From: "Zoltan Boszormenyi" <zboszor@pr.hu>
To: Kamel Bouhara <kamel.bouhara@bootlin.com>,
Jean-Marie Lemetayer <jeanmarie.lemetayer@gmail.com>
Cc: openembedded-core@lists.openembedded.org,
Alexandre Belloni <alexandre.belloni@bootlin.com>,
Thomas Petazzoni <thomas.petazzoni@bootlin.com>,
sandra.tobajas@savoirfairelinux.com
Subject: Re: [OE-core] [PATCH 0/2] Fix issue with npm shrinkwrap fetcher
Date: Wed, 3 Feb 2021 10:55:01 +0100 [thread overview]
Message-ID: <390151a7-070e-4d1f-2ffe-4051e04e1ab1@pr.hu> (raw)
In-Reply-To: <YABmBpmfl8WIK/Jf@kb-xps>
2021. 01. 14. 16:40 keltezéssel, Kamel Bouhara írta:
> On Thu, Jan 14, 2021 at 03:25:50PM +0100, Jean-Marie Lemetayer wrote:
>> Hi Kamel,
>>
>
> Hi Jean-Marie,
>
>> Thanks for your work. I should have fixed this a long time ago but I
>> didn't have the time :(
>>
>
> Well its never too late !
>
>> There is actually an open bug for this issue:
>> https://bugzilla.yoctoproject.org/show_bug.cgi?id=13901
>> I think you can add the reference to your commit messages. And close
>> it if merged.
Please backport this series to Gatesgarth as soon
as possible because it's a problem there, too.
The series applies cleanly.
It fixes some of the issues I see with
https://www.npmjs.com/package/nanomsg which
still fails for a different reason.
Thanks in advance,
Zoltán
>>
>
> OK, let's see if there are other comments then I shall resent with
> proper commit messages.
>
>> Please add some tests to ensure that:
>> 1. devtool / recipetool do not create empty shrinkwrap file
>> 2. the npmsw fetcher allows empty shrinkwrap file
>> 3. the npm class allows empty shrinkwrap
>
> Actually the fix is only to check wether or not the dependencies are
> listed and if not we make sure the npm fetcher will not look for them.
>
> So the empty shrinkwrap file is still created and allowed.
>
> I've already done some test in both case with and w/o dependencies.
>
>>
>> See:
>> https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=22dd46cc34629d0750177fddff2e1c178c854340
>> https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=44b2ab8d5eb4fd1fc9a157fce37aaa7b7a7065e5
>> https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=f6728edb7e022b27223d28bd80ce40c2f2374a13
>>
>> Also Sandra has already worked on it, but it didn't get merged. I
>> don't know why.
>>
>> See:
>> https://lists.openembedded.org/g/bitbake-devel/topic/76459311#11643
>> https://lists.openembedded.org/g/bitbake-devel/message/11648
>>
>
> I see, yet I went a bit further ensuring the npm recipes will not call
> the npmsw fetcher when no dependencies are given.
>
> Thanks for you comments.
>
> Cheers,
> Kamel
>
>> BR
>> Jean-Marie
>>
>> On Thu, Jan 14, 2021 at 8:13 AM Kamel Bouhara <kamel.bouhara@bootlin.com> wrote:
>>>
>>> Hello all,
>>>
>>> This small patch series aims to fix the following build issue faced
>>> with some npm packages:
>>>
>>> DEBUG: Executing python function do_fetch
>>> DEBUG: Executing python function base_do_fetch
>>> DEBUG: Python function base_do_fetch finished
>>> DEBUG: Python function do_fetch finished
>>> ERROR: Failure expanding variable TUNE_FEATURES_tune-armv7at-neon, expression was ${TUNE_FEATURES_tune-armv7at} neon which triggered exception RecursionError: maximum recursion depth exceeded while calling a Python object
>>>
>>> After struggling a lot, I figured out that this only happen for npm
>>> packages not having dependencies listed in their shrinkwrap file (e.g. bcryptjs).
>>>
>>> Yet I still didn't got how is the json parsing impacting
>>> the python context here ?
>>>
>>> Please feel free to comment.
>>>
>>> Thanks.
>>>
>>> Kamel Bouhara (2):
>>> npm.bbclass: make shrinkwrap file optional
>>> recipetool: create: only add npmsw url if required
>>>
>>> meta/classes/npm.bbclass | 31 +++++++++++++++++++++----------
>>> scripts/lib/recipetool/create_npm.py | 6 +++++-
>>> 2 files changed, 26 insertions(+), 11 deletions(-)
>>>
>>> --
>>> 2.11.0
>>>
>>>
>>>
>>>
>
> --
> Kamel Bouhara, Bootlin
> Embedded Linux and kernel engineering
> https://bootlin.com
>
>
>
>
>
next prev parent reply other threads:[~2021-02-03 9:55 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-14 7:12 [PATCH 0/2] Fix issue with npm shrinkwrap fetcher kamel.bouhara
2021-01-14 7:12 ` [PATCH 1/2] npm.bbclass: make shrinkwrap file optional Kamel Bouhara
2021-01-14 7:12 ` [PATCH 2/2] recipetool: create: only add npmsw url if required Kamel Bouhara
2021-01-14 14:25 ` [OE-core] [PATCH 0/2] Fix issue with npm shrinkwrap fetcher Jean-Marie Lemetayer
2021-01-14 15:40 ` Kamel Bouhara
2021-02-03 9:55 ` Zoltan Boszormenyi [this message]
2021-02-05 1:40 ` Anuj Mittal
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=390151a7-070e-4d1f-2ffe-4051e04e1ab1@pr.hu \
--to=zboszor@pr.hu \
--cc=alexandre.belloni@bootlin.com \
--cc=jeanmarie.lemetayer@gmail.com \
--cc=kamel.bouhara@bootlin.com \
--cc=openembedded-core@lists.openembedded.org \
--cc=sandra.tobajas@savoirfairelinux.com \
--cc=thomas.petazzoni@bootlin.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox