This does break the API of being able to import any lib file and get
its libs, however I'm not sure people did this.
I made this while exploring being able to swap out docFn with a stub
in #2305, to avoid functor performance problems. I don't know if that
is going to move forward (or if it is a problem or not,) but after
doing all this work figured I'd put it up anyway :)
Two notable advantages to this approach:
1. when a lib inherits another lib's functions, it doesn't
automatically get put in to the scope of lib
2. when a lib implements a new obscure functions, it doesn't
automatically get put in to the scope of lib
Using the test script (later in this commit) I got the following diff
on the API:
+ diff master fixed-lib
11764a11765,11766
> .types.defaultFunctor
> .types.defaultTypeMerge
11774a11777,11778
> .types.isOptionType
> .types.isType
11781a11786
> .types.mkOptionType
11788a11794
> .types.setType
11795a11802
> .types.types
This means that this commit _adds_ to the API, however I can't find a
way to fix these last remaining discrepancies. At least none are
_removed_.
Test script (run with nix-repl in the PATH):
#!/bin/sh
set -eux
repl() {
suff=${1:-}
echo "(import ./lib)$suff" \
| nix-repl 2>&1
}
attrs_to_check() {
repl "${1:-}" \
| tr ';' $'\n' \
| grep "\.\.\." \
| cut -d' ' -f2 \
| sed -e "s/^/${1:-}./" \
| sort
}
summ() {
repl "${1:-}" \
| tr ' ' $'\n' \
| sort \
| uniq
}
deep_summ() {
suff="${1:-}"
depth="${2:-4}"
depth=$((depth - 1))
summ "$suff"
for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do
if [ $depth -eq 0 ]; then
summ "$attr" | sed -e "s/^/$attr./"
else
deep_summ "$attr" "$depth" | sed -e "s/^/$attr./"
fi
done
}
(
cd nixpkgs
#git add .
#git commit -m "Auto-commit, sorry" || true
git checkout fixed-lib
deep_summ > ../fixed-lib
git checkout master
deep_summ > ../master
)
if diff master fixed-lib; then
echo "SHALLOW MATCH!"
fi
(
cd nixpkgs
git checkout fixed-lib
repl .types
)
It's broken on all versions of Python (I've tried 2.7, 3.4, 3.5, 3.6)
I think the root cause is that PyBrain is not working with numpy >= 1.12.0 as I reported here:
https://github.com/pybrain/pybrain/issues/217
(The relevant release notes may be found here):
https://docs.scipy.org/doc/numpy-1.12.0/release.html#compatibility-notes
The PyBrain github repo does not seem very active (last commit 18 months ago, last release 3 years),
so I have some doubts as to whether this will be fixed any time soon.
I suppose an alternative solution could be to reintroduce the explicit dependency to numpy 1.11. But,
this is not entirely trivial: in c9b4a2f319, the versions 1.10, 1.11, 1.12 were folded into a single version.
Also, the numpy dependency is not a direct one, but is implied via scipy
ftfy package was added for spaCy and is only used by spaCy.
This change downgrades its version to meet the bounds specified by
spaCy (>=4.4.2,<5.0.0).
Relevant to #28643.
10k staging builds are not yet finished on Hydra (mostly darwin),
but we now have a 20k jobs rebuilding directly on master, so we would
never get to merge this way...
The backslash wasn't properly escaped, and "\." is apparently equal to
".". So it's accidentally filtering out these valid file names (in
Nixpkgs):
trace: excluding clfswm
trace: excluding larswm
trace: excluding mkpasswd
While at it, turn the file filter stricter to what it was before
e2589b3ca2. That is, the file name must
start with a dot: '.swp', '.foo.swo' are filtered but 'bar.swf' is not.
An attempt to fix tests under 32bit linux to unblock nix build
(cherry picked from commit 3418e734698793e1272769a6cac2efe1498a0de1)
Signed-off-by: Domen Kožar <domen@dev.si>
Google publishes prebuilt tensorflow whl for python 3.4, 3.5, 3.6,
but nix expression for tensorflow only supported 3.5.
This change adds support for python-3.6.
It doesn't look good when the initial admin user is named
"<hash>-gitolite-admin" and the key stored as
"<hash>-gitolite-admin.pub". Instead, make it simply "gitolite-admin"
and "gitolite-admin.pub".
