Cocoapods: PivotalCoreKit Long Directory Name break
Report
Original issue filed with PivotalCoreKit here: https://github.com/pivotal/PivotalCoreKit/issues/191
What did you do?
Updated our podfile dependencies ('pod update')
What did you expect to happen?
Install all pod dependencies correctly
What happened instead?
We received an error ‘ENAMETOOLONG’ for installing our PivotalCoreKit pod as shown in attached images
mgoovaer
All 16 comments
Could you please share the error reporting, including the used Podfile as text — that way folks won't have to retype it when trying to reproduce
neonichu
on 8 Jun 2016
Maybe @mrackwitz wants to check this one out — I'm thinking beyond a certain length the auto generated name becomes non-human readable anyway and we could switch to a different format (hash of the full name?).
neonichu
on 9 Jun 2016
I _think_ turning off de-duplicating targets removes the fancy names
orta
on 9 Jun 2016
Dang! @orta is right. That's a problem with the names generated by the target deduplication to differentiate the targets. Maybe we should just use numbers all together?
As an alternative workaround you could also share the same subspecs across both targets.
mrackwitz
on 9 Jun 2016
I am having the same issue, and can not share all the subspaces across targets, as different subspecs in the pod target different platform sets, and each target in my podfile is platform specific.
jjamminjim
on 10 Jun 2016
@neonichu attached are the Podfile and the error as text, thanks for looking into this!
mgoovaer
on 13 Jun 2016
@neonichu Is there any update on this?
the input seems to be provided.
juliozatarain
on 20 Jun 2016
We know the problem, but no-one has worked on this yet I'm afraid. This is the crux of the problem:
That's a problem with the names generated by the target deduplication to differentiate the targets.
re:
Maybe we should just use numbers all together?
Could we do it if it's over _x_ chars? E.g. 60?
orta
on 20 Jun 2016
@orta could do that, or take an MD5 sum if it's over x chars as well
segiddins
on 20 Jun 2016
@orta @segiddins any eta on this fix? it looks like you have an idea of what approach to take...
mgoovaer
on 23 Jun 2016
No ETA yet, but I've a work in progress fix, I'll put up soon.
mrackwitz
on 23 Jun 2016
@mrackwitz thanks!
mgoovaer
on 24 Jun 2016
bump @mrackwitz
juliozatarain
on 6 Jul 2016
Hey guys, is there an update on this?
cooldhiman
on 8 Jul 2016
I have a common problem, when use a private repos?
devluyy
on 11 Jul 2016
There's an open PR addressing this now
segiddins
on 11 Jul 2016
Related issues
evermeer
·
3Comments
lzwjava
·
3Comments
steffendsommer
·
3Comments
marzapower
·
3Comments
pronebird
·
3Comments
Most helpful comment
There's an open PR addressing this now