Enhancements to zip_to_disk.frame()

[cchris005]

Hi!
To promote use of disk.frame in our office environment, we'd need two enhancements:
1/ Support for .7z compressed format in zip_to_disk.frame()
2/ Support for .001 multipart archive in same function. This implies looking at the extension prior to .001 (like in "large_stuff.zip.001") and taking into account the linked files (.002, .003 etc) which should appear in the same directory as the .001 leader file, either as compressed (.zip or .7z) or CSV file (any other extension), and which are the other parts of the large file that was split, for instance for easing transfer from a machine to another.
Keep on the good work!
CChris

[xiaodaigh]

Support for .7z compressed format in zip_to_disk.frame()

If you can help me do some research on this. What R packages supports .7z format? And does the package support the multipart .7z that you need?

[cchris005]

Hi, and thanks for the quick return. A quick search did not return any R package handling .7z files without 7-zip being installed prior. However, the Python world seems to have it: https://pypi.org/project/py7zr/ This page has links to project, documentation and, more importantly perhaps, 7-zip file specifications. 7-zip is free software, see https://www.7-zip.org/ . Code is under a composite license (LGPL / BSD 3 / unRAR ), you may wish to read the license terms linked to from that web page. .001 (multipart) is both supported by zip and 7-zip formats. Also, file splitters would cut large files into smaller ones with a similar naming scheme (like tabfile.txt.001 and so on after splitting tabfile.txt). There is, of course, a "lazy" implementation requiring that user has 7z.exe installed, which would system() the compressing/uncompressing job out. The following blog post https://www.r-bloggers.com/2012/04/reproducible-research-running-odfweave-with-7-zip/ talks about it and shows the proper command instructing 7z.exe to perform as desired. The part about the odfWeave package is not relevant to us. Please tell me whether you need more focussed items, as long as I can find them :) . Chris

…

----- Mail d'origine ----- De: evalparse ***@***.***> À: xiaodaigh/disk.frame ***@***.***> Cc: cchris005 ***@***.***>, Author ***@***.***> Envoyé: Wed, 10 Mar 2021 23:49:21 +0100 (CET) Objet: Re: [xiaodaigh/disk.frame] Enhancements to zip_to_disk.frame() (#328) Support for .7z compressed format in zip_to_disk.frame()If you can help me do some research on this. What R packages supports .7z format? And does the package support the multipart .7z that you need?— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[xiaodaigh]

did not return any R package handling .7z files without 7-zip being installed prior.

that would be difficult to test properly. I have other dev items to prioritise at the moment like better NSE and highe test coverage so unlikely to be able to prioritise this anytime soon.

If this is really needed, I do accept consulting funding to prioritise feature if needed.

[cchris005]

Hi again, Thank you for your reply, which I can definitely understand. It is not likely that my organisation would pay for the enhancements, given current priorities, but it is not impossible altogether. Any indications will not be available until mid-May where the committee about package certification convenes. Best regards, and keep the good job rolling! Chris

…

----- Mail d'origine ----- De: evalparse ***@***.***> À: xiaodaigh/disk.frame ***@***.***> Cc: cchris005 ***@***.***>, Author ***@***.***> Envoyé: Tue, 16 Mar 2021 23:36:14 +0100 (CET) Objet: Re: [xiaodaigh/disk.frame] Enhancements to zip_to_disk.frame() (#328) did not return any R package handling .7z files without 7-zip being installed prior.that would be difficult to test properly. I have other dev items to prioritise at the moment like better NSE and highe test coverage so unlikely to be able to prioritise this anytime soon.If this is really needed, I do accept consulting funding to prioritise feature if needed.— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.