[14.0][FIX] date format in bokeh #8
Conversation
| def _format_date(self, date): | ||
| if not date: | ||
| return "" | ||
| return datetime.datetime.strptime(date, "%Y-%m-%d").strftime("W%W %Y") |
There was a problem hiding this comment.
Haha, yeah, this will solve it for week granularity :-) but for the month view for example it will be translated in the customer language, so it will be harder to replicate it like this.
Did you consider the version whereby we just display all the items in the dict in the right order, instead of querying for specific keys?
There was a problem hiding this comment.
yes, I am trying to make as few changes as possible, and mostly this is an exercise to prove to myself that this error had nothing to do with libraries etc. Which it does not, apparently. I am still wondering however how this stopped working out of a sudden.
There was a problem hiding this comment.
My take on it was that it actually came into being because of a change you made - the backporting of the _range thing from 15.0 to 14.0. If we would make a slightly bigger change to not query for exact keys, it means we could scrap that backport, and actually end up with less custom code.
There was a problem hiding this comment.
@thomaspaulb please advise here, because I tried a couple of things and nothing works.
There was a problem hiding this comment.
My rough idea was that if we:
-
make sure that groupdata comes out sorted by date
-
change this regular dict to be an
OrderedDict
Then, here we can change:
forecast = data[employee][project]
plot_data[project] = [
forecast.get(self._format_date(date), 0) for date in dates
]
to
forecast = data[employee][project]
plot_data[project] = [f for f in forecast]
This should work unless forecast contains more dates than what gets placed in dates as returned from _get_time_range(); in that case, it will become more complicated. Still not impossible I think, but maybe first see if this will work.
Is this along the lines of what you tried? If not, what did you do?
There was a problem hiding this comment.
@thomaspaulb it took me a while but the solution was there, right in front of me. Adding an additional key in read_group() result solved everything. @gjotten tested this, let's wait for few more tests and then we can merge
There was a problem hiding this comment.
That also seems like a nice solution!
ntsirintanis
force-pushed
the
14.0-backport-project_forecast_line_bokeh_chart_empty_plot_fix
branch
2 times, most recently
from
e62f5b0
to
a18350c
Compare
ntsirintanis
force-pushed
the
14.0-backport-project_forecast_line_bokeh_chart
branch
from
c38c785
to
cbe5cff
Compare
ntsirintanis
force-pushed
the
14.0-backport-project_forecast_line_bokeh_chart_empty_plot_fix
branch
from
a18350c
to
117dd01
Compare
ntsirintanis
force-pushed
the
14.0-backport-project_forecast_line_bokeh_chart_empty_plot_fix
branch
from
117dd01
to
e020bd7
Compare
| @@ -38,6 +38,9 @@ def _read_group_raw( | |||
| if group.get(df): | |||
| range_from, range_to = group[df][0].split("/") | |||
| group["__range"][field_name] = {"from": range_from, "to": range_to} | |||
| # Inject another key in the result | |||
There was a problem hiding this comment.
... for easy access later when we need to display this date
ntsirintanis
force-pushed
the
14.0-backport-project_forecast_line_bokeh_chart_empty_plot_fix
branch
from
9643327
to
12f4b21
Compare
ntsirintanis
merged commit 27770f7
into
14.0-backport-project_forecast_line_bokeh_chart
No description provided.