Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

How Can I put eps file in latex 2.09 ?

297 views
Skip to first unread message

Jong-Uk Bae

unread,
Nov 10, 1998, 3:00:00 AM11/10/98
to
I use following sequences

\begin{figure}[h]
\centerline
\epsfig{Ms.eps}
\caption{test}
\end{figure}

But i get dvi file including eps file.
In Log file,
! Undefined control sequence.
<argument> \epsfig
l.375 \epsfig
{Ms.eps}
Help me!
How can I do?
Thank you!!

Hans Ulrich Marschall

unread,
Nov 10, 1998, 3:00:00 AM11/10/98
to
hallo

sorry if i am giving you a silly answer, but are you using:

\usepackage{epsfig}

anywhere before \begin{document}
and do you have the file epsfig.sty anywhere in your latex-search-path.
if not have a look at:
http://www.dante.de/cgi-bin/ctan-index
there you can find the .sty-file

uli

Bert Tanner

unread,
Nov 10, 1998, 3:00:00 AM11/10/98
to
You can try:
....
\input{psfig}
......
\begin{figure*} [t]
\vspace{...in}
\special{psfile=***.ps hoffset=-280 voffset=-405 hscale=120 vscale=120}
\caption{***}
\label{**}
\end{figure*}

It has worked in the past.
If you find something better pls send me a mail cause i also have an old
machine running 2.09 and i also look for more elegant ways to include
eps to latex 2.09 files.

Bert.

Jong-Uk Bae wrote:

> I use following sequences
>
> \begin{figure}[h]
> \centerline
> \epsfig{Ms.eps}
> \caption{test}
> \end{figure}
>
> But i get dvi file including eps file.
> In Log file,
> ! Undefined control sequence.
> <argument> \epsfig
> l.375 \epsfig
> {Ms.eps}
> Help me!
> How can I do?
> Thank you!!

--
* Herbert G. Tanner
* Ph.D. Student
* Control Systems Laboratory,
* Dept. of Mechanical Engineering
* National Technical University of Athens

MicroPress, Inc.

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to
What you are doing here does not use psfig at all!.
The idea of psfig was to provide a high-level interface to .eps
inclusion (something graphicx now does much better) so one
does not have to use dvips-specific \special's. I suspect that
a more elegant way can be found by reading psfig's documentation.

(A more interesting question is what it would take to make graphicx
work with LaTeX 2.09.... The graphicx author would surely dissapprove
of doing this, but there may be a pressing need for this
sometimes....)

MicroPress, Inc.

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to
On Tue, 10 Nov 1998 16:55:37 +0100, Hans Ulrich Marschall
<m883...@grz08u.unileoben.ac.at> wrote:

>hallo
>
>sorry if i am giving you a silly answer, but are you using:
>
>\usepackage{epsfig}
>
>anywhere before \begin{document}
>and do you have the file epsfig.sty anywhere in your latex-search-path.
>if not have a look at:
>http://www.dante.de/cgi-bin/ctan-index
>there you can find the .sty-file
>
>uli

One problem with this code could be that \usepackage does not exist in
2.09....

David Carlisle

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to MicroPress, Inc.

> (A more interesting question is what it would take to make graphicx
> work with LaTeX 2.09....

As far as I know it works already in compatibility mode. It's been a
while since I started up a `real' 2.09 but it probably wouldn't take
much, but honestly I see no possible point. If you have an old machine
running old documents and you are happy running your old latex, why
change? But what's the point of saying that you want to include all the
latex features introduced since 1992 _except_ for the one line change
that changed the startup banner from LaTeX 2.09 to LaTeX2e ?

> The graphicx author would surely dissapprove
> of doing this,

Of course he would, although he has already put on ctan a wrapper so you
can run that package under plain TeX, so between that and the normal
LaTeX interface it would not presumably be much work to make a 2.09
one.

> but there may be a pressing need for this
> sometimes....)

Hmph.

David

MicroPress, Inc.

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to
On 11 Nov 1998 10:25:03 +0000, David Carlisle <dav...@nag.co.uk>
wrote:

>
>> (A more interesting question is what it would take to make graphicx
>> work with LaTeX 2.09....
>
>As far as I know it works already in compatibility mode. It's been a
>while since I started up a `real' 2.09 but it probably wouldn't take
>much, but honestly I see no possible point. If you have an old machine
>running old documents and you are happy running your old latex, why
>change? But what's the point of saying that you want to include all the
>latex features introduced since 1992 _except_ for the one line change
>that changed the startup banner from LaTeX 2.09 to LaTeX2e ?

Firstly, this is not an attack on 2E or graphicx (IMHO, single best
piece since Plain TeX).

However, the compatibility mode is the worst thing about 2E: it really
does not work in too many places (particularly, in substituting
another fontset) and thus should not be taken seriously.

However, 2.09 is still out there. Apparently, there are packages that
require it (revtex, for example, and some things people did themselves
over the years) and an informal survey of our users
about a possible removal of 2.09 from our distribution brought back
several "please, no".

I think everybody would benefit if graphicx could totally eliminate
the need for using psfig &| epsf and that would indeed require 2.09
compatibility.

The need is probably not that great as to beg you "please do it", but
IMHO it is something to at least consider.

Just my 2c worth.

David Carlisle

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to MicroPress, Inc.

> Firstly, this is not an attack on 2E or graphicx (IMHO, single best
> piece since Plain TeX).

flattery will get you nowhere:-)

> However, the compatibility mode is the worst thing about 2E: it really
> does not work in too many places (particularly, in substituting
> another fontset) and thus should not be taken seriously.

Actually our experience is the opposite, most reports we have had have
indicated that the vast majority of 209 documents go through 2e
unchanged. One area in which compatibility was explictly documented as
being poor was for 2.09 setups that had messed around with fonts. The
font handling in 2e is so different (that, mainly was the point of
making 2e, to have a rational font interface). In practice this does not
cause many problems as font changing in 209 was hard enough that few
people did it. That doesn't help you much as I guess that your setup
has _always_ provided alternative font sets, so there would be more work
to do there, I agree. I would be surprised though if a latex209.def file
for vtex could not be constructed that made the majority of old vtex
latex documents work unchaged. (let's face it: that must be true,
latex209.def could input lplain.tex....)

> Apparently, there are packages that
> require it (revtex, for example,

I have `very good evidence' that a 2e based revtex is on the cards.

> I think everybody would benefit if graphicx could totally eliminate
> the need for using psfig &| epsf and that would indeed require 2.09
> compatibility.

As I said last time it can't be that hard to do (but I won't do it
myself) I made the wrapper for plain TeX in about ten minutes by
just inputting graphicx.sty into a plain tex document and then defining
commands (copied from latex) until the errors went away.
That version (in generic/graphics on ctan) may already work with 2.09 as
far as I know, if not it would not take much to make it work.

The reason why I won't do it is the same as the reason we stuck
\NeedsTeXFormat{LaTeX2e} at the top of our style files (even though
that was typically a lie and initially they would have worked in 2.09
if it were not for that command). However that doesn't stop anyone else
from doing it.

The longer you try to support obsolete software the longer you prolong
the problems and the longer you hinder document portability.
If you had just stopped supporting latex2.09 in 1994 then people would
have moaned at you but by now that version of latex would just be a
fading memory.

Of course they may not have moaned at you; they may have quit using your
product and switched to a competitor who was evil enough to continue
to support 2.09, and you may have gone bust. I can see that could have
been a problem at the time, but by now it is probably safe:-)


David

Louis Vosloo

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to
David Carlisle wrote:

> Actually our experience is the opposite, most reports we have had have
> indicated that the vast majority of 209 documents go through 2e
> unchanged. One area in which compatibility was explictly documented as
> being poor was for 2.09 setups that had messed around with fonts.

Yes, that is the key issue. And people tend to be surprised when their LaTeX 2.09
material doesn't work in `compatabilty mode' simply because it isn't using CM fonts.

> The font handling in 2e is so different (that, mainly was the point of
> making 2e, to have a rational font interface). In practice this does not
> cause many problems as font changing in 209 was hard enough that few
> people did it.

Well, that is a slight exaggeration, since LucidaBright + LucidaNewMath,
as well as MathTime 1.1 and MathTime Plus, have LaTeX 2.09 support (and
plain TeX support). This includes support for various font encodings.
And yes, bold math doesn't work in LaTeX 2.09 worth a damn, but before
LaTeX 2e there was no other show in town, so quite a few people used fonts
other than CM with LaTeX 2.09. And since *some* people aren't focused on
honing the tools they use --- rather on what they are trying to do ---
the transition is a slow and painful one.

--
Y&Y, Inc. mailto:sup...@YandY.com http://www.YandY.com

MicroPress, Inc.

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to
On 11 Nov 1998 17:25:33 +0000, David Carlisle <dav...@nag.co.uk>
wrote:

>


>> Firstly, this is not an attack on 2E or graphicx (IMHO, single best
>> piece since Plain TeX).
>
>flattery will get you nowhere:-)

No flattery intended.....and I'm not certain where I'm trying to
get... But if you feel I am too positive, let me correct the
impression:

Could you please allow implementation-defined keys in
\includegraphics options? Not having them already lead
to ugliness in vtex.def (global \count's to be set) and things
are likely to get just worse....

If we could have this, you would get more (hated) flattery.


>
>> However, the compatibility mode is the worst thing about 2E: it really
>> does not work in too many places (particularly, in substituting
>> another fontset) and thus should not be taken seriously.
>

>Actually our experience is the opposite, most reports we have had have
>indicated that the vast majority of 209 documents go through 2e
>unchanged. One area in which compatibility was explictly documented as

>being poor was for 2.09 setups that had messed around with fonts. The


>font handling in 2e is so different (that, mainly was the point of
>making 2e, to have a rational font interface). In practice this does not
>cause many problems as font changing in 209 was hard enough that few

>people did it. That doesn't help you much as I guess that your setup
>has _always_ provided alternative font sets, so there would be more work
>to do there, I agree. I would be surprised though if a latex209.def file
>for vtex could not be constructed that made the majority of old vtex
>latex documents work unchaged. (let's face it: that must be true,
>latex209.def could input lplain.tex....)

We tried & failed. The entire thing was not sufficiently important to
put more work into it (after all we did have a working 2.09 all the
time)


>
>> Apparently, there are packages that
>> require it (revtex, for example,
>
>I have `very good evidence' that a 2e based revtex is on the cards.

If this is so, big chunk of my reason is gone. But there are still
people out there who just _prefer_ 2.09. And there are people who
have 2.09 installed on their workstations and they do not have much
say in the matter.


>
>> I think everybody would benefit if graphicx could totally eliminate
>> the need for using psfig &| epsf and that would indeed require 2.09
>> compatibility.
>
>As I said last time it can't be that hard to do (but I won't do it
>myself) I made the wrapper for plain TeX in about ten minutes by
>just inputting graphicx.sty into a plain tex document and then defining
>commands (copied from latex) until the errors went away.
>That version (in generic/graphics on ctan) may already work with 2.09 as
>far as I know, if not it would not take much to make it work.
>
>The reason why I won't do it is the same as the reason we stuck
>\NeedsTeXFormat{LaTeX2e} at the top of our style files (even though
>that was typically a lie and initially they would have worked in 2.09
>if it were not for that command). However that doesn't stop anyone else
>from doing it.
>
>The longer you try to support obsolete software the longer you prolong
>the problems and the longer you hinder document portability.
>If you had just stopped supporting latex2.09 in 1994 then people would
>have moaned at you but by now that version of latex would just be a
>fading memory.
>
>Of course they may not have moaned at you; they may have quit using your
>product and switched to a competitor who was evil enough to continue
>to support 2.09, and you may have gone bust. I can see that could have
>been a problem at the time, but by now it is probably safe:-)

I never thought about the problem in this way (losing users) and
dropping 2.09 is not a big concern for me -- it works, does not take
much space and does not need to be retested all the time. But dropping
psfig and epsf (or at least being able to mark them "do not use" would
have made me much happier.)
>
>
>David

Michael

MicroPress, Inc.

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to
On Wed, 11 Nov 1998 12:54:24 -0500, Louis Vosloo <sup...@YandY.com>
wrote:

>David Carlisle wrote:
>
>> Actually our experience is the opposite, most reports we have had have
>> indicated that the vast majority of 209 documents go through 2e
>> unchanged. One area in which compatibility was explictly documented as
>> being poor was for 2.09 setups that had messed around with fonts.
>

>Yes, that is the key issue. And people tend to be surprised when their LaTeX 2.09

>material doesn't work in `compatabilty mode' simply because it isn't using CM fonts.


>
>> The font handling in 2e is so different (that, mainly was the point of
>> making 2e, to have a rational font interface). In practice this does not
>> cause many problems as font changing in 209 was hard enough that few
>> people did it.
>

>Well, that is a slight exaggeration, since LucidaBright + LucidaNewMath,
>as well as MathTime 1.1 and MathTime Plus, have LaTeX 2.09 support (and
>plain TeX support). This includes support for various font encodings.
>And yes, bold math doesn't work in LaTeX 2.09 worth a damn, but before
>LaTeX 2e there was no other show in town, so quite a few people used fonts
>other than CM with LaTeX 2.09. And since *some* people aren't focused on
>honing the tools they use --- rather on what they are trying to do ---
>the transition is a slow and painful one.

Well, actually it is much easier to implement a new fontset in 2.09
than in 2E. With HV-math and the other forthcoming fontset it took
around 10-15 minutes to add support for Plain or 2.09. It took _a few
hours_ to add the support for 2E. -- not a complain, just a statement
of fact.

MV
http://www.micropress-inc.com


David Carlisle

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to MicroPress, Inc.

Could you please allow implementation-defined keys in
\includegraphics options?


yes

in graphics.def just have

\AtEndOfPackage{
\RequirePackage{keyval}
\define@key{gin}{xyz}{\def\vtex@xyz{#1}}
}

then in vtex.def the code to handle the included graphic can look at the
value of \vtex@xyz which will have whatever the user gave as [xyz=....]
(with a bit of luck:-)

Of course this has the usual warnings about document not being portable to
other drivers etc...

> If we could have this, you would get more (hated) flattery.

I can take it.

> If this is so, big chunk of my reason is gone.

It is so.

> But there are still people out there who just _prefer_ 2.09.

Apply violence until they change their minds. That usually works.

> And there are people who have 2.09 installed on their workstations and
> they do not have much say in the matter.

Depends on the size of their disk quota:-)

David

David Carlisle

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to MicroPress, Inc.

Would you (or anyone else) prefer if I made a one line change
to keyval.sty equivalent to the test below, which makes it give
warnings rather than errors on spurious input.
(Or it could be an option to chose, but then which should be the
default.) (current def is \def\KV@err#1{\PackageError{keyval}{#1}{}})


\documentclass{article}

\usepackage{graphicx}
\makeatletter
\def\KV@err#1{\PackageWarning{keyval}{#1}{}}
\begin{document}

\includegraphics[rubbish,bb]{foo}

\end{document}

Package keyval Warning: rubbish undefined on input line 8.


Package keyval Warning: No value specified for bb on input line 8.


David Kastrup

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to
sup...@micropress-inc.com (MicroPress, Inc.) writes:

> Well, actually it is much easier to implement a new fontset in 2.09
> than in 2E. With HV-math and the other forthcoming fontset it took
> around 10-15 minutes to add support for Plain or 2.09. It took _a few
> hours_ to add the support for 2E. -- not a complain, just a statement
> of fact.

On the other hand, if the new fontset has a different font encoding
than the input encoding, and inparticular if you want to mix old and
new fonts in one document, LaTeX2.09 is going to break an arm and
both legs.

It is not difficult to replace the entire font stuff of LaTeX2.09 (or
plain TeX) by an entirely different, consistent one. But having more
than one scheme active and working at the same time is pretty hard.


--
David Kastrup Phone: +49-234-700-5570
Email: d...@neuroinformatik.ruhr-uni-bochum.de Fax: +49-234-709-4209
Institut für Neuroinformatik, Universitätsstr. 150, 44780 Bochum, Germany

MicroPress, Inc.

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to
Vote for putting it -> keyval with the default set to warning and an
option [strict available for keyval.

Thank you!

M.

On 11 Nov 1998 22:06:10 +0000, David Carlisle
<da...@dcarlisle.demon.co.uk> wrote:

>
>Would you (or anyone else) prefer if I made a one line change

>to keyval.sty equivalent to the test elow, which makes it give


>warnings rather than errors on spurious input.
>(Or it could be an option to chose, but then which should be the
>default.) (current def is \def\KV@err#1{\PackageError{keyval}{#1}{}})
>
>
>\documentclass{article}
>
>\usepackage{graphicx}
>\makeatletter
>\def\KV@err#1{\PackageWarning{keyval}{#1}{}}
>\begin{document}
>

>\includegraphics[rubbish,bbc{foo}

MicroPress, Inc.

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to
true. I wa referring to the case where we copied the encoding eactly.

On 11 Nov 1998 23:08:32 +0100, David Kastrup

Louis Vosloo

unread,
Nov 11, 1998, 3:00:00 AM11/11/98
to
David Kastrup wrote:

> sup...@micropress-inc.com (MicroPress, Inc.) writes:

> > Well, actually it is much easier to implement a new fontset in 2.09
> > than in 2E. With HV-math and the other forthcoming fontset it took
> > around 10-15 minutes to add support for Plain or 2.09. It took _a few
> > hours_ to add the support for 2E. -- not a complain, just a statement
> > of fact.

> On the other hand, if the new fontset has a different font encoding
> than the input encoding,

I don't know, we have been doing this for a very long time (pre 1992 in
any case). It is as simple as \input texnansi, \input ansinew,
\input dcaccen etc. using files we provide with our fonts. Without
this the fonts would have been much less useful, particularly since OT1
does not provide a good match for anything but CM. Keep in mind
that support for plain TeX and LaTeX 2.09 was crucial for LucidaBright
and MathTime since they predate LaTeX 2e (and may have influenced some
of the font machinery in LaTeX 2e as a result).

> and in particular if you want to mix old and new fonts in one document,


> LaTeX2.09 is going to break an arm and both legs.

Using text fonts in one job with more than one encoding is mighty
confusing no matter what format you are using.

Hans Ulrich Marschall

unread,
Nov 12, 1998, 3:00:00 AM11/12/98
to
hallo

> >\usepackage{epsfig}


>
> One problem with this code could be that \usepackage does not exist in
> 2.09....

i think it should exist in 2.09 because i have found it in the following
directory on ctan:
macros/latex209/contrib/epsfig/epsfig.sty

uli


MicroPress, Inc.

unread,
Nov 19, 1998, 3:00:00 AM11/19/98
to

epsfig sure exists, but "\usepackage" ?!

0 new messages