You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'd like to be able to generate wheels for all dependencies in a Pipfile.lock while taking advantage of hash verification. Prior to 11.1.2 you could accomplish this with:
pipenv lock -r > reqs.txt
pip wheel -r reqs.txt
(#1417 was problematic for this workflow, but otherwise it worked)
In 11.1.2, hashes were removed from the requirements-style output which (afaik) means there is no way to build/download hash-verified wheels from a lock file.
The use-case is to package Python modules for offline installs or installs without compilation/development tools on the server. The wheels can be built, then distributed and installed in a minimal Python environment.
Is this functionality that you'd consider including in the project? If not, is there some way to support this workflow again?
The text was updated successfully, but these errors were encountered:
Just in case anybody looks here for a solution, it's pretty easy (assuming the pipenv internals don't change):
frompipenv.utilsimportconvert_deps_to_pipwithopen('Pipfile.lock') asf:
deps=json.load(f)['default']
# remove local project which wouldn't have a hashfork, vinlist(deps.items()):
ifv.get('path') =='.':
del(deps[k])
path_to_requirements_file_with_hashes=convert_deps_to_pip(deps)
I'd like to be able to generate wheels for all dependencies in a
Pipfile.lock
while taking advantage of hash verification. Prior to 11.1.2 you could accomplish this with:(#1417 was problematic for this workflow, but otherwise it worked)
In 11.1.2, hashes were removed from the requirements-style output which (afaik) means there is no way to build/download hash-verified wheels from a lock file.
The use-case is to package Python modules for offline installs or installs without compilation/development tools on the server. The wheels can be built, then distributed and installed in a minimal Python environment.
Is this functionality that you'd consider including in the project? If not, is there some way to support this workflow again?
The text was updated successfully, but these errors were encountered: