Missing <repo>/src/sc/src/jq/disp/resolve.jq about sc HOT 13 CLOSED

As far as an api key is concerned, I thought the same as you did until I read this. I meant to link to it in the readme, but hadn't yet because I haven't decided the best way to go about it. I could also ship with a few keys to use as fallbacks.

from sc.

Sorry, it isn't really finished, but it was working alright for me. To be honest I didn't expect anyone else to try it out to soon. I'm working on adding tests at the moment.

For the time being though, can you tell me what command you were running?

from sc.

Also, I just pushed up an env command that prints out the versions of all the dependencies. If you could pull down that and run

$ sc env

tnat would be awesome!

from sc.

yeah, I saw it added to the JSON tooling github repo and thought I'd check it out. So here's some of the info you requested:

 $  sc env
find (GNU findutils) 4.6.0.187-5dae
git version 2.19.1
jq-1.5-1-a5b5cbe
mktemp (GNU coreutils) 8.28
tree v1.7.0 (c) 1996 - 2014 by Steve Baker, Thomas Moore, Francesc Rocher, Florian Sesser, Kyosuke Tokoro 
wcache not found
zsh 5.5.1 (x86_64-ubuntu-linux-gnu)
sc 0.1.0

Now this is a good thing to have. I am not able to find "wcache" as something I can install on my Ubuntu machine (18.10). The closest thing I have to it in apt is ccache, which I installed. That isn't it. I even searched all the repos on my system for /usr/bin/wcache -- no dice. I googled it. Nothing. I checked for ppas. Nothing. So some instructions on how to install that would be helpful. The command I'm running is sc resolve earthlibraries

from sc.

I symlinked (band-aid) wcache to ccache btw...that's when I get the missing resolve.jq error btw. It's doubtful that the missing wcache/ccache would have any bearing on the missing resolve.jq.

from sc.

Okay, the problem is that wcache is another tool I made, and it is doubtful that ccache would be a viable replacement. wcache is a wget wrapper that basically downloads files to a centralized cache, and upon successive requests for the same url, will give you the cached version, reducing unnecessary requests for relatively static content. There is a flag in wcache that causes it to echo the path to the cached file and it is this path that is used in the call to jq. Therefore, the reason sc is looking for a file named resolve.jq is because whenever ccache gets run, its output is being used as part of a path.

Here is the repo for wcache: http://github.com/annacrombie/wcache.

I didn't really explain this requirement at all in the README.md, and I doubt many people would want to install both sc and wcache separately, so to fix the problem, I bundled wcache as a submodule of sc and added a wrapper that uses the bundled version (a2f771a). Since you already cloned this repo, you will have to manually initialize the new submodules. Or you can just re-clone:

git clone --recurse-submodules https://github.com/annacrombie/sc

Hopefully that will fix the problem.

from sc.

OK, cool. That got me further. Looks like there's another dependency problem. Your column args don't seem to match what my column args are. This could be a version disparity. Are you using Mac or some builtin function? My column doesn't have a -N argument.:

[sc log] [debug] checking /home/<user>/<repo>/sc/data/cache/api.soundcloud.com/resolve/earthlibraries@ (36s old), best-by 86400
[sc log] [debug] its fresh!
column: invalid option -- 'N'
usage: column [-txne] [-c columns] [-s sep] [file ...]
[sc log] [error] jq error in /home/<user>/<repo>/sc/src/jq/disp/user.jq
while parsing /home/<user>/<repo>/sc/data/cache/api.soundcloud.com/users/509984022$

You might want to document the fact that an API client id is required to use this tool and that a problem exists where soundcloud is no longer offering API Client IDs (until further notice it seems). There's a way around this but it requires seeking someone else's client ID from github.

from sc.

In regard to column mine reports column from util-linux 2.33,

       -N, --table-columns names
              Specify  the  columns  names  by comma separated list of names. The names are used for the table header or to address column in
              option arguments.

I was actually planning on implementing a change that would avoid the need for me to use -N at all, so I will go ahead and do that.

from sc.

Cool! So yeah, you're one version ahead of me on util-linux

$ sudo dpkg -l util-linux
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name                                         Version                     Architecture                Description
+++-============================================-===========================-===========================-=============================================================================================
ii  util-linux                                   2.32-0.1ubuntu2             amd64                       miscellaneous system utilities

and for some reason, columns doesn't even come in util-linux for me so it sounds like you're probably using 19.04?

$ sudo apt-file search $(which column)
autogen: /usr/bin/columns
bsdmainutils: /usr/bin/column

from sc.

I'm actually running Gentoo and my sys-apps/util-linux-2.33-r1 does come with column. Idk, maybe relying on column at all is a bad idea.

from sc.

Yeah, that's a nuanced tool. It's a cool one, but I guarantee there will be problems with it at least in the short term.

from sc.

Okay, the latest commit no longer requires column to have a -N flag :).

from sc.

Yup! That worked!

from sc.