Commit Graph

7051 Commits

Author SHA1 Message Date
Avital Oliver
2f7724ac2a New failing test for re-running helpers in #each within #with 2014-03-18 21:53:07 -07:00
David Greenspan
53c83be0b8 Make #each stop sooner; observe inside Deps comp
By doing the #each observe inside the current Deps computation from materialization, it will stop when the parent computation is invalidated.
2014-03-18 14:59:42 -07:00
David Greenspan
20db762c6d Modify test 2014-03-18 14:38:40 -07:00
David Greenspan
af9bc84ecc Test for helper re-run after template removed
(some failures)
2014-03-18 14:32:07 -07:00
David Greenspan
42258895be Hack to stop emboxed .data() sooner
This fixes cmather's test case for real this time (https://github.com/cmather/meteor-emboxed-issue)
2014-03-17 20:37:10 -07:00
David Greenspan
75b34dff39 rewrite emboxValue (doesn't fix everything yet) 2014-03-17 20:22:22 -07:00
Nick Martin
d8575d0a24 Merge branch 'devel' into shark release/0.7.2-rc1-with-blaze 2014-03-17 15:51:18 -07:00
David Glasser
8a16f50138 Ensure that hook.each is called from a Fiber
That gives us enough guarantees that we can simplify the error-handling
code.

We also ensure that the "new connection" handler in livedata_server runs
in a Fiber, which fixes a bug introduced in d049bf7506 where connections
from pre-pre1 clients would crash (due to Meteor.setTimeout which only
works in a Fiber).
2014-03-17 15:49:27 -07:00
Avital Oliver
6282e42b1d Test that helpers are called only once when invalidated.
(This was fixed by the recent commit eliminating Deps.isolateValue)
2014-03-17 15:24:14 -07:00
Avital Oliver
b97d9e143a Eliminate Deps.isolateValue
Unfortunately, using it always led to reactive functions
firing twice. Instead, we now use emboxValue in the two
cases we originally had Deps.isolateValue.
2014-03-17 15:09:52 -07:00
David Glasser
da96352c65 Make some white-box tests less flaky
If we want the *internal* state of the oplog driver to be consistent, we
need to make sure that we start our observes at a consistent point in
the oplog.  (ie, initial inserts need to have been fully processed, so
that we don't process them during the original QUERYING phase, which can
give us a different unpublished buffer.)

Note that only the white-box tests (looked at _unpublishedBuffer, eg)
appeared to be flaky: the actual docs published seemed to be correct in
all cases.
2014-03-17 12:11:21 -07:00
David Glasser
5486fd0b33 Merge branch 'sort-selector' into devel 2014-03-17 02:56:12 -07:00
David Glasser
93e1969560 history update 2014-03-17 02:56:07 -07:00
David Glasser
17297d98d7 make LocalCollection.Cursor easier to read
and whitespace cleanup
2014-03-17 02:17:06 -07:00
David Glasser
15fa6b2ab7 Pass Matcher to Sorter constructor 2014-03-17 02:17:06 -07:00
David Glasser
cda5152491 Refactor a test to remove magic numbers 2014-03-17 02:17:06 -07:00
David Glasser
d4241aa998 Improve ELEMENT_OPERATORS comment.
Make there be one consistent syntax for element operators.
2014-03-17 02:17:06 -07:00
David Glasser
8488047c13 support non-initial keys in sort selector filter 2014-03-17 02:17:06 -07:00
David Glasser
29cfa27ef3 support regexps in sort key filter
also in EJSON.clone
2014-03-17 02:17:06 -07:00
David Glasser
125051f2f8 more tests for sort/select, and fix a silly bug 2014-03-17 02:17:06 -07:00
David Glasser
17ec29603c Initial stab at fixing the sort/selector issue 2014-03-17 02:17:06 -07:00
David Glasser
8822386d7b test for affectedByModifier for $elemMatch 2014-03-17 02:17:06 -07:00
David Glasser
9968f19f8a Failing test for sort/selector issue 2014-03-17 02:17:05 -07:00
Nick Martin
4b2696eda4 Make test-only package internal. 2014-03-17 00:05:26 -07:00
Nick Martin
82fd3201a1 Fix CSS error in Firefox. default is not a good thing to put here. sans-serif makes it look better. 2014-03-16 22:46:02 -07:00
Nick Martin
0bb6b0bb4c Fix hot code reload in Private Browsing mode in Safari. Doesn't actually save the migration data, just log an exception. But better than never reloading. 2014-03-16 21:18:29 -07:00
Emily Stark
70689568f6 Make final logout unconditional for logs-mongo tests release/0.7.2-rc0 2014-03-14 18:50:48 -07:00
Emily Stark
ff8abb4b04 Call logout in selftests that log in.
This avoids wracking up huge numbers of login tokens for the test
user. It would be nice to have these automatically cleaned up, but this
will do for now.
2014-03-14 18:50:48 -07:00
Emily Stark
85c7fa93e6 Add missing selftest 'var' 2014-03-14 18:50:48 -07:00
Emily Stark
4e19bf5e1d add semicolon 2014-03-14 18:50:48 -07:00
Nick Martin
ac9d47040e Credit contributors. 2014-03-14 18:00:48 -07:00
David Glasser
21e7c961b4 Use polling driver for {sort: {$natural: 1}} 2014-03-14 17:23:06 -07:00
David Glasser
76f03776b6 disallow {sort: {$natural: 1}} in minimongo 2014-03-14 17:08:44 -07:00
David Glasser
b161d3c0ec Merge branch 'limit-sort-overlap' into devel 2014-03-14 17:04:43 -07:00
David Glasser
c9c75fee84 oplog-limit: don't ignore modifiers to sort key
Previously we would ignore modifier updates to documents outside our
current buffer if they did not affect the selector but did affect the
sort key.
2014-03-14 17:04:31 -07:00
David Glasser
9e6c07c57f failing (with oplog) test for updating a sort key 2014-03-14 17:04:31 -07:00
David Glasser
22f2c6e4df expand an oplog/limit test a little
- put more info in state comments
- add a few more assertions
- stop requiring "usesOplog &&" before some helpers
2014-03-14 17:04:31 -07:00
Nick Martin
0a5800eb67 History pass 2014-03-14 16:29:24 -07:00
David Glasser
0d4b220111 Better error if non-doc passed to documentMatches 2014-03-14 12:19:44 -07:00
David Glasser
b855f2d7c5 Revert "Update docs around storing EJSON in MongoDB"
This reverts commit f91d870778.

It turns out that while we did implement "storing EJSON in MongoDB", the
technique we chose is very flaky.  Some MongoDB commands (insert,
updates that don't increase the doc size?) let you store fields with
non-starting $, and others (update that does increase the doc size?)
don't.
2014-03-13 20:41:47 -07:00
David Glasser
f91d870778 Update docs around storing EJSON in MongoDB 2014-03-13 19:46:01 -07:00
David Glasser
728ce8d251 Document one-Collection-per-collection rule 2014-03-13 17:38:20 -07:00
Dan Dascalescu
adf3f6553f Clearer docs on avoiding initial changed() calls
Explain `initializing` specifically, and move the comments closer to its
definition.
2014-03-13 17:38:20 -07:00
Dan Dascalescu
22854bee9b Some docs tweaks
- get rid of the sole use of "attribute set"
- add some links
- add missing "the"
2014-03-13 17:38:20 -07:00
Tim Haines
fcf08c966a Update Google oauth scopes. The previous ones were deprecated. 2014-03-13 17:27:36 -07:00
David Glasser
a46f5591a6 Better error for new Meteor.Collection({})
Fixes #1879.
2014-03-13 17:22:53 -07:00
Avital Oliver
1d3a840372 Update Spacebars README
{{> content}} and {{> elseContent}} are now
{{> UI.contentBlock}} and {{> UI.elseBlock}}
respectively.
release/blaze-rc1-1 release/blaze-rc1-0 release/blaze-rc1
2014-03-13 14:45:29 -07:00
Avital Oliver
b2bb353a10 Implement UI.getElementData
Replaces Spark.getDataContext
2014-03-13 14:06:24 -07:00
Emily Stark
dd4a4f0734 remove trailing comma 2014-03-13 11:04:27 -07:00
Emily Stark
d737c06fc6 no need for loginHandlers to be global 2014-03-13 11:03:13 -07:00