[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

pigx-rnaseq Re: merge request opened Re: libjs-jquery-datatables-extensions: pandoc needs vfs_fonts.js



Hi all,

pigx-rnaseq runs a set of RNAseq tools within snakemake and then
prepares reports with pandoc that look neat since the tables are all
jquery-javascripted dynamic. Just, when pandoc expected the .js/.css
files in a location that is not matching reality, then the whole process
comes to an end. When removed all the embedded *js files from our R
packages, our job with symlinks had imperfections. Particularly so, when
a script uses the symlinks with source and dest exchanged, this has
weird effects that are not repared by just updating a package. One needs
to manually remove the link. That hurt. And with a symlink hidden in a
post-inst script, this does not pop up in "dpkg -c" or so, as would
anything that goes through dh_link. I need to make up my mind about
this, but somehow I feel this is something that lintian and friends
should in part address.

My merge request on
https://salsa.debian.org/js-team/datatables-extensions/-/merge_requests
I sense is the last technical hurdle before an upload to main. And we
then indeed have a checkpointed parallel workflow for RNAseq with snakemake.

Next on my list now is pigx-scrnaseq. Then, hm, I think I will look into
the Nanopore a bit more. The LamPORE
https://www.biotechniques.com/diagnostics-preclinical/oxford-nanopore-makes-deal-with-the-uk-government-to-provide-450000-covid-19-tests/
looks interesting.

Best,

Steffen


On 24.08.20 16:40, Steffen Möller wrote:
> Hi Andreas,
>
> On 24.08.20 15:16, Steffen Möller wrote:
>> On 24.08.20 11:08, Andreas Tille wrote:
>>> Hi Steffen,
>>>
>>> On Mon, Aug 24, 2020 at 04:17:37AM +0200, Steffen Möller wrote:
>>>> @Andreas,
>>>>
>>>> good news: I made quite some inroads on why pigx-rnaseq does not work.
>>>> But that is for another email.
>>>>
>>>> My current hurdle is that vfs_fonts.js is missed by pandoc:
>>>>
>>>> pandoc:
>>>> /usr/lib/R/site-library/DT/htmlwidgets/lib/datatables-extensions/Buttons/js/vfs_fonts.js:
>>>> openBinaryFile: does not exist (No such file or directory)
>>>>
>>>> This file should have been located next to pdfmake in
>>>>
>>>> /usr/share/javascript/jquery-datatables-extensions/pdfmake/build
>>>>
>>>> but
>>>>
>>>> Files-Excluded: */buttons.flash*
>>>>                 */*.swf
>>>>                 */vfs_fonts.js
>>>>
>>>> does not want that. There are suggestions on
>>>> https://stackoverflow.com/questions/48835055/pdfmake-cant-find-the-font-files-on-vfs
>>>> to create one's own variant, but this should then happen within that
>>>> JavaScript package, right?
>>> May be you file a bug report to datatables-extensions upstream to provide
>>> a free font version?
>> Here is the original
>> https://raw.githubusercontent.com/naoa/pdfmake/master/build/vfs_fonts.js
>>
>> and this GenShinGothic has a "non-commercial use only" license, indeed.
>>
>> https://en.m.fontke.com/font/23789742/detail/
>>
>> I just submitted https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=968944.
>>
>> On a sidenote, Rebecca reopened
>> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=927787 since
>> DataTables apparently has a new repository on github where each
>> subdirectory is presented as its very own package, like
>> https://github.com/DataTables/Select or (more prominently)
>> https://github.com/DataTables/Buttons .
>>
>> I also checked
>> https://github.com/DataTables/pdfmake/blob/master/build/vfs_fonts.js
>> which now just ships with a an embedded "LICENSE.txt" and no font which
>> I presume to be fine to redistribute. I'll see if this also works with
>> the current version and am not unlikely to close my own bug that way.
>>
>> As Rebecca stated, this new DataTables repository also affects
>> snakemake. I consider it remote to our cause for now. Would not mind a
>> "JavaScript" Tab in our Spreadsheet.
> I added a patch that adds the future free version of vfs_fonts.js.
> Please kindly review and merge.
>
> Best,
>
> Steffen
>
>
>


Reply to: