Re: LAPACK vs. OpenBLAS

[Paul Garlick]

Hi Ludo,

> A surprisingly large number of packages depend on ‘lapack’:

> Perhaps we could have a lint checker warning against the use of
lapack.

Good idea.  Possibly with a helpful message along the lines of 'the
openblas package provides a LAPACK interface'.

I encountered this issue when packaging an optimization package [0]
recently.  The build system, cmake, requires a path to the BLAS_LIBRARY
and also a path to the LAPACK_LIBRARY.  One can see how the shared
library liblapack.so, provided by the lapack package, could be the
first (but mistaken) choice for the packager.

I notice that Debian [1] use NO_LAPACK=1 as an extra make option for
openblas.  This  has the effect of generating a liblapack.so file.

For the case of optizelle, the package I was working on, I labelled the
openblas input as "blas/lapack" to make it clear that the package has a
dual purpose.

Best regards,

Paul.

[0] https://hpc.guix.info/package/optizelle
[1] https://sources.debian.org/src/openblas/0.3.13+ds-3/debian/rules/

Re: git-fetch for emacs-auctex?

[Paul Garlick]

Hi Nicolas,

On Tue, 2021-05-25 at 22:10 +0200, Nicolas Goaziou wrote:
> Hello,
> 
> There are no tags in the ELPA repository, but new releases are
> triggered by a bump of "Version:" keyword. So, it is technically
> possible to map a version to a commit hash by looking for such
> changes. This has been suggested already in bug#46489.
> 

Thank you for pointing to the bug report 
https://issues.guix.gnu.org/issue/46849

It is the same issue as found in Nix.  This does not seem to be fixed
yet.  

I imagine there is a choice between an auctex-only fix and a general
ELPA fix.

Best regards,

Paul.

Re: git-fetch for emacs-auctex?

[Paul Garlick]

Hi Leo,

On Tue, 2021-05-25 at 22:00 +0200, Leo Prikler wrote:
> 
> What it this auctex and how does it differ from the one packaged in
> ELPA?
> 
This is the repository for the AUCTeX project.  The home page is 
https://www.gnu.org/software/auctex/

The ELPA package is derived from the upstream version.  It has its own
version numbering scheme.  For example, the current version on ELPA is
13.0.11.  The current upstream version is 12.3.

Every so often the upstream development branch is merged into the ELPA
branch and a new ELPA version is released.

One option for the Guix package would be to bypass ELPA and use the
upstream repository directly.  However, the version string would change
from "13.0.11" to something like "12.3-commit", where 'commit'
identifies the git reference of the upstream branch.

This would solve the reproducibility problem.  However, a one-off
update resulting in a smaller version number might cause confusion.

Are there views on this point?

Best regards,

Paul.

git-fetch for emacs-auctex?

[Paul Garlick]

Hi Guix,

I recently attempted to reproduce a profile containing emacs-auctex,
only to find a '404: not found' error.

The reason is that elpa.gnu.org archives previous auctex versions with
a .tar.lz extension.   This throws out the Guix package source
definition, which expects a .tar extension.

I have considered a few ideas to solve this:

i) track the last-but-one auctex version in Gnu Elpa.  This way all
versions would have the same extension.  However, would the url-fetch
method cope with the .tar.lz format?

ii) use the git-fetch method with the upstream repository.  However,
the upstream version tag at http://git.savannah.gnu.org/cgit/auctex.git
 is a smaller number than the Gnu Elpa version tag.  This would result
in an apparent downgrade.  Would that be permissible?

iii) use the git-fetch method to clone the auctex branch of the
elpa.git repository (along the lines of 'git clone --single-branch --
branch externals/auctex .../elpa.git).  As far as I can see this is not
currently possible.

Are there any insights or suggestions?

Best regards,

Paul.

Re: Adding Trilinos to dealii package

[Paul Garlick]

Dear Paul,

> I would like to eventually package the Lethe CFD library

Nice!

Trilinos has many bells and whistles. I imagine that there will be some
significant effort required to package it.  One option would be to
subdivide the library and initially package only the elements you need
for Lethe.  See the Debian approach to packaging Trilinos for example
[1].

If Lethe does not use PETSc there would be no risk of conflict.  The
choice between including the new Trilinos (sub)package as a dependency
of dealii and dealii-openmpi or creating new dealii-trilinos and
dealii-trilinos-openmpi packages would be marginal.  How many models
need to use PETSc and Trilinos at the same time?

> This assumes people won't mind having to build Trilinos as a
> dependency for dealii even if they don't need it.

The model run times are usually very much longer than the library build
times.  I expect this would be true for Trilinos and Lethe too.

Best regards,

Paul.

[1]: https://packages.debian.org/source/buster/trilinos

Re: configure: error: A recent Guile-zlib could not be found; please install it.

[Paul Garlick]

Hi Roel,

> How can I get a working development environment to work on Guix?

A 'guix pull' within your profile will update the guile-zlib version
that is used by 'guix environment ...'.  Then the configure script
requirement will be met.

Best regards,

Paul.

Re: gpg key expiration time

[Paul Garlick]

Hi Tobias,

On Mon, 2021-03-15 at 18:24 +0100, Tobias Geerinckx-Rice wrote:

> The expiry date is not embedded in your cryptographic key, merely 
> attached to it as a signed ‘packet’.

Thanks. Got it!

Best regards,

Paul.

gpg key expiration time

[Paul Garlick]

Hi Guix,

After renewing the expiration time of one's gpg key, and uploading the
updated key to Savannah, is it also necessary to update the
corresponding file in the keyring branch?

I notice from the log that, so far, none of the original files have
been updated.  Does this mean that the original validation is accepted
for new commits signed with an updated key?

Best regards,

Paul.

Re: bug#45069: BUG: Re: guix environment: error: cannot create container: unprivileged user cannot create user namespaces

[Paul Garlick]

Hi Pierre,

> # echo 1 > /proc/sys/kernel/unprivileged_userns_clone
> -bash: /proc/sys/kernel/unprivileged_userns_clone: No such file or
> directory

Thanks, that gives us a clue.  So all or part of the path
'/proc/sys/kernel' is missing?

Best regards,

Paul.

Re: bug#45069: BUG: Re: guix environment: error: cannot create container: unprivileged user cannot create user namespaces

[Paul Garlick]

Hi Pierre,

Can you try, as root on Guix System:

$ echo 1 > /proc/sys/kernel/unprivileged_userns_clone

If you could report success or failure that would be helpful; the
unprivileged-user-namespace-supported? test in gnu/build/linux-
container.scm should be the same irrespective of the underlying
distribution (Debian, CentOS, Guix System ...).

Best regards,

Paul.

On Mon, 2020-12-07 at 12:57 +0100, Pierre Neidhardt wrote:
> Hi!
> 
> I can reproduce the issue since I 'recondigure'd my Guix System.
> I'm on cebfb29abb151ede95696181d2446c63504593d7.
> 
> Guix' bug?
> 
> 

Re: Licensing issue with Gmsh?

[Paul Garlick]

Hi Simon,

> There is only a parmetis.c file included in the metis directory.
> However, it is commented out in CMakeLists.txt.

> To be "Guix compliant", this parametis.c file should be removed,
> right?

The file is part of METIS and thereby covered by the Apache license. 
It only provides entry points for ParMETIS, so ignoring it in the build
stage, in the way that Gmsh does, seems to be appropriate.

However, I notice from the patch that the Guix package metis is
included as an input.  This means that the whole of the bundled
'contrib/metis' directory could be deleted after all.  If I remember
correctly CMake gives preference to the 'system' packages ahead of the
bundled packages.

Best regards,

Paul.

Re: Licensing issue with Gmsh?

[Paul Garlick]

Hi Simon,

Thank you for working on the Gmsh package.  There have been a great
many changes since the last update.

> I am inclined to enable METIS because it appears to me
fine.  However,
> since freedom is essential, it would like another advice and/or
comment
> on.

There is a distinction between METIS (free) and ParMETIS (non-free)
which may have influenced the decision to remove the whole metis
directory.  As far as I can see Gmsh does not use any files from the
ParMETIS library.  

There is only a parmetis.c file included in the metis directory. 
However, it is commented out in CMakeLists.txt.

Best regards,

Paul.

Re: getting started with the texlive importer

[Paul Garlick]

Hi Ricardo, Hi John,

> Looks good to me.  Please push!

Cheers!

Pushed as 735808b12cc23909b421e10e212a07e7aa69a5eb 

Best regards,

Paul.

Re: getting started with the texlive importer

[Paul Garlick]

Hi Guix,

Ok, I have found a fix.

It turns out that the 'svn export' command throws an error if the
target directory already exists.  

Initially I set the 'log' argument of 'download-svn-to-store' to
'current-output-port' to see the error message from svn:

svn: E155000: Destination directory exists; please remove the directory
or use - -force to overwrite

This error is generated because a temporary directory is created by the
'call-with-temporary-directory' procedure (from svn-download.scm).  The
name of the directory is used as an argument for the 'svn export'
command.

The fix I have tested is the following:

--- a/guix/svn-download.scm
+++ b/guix/svn-download.scm
@@ -159,10 +159,11 @@ reports to LOG."
 (parameterize ((current-output-port log))
   (build:svn-fetch (svn-reference-url ref)
(svn-reference-revision ref)
-   temp
+   (string-append temp "/svn")
#:user-name (svn-reference-user-name
ref)
#:password (svn-reference-password
ref)
(and result
-(add-to-store store name #t "sha256" temp))
+(add-to-store store name #t "sha256"
+  (string-append temp "/svn")))
 
 ;;; svn-download.scm ends here

The effect is to add a subdirectory to the temporary directory.  This
is used as the target to download the files to.  It does not exist
until created by the 'svn export' command.

WDYT?  If there are no objections I will push a commit early next week.

Best regards,

Paul.

Re: getting started with the texlive importer

[Paul Garlick]

Hi Guix,

I have been able to run the texlive importer after making some changes
to guix/import/texlive.scm:

@@ -148,20 +148,22 @@ expression describing it."
  ((lst ...) (map string->license lst
(home-page  (string-append "http://www.ctan.org/pkg/"; id))
(ref(texlive-ref component id))
-   (checkout   (download-svn-to-store store ref)))
+   ;; (checkout   (download-svn-to-store store ref))
+   )
   `(package
  (name ,(guix-name component id))
  (version ,version)
  (source (origin
(method svn-fetch)
(uri (texlive-ref ,component ,id))
-   (sha256
-(base32
- ,(bytevector->nix-base32-string
-   (let-values (((port get-hash) (open-sha256-
port)))
- (write-file checkout port)
- (force-output port)
- (get-hash)))
+   ;; (sha256
+   ;;  (base32
+   ;;   ,(bytevector->nix-base32-string
+   ;; (let-values (((port get-hash) (open-sha256-
port)))
+   ;;   (write-file checkout port)
+   ;;   (force-output port)
+   ;;   (get-hash)
+   ))
  (build-system texlive-build-system)
  (arguments ,`(,'quote (#:tex-directory ,(string-join (list
component id) "/"
  (home-page ,home-page)

This successfully produces a definition, albeit without the hash field.

So it seems that the procedure 'download-svn-to-store ' is the source
of the original error.  Any ideas on how to fix?

Paul.  

getting started with the texlive importer

[Paul Garlick]

Hi Guix,

I have been trying to use the texlive importer to package some TeXLive
packages.

Using the example from the manual, I try:

$ guix import texlive fontspec

I get:

following redirection to `https://ctan.org/xml/1.2/pkg/fontspec'...
Backtrace:
  13 (primitive-load "/usr/local/bin/guix")
In guix/ui.scm:
  2050:12 12 (run-guix-command _ . _)
In guix/scripts/import.scm:
   116:11 11 (guix-import . _)
In guix/scripts/import/texlive.scm:
91:19 10 (guix-import-texlive . _)
In guix/memoization.scm:
 98:0  9 (mproc "fontspec" "latex")
In unknown file:
   8 (_ #
…)
In ice-9/boot-9.scm:
  1736:10  7 (with-exception-handler _ _ #:unwind? _ # _)
In guix/store.scm:
   631:37  6 (thunk)
In guix/import/texlive.scm:
   151:23  5 (_ #)
In guix/utils.scm:
705:8  4 (call-with-temporary-directory #)
In guix/svn-download.scm:
   160:14  3 (_ "/tmp/guix-directory.nfum2k")
In guix/build/svn.scm:
 39:2  2 (svn-fetch _ _ _ #:svn-command _ #:recursive? _ # _ # _)
In guix/build/utils.scm:
654:6  1 (invoke _ . _)
In ice-9/boot-9.scm:
  1669:16  0 (raise-exception _ #:continuable? _)

ice-9/boot-9.scm:1669:16: In procedure raise-exception:
ERROR:
  1. &invoke-error:
  program: "svn"
  arguments: ("export" "--non-interactive" "--trust-server-cert" "-
r" "51265" "svn://
www.tug.org/texlive/tags/texlive-2019.3/Master/texmf-dist/source/latex/fontspec
" "/tmp/guix-directory.nfum2k")
  exit-status: 1
  term-signal: #f
  stop-signal: #f

So the 'svn export' command is causing an error.  Interestingly, if I
run the same command from the shell it succeeds; the files are
downloaded to the temporary directory.

Are others seeing the same thing?  I wonder whether a recent change in
guix has jiggered the importer.

Best regards,

Paul.

Re: Reproducible Research Hackathon: Friday, July 3rd

[Paul Garlick]

Hi Simon,

On Fri, 2020-07-03 at 20:19 +0200, zimoun wrote:
> 
> I do not know if it is the best but I do:
> 
>   guix build -L path/to/clone the-package
>   guix install -L path/to/clone the-package -p /tmp/test
>   /tmp/test/bin/the-package

This works for me if I use the command:

$guix build -L /path/to/guix-past-checkout/modules the-package

Many thanks for the tip.

Best regards,

Paul.

Re: Reproducible Research Hackathon: Friday, July 3rd

[Paul Garlick]

Hi Ricardo,

Many thanks for this tip.

> I did this from the checkout of “guix-past”:
> 
> GUIX_PACKAGE_PATH=$PWD guix build my-package
> 

First, I changed directory to the 'modules' directory in the guix-past
checkout.  Then   the 'guix build' command can follow the
subdirectories and find the package definitions.

Best regards,

Paul.

Re: Reproducible Research Hackathon: Friday, July 3rd

[Paul Garlick]

Hi Simon,

Thank you for organising the day.

> We are interested to hear your feedback.  Especially about what pass,
> what fail and what you have learnt, if you enjoyed the experience, or on
> the contrary if you not, what could be improved for the next round.

One outstanding puzzle for me is to figure out how to test a package
definition in a new channel.  For example, I have cloned the guix-past
repository and started to add a new package.  However, what is the best
workflow to attempt a build from the channel?  Do you add the channel
to the git checkout of guix?

Best regards,

Paul.

Re: 01/03: gnu: ikiwiki: Revert to standard wrapper.

[Paul Garlick]

Hi Chris,

> I don't think there was ever a bug report for this, but I think I've
> found the IRC conversation for context:
> 
>   http://logs.guix.gnu.org/guix/2017-10-17.log#203630

Thank you for this link and taking a look at this patch.

I have been adding the missing inputs to ikiwiki.  In the process I
noticed that the cgi script needs a way of finding the perl
dependencies.  The Debian way of packaging the perl modules allows
ikiwiki to see its dependencies immediately.  In Guix, as you said, it
needs some help via PERL5LIB.

In using ikiwiki I found that I needed to add the PERL5LIB variable to
the setup file anyway, so that the cgi script can see extra modules
that are not needed at build time.  In my setup file there is an 'ENV'
section which contains: 

PERL5LIB: /home/paul/.guix-profile/lib/perl5/site_perl

This means that there is no need to have PERL5LIB in the wrapper. 
Instead the dependencies can be propagated in the normal way.  This
fits in with the PREREQ_PM section of ikiwiki's Makefile.PL.

So far,everything I have tested works as expected.  After the latest
commit 
e642f78f2eaf1d53d656ff3cd86d4dd186e8eb52 all the main tests pass.

Best regards,

Paul.

Re: Building R Packages

[Paul Garlick]

Hi Daniela,

Have you tried './pre-inst-env guix build ...' instead of 'guix build
...'?

The difference is that the former command will look for package
definitions in your checked out, and modified, version of Guix.  

The latter command looks for package definitions in your installed
version of Guix.  Your modifications will be seen by this command after
you submit a patch, it is committed, then you update your installed
version with a 'guix pull'.

Best regards,

Paul.

Re: Packaging FreeCAD

[Paul Garlick]

Hi John,

> I'm excited to announce that I opened FreeCAD for the first time this
> evening thanks most recently to support on the FreeCAD forum! 

Great news!

> Now where do I put the package definition in gnu/packages? 

How about engineering.scm for FreeCAD?  This module already contains
other CAD-related packages such as LibreCAD and KiCad.

Best regards,

Paul.

Re: Video license

[Paul Garlick]

Hello Björn,

> For the code I would suggest GPL v3+, for the voices (currently Paul) 
> and videos I would suggest CC-BY-SA 4.0
> 
> What do you think? Does every contributor agree to that?

That looks good to me.  

For the source code we could add copyright statements at the top of the
files in a similar way to the rest of the Guix code.

Also, would you like to take a look at the CREDITS file and add entries
for the work that you have done?.  The idea is twofold: to acknowledge
the contributions made and also to make the steps more visible. 
Potential future contributors could then say ' hey, I could do that in
my language'.

Best regards,

Paul. 

Re: Documentation videos are being uploaded!

[Paul Garlick]

Hi Laura,

> Yes, sure! I also agree that it is important giving credits :)
> Would you mind asking them if they want to have their name there?

I have made a start and pushed a new CREDITS file to the repository. 
There are entries in the list for the contributors that I know about
(thanks, Ricardo!) but I suspect that there has been more work done
than I am aware of.

Would you like to take a look and fill in the gaps?  I have seen the
names of Gabor and Bjorn in the logs, for example.

Best regards,

Paul.

Re: Packaging FreeCAD

[Paul Garlick]

Hi John,

> Another question: do we tend to try to use guix packages whenever a
package 
> ships bundled with some third party source?

Yes.  The preference is to use Guix packages where possible.

Best regards,

Paul.

Re: Packaging FreeCAD

[Paul Garlick]

Hi John,

It is good to hear that you have made such good progress on packaging
FreeCAD.

> we would need to also package opencascade-occt 

I can help with this part.  I am planning to introduce the opencascade-
occt package in order to upgrade gmsh, which also depends on
OpenCASCADE.  

All being well, I should be able to start on this within the next two
weeks.  I will update this thread when the package is ready.

Best regards,

Paul.

Re: Documentation videos are being uploaded!

[Paul Garlick]

Hi All,

> Kudos Laura & Paul!
> I am cc'ing Paul :)

It is good to hear that the end result is being well received :)

I am thinking of adding a CREDITS file to the repository, primarily to
acknowledge the help given by the sound engineer.  The use of the
studio equipment and the editing work made this part much easier than
it would otherwise have been.

Also, if all the other contributors would like to add in a record for
their work on workflow design, scripting, slide production and so on it
would help future contributors understand how the videos are made.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> I found the reason for this.

Great.

> I wanted to take a look at how totem was packaged but I can't find
> it.

The method I use to track packages is to search at 
https://guix-hpc.bordeaux.inria.fr/browse

In this case a search for 'totem' points to the definition at
gnu/packages/gnome.scm, line 3845.

If you take a look at the definition you can see how the dependencies
are specified.  I am not very familiar with the Gnome packages but I
could imagine there may be a difference between the Debian list of
dependencies and the Guix list.  

Perhaps your test has revealed a bug in the Guix packaging of totem.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> BTW, I finished generating the videos.
> Sth weird happens with totem. I end up having a segmentation fault,

Did this problem resolve itself?

I tested 01-installation-from-script on my system without seeing any
interruption.  I am testing with totem version 3.26.0 on Ubuntu 18.04.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> Could you re-build the other videos and check the internal timings?
> This is the fine adjustment that synchronizes the audio with the
> command line output.
> By this, do you  mean checking that the audio matches the video,
> taking into account that if I have to add some silence in between I
> have to remove the same amount of time somewhere else?

Yes.  I think that the total duration of each cli session is now good. 
However, there may be improvements in a few places to further
synchronize the narration with the screen output.  

For example, if a particular output is displayed slightly ahead of the
narration one could add some extra sleep before the section and remove
the same amount after.

That way the total duration, of each session, will stay the same.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> I see the only one that needs to be finished is 04-packaging3.
> Would you like me to adjust it? 

I can do the coarse adjustment on that one, to set the overall duration
and make sure that the audio streams are not truncated.

Could you re-build the other videos and check the internal timings? 
This is the fine adjustment that synchronizes the audio with the
command line output.  

I have briefly checked that this is sensible but there could be
improvements to be made.  The overall duration should stay the same. 
So if you remove 1000 ms of sleep from one section, for example, it
needs to be added back in elsewhere.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> Now I am generating all the videos and will be pushing the changes
> asap.

Great stuff.

I have made a start, setting the durations for the first three videos. 
If you pull the changes you can check the inner timings as well.  The
sum of the sleep commands should now be correct for these videos.

I plan to finish the coarse level sleep settings for the remaining
videos in the next few days.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Ricardo,

> Are you saying you want to take snapshots in larger intervals and pad
> the intermediate frames with duplicates of previous frames?

Thank you for writing the snap! procedure.  It is writing the output as
intended.  

The problem we ran into was to do with the timing.  If you go back in
this thread you will see the history.  

In order to produce .webm files that are compatible with the Firefox
browser it is necessary to eliminate gaps in the audio streams.  ffmpeg
can do this automatically using the 'apad' filter.  However, it will
truncate the audio stream if the video stream is shorter in duration
than the audio stream.  

So, in preparing the cli session files we need to be able to accurately
set the duration of the last 'sleep' command.  Since we know the frame
rate (25 fps) it is straightforward to calculate the number of repeated
frames we need for any given duration.  This is the job of the new
'repeat' procedure.  

In this way we effectively replace the timer with a counter.

The session files now need to be checked to tune the sleep commands to
synchronize with the audio stream.  I will be able to look at 02-daily-
use1 again tomorrow.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> My good news are that I get the 10 secs difference, the bad ones that
> the timing is different:
> 00:01:18.40800
> and
> 00:01:28.40700
> Any clue?

Great.  These are the same, my timings expressed in seconds yours in
minutes and seconds.

I think we can live the millisecond difference in the last significant
digit!

So we should be able to finish from here without major obstacles. 
However, I suspect that the cli timings that you did previously will
need to be re-checked as we will now be using the accurate 'sleep'
command.

If you could check the inner sleep commands in the way that you did
before I can adjust the ending sleep to make sure that the 'apad'
filter works correctly.  How about that?

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> I like testing (and trying to break) stuff - I
> wished this worked well though :/

I have a new version to test.  It's unbreakable though, this one!

I have implemented the repeat procedure and pushed the commit to the
repository.

> I definitely HAVE TO learn Guile. 

This is a good example.  

There is some arithmetic needed to convert the sleep durations, in
milliseconds, to frames, using the frame rate.  If you have seen
reverse polish notation before you will recognise the way the
arithmetic expressions are built up.  

The function comes first and the arguments afterwards.  So (/ wait
1000), for example, means divide the variable 'wait' by one thousand.

The repeat procedure uses a 'named let' for recursion.  This calls the
snap! procedure the correct number of times to generate the extra
frames.

The slightly tricky part was making sure that the snap! procedure and
the repeat procedure are in the same scope.  Otherwise, an 'unbound
variable' error is generated.  I was able to do this by moving the
repeat procedure, plus the other procedures that are needed for
processing the session files, inside the same let construct in the main
body.

To test I used the firstCli session file from 02-daily-use1.  I changed
the last line from '#:sleep 3' to '#:sleep 4'.  The time
durations were:

before:  78.408000
after:  88.408000

That's more like it!   a delay of ten seconds, as it should be.

Can you confirm that you get  the same result?

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> firstCli originally gives me a length of 00:01:54.84700
> Then, adding the sleep metacommand I get 00:02:18.24800
> 
> As regards secondCli, originally: 00:00:59.60800
> And adding the sleep metacommand: 00:01:11.64800

Thank you for testing.  The results are different from my system.  

I think we are seeing a difference between wall-clock time and cpu
time.  In the screen file I can see there is a snap! procedure that
writes a frame to disk after each interval.  It looks as if the disk
i/o is interrupting the sleep counter and introducing a delay that
depends on the system activity.

We may need to change tack to get accurate time delays.  One thought is
to replace the timer with a 'repeat' procedure that repeats the
previous frame.  If we know the sleep duration, s, and the interval, i,
we can calculate the number of repeats, N, needed:

N = s / i

This method would be unaffected by disk write times or rounding errors.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> With the new changes, right?

Yes.

In my case I was noticing erratic intervals produced by the sleep
commands.  So '#:sleep 1000' gave a 6 s pause and '#:sleep 13000' gave
a 26 s pause, for example.  After the changes the commands produce the
expected results.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> I see. I was testing with the current screen file. Please, let me
> know when you push the changes so that I make all the videos again
> and adjust what is necessary. 

Many thanks.  I have pushed the changes to the repository.

It would be great if you can test the accuracy on your system too.  If
you add '#:sleep 1' to one of the session files, for example, does
it make the cli video 10 seconds longer?

If so, I think it will make finishing off the timings and checking the
durations much easier.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> > I have made a start with the 01-installation-from-script video.  I
> > am using Firefox version 66.0.3 to test.

> Ok I will go on with the followings then!

Working on the timings for 02-daily-use1 I have found that I have
needed to make a couple of changes to the 'hesitate' procedure in the
screen script.  I have been finding that the pause introduced by the
sleep command is not accurate, making it difficult to set the right
values in the sessions files.

I wonder whether you have noticed the same effect?

To improve the accuracy I have changed the procedure definition to use
a scaling factor of one thousand, to go from milliseconds to
microseconds, and have increased the chunk size to 2 ms.  The Guile
documentation for usleep notes that many systems round the interval
specified up or down and are unable to achieve accuracy to the
microsecond level.  

After some testing on the system I am using to build the videos I found
that 2 ms was the optimum value.  With this value '#:sleep 3', for
example, produces a pause of 30 s.

I will commit the change so that you can test it too.  If we keep the
change it will mean re-checking the timings in the session files. 
However, I think that this will be necessary to make the adjustments in
the total durations for Firefox.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura and All,

I have determined the cause of the playback problems with Firefox and
pushed a fix to the  videos.git repository.

Essentially, a limitation of the treatment of opus audio streams in
Firefox means that videos with gaps in the audio stream are not played
properly.

There were two parts to the fix.  First, the audio streams need to be
padded with silence to match the duration of the video streams.  Then,
the concatenation of the audio streams needs to be done separately from
the video streams.  The two streams are the same duration and may be
joined without re-encoding.

A consequence of the new approach is that the cli sessions need to be
checked to ensure that the duration of the video streams are (slightly)
longer than the audio streams.  Otherwise, the audio streams will be
truncated.  

I have made a start with the 01-installation-from-script video.  I am
using Firefox version 66.0.3 to test.

Can others confirm that the new method works with their browsers and
players? 

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> I deleted all the videos and made them again
> and now I don't see that issue anymore.

Ah, good news!

> For this, what I could do is generate the cli sesison video with the
> "silence1.mp3" to have duration1, and then compare it with the
> duration of cli1.mp3 (for instance).  Then, run a script comparing
> the durations while duration2 < duration1 append silence1.mp3 to
> cli1.mp3.  And finally use the generated cli1.mp3 as the input
> matching the video stream.
> I find a little bit manual, but I don't mind doing it if you find it
> OK.
> WDYT?

We may have to experiment a little to solve the question of combining
the cli and nocli videos in a manner than can be replayed by
Firefox/Icecat.

One test I can try tomorrow is to concatenate the nocli videos on their
own.  Will the combined video play in Firefox?  If so, it would suggest
that the mismatch of the audio and video stream durations is the cause
of the stoppage.  

There is a way to concatenate audio files with ffmpeg.  So if we know
we need an extra  seconds of silence, say, we can generate the file
with the command I previously posted and join the silence to the end of
the narration.

This would be a script-free alternative method.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura, 

> Now, as regards the colours, it works with totem too :)
> But the cli sessions are at kind of super speed now, it happens
> watching the full video, or the separate cli session videos. I've
> tried it with totem, mvp and vlc. Any clue about it?

I have not seen this effect on the playback speed.  I am using the
first video for testing.  I do:

$  make clean VIDEO=01-installation-from-script

then,

$ ./build-video.sh 01-installation-from-script

This plays at the right speed for me in Videos/Totem (version 3.26).

However, the concatenation question (for Firefox) may also be related
to the speed question.  Is super-speed very different from the expected
speed?  In the case of the first cli session for the first video there
is a difference between the duration of the audio and video streams:

duration1 (video stream):  1 minute 29 seconds
duration2 (audio stream):  1 minute 18 seconds

This difference may be treated in a different manner by different
players.  It may be necessary to add silence to the audio stream to
equalize the durations and allow reliable concatenation without re-
encoding.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura and Ludo,

> the colours are changed. Ie: the dark gray turns
> green, and the white turns somewhat pink.

Commit 4bd2e78b893fef5ce1f12bec895ee8234cabaf1f fixes the colors. 
ffmpeg needed a different pixel format to set the chroma subsampling
correctly [1].

I have been able to play the newly built cli and nocli videos in
Firefox.  Also, the combined video now starts in Firefox but stops just
before the first transition.  I think the concatenation step in the
Makefile could be the cause.

Best regards,

Paul.

1. https://trac.ffmpeg.org/wiki/Encode/VP9

Re: Video narration

[Paul Garlick]

Hi Laura,

> Can you tell me which players work for you? I tried with my browser
> and it still tells me that the file is corrupt :/

I have tested Parole and Videos media players and Firefox and Midori
browsers.  All work except Firefox, which gives the message:

'Video can't be played because the file is corrupt'

Is Firefox the browser you are testing with?  

Possibly another tweak to the ffmpeg encoding might be needed to
support Firefox.  

> I didn't notice this. Is it possible to make the audio shorter than
> the first duration?

Ah, the expression for duration1 should be:

duration1 (in seconds) = number of frames / 25

What I was thinking was that the cli sessions should end with a short
pause.  This will be the case as long as either duration1 is greater
than duration2 or the audio file has a silent section at the end.

If any of the transitions are too abrupt we can add some silence to the
relevant audio file.  The command I used for the silent slide in 02-
daily-use2 was:

$ ffmpeg -f lavfi -i anullsrc=r=4800:cl=mono -t  -q:a 9 -c:a
libmp3lame .mp3

This creates an mp3 file with  seconds of silence.

So a one second pause, say, could be concatenated (with ffmpeg) at the
beginning or end of an audio file to improve the transition.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,


> Thank you so much for fixing this. It was in my TODO list, but it is
> GREAT and if you can share your knowledge we can work and make the
> videos available asap :) 

I have been able to match the aspect ratio of the cli videos to the
nocli videos.  With ffmpeg there is a filtergraph that can be added to
pad the images.  The result is a smooth transition between the two
types.  There is no longer flickering or crashing in the players I have
tested.  

The fix is commit 7c5db139081c37cbd4896628e38f5bff8dc5427e.

>  I will go back tomorrow with fixing the timing

Great.  One thought I had on the timing of the cli videos is that there
are two durations at work.  Firstly, the frame rate and the number of
frames:

duration1 (in seconds) = number of frames x 25

Secondly, there is the duration of the audio file (duration2).  

Should we check that duration1 approximately equals duration2?  Or
possibly the audio file should always be slightly shorter than the
duration of the frame display.

WDYT?

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Ricardo,

> This is probably my mistake.  I didn’t pay much attention to frame
> rates when I first built the scripts for generating the CLI videos.
> 

Ok, thanks.  This is fixed in commit
7180fff4ecb46cfed41c6214579a53af6a636a21.

The new frame rate is 25 fps.  This is the European standard for PAL/HD
video.

The file sizes are not dramatically affected by the change.  The first
video is reduced in size from 11MB to 10 MB, for example.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi All,

> Yes! I am working on them to have the best match we can :)
Nice, thanks Laura!

I have pushed all of the new audio recordings to guix/videos.git.

There were two extra svg images needed (for 02-daily-use2 and 04-
packaging1) to match up with the audio.  I have added templates for
them.

One question that cropped up was about the frame rate of the videos.  I
checked the cli and nocli videos and noticed that the nocli videos have
a frame rate of 30 frames per second (fps) and the cli videos have a
frame rate of 20 fps.  

Was there a reason for the difference?  

It may make it easier for the players to have a constant frame rate
throughout.  I gather that 25 fps is a common standard.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

The recording session went very well yesterday.  We were able to make
all the recordings in a studio without having to worry about background
noise.

I had help from a sound engineer who is familiar with the microphone
set-up and the recording process.

> Great!  :)  Let me know when you have time to push them  - no rush,
> just to test  :) -

I have pushed the first set of audio files, for the 01-installation-
from-script video.  I have been able to re-build the video to check
that it works.  It does!

For the cli sessions we used the timings in your recordings as a guide
to synchronize my voice with the screen.  I notice that there are
'sleep' commands in the session files so it might be possible to fine-
tune the timing if necessary.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,


> Great! Could you remind me in UTC at which time of the day? Because I
> generally have to record them at 00 UTC to reduce the environmental
> noise. And then match them.
> 
Thank  you for uploading the cli audios for the packaging3 video.  That
completes the set!

I will be going to the studio in half an hour (11.00am UTC).  They have
a space with low background noise.  However, I will ask for a short
'silence' recording too, so that we can use it for joining sections if
we need to.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

I have finished updating the transcripts and I am preparing to do the
recordings tomorrow.

One question about the makefile: am I right in thinking that the
makefile automatically adjusts the duration of each slide according to
the duration of the corresponding audio file?  In other words it would
not matter if I supply audio files that are slightly longer (or
shorter) than your originals?

> Shouldn't the 'guix environment' step be later in the list of steps?
> If I was packaging the aspi package I would do:
> 
> $ guix environment --pure r-aspi
> The environment was set up in the first packaging video. As far as I
> am concerned, the idea is to have a clean guix environment and
> package
> everything there. If you are using a foreign distro then I believe
> that it could not be reproducible if you package it first.
> Please, let the others correct me if I am wrong.

I think that this question has to do with the complexity of the
package.  The aspi example is a simple case that does not have any
dependencies outside the R build system.  For a more complex package
(newPackage, for example) using:

$ guix environment --pure guix
$ guix build newPackage

would fail if newPackage has extra inputs.  Instead, the commands
should be:

$ guix environment --pure newPackage
$ guix build newPackage

I think this would be easy to communicate to new users if steps 1 and 2
are switched on Slide 2 and I add some extra words to the transcript.

Do you (and others) agree?  

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

I am working on the 04-packaging2 transcript.  I have a question on
Slide 2.

Shouldn't the 'guix environment' step be later in the list of steps? 
If I was packaging the aspi package I would do:

$ guix environment --pure r-aspi

after defining the package first.  Then I would attempt to build the
package.

WDYT?

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

I am 4/7ths of the way through the transcripts :)

On the latest one , 02-daily-use2, line 9, there is the URL:

https://audio.video.gnu.org/guix/everyday-use-part1.webm

Is this correct?  The output in the videos subdirectory is '02-daily-
use1.webm'

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> If others could test with their own players that may give us  clue.
> Sure, as many players, the better.

I have found that the flickering is caused by the change in aspect
ratio as the video moves from the slide view to the screen view (and
back).  Parole and Videos have a default 'Auto' setting for aspect
ratio that re-sets the aspect ratio during playback.  If I change to a
constant aspect ratio (4:3, for example) the flickering stops.

The snag with a constant aspect ratio is that one or other of the views
then looks stretched.  

Padding on the left and right of the screen view might be the best fix.
Then the two views could share the same aspect ratio.

Is this something that you could look into?

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> I have just thought, may you see the transcripts of all the videos,
> and arrange them if you find a better way of saying things in native
> English, and then I record the CLI session videos and create
> subtitles for them? Do you find this workflow fine?

Yes, sounds a good plan.  I will aim to work through the videos to
build each one, then listen to the playback and then edit the
transcript at a rate of one-per-day this week.  That should mean
finishing them all by the end of this week.

If you can do the cli sessions as well that would fit in perfectly.  I
have put back the recording session to next Tuesday 2nd April so that
there is a little more time to get everything finished.

The question about flickering could take some investigation.  Do we
know yet whether this is consistent across all players? or are there
examples where flickering does not happen?   

If others could test with their own players that may give us a clue.  

One other suggestion was to check the codecs.  Do the slides and the
cli sessions use the same codec?

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> Which player are you using?

I have tried with Parole (XFCE) and Videos (GNOME).  The flicker seems
to display something like slide-screen-slide-screen in rapid
succession.  I would expect a simple slide->screen jump at the
beginning of the session and a screen->slide jump at the end.

Have you seen this too?

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

I have updated the transcript for the 03-help video and the changes are
ready to view in the repo.

Next I have built the 01-installation-from-script video.  I notice that
there is no sound for the cli sessions.  Will you adding transcripts
for these?

Also, there is a brief flicker in the transition from slide view to
screen view.  Can anything be done about that?

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> I have just pushed the last video.
> Please, let me know if everything is fine.

'git pull' and 'git push' are working for me.  I have updated the
README file as an initial test commit.  Please check that you are happy
with the changes.

Also, I have provisionally booked next Friday 29th March to do the
recordings.  I will be getting some help from a sound engineer to use
the gear in the studio.  All being well it should be possible to do all
the recordings in one session :)

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> I have realized that I have a typo and have to make a minor change in
> 04-packaging2, that will affect the audio and of course the
> transcript. Just wanted to let you know. Will push it ASAP.

Thank you.  I will also check through the transcripts and make any
slight changes needed to make the narration more-natural-for-me to
speak.  I will be careful not to alter the timings or meanings.

Also, I will have more information on Wednesday about the time slots
available for the recording process.

Best regards,

Paul.

Re: Video narration

[Paul Garlick]

Hi Laura,

> If you want, you can build them by cloning
> git.savannah.gnu.org/git/guix/videos.git.

I have cloned the repository and built the first video with:

$ ./build-video.sh 03-help

This generates 03-help.webm which I can play with Parole, for example. 
Very nice work!

I can see that there is an 'audios' subdirectory with one audio file
per svg file.  So my task, if I understand correctly, is to move
'audios' to 'audios_laura' and create a new 'audios' subdirectory with
new audio files and then re-build.

On the question of total speaking time, 03-help.webm is about 4 minutes
long.  There are six videos, so is 6x4=24mins a reasonable estimate of
the total speaking time?

Also, are the transcripts available for all the videos?  I will need
these before I go in to do the recordings.

Best regards,

Paul.

Re: Request for commit access

[Paul Garlick]

Hi Laura,

I am looking forward to viewing the videos after I have cloned the
repository as a member.  Initially, I am not planning to make any
changes, just to make an estimate of the total speaking time.

I will update on progress and post any questions or comments on the
'Video narration' thread.

Best regards,

Paul. 

Re: Request for commit access

[Paul Garlick]

Hi Ludo,

> welcome aboard!  :-)

Great, many thanks.

I will use the attached OpenPGP subkey  for signing commits.

Best regards,

Paul.


signature.asc
Description: This is a digitally signed message part

Request for commit access

[Paul Garlick]

Hi Guix,

May I request commit access for the Guix repositories?

This access would be primarily for maintaining and addding packages in
the (gnu packages maths/engineering/simulation) modules.

Also, I am preparing to do the narration for the Guix videos.  With
commit access, I will be able to push the audio files directly to
videos.git.

I  have an account on Savannah and have uploaded my OpenPGP key.

Best regards,

Paul Garlick.

Re: Video narration

[Paul Garlick]

Hi Laura,

> I just wanted to ask you if you are still interested in contributing
> being the narrator of our videos.

Sure.  Can you give me a few more details about what is required:

i) how many videos are there?
ii) are there two types of video, one for command line sessions and one
for slides?
iii) how big are the files?

Best regards,

Paul.

Re: Packaging FreeCAD

[Paul Garlick]

Hi John,

> I have been getting a little stuck building the pyside2 dependencies

There has been an effort to package pyside2 for Debian.  This has been
completed in the last six months. 

A good place to look for information is 
https://tracker.debian.org/pkg/pyside2

You can browse the source code and follow the links to the 'debian'
directory, which contains the files that govern the packaging process. 
In general for Debian packages, the 'rules' file is worth reading and
the 'patches' directory has the changes to the upstream code.  

One element that could be important in Guix is an update of patchelf to
a recent commit (see 'update-patchelf.patch' in the patches directory).

Best regards,

Paul.

Re: English Voice Wanted

[Paul Garlick]

Hi Laura,

> Paul me know if a .srt file is fine.

Sure, .srt should be fine.  I will find out more details of the
recording equipment and the process.

Best regards,

Paul.

Re: English Voice Wanted

[Paul Garlick]

Hi Bjorn,

Yes, I can be a candidate.  I was born in London and I have a native
English voice.

Also, I can arrange access to professional recording equipment at the
local university.  The voiceover could be supplied in the format of
your choice (AIFF, WAV, ...).

Best regards,

Paul.

Re: Packaging FreeCAD

[Paul Garlick]

Hi John,

> I did not know Debian changed their version of opencascade

There are two versions of OpenCASCADE in Debian, named 'liboce' and
'libocct'.  'occt' is the upstream variant, 'oce' is the community
maintained variant.  Historically, Debian has switched between the two
as licence requirements dictated.  

Currently, both are packaged and FreeCAD version 0.17 depends on
liboce.  However, libocct is being more actively maintained and I
believe the intention for Debian FreeCAD version 0.18 is for the
dependency to switch to libocct.

Best regards,

Paul.

Re: Packaging FreeCAD

[Paul Garlick]

Hi John,

This is good news indeed for Guix-using engineers and architects.  I
shall follow developments with interest.

I can look after the OpenCASCADE dependency too.  This package is due
an update but upstream development has been slow of late.  Debian have
switched to the OCCT version for FreeCAD and I am thinking that it
would be best for Guix to follow suit.  This would entail the
introduction of a new opencascade-occt package, which I plan to do
soon, if no-one beats me to it!

Best regards,

Paul.

Re: Heads-up: New dependency on Guile-Gcrypt

[Paul Garlick]

Hi Ludo,

> Or am I missing something?
> 

No, I don't think so.  Everything is working again after your bug fix.
 

I have just needed to introduce a couple of concepts in order to use
the ./pre-inst-env commands in the git checkout.  Firstly, I use a
development environment created by:

$ guix environment --ad-hoc guile gnutls guile-gcrypt guile-json nss-
certs

This adds the guile-gcrypt dependency and prevents the 'no code for
module (gcrypt hash)' error.

Then, for working on a new package ('my-new-package'), I create a new
environment:

$ ./pre-inst-env guix environment --pure my-new-package --ad-hoc guile-
gcrypt

Lastly, to build the package I need to define GUILE_LOAD_PATH:

$ GUILE_LOAD_PATH=$GUIX_ENVIRONMENT/share/guile/site/2.2 ./pre-inst-env 
guix build -K my-new-package

An alternative would perhaps be to install guile-gcrypt in the user
profile (or a separate profile as suggested by Pjotr).  Then use
GUILE_LOAD_PATH=/home/paul/.guix-profile/share/guile/site/2.2.

Best regards,

Paul.

Re: Heads-up: New dependency on Guile-Gcrypt

[Paul Garlick]

Hi Pjotr,

Thank you.

> Guix is built with Guile and Guile has trouble finding the gcrypt
> package.  

Having successfully upgraded my root and user Guix installations I now
find that my development installation is unable to locate guile-gcrypt:

$ ./pre-inst-env guix --version
ERROR: In procedure scm-error:
no code for module (gcrypt hash)

If I use my root guix profile, in a simlar way to your suggestion, it
works:

$ sudo GUILE_LOAD_PATH=/root/.guix-profile/share/guile/site/2.2 ./pre-
inst-env guix --version
guix (GNU Guix) 0.15.0.1558-b0632

Is there a best way to notify the ./pre-inst-env commands about guile-
gcrypt?  Using sudo seems like the wrong approach.  Would installing
guile-gcrypt as the development user suffice?

Best regards,

Paul. 

Re: Heads-up: New dependency on Guile-Gcrypt

[Paul Garlick]

Hi Ludo,

Success! I have been able to upgrade the root Guix installation.

The steps I have taken were:

i) remove the 'latest' symlink
ii) re-create the symlink to point to the previous installation
iii) use 'guix gc --delete' to delete the guile-gcrypt package
components
iv) run 'GUILE_LOAD_PATH=/root/.guix-profile/share/guile/site/2.2 guix
pull'
v) run
'GUILE_LOAD_COMPILED_PATH=/gnu/store/3yvsd7w79a8rl42rxfg0xqzh9s14f3si-
guile-gcrypt-0.1.0/lib/guile/2.2/site-ccache guix package --upgrade'

This time the '... guix pull' command did not generate the '
Failed to autoload make-session in (gnutls)' errors.  In retrospect, I
think the only missing element from the first attempt was the
GUILE_LOAD_PATH variable.

The last piece of the jigsaw was setting GUILE_LOAD_COMPILED_PATH so
that the gcrypt module could be found.  This was required for all of
the guix commands that I tried.

Now:

$ guix package --list-installed
guix0.15.0-
3.3d43017out /gnu/store/sa9xj13i0568xajsjh78g3pnlsww8x8g-
guix-0.15.0-3.3d43017
...

Best regards,

Paul.

Re: Heads-up: New dependency on Guile-Gcrypt

[Paul Garlick]

> 
> Pushed as 3ffcad7df3ab8947010814f61b32ce14ea80e780, thanks!
> 

Hi Ludo,

I tried a 'guix pull' as root and this time it completes.  However,
there were warnings generated:

WARNING: compilation of /gnu/store/38553wfz0jwlgbw13pk99xl79pbfx58d-
guile-2.2.3/bin/guild failed

and

Failed to autoload make-session in (gnutls)

I think the GUILE_LOAD_PATH variable needed to be set before running
'guix pull'

Now all the guix commands I have tried generate:

ERROR: In procedure scm-error:
no code for module (gcrypt hash)

How best to recover?  Ideally I would undo the 'guix pull' and re-do
the command with GUILE_LOAD_PATH set.

Best regards,

Paul.

Re: Heads-up: New dependency on Guile-Gcrypt

[Paul Garlick]

On Wed, 2018-09-05 at 22:58 +0200, Ludovic Courtès wrote:
> 
> Indeed, that may well be the reason.  Can you try to apply the patch
> below in a local branch, and then run “make as-derivation”?
> 

Hi Ludo,

Many thanks.  I have applied the patch in a new branch.

$ git describe
v0.15.0-1390-gca71942

However, 'make as-derivation' is not working yet.  Do I need an ad-hoc
environment to run this command?  

So far, I have tried using the system (Centos 7) make but this fails
since system guile (v2.0.9) is not installed.

Best regards,

Paul.

Re: Heads-up: New dependency on Guile-Gcrypt

[Paul Garlick]

Hi Ludo,

> ‘guix pull’ will happily perform the transition.

For me, I have an older Guix (16th May) that fails to install guile-
gcrypt.  As root:

$ guix --version
guix (GNU Guix) 6f84dc4314cd84550d9fc7e7afa11c495edc45a5

When I try 'guix pull' there is an attempt to build guile-gcrypt
locally that fails with the message './configure:  No such file or
directory'.  I can see that the build procedure moves from the 'unpack'
phase straight to the 'patch-usr-bin-file' phase with no 'bootstrap'
phase in between.

Aha, another chicken-and-egg, I thought; my older Guix does not have a
'bootstrap' phase.  However, I can see that the commit for adding the
new phase was made on 11th March, so perhaps 'guix pull' is failing for
a different reason.

Maybe there is a workaround for this scenario.  WDYT?

Best regards,

Paul.

Re: [PATCH] gnu: Add openfoam

[Paul Garlick]

Hello Ludo,
> I think it might make sense to create a new “simulation” module, and
> eventually move OpenCascade there as well, WDYT?
> 
Sure.  I will create a new module and put the OpenFOAM package
definition in there as the first one.
> Some comments:
> 
> 
> > 
> > +(build-system trivial-build-system)
> > 

> 
> 
> Did you consider using ‘gnu-build-system’?  I think this would save
> quite a few lines corresponding to the initial setup (set-paths, unpack,
> patch-shebang, etc.), and also ensure that the final phases (strip,
> compress-documentation, etc.) are run.  In addition you wouldn’t need to
> list all the usual build inputs (GCC, Coreutils, Make, etc.).  WDYT?
> 

I did consider both options, removing and modifying stages in the gnu-
build-system or starting from scratch with the trivial-build-system.  I
can give the gnu-build-system option a go, reverting to the trivial-
build-system if the new attempt proves too troublesome.  
One aspect I will need to investigate in the gnu-build-system is using
the "./Allwmake" command to trigger the build process.  In OpenFOAM,
wmake is short for "wrapped-make" and the package has its own
configuration step.  In brief, it has its own sequence and does not
follow the standard "./configure && make && make install" steps.
Using the trivial-build-system the patch-shebang section is indeed a
little extended.  I found this was necessary to avoid a failure of the
patch-shebang command using a simple 'find-files "."' from top-level
directory.  There is a scenario where sub-directories named '0', in the
'tutorials' directory, identified by the find-files command, being
passed to 'patch-shebang', which should only receive files, not
directories.  This causes 'patch-shebang' to stop and the build fails.
> > 
> > +  )
> > +)
> > +   )
> > + )
> > 

> 
> 
> Please listen to what ‘guix lint’ has to say about these.  :-)
> 

Interestingly, 'guix lint' let me get away with this without making
comment.  However, I shall condense things accordingly in the new
patch.
> What about wrapping the resulting binaries so that they have
> ‘FOAM_INST_DIR’ set to %output?
> 
> 

In fact, FOAM_INST_DIR is used in multiple places to navigate around
the installed files (not just the binaries), so I think this variable
may need to remain available as is.
> 
> You could make it “if true”, thereby avoiding the need to define
> $READLINE_ROOT.
> 
> 

Could you elaborate on this idea a little?  At the moment the test is
"if file exists".
> Could you send an updated patch to guix-patc...@gnu.org, where it will
> , where it will
> be visible in the patch tracker?
> 
> 
> 

Aha, a new system!
Best,
Paul.

[PATCH] gnu: Add openfoam

[Paul Garlick]

tch))
+(setenv "METISVERSION" ,(package-version metis))
+;; set environment variables to define library paths
+(map (lambda (library)
+  (setenv (string-append
+   (string-upcase library) "_ROOT")
+   (assoc-ref %build-inputs library))) libraries)
+(setenv "LIBRARY_PATH" (string-append
+ (getenv "GLIBC_ROOT") "/lib" ":"
+ (getenv "ZLIB_ROOT") "/lib"))
+;; set environment variables to define header paths
+(setenv "CPATH" (string-append
+ (getenv "LINUX-LIBRE-HEADERS_ROOT")
+ "/include"))
+(setenv "CPLUS_INCLUDE_PATH" (string-append
+ (getenv "METIS_ROOT") "/include" ":"
+ (getenv "FLEX_ROOT") "/include" ":"
+ (getenv "ZLIB_ROOT") "/include"))
+;; compile OpenFOAM libraries and applications
+(system (format #f "source ./etc/bashrc && ./Allwmake -j~a"
+   (parallel-job-count)))
+;; update configuration files
+(system (format #f "sed -i 's|~a|'~:*~s'|' ~a"
+   "$CGAL_ROOT" "etc/config.sh/CGAL"))
+(system (format #f "sed -i 's|~a|'~:*~s'|' ~a"
+   "$BOOST_ROOT" "etc/config.sh/CGAL"))
+(system (format #f "sed -i 's|~a|'~:*~s'|' ~a"
+   "$METIS_ROOT" "etc/config.sh/metis"))
+(system (format #f "sed -i 's|~a|'~:*~s'|' ~a"
+   "$METISVERSION" "etc/config.sh/metis"))
+(system (format #f "sed -i 's|~a|'~:*~s'|' ~a"
+   "$SCOTCH_ROOT" "etc/config.sh/scotch"))
+(system (format #f "sed -i 's|~a|'~:*~s'|' ~a"
+   "$SCOTCHVERSION" "etc/config.sh/scotch"))
+(system (format #f "sed -i 's|~a|'~:*~s'|' ~a"
+   "$GMP_ROOT" "etc/config.sh/settings"))
+(system (format #f "sed -i 's|~a|'~:*~s'|' ~a"
+   "$MPFR_ROOT" "etc/config.sh/settings"))
+;; set write permission on tutorial files
+(system (format #f "chmod -R u+w ~a" "tutorials"))
+;; create install directory
+(mkdir-p install-dir)
+;; move contents of build directory to install directory
+(copy-recursively "." install-dir)
+  )
+)
+   )
+ )
+(native-search-paths
+ ;; define the FOAM_INST_DIR variable
+ (list (search-path-specification
+(variable "FOAM_INST_DIR")
+(files '(".")
+(synopsis "Framework for numerical simulation of fluid flow")
+(description "OpenFOAM provides a set of solvers and methods for tackling
+problems in the field of Computational Fluid Dynamics (CFD).  It is written in
+C++.  Discrete versions of governing equations such as the Navier-Stokes
+equations can be solved in integral form.  OpenFOAM implements the Finite
+Volume Method for discretizing the spacial dimensions.  Grids are accepted that
+define complex geometries in two- and three-dimensional domains.  Grids may be
+unstructured in layout and may contain any combination of polyhedral cells.
+Large problems can be split into connected smaller problems for efficient
+solution on parallel systems")
+(license license:gpl3+)
+(home-page "https://openfoam.org";)))
+
 (define-public maxflow
   (package
 (name "maxflow")
diff --git a/gnu/packages/patches/openfoam-cleanup.patch 
b/gnu/packages/patches/openfoam-cleanup.patch
new file mode 100644
index 0..e7760b7a5
--- /dev/null
+++ b/gnu/packages/patches/openfoam-cleanup.patch
@@ -0,0 +1,267 @@
+# This patch removes all need for the ThirdParty files of OpenFOAM
+
+# Derived from easybuild patch:  Ward Poelm

[PATCH] gnu: gmsh: Add opencascade-oce to propagated-inputs

[Paul Garlick]

* gnu/packages/maths.scm (gmsh)
[propagated-inputs]: Add opencascade-oce
[arguments]: Remove #:phases parameter for component8_in_a_box test
---
 gnu/packages/maths.scm | 11 ++-
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/gnu/packages/maths.scm b/gnu/packages/maths.scm
index d479472..c3cedbb 100644
--- a/gnu/packages/maths.scm
+++ b/gnu/packages/maths.scm
@@ -1035,6 +1035,7 @@ Open CASCADE library.")
("lapack" ,lapack)
("mesa" ,mesa)
("glu" ,glu)
+   ("opencascade-oce" ,opencascade-oce)
("libx11" ,libx11)
("libxext" ,libxext)))
 (inputs
@@ -1043,15 +1044,7 @@ Open CASCADE library.")
 (arguments
  `(#:configure-flags `("-DENABLE_METIS:BOOL=OFF"
"-DENABLE_BUILD_SHARED:BOOL=ON"
-   "-DENABLE_BUILD_DYNAMIC:BOOL=ON")
-   #:phases (modify-phases %standard-phases
-  (replace
-   'check
-   (lambda _
- (zero? (system* "make" "test"
- ;; Disable this test.  See
- ;; https://geuz.org/trac/gmsh/ticket/271
- "ARGS=-E component8_in_a_box")))
+   "-DENABLE_BUILD_DYNAMIC:BOOL=ON")))
 (home-page "http://www.geuz.org/gmsh/";)
 (synopsis "3D finite element grid generator")
 (description "Gmsh is a 3D finite element grid generator with a built-in
-- 
2.7.4

[PATCH] gnu: gmsh: Update to 2.16.0

[Paul Garlick]

* gnu/packages/maths.scm (gmsh): Update to 2.16.0
---
 gnu/packages/maths.scm | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/gnu/packages/maths.scm b/gnu/packages/maths.scm
index 11b8181..d479472 100644
--- a/gnu/packages/maths.scm
+++ b/gnu/packages/maths.scm
@@ -1014,14 +1014,14 @@ Open CASCADE library.")
 (define-public gmsh
   (package
 (name "gmsh")
-(version "2.15.0")
+(version "2.16.0")
 (source
  (origin
   (method url-fetch)
   (uri (string-append "http://gmsh.info/src/gmsh-";
   version "-source.tgz"))
   (sha256
-   (base32 "02h7fk4vv8qwnq3ymm409c5sp4nksd0m9h2vkxqmy42l0ic4nalr"))
+   (base32 "1slf0bfkwrcgn6296wb4qhbk4ahz6i4wfb10hnim08x05vrylag8"))
   (modules '((guix build utils)))
   (snippet
;; Remove non-free METIS code
-- 
2.7.4

[PATCH] fix test-skip typo

[Paul Garlick]

---
 tests/syscalls.scm | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/syscalls.scm b/tests/syscalls.scm
index 92e02f3..1934704 100644
--- a/tests/syscalls.scm
+++ b/tests/syscalls.scm
@@ -456,7 +456,7 @@
   (eof-object? (read-utmpx (%make-void-port "r"
 
 (unless (access? "/var/run/utmpx" O_RDONLY)
-  (tes-skip 1))
+  (test-skip 1))
 (test-assert "read-utmpx"
   (let ((result (call-with-input-file "/var/run/utmpx" read-utmpx)))
 (or (utmpx? result) (eof-object? result
-- 
2.7.4

Re: [PATCH] gnu: Add opencascade-oce

[Paul Garlick]

On Sat, 2017-01-21 at 16:13 +0100, Ludovic Courtès wrote:
> Marius Bakke  skribis:
> 
> > 
> > Paul Garlick  writes:
> [...]
> 
> > 
> > Libtool and file should likely be native-inputs, not sure about the
> > others. Could you check whether they are required and send an
> > updated
> > patch?
> ‘file’ is already an implicit input, so I’m not why it needs to be
> added
> (perhaps OpenCascade depends on libmagic?).
> 
> As for Libtool, I wonder why it is needed at all, especially given
> that
> OpenCascade is built with CMake.
> 
> Ludo’.
Hi Marius and Ludo,
Thank you for your comments.  I have checked the dependencies and here
are the results:
ftgl: not needed
qt: not needed
libtool: not needed
file: not needed
tcl: required by the DRAW test harness
tk: required by the DRAW test harness
I have amended the patch to remove the extra dependencies.  The public
domain licence information is now included.  
Also, there is an extra configuration flag required to successfully
build the DRAW libraries and executable.  The '-UCMAKE_INSTALL_RPATH'
flag is required for the validate-runpath stage to succeed.  This is to
do with the way the global variable is initialized.  There is a
distinction in CMake between lists and strings, which was new to me so
I will document it here.  
Essentially, if a variable is initialized as a single string it cannot
subsequently be overriden by a list of strings.  The cmake instructions
in OpenCascade (in adm/cmake/BuildToolKit.cmake) attempt to do this at
the install stage.  The DRAW harness needs a RUNPATH including both
.../lib and .../lib/oce-0.17.  With the standard definition of
CMAKE_INSTALL_RPATH (from cmake-build-system.scm) the substitution of
the INSTALL_RPATH property only includes the first string.
 Consequently, the validate-runpath phase fails (error: library ... not
in RUNPATH ...).  The configuration flag -UCMAKE_INSTALL_RPATH unsets
the global variable and allows the opencascade cmake rules to do the
work instead, setting the RUNPATH correctly for each target.
On the subject of the 0.18 version, perhaps this could be the subject
of a separate patch.  There could be an argument to just update in the
normal way (instead of having two packages at once).  The issue is
related to the hardware requirements of recent versions of OpenGL.  It
is useful to start with 0.17 as this imposes the least restriction on
older graphics card.  I can do some testing on the new version, later
on, to determine how important this is.
The amended patch is attached.
Best,
Paul.
From a5f64ac9b488249027fc92c46d737d4572e57d02 Mon Sep 17 00:00:00 2001
From: Paul Garlick 
Date: Wed, 18 Jan 2017 10:01:36 +
Subject: [PATCH] gnu: Add opencascade-oce

* gnu/packages/maths.scm: Add opencascade-oce
---
 gnu/packages/maths.scm | 48 
 1 file changed, 48 insertions(+)

diff --git a/gnu/packages/maths.scm b/gnu/packages/maths.scm
index 313f6ac..c8e280f 100644
--- a/gnu/packages/maths.scm
+++ b/gnu/packages/maths.scm
@@ -15,6 +15,7 @@
 ;;; Copyright © 2016 Ludovic Courtès 
 ;;; Copyright © 2016 Leo Famulari 
 ;;; Copyright © 2016 Thomas Danckaert 
+;;; Copyright © 2017 Paul Garlick 
 ;;;
 ;;; This file is part of GNU Guix.
 ;;;
@@ -956,6 +957,53 @@ Work may be performed both at the interactive command-line as well as via
 script files.")
 (license license:gpl3+)))
 
+(define-public opencascade-oce
+  (package
+(name "opencascade-oce")
+(version "0.17.2")
+(source
+  (origin
+(method url-fetch)
+(uri (string-append
+   "https://github.com/tpaviot/oce/archive/OCE-";
+   version
+   ".tar.gz"))
+(sha256
+  (base32
+"0vpmnb0k5y2f7lpmwx9pg9yfq24zjvnsak5alzacncfm1hv9b6cd"
+(build-system cmake-build-system)
+(arguments
+ '(#:configure-flags
+(list "-DOCE_TESTING:BOOL=ON"
+  "-DOCE_USE_TCL_TEST_FRAMEWORK:BOOL=ON"
+  "-DOCE_DRAW:BOOL=ON"
+  (string-append "-DOCE_INSTALL_PREFIX:PATH="
+(assoc-ref %outputs "out"))
+  "-UCMAKE_INSTALL_RPATH")))
+(inputs
+  `(("freetype" ,freetype)
+("glu" ,glu)
+("libxmu" ,libxmu)
+("mesa" ,mesa)
+("tcl" ,tcl)
+("tk" ,tk)))
+(native-inputs
+  `(("python" ,python-wrapper)))
+(home-page "https://github.com/tpaviot/oce";)
+(synopsis
+  "Libraries for 3D modeling and numerical simulation")
+(description "OCE stands for Open CASCADE Community Edition.  The project
+gathers patches/improvements/experiments contributed by users over the official
+Open CASCADE libr

[PATCH] gnu: Add opencascade-oce

[Paul Garlick]

* gnu/packages/maths.scm: Add it
---
 gnu/packages/maths.scm | 49 +
 1 file changed, 49 insertions(+)

diff --git a/gnu/packages/maths.scm b/gnu/packages/maths.scm
index 313f6ac..62415e7 100644
--- a/gnu/packages/maths.scm
+++ b/gnu/packages/maths.scm
@@ -15,6 +15,7 @@
 ;;; Copyright © 2016 Ludovic Courtès 
 ;;; Copyright © 2016 Leo Famulari 
 ;;; Copyright © 2016 Thomas Danckaert 
+;;; Copyright © 2017 Paul Garlick 
 ;;;
 ;;; This file is part of GNU Guix.
 ;;;
@@ -43,6 +44,7 @@
   #:use-module (guix build-system gnu)
   #:use-module (guix build-system r)
   #:use-module (gnu packages algebra)
+  #:use-module (gnu packages autotools)
   #:use-module (gnu packages bison)
   #:use-module (gnu packages boost)
   #:use-module (gnu packages check)
@@ -52,6 +54,7 @@
   #:use-module (gnu packages cyrus-sasl)
   #:use-module (gnu packages documentation)
   #:use-module (gnu packages elf)
+  #:use-module (gnu packages file)
   #:use-module (gnu packages flex)
   #:use-module (gnu packages fltk)
   #:use-module (gnu packages fontutils)
@@ -78,6 +81,7 @@
   #:use-module (gnu packages perl)
   #:use-module (gnu packages pkg-config)
   #:use-module (gnu packages python)
+  #:use-module (gnu packages qt)
   #:use-module (gnu packages readline)
   #:use-module (gnu packages tbb)
   #:use-module (gnu packages shells)
@@ -2843,3 +2847,48 @@ instruction sets.  Thus, an application written with Vc 
can be compiled for:
 @end enumerate\n")
 (home-page "https://github.com/VcDevel/Vc";)
 (license license:bsd-3)))
+
+(define-public opencascade-oce
+  (package
+(name "opencascade-oce")
+(version "0.17.2")
+(source
+  (origin
+(method url-fetch)
+(uri (string-append
+   "https://github.com/tpaviot/oce/archive/OCE-";
+   version
+   ".tar.gz"))
+(sha256
+  (base32
+"0vpmnb0k5y2f7lpmwx9pg9yfq24zjvnsak5alzacncfm1hv9b6cd"
+(build-system cmake-build-system)
+(arguments
+ '(#:configure-flags
+(list "-DOCE_TESTING:BOOL=ON"
+  (string-append "-DOCE_INSTALL_PREFIX:PATH="
+(assoc-ref %outputs "out")
+(inputs
+  `(("freetype" ,freetype)
+("ftgl" ,ftgl)
+("glu" ,glu)
+("qt" ,qt)
+("libtool" ,libtool)
+("libxmu" ,libxmu)
+("file" ,file)
+("tk" ,tk)
+("tcl" ,tcl)
+("mesa" ,mesa)))
+(native-inputs
+  `(("python" ,python-wrapper)))
+(home-page "https://github.com/tpaviot/oce";)
+(synopsis
+  "Libraries for 3D modeling and numerical simulation")
+(description "OCE stands for Open CASCADE Community Edition.  The project
+gathers patches/improvements/experiments contributed by users over the official
+Open CASCADE library.")
+(license (list license:lgpl2.1; OCE libraries, with an exception for the
+  ; use of header files; see
+  ; OCCT_LGPL_EXCEPTION.txt
+   license:expat; file src/OpenGl/OpenGl_glext.h
+   license:bsd-3; test framework gtest
-- 
2.7.4

Re: [PATCH] gnu: Add opencascade-oce

[Paul Garlick]

Hello Marius,
Thank you for your review.
> I wonder if this could fit in "engineering.scm". Or maybe
> "geometry.scm"
> or "3d.scm". We try to avoid "single program" modules. What do you
> think?
Perhaps maths.scm is the best place, in order to avoid creating a new
module.  The Gmsh package is already there.  To check; am I right in
saying that a library in one module can be used as an input for a
package in another module?  For example, if FreeCAD were to be
packaged, and placed in engineering.scm, could it use opencascade-oce
as an input from maths.scm?
> > 
> > +(define-public opencascade-oce
> > +  (package
> > +(name "opencascade-oce")
> > +(version "0.17.2")
> > 

> 
> 
> Version 0.18 was released a few days ago!
> 

It would be useful to package both versions.  There is a difference in
the OpenGL graphics card requirements in going from version 0.17.2 to
0.18.
> > 
> > +(license license:lgpl2.1))) ; plus header files exception
> > 

> 
> 
> Please expand on the exception comment and mention
> OCCT_LGPL_EXCEPTION.txt. There are also some public domain files, and
> the bundled "gtest" is bsd-3. src/OpenGl/OpenGl_glext.h is distributed
> under the expat license. Some files generated by "bison" has GPL3+.
> 

I have added the information about the expat and BSD-3 licenses.  Can
you elaborate on the public domain files and Bison?
Best,
Paul.

[PATCH] gnu: Add opencascade-oce

[Paul Garlick]

* gnu/packages/opencascade.scm: New file
* gnu/local.mk (GNU_SYSTEM_MODULES): Add it
---

The OpenCASCADE library is used for defining 3D geometry.  It can be
incorporated as a build option in other packages.  Examples of such packages
(and the subject domains in which they are used) are Gmsh (mathematics),
IfcOpenShell (architecture) and FreeCAD (engineering).

There are two versions of OpenCASCADE; the upstream version and the community-
maintained version.  OCE is the community-maintained version.

 gnu/local.mk |  1 +
 gnu/packages/opencascade.scm | 75 
 2 files changed, 76 insertions(+)
 create mode 100644 gnu/packages/opencascade.scm

diff --git a/gnu/local.mk b/gnu/local.mk
index c6cb55b..c879f32 100644
--- a/gnu/local.mk
+++ b/gnu/local.mk
@@ -281,6 +281,7 @@ GNU_SYSTEM_MODULES =\
   %D%/packages/ocr.scm \
   %D%/packages/onc-rpc.scm \
   %D%/packages/openbox.scm \
+  %D%/packages/opencascade.scm \
   %D%/packages/openldap.scm\
   %D%/packages/openstack.scm   \
   %D%/packages/orpheus.scm \
diff --git a/gnu/packages/opencascade.scm b/gnu/packages/opencascade.scm
new file mode 100644
index 000..3c4b9b5
--- /dev/null
+++ b/gnu/packages/opencascade.scm
@@ -0,0 +1,75 @@
+;;; GNU Guix --- Functional package management for GNU
+;;; Copyright © 2017 Paul Garlick 
+;;;
+;;; This file is part of GNU Guix.
+;;;
+;;; GNU Guix is free software; you can redistribute it and/or modify it
+;;; under the terms of the GNU General Public License as published by
+;;; the Free Software Foundation; either version 3 of the License, or (at
+;;; your option) any later version.
+;;;
+;;; GNU Guix is distributed in the hope that it will be useful, but
+;;; WITHOUT ANY WARRANTY; without even the implied warranty of
+;;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+;;; GNU General Public License for more details.
+;;;
+;;; You should have received a copy of the GNU General Public License
+;;; along with GNU Guix.  If not, see <http://www.gnu.org/licenses/>.
+
+(define-module (gnu packages opencascade)
+  #:use-module ((guix licenses) #:prefix license:)
+  #:use-module (guix packages)
+  #:use-module (guix download)
+  #:use-module (guix build-system cmake)
+  #:use-module (gnu packages)
+  #:use-module (gnu packages autotools)
+  #:use-module (gnu packages file)
+  #:use-module (gnu packages fontutils)
+  #:use-module (gnu packages gl)
+  #:use-module (gnu packages python)
+  #:use-module (gnu packages qt)
+  #:use-module (gnu packages tcl)
+  #:use-module (gnu packages xorg))
+
+(define-public opencascade-oce
+  (package
+(name "opencascade-oce")
+(version "0.17.2")
+(source
+  (origin
+(method url-fetch)
+(uri (string-append
+   "https://github.com/tpaviot/oce/archive/OCE-";
+   version
+   ".tar.gz"))
+(sha256
+  (base32
+"0vpmnb0k5y2f7lpmwx9pg9yfq24zjvnsak5alzacncfm1hv9b6cd"
+(build-system cmake-build-system)
+(arguments
+ '(#:configure-flags
+(list
+  (string-append
+"-DOCE_INSTALL_PREFIX:PATH=" (assoc-ref %outputs "out"))
+  "-DOCE_TESTING:BOOL=ON")
+   #:tests? #t))
+(inputs
+  `(("freetype" ,freetype)
+("ftgl" ,ftgl)
+("glu" ,glu)
+("qt" ,qt)
+("libtool" ,libtool)
+("libxmu" ,libxmu)
+("file" ,file)
+("tk" ,tk)
+("tcl" ,tcl)
+("mesa" ,mesa)))
+(native-inputs
+  `(("python" ,python-wrapper)))
+(home-page "http://www.opencascade.org/";)
+(synopsis
+  "Open CASCADE OCE; libraries for 3D modeling and numerical simulation")
+(description "OCE stands for Open CASCADE Community Edition.  The project
+gathers patches/improvements/experiments contributed by users over the official
+Open CASCADE library.")
+(license license:lgpl2.1))) ; plus header files exception
-- 
2.7.4

[PATCH] gnu: gmsh: Update to 2.15.0

[Paul Garlick]

* gnu/packages/maths.scm (gmsh): Update to 2.15.0
---
 gnu/packages/maths.scm | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/gnu/packages/maths.scm b/gnu/packages/maths.scm
index fc98eae..313f6ac 100644
--- a/gnu/packages/maths.scm
+++ b/gnu/packages/maths.scm
@@ -959,14 +959,14 @@ script files.")
 (define-public gmsh
   (package
 (name "gmsh")
-(version "2.14.1")
+(version "2.15.0")
 (source
  (origin
   (method url-fetch)
   (uri (string-append "http://gmsh.info/src/gmsh-";
   version "-source.tgz"))
   (sha256
-   (base32 "1vsxp47j6srmy8kqb3p1z9pmlm42whhhz7r0vzpa2a86gga4zx17"))
+   (base32 "02h7fk4vv8qwnq3ymm409c5sp4nksd0m9h2vkxqmy42l0ic4nalr"))
   (modules '((guix build utils)))
   (snippet
;; Remove non-free METIS code
-- 
2.7.4

[PATCH] gnu: gmsh: Update to 2.14.1

[Paul Garlick]

* gnu/packages/maths.scm (gmsh): Update version, hash and uri
---
 gnu/packages/maths.scm | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/gnu/packages/maths.scm b/gnu/packages/maths.scm
index 64e53e2..c9792a1 100644
--- a/gnu/packages/maths.scm
+++ b/gnu/packages/maths.scm
@@ -949,14 +949,14 @@ script files.")
 (define-public gmsh
   (package
 (name "gmsh")
-(version "2.11.0")
+(version "2.14.1")
 (source
  (origin
   (method url-fetch)
-  (uri (string-append "http://www.geuz.org/gmsh/src/gmsh-";
+  (uri (string-append "http://gmsh.info/src/gmsh-";
   version "-source.tgz"))
   (sha256
-   (base32 "1ilplibvjgf7a905grpnclrvkmqy9fgrpl7xyp3w4yl1qc682v9b"))
+   (base32 "1vsxp47j6srmy8kqb3p1z9pmlm42whhhz7r0vzpa2a86gga4zx17"))
   (modules '((guix build utils)))
   (snippet
;; Remove non-free METIS code
-- 
2.7.4