Bug#384958: tex4ht: some grouping issues with oolatex math
Kapil Hari Paranjape schrieb: Since I do not use OO, I am not sure whether this issue has been resolved. Could you please provide some information on the current state? OOo 2.2.1-7 has not fixed this bug. According to the OOo issue tracker it's not fixed up to now. BTW: it's Issue 69088 there. http://qa.openoffice.org/issues/show_bug.cgi?id=69088 Tobias. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#384958: tex4ht: some grouping issues with oolatex math
tags 384958 + upstream tags 384958 - moreinfo retitle 384958 openoffice.org: incorrect import of mathml into oomath reassign 384958 openoffice.org thanks Hello, Based on this response from the upstream author of tex4ht: On Wed, 26 Sep 2007, Eitan Gurari wrote: The tex4ht utility produces OpenDocument format, where math is represented in mathml. When loading an OpenDocument file, OpenOffice translates the mathml code into an internal oomath format. The mathml code produced by tex4ht for the given example is correct, and it displays fine both under Mozilla and MSIE+MathPlayer. The behavior of OpenOffice is incorrect. and this response from the original submitter of the bug: On Wed, 26 Sep 2007, Tobias Schlemmer wrote: OOo 2.2.1-7 has not fixed this bug. According to the OOo issue tracker it's not fixed up to now. BTW: it's Issue 69088 there. http://qa.openoffice.org/issues/show_bug.cgi?id=69088 I'm reassigning the bug to openoffice.org and tagging it upstream. Thanks to both of them for their prompt responses. Regards, Kapil. -- signature.asc Description: Digital signature
Bug#384958: tex4ht: some grouping issues with oolatex math
tags 384958 + moreinfo
thanks
Hello,
On Mon, 28 Aug 2006, Tobias Schlemmer wrote:
mk4ht oolatex test.tex seems to produce incorrect output.
At least my oowriter (2.0.3) does not like the double indices $x_{i_j}$
without further grouping.
In oomath input syntax it produces x_i_j, which should be x_{i_j}.
On Tue, 29 Aug 2006, Eitan Gurari wrote:
I'm not sure, however, what effort it will take for non-trivial
cases, and I don't see where I can find the time for this job in
the near futute. I'm also wondering whether such a fix is not more
appropriate for the OO people to do.
Since I do not use OO, I am not sure whether this issue has been
resolved. Could you please provide some information on the current
state?
Thanks and regards,
Kapil.
--
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#384958: tex4ht: some grouping issues with oolatex math
mk4ht oolatex test.tex seems to produce incorrect output.
At least my oowriter (2.0.3) does not like the double indices $x_{i_j}$
without further grouping.
In oomath input syntax it produces x_i_j, which should be x_{i_j}.
On Tue, 29 Aug 2006, Eitan Gurari wrote:
I'm not sure, however, what effort it will take for non-trivial
cases, and I don't see where I can find the time for this job in
the near futute. I'm also wondering whether such a fix is not more
appropriate for the OO people to do.
Since I do not use OO, I am not sure whether this issue has been
resolved. Could you please provide some information on the current
state?
The tex4ht utility produces OpenDocument format, where math is
represented in mathml. When loading an OpenDocument file, OpenOffice
translates the mathml code into an internal oomath format. The mathml
code produced by tex4ht for the given example is correct, and it
displays fine both under Mozilla and MSIE+MathPlayer. The behavior
of OpenOffice is incorrect.
A possible `solution' could have been for tex4ht to complement the
math in the OpenDocument code with oomath code, and by doing so help
with bugs within the OpenOffice engine. It is a task I unlikely to
work on.
I don't think this issue should be the concern of tex4ht or the debian
distribution. I hope before long it will be resolved within
OpenOffice.
-eitan
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#384958: tex4ht: some grouping issues with oolatex math
Eitan Gurari wrote:
The problem is with a broken mathml engine in OpenOffice 2.
A manual editing of x_i_j into {x}_{{i}_{j}} or x_{i_j} provides the
proper display (and identical mathml code). I don't know what kind of
mathml code tex4ht should produce to obtain correct display in OO2.
As a workaround use the math:annotation tag with StarMath5 syntax as
OO does.
Deleting this tag looks right at the first glance as OOo saves a binary
copy of it inside the document. But when you try to edit the formula it
gets confused again.
Tobias
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#384958: tex4ht: some grouping issues with oolatex math
As a workaround use the math:annotation tag with StarMath5 syntax as OO does. Deleting this tag looks right at the first glance as OOo saves a binary copy of it inside the document. But when you try to edit the formula it gets confused again. Might be a good idea for simple formulas. I'm not sure, however, what effort it will take for non-trivial cases, and I don't see where I can find the time for this job in the near futute. I'm also wondering whether such a fix is not more appropriate for the OO people to do. -eitan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#384958: tex4ht: some grouping issues with oolatex math
Package: tex4ht
Version: 20060619-1
Severity: normal
Hi,
mk4ht oolatex test.tex seems to produce incorrect output.
At least my oowriter (2.0.3) does not like the double indices $x_{i_j}$ without
further grouping.
In oomath input syntax it produces x_i_j, which should be x_{i_j}.
A similar effect produces
\begin{eqnarray}a=b\end{eqnarray}, where the = expects two operators, but
a # = # b is produced. Both '{}={}' and '=' (oowriter syntax) produce
useful output.
Example:
\documentclass{article}
\begin{document}
\[\bar\rho_{s_i} = \sum_{j\neq i}\rho_{s_j}(r_{ij})\]
\begin{eqnarray}
\label{eq:1}
\bar\rho_{s_i} = \sum_{j\neq i}\rho_{s_j}(r_{ij})
\end{eqnarray}
\end{document}
Note: xtpipes won't accept the eqnarray. So, please try it with commenting it
out.
Note: I didn't check against the newer versions of tex4ht mentioned in
Bug#384578.
-- System Information:
Debian Release: testing/unstable
APT prefers dapper-updates
APT policy: (500, 'dapper-updates'), (500, 'dapper-security'), (500,
'dapper-proposed'), (500, 'dapper-backports'), (500, 'dapper'), (500,
'breezy'), (500, 'testing'), (500, 'stable')
Architecture: amd64 (x86_64)
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.15-1-amd64-generic
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)
Versions of packages tex4ht depends on:
ii libc6 2.3.6-15 GNU C Library: Shared libraries
ii libkpathsea4 3.0-17 path search library for teTeX (run
ii tetex-bin 3.0-17 The teTeX binary files
ii tex4ht-common 20060619-1 LaTeX and TeX for Hypertext (HTML)
tex4ht recommends no packages.
-- no debconf information
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#384958: tex4ht: some grouping issues with oolatex math
Hello,
On Mon, 28 Aug 2006, Tobias Schlemmer wrote:
Package: tex4ht
Version: 20060619-1
Severity: normal
mk4ht oolatex test.tex seems to produce incorrect output.
At least my oowriter (2.0.3) does not like the double indices $x_{i_j}$
without further grouping.
In oomath input syntax it produces x_i_j, which should be x_{i_j}.
Perhaps this is the same bug which has already been noted in the
documentation. Generally speaking HyperText documents are more strict
about subscript grouping than TeX. So authors should prefer
$x_{a_{b}}$ while preparing a document to be processed by TeX4ht
regardless of how simple a and b are.
Kapil.
--
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#384958: tex4ht: some grouping issues with oolatex math
Kapil Hari Paranjape wrote:
So authors should prefer
$x_{a_{b}}$ while preparing a document to be processed by TeX4ht
regardless of how simple a and b are.
Which documentation?
The problem is not b, but the grouping around a_b is lost. TeX needs
that grouping, which is not translated into the math.
Tobias
begin:vcard
fn:Tobias Schlemmer
n:Schlemmer;Tobias
org:Forschungszentrum Rossendorf;FWSM
adr;quoted-printable:;;Bautzner Landstra=C3=9Fe 128;Dresden;SN;01328;Germany
email;internet:[EMAIL PROTECTED]
title:Dipl.-Math.
tel;work:(+49 3 51) 2 60 - 21 27
x-mozilla-html:FALSE
url:http://www.fz-rossendorf.de/
version:2.1
end:vcard
signature.asc
Description: OpenPGP digital signature
Bug#384958: tex4ht: some grouping issues with oolatex math
The problem is with a broken mathml engine in OpenOffice 2.
eqnarray
The \begin{eqnarray}a=b\end{eqnarray} is translated by tex4ht into
mia/mimo=/momib/mi
and is loaded as
matrix {a # = # b}
by OO2 into a broken display. The xtpipe phase `fixes' the problem by
producing improper mathml output
mia/mimtext=/mtextmib/mi
which OO2 loads as
matrix {a # = # b}
and provides proper display.
double subscript
The output of tex4ht on $x_{i_j}$ is
math:msub
math:mrow
math:mix/math:mi
/math:mrow
math:mrow
math:msub
math:mrow
math:mii/math:mi
/math:mrow
math:mrow
math:mij/math:mi
/math:mrow
/math:msub
/math:mrow
/math:msub
and OO2 loads the code into a broken format and view x_i_j. The same
outcome occurs when tex4ht is modified to produce the following output.
math:msub
math:mix/math:mi
math:msub
math:mii/math:mi
math:mij/math:mi
/math:msub
/math:msub
A manual editing of x_i_j into {x}_{{i}_{j}} or x_{i_j} provides the
proper display (and identical mathml code). I don't know what kind of
mathml code tex4ht should produce to obtain correct display in OO2.
mk4ht oolatex test.tex seems to produce incorrect output.
At least my oowriter (2.0.3) does not like the double indices $x_{i_j}$
without further grouping.
In oomath input syntax it produces x_i_j, which should be x_{i_j}.
A similar effect produces
\begin{eqnarray}a=b\end{eqnarray}, where the = expects two operators, but
a # = # b is produced. Both '{}={}' and '=' (oowriter syntax) produce
useful output.
Note: xtpipes won't accept the eqnarray. So, please try it with commenting
it out.
Note: I didn't check against the newer versions of tex4ht mentioned in
Bug#384578.
-- System Information:
Debian Release: testing/unstable
APT prefers dapper-updates
APT policy: (500, 'dapper-updates'), (500, 'dapper-security'), (500,
'dapper-proposed'), (500, 'dapper-backports'), (500, 'dapper'), (500,
'breezy'), (500, 'testing'), (500, 'stable')
Architecture: amd64 (x86_64)
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.15-1-amd64-generic
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)
Versions of packages tex4ht depends on:
ii libc6 2.3.6-15 GNU C Library: Shared libraries
ii libkpathsea4 3.0-17 path search library for teTeX
(run
ii tetex-bin 3.0-17 The teTeX binary files
ii tex4ht-common 20060619-1 LaTeX and TeX for Hypertext
(HTML)
tex4ht recommends no packages.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
