Received: by nummer-3.proteosys id <01C19443.9A755B2C@nummer-3.proteosys>; Thu, 3 Jan 2002 11:44:21 +0100 Return-Path: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C19443.9A755B2C" x-vm-v5-data: ([nil nil nil nil nil nil nil nil nil][nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil]) X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message Subject: sorry, resend fmtname Date: Tue, 11 Feb 1992 21:23:45 +0100 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: From: "Frank Mittelbach" Sender: "LaTeX-L Mailing list" To: "Rainer M. Schoepf" Reply-To: "LaTeX-L Mailing list" Status: R X-Status: X-Keywords: X-UID: 573 This is a multi-part message in MIME format. ------_=_NextPart_001_01C19443.9A755B2C Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable > It happens from time to time that a (La)TeX designer has to know in > which basic software his code was loaded and is executing. In order > to obtain this information we test "basically" \fmtname and we decide > that we are in LaTeX if \fmtname=3Dlplain. Unfortunately people = change > this name. We can't do nothing against this practice unless to say > that it's not a good thing. But do we admit really that \fmtname is > showing the name of a basic format-code and must not reflect user > modifications ? The custom of calling a LaTeX format file lplain is somewhat funny >from the beginning since the format contains much more then only macros >from lplain. Therefore I agree in principle if people try to use \fmtname as the place to record all information, i.e., all packages that are present in the format. This should perhaps also include the hyphen patterns used etc. etc. Therefore in my feeling such a record is quite good since its the only way to check, in some sense, what is in a format if you don't have the log file from the format generation. However, since there is no accepted procedure so far what to place into this cs and how, it isn't of much use at the moment. > Is-it a better and well admitted way to test if a code is running in > LaTeX or not ? To answer Bernard's question I would suggest to use a different control sequence to test which is very `latexly' and more or less safe >from being part of copied coded. Best guess would be a long internal name from latex's output routine:-). Testing for NFSS I normally use \ifx\selectfont\undefined@ .., but I agree that any such solution is neither perfect nor clean. > And in the LaTeX 3 project? For LaTeX3 we have considered a proper interface where every style file, additional code, and, and, has to identify itself to the kernel. Via this interface other modules then can test for the existence of a package if necessary. But providing something like this for the current LaTeX is too difficult to be worth in my opinion. cheers Frank =1A ------_=_NextPart_001_01C19443.9A755B2C Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable sorry, resend fmtname

> It happens from time to time that a (La)TeX = designer has to know in
> which basic software his code was loaded and is = executing. In order
> to obtain this information we test = "basically" \fmtname and we decide
> that we are in LaTeX if \fmtname=3Dlplain.  = Unfortunately people change
> this name. We can't do nothing against this = practice unless to say
> that it's not a good thing. But do we admit = really that \fmtname is
> showing the name of a basic format-code and must = not reflect user
> modifications ?

The custom of calling a LaTeX format file lplain is = somewhat funny
>from the beginning since the format contains much = more then only macros
>from lplain.  Therefore I agree in principle = if people try to use
\fmtname as the place to record all information, = i.e., all packages
that are present in the format. This should perhaps = also include the
hyphen patterns used etc. etc.  Therefore in my = feeling such a record
is quite good since its the only way to check, in = some sense, what is
in a format if you don't have the log file from the = format generation.
However, since there is no accepted procedure so far = what to place
into this cs and how, it isn't of much use at the = moment.


> Is-it a better and well admitted way to test if a = code is running in
> LaTeX or not ?

To answer Bernard's question I would suggest to use a = different
control sequence to test which is very `latexly' and = more or less safe
>from being part of copied coded. Best guess would = be a long internal
name from latex's output routine:-). Testing for NFSS = I normally use
\ifx\selectfont\undefined@ .., but I agree that any = such solution is
neither perfect nor clean.

> And in the LaTeX 3 project?

For LaTeX3 we have considered a proper interface where = every style
file, additional code, and, and, has to identify = itself to the kernel.
Via this interface other modules then can test for = the existence of a
package if necessary. But providing something like = this for the current
LaTeX is too difficult to be worth in my = opinion.

cheers
Frank
=1A

------_=_NextPart_001_01C19443.9A755B2C--