Refactor iterator to separate into HTTP/GAX iterators #2617
Conversation
For this commit, moving some features of HTTPIterator constructor onto base class.
This was HTTP/JSON specific and belongs in the HTTP subclass.
Also updating the _GAXPageIterator mock to allow multiple pages.
Instead, using the page token directly from the page iterator passed in (this may occasionally be strange to a user, e.g. if the token is INITIAL_PAGE).
googlebot
added
the
cla: yes
This somewhat of a cart-before-the-horse change, but is done this way to make the commit easier to understand, before unifying the two approaches via _next_page().
Also moving the token setting behavior from HTTPIterator._get_next_page_response() into _next_page().
Also a lint fix for an unimorted member and a unit test fix adding a page token to allow more paging.
|
This conceptually LGTM, I'll defer @tseaver for implementation review. |
|
@tseaver Can I get an amen?
|
tseaver
left a comment
tseaver
left a comment
There was a problem hiding this comment.
Ouch! Commit-by-comit was a hard row to hoe, especially on the last commit, which blew away so much of what had been done before
|
|
||
| class Iterator(object): | ||
| class HTTPIterator(object): | ||
| """A generic class for iterating through Cloud JSON APIs list responses. |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
core/unit_tests/test_iterator.py
Outdated
| self.assertIs(iterator.client, client) | ||
| self.assertEqual(iterator.max_results, max_results) | ||
| self.assertEqual(list(iterator._page_iter), []) | ||
| self.assertFalse(iterator._page_increment) |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
| page = Page(iterator, items, _item_to_topic) | ||
| iterator.next_page_token = page_iter.page_token or None | ||
| iterator.num_results += page.num_items | ||
| yield page |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
core/google/cloud/iterator.py
Outdated
| except StopIteration: | ||
| return None | ||
|
|
||
| def _wrap_gax(self, page_iter): |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
| response = iterator._get_next_page_response() | ||
| self.assertEqual(response['items'], [{'name': key1}, {'name': key2}]) | ||
| self.assertEqual(iterator.page_number, 1) | ||
| self.assertEqual(iterator.next_page_token, token) |
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.
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
|
@tseaver Sorry for the bait and switch at the end. It didn't become clear that a |
…ctor Refactor iterator to separate into HTTP/GAX iterators
Refactor iterator to separate into HTTP/GAX iterators
5 participants
NOTE: I'm happy to split this PR into tiny PRs if it makes review easy, though the commits are all pretty compact, so it should be reviewable commit-by-commit.