Dependency stops getting locked if switching to Git install

[Gallaecio]

All works as expected when using regular packages:

$ cat requirements.in 
scrapinghub-entrypoint-scrapy
scrapy==2.5.1
$ pip-compile |& grep twisted
    #   twisted
    # via twisted
    # via twisted
    #   twisted
    # via twisted
    # via twisted
    # via twisted
twisted[http2]==22.1.0
    # via twisted
    #   twisted

However, if I install the very same Scrapy version from Git instead of from the Python Package Index, twisted is no longer frozen:

$ cat requirements.in 
scrapinghub-entrypoint-scrapy
git+https://github.com/scrapy/scrapy.git@2.5.1#egg=scrapy
$ pip-compile |& grep twisted

Note: It would not surprise me if this is a known issue, but I have checked for possible existing items, but while there are multiple issues about extras, I could not find any that specifically mention the combination of extras and installation from Git being an issue.

[AndydeCleyre]

This looks to me to be fixed in PR #1519. Hopefully that (and a handful of other PRs) will soon make it to master.

[atugushev]

New resolver from #1539 also fixes this issue.

Details

$ pip-compile --resolver backtracking
#
# This file is autogenerated by pip-compile with python 3.8
# To update, run:
#
#    pip-compile --resolver=backtracking
#
attrs==21.4.0
    # via
    #   automat
    #   service-identity
    #   twisted
automat==20.2.0
    # via twisted
certifi==2021.10.8
    # via requests
cffi==1.15.0
    # via cryptography
charset-normalizer==2.0.12
    # via requests
constantly==15.1.0
    # via twisted
cryptography==36.0.1
    # via
    #   pyopenssl
    #   scrapy
    #   service-identity
cssselect==1.1.0
    # via
    #   parsel
    #   scrapy
h2==3.2.0
    # via
    #   scrapy
    #   twisted
hpack==3.0.0
    # via h2
hyperframe==5.2.0
    # via h2
hyperlink==21.0.0
    # via twisted
idna==3.3
    # via
    #   hyperlink
    #   requests
incremental==21.3.0
    # via twisted
itemadapter==0.4.0
    # via
    #   itemloaders
    #   scrapy
itemloaders==1.0.4
    # via scrapy
jmespath==0.10.0
    # via itemloaders
lxml==4.8.0
    # via
    #   parsel
    #   scrapy
parsel==1.6.0
    # via
    #   itemloaders
    #   scrapy
priority==1.3.0
    # via twisted
protego==0.2.1
    # via scrapy
pyasn1==0.4.8
    # via
    #   pyasn1-modules
    #   service-identity
pyasn1-modules==0.2.8
    # via service-identity
pycparser==2.21
    # via cffi
pydispatcher==2.0.5
    # via scrapy
pyopenssl==22.0.0
    # via scrapy
queuelib==1.6.2
    # via scrapy
requests==2.27.1
    # via scrapinghub
retrying==1.3.3
    # via scrapinghub
scrapinghub==2.3.1
    # via scrapinghub-entrypoint-scrapy
scrapinghub-entrypoint-scrapy==0.12.1
    # via -r requirements.in
scrapy @ git+https://github.com/scrapy/scrapy.git@2.5.1
    # via
    #   -r requirements.in
    #   scrapinghub-entrypoint-scrapy
service-identity==21.1.0
    # via scrapy
six==1.16.0
    # via
    #   automat
    #   parsel
    #   protego
    #   retrying
    #   scrapinghub
    #   service-identity
    #   w3lib
twisted[http2]==22.1.0
    # via scrapy
typing-extensions==4.1.1
    # via twisted
urllib3==1.26.8
    # via requests
w3lib==1.22.0
    # via
    #   itemloaders
    #   parsel
    #   scrapy
zope-interface==5.4.0
    # via
    #   scrapy
    #   twisted

# The following packages are considered to be unsafe in a requirements file:
# setuptools

[AndydeCleyre]

Closing as fixed by #1519, thanks!