300 -o 6 "l2r_0001"
which causes the batch file to try processing "l2r_0001".tex which 
obviously doesn't exist.
Don't ask me why the quotes aren't removed by the command processor.

2.) Having fixed this (and the call to pdf2pnga), there remains the 
problem that the bitmaps resulting from equation conversion aren't 
inserted in the rtf file while the bitmaps resulting from graphics 
conversion are.
I haven't found why.


-- 
Wilfried Hennings c./o.
Forschungszentrum (Research Center) Juelich GmbH
INB-MUT
52425 Juelich, Germany
Tel. (+49) 2461 61 5570 ; Fax (+49) 2461 61 8254
email: W.H...@fz...
PGP key available from key servers, e.g. from
     http://pgpkeys.pca.dfn.de/
work homepage: http://www.fz-juelich.de/mut/mut_home
All opinions mentioned are strictly my own, not my employer's



-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich

Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv. 
Vorsitzender)
-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------

Re: [Latex2rtf-developers] pdf_to_png()

[Adam R. M. <ama...@ma...>]

Hi Wilfried,

On Sep 28, 2007, at 01:30, Wilfried Hennings wrote:

> Hello Scott and Adam,
> I downloaded subversion 706, changed the call to gs into call of
> pdf2pnga, system independent, and added two scripts, for DOS/Win and  
> Unix.
> This results in subversion 707.
> So anyone can modify the script to specify the processor he likes  
> best.

Sounds reasonable to me...

> But in graphics.c the function "round" is called which is undefined -
> compiler messages:
> graphics.c: In function `AdjustScaling':
> graphics.c:549: warning: implicit declaration of function `round'

I noticed that warning on RedHat as well.  I didn't check math.h  
there, since it linked successfully.

> ...
> graphics.o(.text+0x1154): In function `AdjustScaling':
> graphics.c:549: undefined reference to `_round'
> graphics.o(.text+0x117a):graphics.c:551: undefined reference to  
> `_round'
> graphics.o(.text+0x11d5):graphics.c:556: undefined reference to  
> `_round'

Is there a Windows equivalent of -lm missing?  I thought round () was  
part of POSIX.

-- 
adam

Re: [Latex2rtf-developers] pdf_to_png()

[Wilfried H. <W.H...@fz...>]

Hello Scott and Adam,
I downloaded subversion 706, changed the call to gs into call of 
pdf2pnga, system independent, and added two scripts, for DOS/Win and Unix.
This results in subversion 707.
So anyone can modify the script to specify the processor he likes best.

But in graphics.c the function "round" is called which is undefined - 
compiler messages:
graphics.c: In function `AdjustScaling':
graphics.c:549: warning: implicit declaration of function `round'
...
graphics.o(.text+0x1154): In function `AdjustScaling':
graphics.c:549: undefined reference to `_round'
graphics.o(.text+0x117a):graphics.c:551: undefined reference to `_round'
graphics.o(.text+0x11d5):graphics.c:556: undefined reference to `_round'


At 27.09.2007 16:56, you wrote:
>>I have problems with this pdf file, too, because convert doesn't
>>crop it.
>
>But everything else still works?  I was worried about regressions.
>
>>I probably will end up by calling a "pdf2png" script, which isn't
>>supplied by gs and would have to be included in the latex2rtf
>>distribution.
>
>This is ok.  No rush on resolving this problem.  I think that it only
>affects a small fraction of pdf files and then only on XP and then
>the only downside is that the whitespace is not cropped.


-- 
Dipl.-Ing.(=M.Eng.) Wilfried Hennings c./o.
Research Center Juelich
INB-MUT
52425 Juelich
Germany
Tel. (+49) 2461 61 5570 ; Fax (+49) 2461 61 8254
email: W.H...@fz...
PGP key available from key servers, e.g. from
     http://pgpkeys.pca.dfn.de/
work homepage: http://www.fz-juelich.de/mut/mut_home
All opinions mentioned are strictly my own, not my employer's



-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich

Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv. 
Vorsitzender)
-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------

Re: [Latex2rtf-developers] PDF graphics conversion

[Adam R. M. <ama...@ma...>]

 
On Thursday, September 27, 2007, at 03:34PM, "Adam R. Maxwell" <ama...@ma...> wrote:
>On OS X, using sips might be interesting, since it generally has pretty good PDF support, and wouldn't require any 3rd party software (gs, ImageMagick).  I might mess with that in my Copious Free Time.

Since I created a PDF graphic today that crashes gs, I had to try this.  In case someone else is interested in the hack, here it is for PDF.

Background:  I use pdflatex exclusively, and I have some valid PDF files that cause GhostScript to barf when converting to png.  On others, gs gets the crop box wrong in comparison with Adobe and Apple's PDF.  However, OS X 10.3 and later have a command-line tool called sips that processes images, and conveniently accepts PDF as input.  With this change and the scaling fixes in svn, I finally don't have to tweak graphics at all in the RTF file, except for the subfigure environment.

static char *SysGraphicsCommand(char *in, char *out, int specify_density_and_crop)
{
    char cmd[512];
	char format[]  = "convert -crop 0x0 -units PixelsPerInch -density %d \"%s\" \"%s\"";
	char format2[] = "convert \"%s\" \"%s\"";
	
	if (specify_density_and_crop == TRUE)
    	snprintf(cmd, 511, format, g_dots_per_inch, in, out);
    else
    	snprintf(cmd, 511, format2, g_dots_per_inch, in, out);
	
#ifdef UNIX
	/* convert works nicely, except for some pdfs, so use gs directly for those */
	if ((strstr(in, ".pdf") != NULL) || (strstr(in, ".PDF") != NULL)) {
#ifdef __APPLE__
        /* if Quartz is available, it does a better job than ghostscript in most cases */
        char form[] = "/usr/bin/sips -s format png -s dpiHeight %d -s dpiWidth %d --out \"%s\" \"%s\"";
    	snprintf(cmd, 511, form, g_dots_per_inch, g_dots_per_inch, out, in);
#else
		char form[] = "gs -q -dNOPAUSE -dSAFER -dBATCH -sDEVICE=pngalpha -r%d -sOutputFile=\"%s\" \"%s\"";
    	snprintf(cmd, 511, form, g_dots_per_inch, out, in);
#endif
    }
#endif

	return strdup(cmd);
}

Re: [Latex2rtf-developers] PDF graphics conversion

[Adam R. M. <ama...@ma...>]

 
On Wednesday, September 26, 2007, at 02:42PM, "Adam R. Maxwell" <ama...@ma...> wrote:
>As far as I could tell last time I looked at this, ImageMagick reads the /MediaBox instead of the /CropBox of the PDF file, then composites the ghostscript output to a bitmap of the media box size.

A (hopefully) final note on this topic: the problem doesn't occur with 5.5.x versions of ImageMagick; I think it came about after version 6 was released.  Anyway, 6.0.7 shipped with RHEL has the problem.

The ImageMagick docs confirm my statement about the media box, and say that you can use e.g. `convert file.pdf -define pdf:use-cropbox=true file.png`, but I can't get this to work here.  The -define flag requires a recent version of ImageMagick, so putting that in latex2rtf is probably not the best option.

On OS X, using sips might be interesting, since it generally has pretty good PDF support, and wouldn't require any 3rd party software (gs, ImageMagick).  I might mess with that in my Copious Free Time.

-- 
Adam

Re: [Latex2rtf-developers] nesting error

[Adam R. M. <ama...@ma...>]

 
On Thursday, September 27, 2007, at 12:44PM, "Scott Prahl" <pr...@bm...> wrote:
>
>On Sep 27, 2007, at 11:30 AM, Adam R. Maxwell wrote:
>
>> I'm getting a brace nesting error in one of my files (minimal  
>> example follows).  I'd take a shot at fixing it, but I'm not sure  
>> where to look.  Any ideas?  If I remove the begin/end doublespace,  
>> the problem goes away.
>
>Resolved in revision 703 (no double space, but no error any more).

Cool, thanks!  I don't care if it's double spaced; I'm just trying to minimize the stuff that has to be special cased for rtf conversion.

>Basically, somewhere in the bowels of the horrific subroutine  
>getSection(), nested environments fail to work.  There is probably a  
>general fix, but this patch should get you going.

After that description, I won't even bother looking for it ;).

Apologies for the double post...webmail is flaky when it times out.

regards,
Adam

Re: [Latex2rtf-developers] nesting error

[Scott P. <pr...@bm...>]

On Sep 27, 2007, at 11:30 AM, Adam R. Maxwell wrote:

> I'm getting a brace nesting error in one of my files (minimal  
> example follows).  I'd take a shot at fixing it, but I'm not sure  
> where to look.  Any ideas?  If I remove the begin/end doublespace,  
> the problem goes away.

Resolved in revision 703 (no double space, but no error any more).

Basically, somewhere in the bowels of the horrific subroutine  
getSection(), nested environments fail to work.  There is probably a  
general fix, but this patch should get you going.

Scott

>
> \documentclass[10pt]{article}
> \usepackage{setspace}
>
> \begin{document}
>
> \begin{doublespace}
>
> \section{Introduction}
> Test.
>
> \end{doublespace}
>
> \end{document}


Index: commands.c
===================================================================
--- commands.c  (revision 702)
+++ commands.c  (working copy)
@@ -891,6 +891,7 @@
      {"theindex", CmdIgnoreEnviron, 0},
      {"landscape", CmdTolerateEnviron, 0},
      {"sloppypar", CmdTolerateEnviron, 0},
+    {"doublespace", CmdTolerateEnviron, 0},
      {"", NULL, 0}
  };                              /* end of list */

[Latex2rtf-developers] nesting error

[Adam R. M. <ama...@ma...>]

I'm getting a brace nesting error in one of my files (minimal example follows).  I'd take a shot at fixing it, but I'm not sure where to look.  Any ideas?  If I remove the begin/end doublespace, the problem goes away.

\documentclass[10pt]{article}
\usepackage{setspace}

\begin{document}

\begin{doublespace}

\section{Introduction}
Test.

\end{doublespace}

\end{document}

[Latex2rtf-developers] nesting error

[Adam R. M. <ama...@ma...>]

I'm getting a brace nesting error in one of my files (minimal example follows).  I'd take a shot at fixing it, but I'm not sure where to look.  Any ideas?  If I remove the begin/end doublespace, the problem goes away.

\documentclass[10pt]{article}
\usepackage{setspace}

\begin{document}

\begin{doublespace}

\section{Introduction}
Test.

\end{doublespace}

\end{document}

[Latex2rtf-developers] PDF graphics conversion

[Adam R. M. <ama...@ma...>]

Hi Scott et al., 

I just saw that Wilfried reverted my patch for PDF graphics conversion, since it apparently fails under WinXP.  Unfortunately, his fix of using convert -crop 0x0 doesn't work for me, and all of my PDF graphics are screwed up again.  I'm using ImageMagick 6.3.4 and GhostScript 8.60 prerelease, both from May 2007.  Does anyone else have this problem?   Using gs directly works fine on Mac OS X and Linux.

As far as I could tell last time I looked at this, ImageMagick reads the /MediaBox instead of the /CropBox of the PDF file, then composites the ghostscript output to a bitmap of the media box size.  I posted an example PDF at <http://homepage.mac.com/amaxwell/.cv/amaxwell/Sites/.Public/SequimBayBathymetry.pdf-zip.zip>.

thanks, 
Adam

Re: [Latex2rtf-developers] Release candidate for latex2rtf

[Wilfried H. <W.H...@fz...>]

At 17.09.2007 19:56, Scott Prahl wrote:
>...
>
>It would be good to document unicode a bit better. Perhaps Wilfried
>could repeat his limited testing in latex2rtf.texi with a few unicode
>encoded files ([utf8x]{inputenc} and related matters).

I added a few sentences to the docu regarding utf-8 support.
I think that the users who use utf-8 encoded TeX input do know better 
than I how to use this, so I'd like to wait if something is not clear 
enough. utf-8 support was just recently added to latex2rtf and is 
correctly described as experimental, so anyone using it should be 
prepared to find bugs.


>It would be good to talk about the realities of equations and the poor
>conversion to equation editor and slightly better conversions with the
>MathType Editor.  If either of you could lift verbiage from the recent
>emails about this, then that would be good.

I also added some more sentences to the equations chapter.

>Here is a draft of the release document:
>
>
>latex2rtf 1.9.17 is a new release of a semantic converter
>from latex to rtf.  It has a number of new features and bug
>fixes and is recommended for all users of earlier versions
>of latex2rtf.
>
>Before installing you should be aware that any changes that
>you may have made to files in the cfg/ directory will be lost.
>These files will be overwritten in the installation process.
>
>New Features
>
>         much better handling of graphics
>         improved support of symbolic and combined characters
>         support for \iflatextortf ... \else ... \fi
>         better equation translation
>         support geometry package
>         \doublespacing support
>         full support for natbib package
>         support .aux files that include other .aux files
>         babel support for Ukranian
>         basic endnote functionality
>         support \doublespacing
>         support \underbar
>         support \vref
>
>Experimental Features
>         support for unicode in input files
>         direct output of equations as 'raw latex' in the rtf file
>
>Bug fixes
>         terminate processing when \end{document} encountered
>         support \dots
>         eliminate psfig crasher
>         support \quotedblbase
>         fix for non-ascii in verbatim environment
>         properly ignore small, sloppypar, and landscape environments
>         many other bug fixes (see changelog for details)

I like this.

If there is nothing else to fix, I am ready to upload the DOS version 
of this release.


-- 
Wilfried Hennings c./o.
Forschungszentrum (Research Center) Juelich GmbH
INB-MUT
52425 Juelich, Germany
Tel. (+49) 2461 61 5570 ; Fax (+49) 2461 61 8254
email: W.H...@fz...
PGP key available from key servers, e.g. from
     http://pgpkeys.pca.dfn.de/
work homepage: http://www.fz-juelich.de/mut/mut_home



-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich

Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv. 
Vorsitzender)
-----------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------

[Latex2rtf-developers] New version of latex2rtf

[Scott P. <pr...@bm...>]

latex2rtf 1.9.17 is a new release of a semantic converter
from latex to rtf.  It has a number of new features and bug
fixes.  It is recommended for all users of latex2rtf.

Before installing you should be aware that any changes that
you may have made to files in the cfg/ directory will be lost.
These files will be overwritten in the installation process.

New Features

	much better handling of graphics
	improved support of symbolic and combined characters
	support for \iflatextortf ... \else ... \fi
	better equation translation
	support margin settings in the geometry package
	full support for natbib package, including \bibpunct
	support .aux files that include other .aux files
	babel support for Ukranian
	basic endnote functionality
	support \doublespacing
	support \underbar
	support \vref

Experimental Features
	support for unicode in input files
	direct output of equations as 'raw latex' in the rtf file

Bug fixes
	terminate processing when \end{document} encountered
	support \dots
	eliminate psfig crasher
	support \quotedblbase
	fix for non-ascii in verbatim environment
	properly ignore small, sloppypar, and landscape environments
	many other bug fixes (see Changelog for details)