BigQuery: docs for Row.keys, items, values, and get functions. #4410
Conversation
googlebot
added
the
cla: yes
|
|
||
| def values(self): | ||
| """Return the values included in this row. | ||
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
| ['x', 'y'] | ||
| """ | ||
| keys = self._xxx_field_to_index.keys() | ||
| return keys |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
| for k, i | ||
| in self._xxx_field_to_index.items() | ||
| ] | ||
| return items |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
| >>> Row(('a', 'b'), {'x': 0, 'y': 1}).get('z', default = '') | ||
| '' | ||
| """Return value for ``key`` if ``key`` is in the row, else | ||
| ``default``. |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
| Args: | ||
| key (str): The key of the column to access | ||
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
| >>> Row(('a', 'b'), {'x': 0, 'y': 1}).get('x') | ||
| 'a' | ||
| The default value is ``None`` when the key does not exist. |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
- Refactors items, keys to return iterators. - Copies values to prevent mutating the original row object.
|
@tswast For future PRs, can you squash while merging the PR? It's harder to review a PR when I can't tell what you changed when squashing. (Though I definitely feel the seem way, I squash locally a lot because for some reason I care about commit log clutter, even though we are 5500+ commits in to this project.) |
| Return keys as of a dict: | ||
| >>> Row(('a', 'b'), {'x': 0, 'y': 1}).keys() | ||
| ['x', 'y'] | ||
| return copy.deepcopy(self._xxx_values) |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
| ] | ||
| return items | ||
| for key, index in six.iteritems(self._xxx_field_to_index): | ||
| yield (key, copy.deepcopy(self._xxx_values[index])) |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
dhermes
left a comment
dhermes
left a comment
There was a problem hiding this comment.
LGTM, though I'm a bit "sad" about the copies. (They seem necessary evils, but still evils.)
|
Yeah, it is a bit sad about copies. There's definitely a trade-off between safety and performance. Probably better to err on the side of safety. |
|
@tswast Looks like the generators broke the build: https://circleci.com/gh/GoogleCloudPlatform/google-cloud-python/4288 Shall we revert, then re-submit this PR but push the branch to the GoogleCloudPlatform remote so system tests run? |
|
Yeah, let's revert. I didn't think to check the system tests. |
3 participants
I forgot to ask for these in #4393 before I merged.