FriCAS 1.3.11 is released

[Waldek Hebisch]

I have put sources and binaries of FriCAS 1.3.11 on Sourceforge.

Sources correspond to r1.3.11 branch and 1.3.11 tag on Github.

--
Waldek Hebisch

[Qian Yun]

I've created release page at
https://github.com/fricas/fricas/releases/tag/1.3.11

And uploaded windows and macos binary built by CI, and reference book.

- Qian

[Andrey G. Grozin]

Thank you for the release,

After compiling fricas I did

cd src/doc
make book

and got (after some time)

tmp/ug10.tex:339: LaTeX Error: File `ribbons5' not found.

What can be the reason?

Andrey

[Qian Yun]

Can you show me the result of

ls src/doc/tmp/ribbon*

Also is this reproducible?

- Qian

[Andrey G. Grozin]

On Sun, 30 Jun 2024, Qian Yun wrote:
> ls src/doc/tmp/ribbon*
bilbo
/var/tmp/portage/sci-mathematics/fricas-1.3.11/work/fricas-1.3.11/src/doc
# ls tmp/ribbon*
tmp/ribbon1.pdf tmp/ribbon2.pdf tmp/ribbon2r.pdf tmp/ribbons2b.pdf
tmp/ribbons2.pdf tmp/ribbons.pdf

tmp/ribbon1.VIEW:
data image.ps

tmp/ribbon2r.VIEW:
data image.ps

tmp/ribbon2.VIEW:
data image.ps

tmp/ribbons2b.VIEW:
data image.ps

tmp/ribbons2.VIEW:
data image.ps

tmp/ribbons.VIEW:
data image.ps
bilbo
/var/tmp/portage/sci-mathematics/fricas-1.3.11/work/fricas-1.3.11/src/doc
#

Andrey

[Andrey G. Grozin]

On Sun, 30 Jun 2024, Qian Yun wrote:

> Also is this reproducible?
Dit

make clean
make book

The result is the same
Andrey

[Ralf Hemmecke]

Hi Andrey,

I assume you do "make clean book" in the build directory and you are not
building in the source directory.

Anyway, can you do the following

cd FRICASBUILDDIR
cd src/doc
rm -rf tmp/ribbon* stamp-ug* vp-*
make book | tee make-book.log

Please make this log available to us.

I guesss you are building on Windows? Under WSL2 or how exactly?

Thanks
Ralf

[Qian Yun]

Hi Ralf,

Looks like the problem is that target "${INPUT_EXTRA_INPUT}"
is not run before target "${HTEX_VIEWPORTS_STAMP}".

I missed this bug during testing because this happens with
"make -j1 book", not "make -j8 book".

Do you agree that following patch fix this?

- Qian

diff --git a/src/doc/Makefile.in b/src/doc/Makefile.in
index be40c60c..215dfa32 100644
--- a/src/doc/Makefile.in
+++ b/src/doc/Makefile.in
@@ -666,7 +666,7 @@
tmp/vp-knot3.input: knot3.input ${INPUT_EXTRA_INPUT}
cp $< $@

-${HTEX_VIEWPORTS_STAMP}: tmp/stamp-%: tmp/vp-%.input
+${HTEX_VIEWPORTS_STAMP}: tmp/stamp-%: tmp/vp-%.input ${INPUT_EXTRA_INPUT}
unset DAASE; FRICAS=${FRICAS}; export FRICAS; \
echo ")read $<" | FRICAS_INITFILE='' ${XVFB} \
${FRICAS}/bin/fricas -noht -noclef > /dev/null

[Waldek Hebisch]

On Sun, Jun 30, 2024 at 08:35:00PM +0800, Qian Yun wrote:
> Hi Ralf,
>
> Looks like the problem is that target "${INPUT_EXTRA_INPUT}"
> is not run before target "${HTEX_VIEWPORTS_STAMP}".
>
> I missed this bug during testing because this happens with
> "make -j1 book", not "make -j8 book".

AFAICS there is no problem when you make HyperDoc pages first.
And HyperDoc pages are build during normal build. So the
trouble can appear only when building from pre-generated
files.

> Do you agree that following patch fix this?

Certainly it is needed and with it book build seem to work.

> --
> You received this message because you are subscribed to the Google Groups "FriCAS - computer algebra system" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to fricas-devel...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/fricas-devel/cca4e87f-6d58-4752-9cdf-e2c2084c2487%40gmail.com.

--
Waldek Hebisch

[Andrey G. Grozin]

Dear Qian,

On Sun, 30 Jun 2024, Qian Yun wrote:

> Do you agree that following patch fix this?
>
> - Qian
>
> diff --git a/src/doc/Makefile.in b/src/doc/Makefile.in
> index be40c60c..215dfa32 100644
> --- a/src/doc/Makefile.in
> +++ b/src/doc/Makefile.in
> @@ -666,7 +666,7 @@
> tmp/vp-knot3.input: knot3.input ${INPUT_EXTRA_INPUT}
> cp $< $@
>
> -${HTEX_VIEWPORTS_STAMP}: tmp/stamp-%: tmp/vp-%.input
> +${HTEX_VIEWPORTS_STAMP}: tmp/stamp-%: tmp/vp-%.input ${INPUT_EXTRA_INPUT}
> unset DAASE; FRICAS=${FRICAS}; export FRICAS; \
> echo ")read $<" | FRICAS_INITFILE='' ${XVFB} \
> ${FRICAS}/bin/fricas -noht -noclef > /dev/null

Yes! Now it works.

I'm the maintainer of the fricas package in gentoo linux. Now I'm trying
to add the USE flag "doc" to fricas-1.3.11.ebuild.

I'll add this patch. The next problem is that X is needed. I hope that the
eclass virtualx (the function virtx in which startx a new Xvfb session and
runs a command in it) will help.

Many thanks,
Andrey

[Ralf Hemmecke]

On 6/30/24 14:35, Qian Yun wrote:
> I missed this bug during testing because this happens with
> "make -j1 book", not "make -j8 book".
>
> Do you agree that following patch fix this?

I haven't tested, but it looks good to me.

I hope for the next release, I'll have code to generated these
INPUT_EXTRA files from the .htex files.

Ralf

PS: In fact, there is also a missed version change in citation.rst.

@Misc{FriCAS,
key = {FriCAS},
author = {{FriCAS team}},
year = {2024},
title = {{FriCAS} 1.3.10---an advanced computer algebra system},
note = {Available at \url{http://fricas.github.io}}
}

I'll also try to set this up for the next release so that 1.3.10 becomes
|PACKAGE_VERSION| and thus will be automatically be generated correctly.

[Ralf Hemmecke]

Hi Waldek,

there is actually a little issue with the sphinx stuff.

I can certainly write code to pick up the year from the commit date at
compile time and put it into the year for the BibEntry and also in the
copyright string that appears at the bottom of every generated html-page.

However, we certainly also want the year to appear correctly, if the
user is running 'make doc' or 'make localdoc'.

So, probably no problem, if there is a .git directory, but if a user
gets the sources from the tarball, then I do not see any reliable source
for the year inside the actual content of the tarball.

To resolve this "year problem", an option is that you include

src/doc/sphinx/source/citation.rst and
src/doc/sphinx/source/conf.py

https://github.com/fricas/fricas/blob/master/src/doc/sphinx/source/citation.rst?plain=1#L15
https://github.com/fricas/fricas/blob/master/src/doc/sphinx/source/conf.py#L47

into your list of things to set appropriately before the release is done.

Honestly, I'd prefer a variable, for example PACKAGE_YEAR or
PACKAGE_RELEASE_DATE (maybe in configure.ac, which you change for the
release) that distributes to the Makefiles. Such a variable can then be
used to automate that sphinx stuff. So all version stuff just in
configure.ac.

What do you prefer?

Ralf

[Waldek Hebisch]

On Sun, Jun 30, 2024 at 09:12:03PM +0200, Ralf Hemmecke wrote:
> Hi Waldek,
>
> there is actually a little issue with the sphinx stuff.
>
> I can certainly write code to pick up the year from the commit date at
> compile time and put it into the year for the BibEntry and also in the
> copyright string that appears at the bottom of every generated html-page.
>
> However, we certainly also want the year to appear correctly, if the user is
> running 'make doc' or 'make localdoc'.
>
> So, probably no problem, if there is a .git directory, but if a user gets
> the sources from the tarball, then I do not see any reliable source for the
> year inside the actual content of the tarball.

I am not sure what the problem is. Why do you want copyright strings
on generated pages? If you want overall year for given release, then
change log should be good enough. And it would work the same for
snapshots.

--
Waldek Hebisch

[Qian Yun]

On 6/30/24 23:41, Andrey G. Grozin wrote:
> Dear Qian,

>
>
> I'm the maintainer of the fricas package in gentoo linux. Now I'm trying
> to add the USE flag "doc" to fricas-1.3.11.ebuild.
>
> I'll add this patch. The next problem is that X is needed. I hope that
> the eclass virtualx (the function virtx in which startx a new Xvfb
> session and runs a command in it) will help.
>
> Many thanks,
> Andrey
>

I don't think "virtualx" is needed. FriCAS already uses "xvfb-run"
automatically under the hood. If "xvfb-run" is installed, everything
should be fine.

- Qian

[Grégory Vanuxem]

Hello,

I forgot to say thank you. Awesome job!

If I had something to say about the PDF book it's about the use of red
colored terms. Studies show, as far as I know, that most mammals
process this color very specifically. Direction of gaze, more
generally orientation reaction, movement of the head for example, time
to process the stimulus and much more. One could say that we interpret
it as: maybe something bad will happen. Not only on roads ;) I can
understand the emphasis on those terms in the FriCAS book but as a
personal point of view I am not a big fan of how this color is used.

- Greg

> --
> You received this message because you are subscribed to the Google Groups "FriCAS - computer algebra system" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to fricas-devel...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/fricas-devel/ZoC2SLwGSMTgXtPo%40fricas.org.

[Ralf Hemmecke]

> I am not a big fan of how this color ((red)) is used.

Greg, can you be a bit more specific?

All this color stuff in my invention and certainly not a very clever
one. There are, for example \spadfun, \spadop, that use different
colors. I'am not sure myself that this is a good choice. Then categories
in this color and Domains in another. I am not an artist. If anyone
tells me a good choice of colors that highlights the different types and
looks visually nice, I am open to change the color scheme. That's pretty
easy. (Actually, anyone who does not like it, can easily change it by
just changing the settings in fricas.sty and recompiling.)

Note that in the Jenks and Sutor book, they used italics and bold and
other things instead of colors. We could also to that.


Ralf

[Grégory Vanuxem]

Hello Ralf,

Le mer. 3 juil. 2024 à 15:54, Ralf Hemmecke <ra...@hemmecke.org> a écrit :
>
> > I am not a big fan of how this color ((red)) is used.
>
> Greg, can you be a bit more specific?

Beside the fact that the red color is somewhat special, personally I
would use it to draw attention on hazardous things and I prefer, for
example, the caution symbol ☡ from N. Bourbaki (yes, I know ;-)) to
say the truth. If I agree on HTML "typesetted" documents that colored
text can be of use, with a à la Knuth typesetted document I prefer
grayscale. But it is just a matter of taste.

> All this color stuff in my invention and certainly not a very clever
> one. There are, for example \spadfun, \spadop, that use different
> colors. I'am not sure myself that this is a good choice. Then categories
> in this color and Domains in another. I am not an artist. If anyone
> tells me a good choice of colors that highlights the different types and
> looks visually nice, I am open to change the color scheme. That's pretty
> easy. (Actually, anyone who does not like it, can easily change it by
> just changing the settings in fricas.sty and recompiling.)

Thanks for the hint, I will keep the one chosen here though. And in
fact I am not against color, just TeX typesetted documents are like in
books for me and I prefer reading a book without a lot of colors
unless it contains images or "visual" things.

> Note that in the Jenks and Sutor book, they used italics and bold and
> other things instead of colors. We could also to that.

Personally I also prefer, but at that time, when I think of printed
documents, it was different I guess also.

Again, just a matter of taste. This was just for the color red.

- Greg



>
>
> Ralf

>
> --
> You received this message because you are subscribed to the Google Groups "FriCAS - computer algebra system" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to fricas-devel...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/fricas-devel/df2e42a8-70d9-472a-aeec-d2ac659bbc30%40hemmecke.org.