There's an assertion in StringTree.py.__set(), assert self.content !=
str(content), which often triggers. Not sure what the idea behind the
assertion was, disable it.
Signed-off-by: Jan Lindemann <jan@janware.com>
Run
monkeytype run jw-acc.py --log-level debug db import --format mdb-json-dir dbs
monkeytype list-modules | grep ^jw | while read m; do
monkeytype apply $m
done
and fix the fallout.
Signed-off-by: Jan Lindemann <jan@janware.com>
child_list() throws an exception saying that <str> has no to_list()
method. No, it hasn't. No idea how this has even worked at all, ever.
Signed-off-by: Jan Lindemann <jan@janware.com>
As opposed to the C++ variant, StringTree.value() doesn't take a path
argument, but only returns the value of the node it's called on.
Change this. Returns None if the path is not found.
Signed-off-by: Jan Lindemann <jan@janware.com>
Add type annotations as generated by monkeytype and jw-devops/test, plus some
hand editing to satisfy both monkeytype and mypy.
Signed-off-by: Jan Lindemann <jan@janware.com>
StringTree.__getitem__() returned the node, now return its value
child instead. That's nicer and closer to the expected behaviour, as
it supports StringTree usage as a key-value store for configuration
or so. It breaks the existing API, but I think I have fixed the only
pace where it was used: jw-grammar/grammar.py.
Signed-off-by: Jan Lindemann <jan@janware.com>
Add support for adding children of type StringTree. If a StringTree
argument is passed to .add(), this should do the right thing. It
makes use of the newly added .__add_children() method and involves
quite a bit of hairy case distinctions.
Signed-off-by: Jan Lindemann <jan@janware.com>
This commit tries to add support for the
[section]
first line
second line
syntax, i.e. non-assignments, without equal sign. Half-baked, but
still better than before. Also support some more Python 3, i.e.
dict.items() instead of iteritems()
Signed-off-by: Jan Lindemann <jan@janware.com>
