1069 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
Ruben Bridgewater	4d1390318c	Merge pull request #853 from fintura/eval ... Remove bad .eval implementation	2015-09-18 00:17:35 +02:00
Ruben Bridgewater	2578aba079	Merge pull request #848 from fintura/new_auth ... Fix auth callback being called more than once	2015-09-18 00:10:42 +02:00
Ruben Bridgewater	d61d97e24e	Fix .auth not working properly ... The arguments parameter was faulty andthe callback could have been triggered twice	2015-09-18 00:05:10 +02:00
Ruben Bridgewater	8b7d4a8449	Remove bad .eval implementation ... The implementation is not really good as mentioned in #722 and we pipline our commands. That way we can't just replace the eval function as it was. This could result in violating the order of execution! If we want to include a function like this we should not break the order of execution and also not recalculate the sha1 hash each time.	2015-09-17 22:51:42 +02:00
Ruben Bridgewater	21d8bdbbcb	Refactor multi to have a consistent error handling ... Ignore *.log files	2015-09-17 20:30:11 +02:00
Ruben Bridgewater	13635c9c8c	Merge pull request #840 from fintura/refactor-exec ... This fixes two TODOs in combination with exec	2015-09-16 06:59:08 +02:00
Ruben Bridgewater	9e2c665d5c	Refactor exec	2015-09-16 06:34:36 +02:00
Ruben Bridgewater	95a2c373a2	Fix broken build. The merge had a small conflict	2015-09-16 05:58:29 +02:00
Ruben Bridgewater	f92f4d0df9	Merge pull request #850 from fintura/fix-late-commands ... Fix commands not being rejected after calling .quit Fixes #791	2015-09-16 05:43:55 +02:00
Ruben Bridgewater	bfcc0ae263	Merge pull request #843 from fintura/reduce-timeout ... Reduce the reauthenticate timeout to 333ms	2015-09-16 05:43:07 +02:00
Ruben Bridgewater	57b5c7f76f	Merge pull request #829 from fintura/broken-mode ... Implement redis connection broken mode and more shiny things Fixes #569 Fixes #587 Fixes #566 Fixes #586 Fixes #280 This includes the fixes as suggested in #671, #615 and #533. Thx a lot to @qdb, @tobek and @chrishamant Closes #675, #463, #362, #438 and #724	2015-09-16 05:35:50 +02:00
Ruben Bridgewater	a0c906256c	Fix commands not being rejected after calling .quit as reported in #791	2015-09-16 05:29:34 +02:00
Ruben Bridgewater	89c8dd056b	Do not emit the broken mode twice if exec is called ... Add more tests	2015-09-16 05:22:43 +02:00
Ruben Bridgewater	30ec1cd6a2	shift in the while loop	2015-09-16 05:22:42 +02:00
Ruben Bridgewater	55d0036eae	Add test and fix keeping the offline queue ... Use a new delay after reconnecting	2015-09-16 05:22:07 +02:00
Ruben Bridgewater	3c2ba8c373	Try exactly until the connection timeout has been reached ... Fixes #587	2015-09-16 05:22:07 +02:00
Chris Hamant	a9e7663aff	removing flush_and_error from on_error handler	2015-09-16 05:22:07 +02:00
Ruben Bridgewater	03e8c03503	Add connection timeout and max attempts tests	2015-09-16 05:22:07 +02:00
Ruben Bridgewater	0b8705abe9	Do not run all tests with every single connection (if one connection works, the others are going to be fine too)	2015-09-16 05:22:07 +02:00
Ruben Bridgewater	1e0421ac3b	Emit errors if the connection timeout / maximum retry attempts have been exceeded ... Accept setting max_attempts to zero. The reconnection event is now emitted when trying to reconnect instead of earlier. The connection timeout is now going to trigger once after exceeding the maximum timeout instead of stopping earlier.	2015-09-16 05:22:07 +02:00
Ruben Bridgewater	f2ee8dbc9e	Add a connection timeout of 24h as new default for maximum reconnecting	2015-09-16 05:22:07 +02:00
Ruben Bridgewater	04c986a4cd	Lowering the first retry_delay and begin from that value ... Earlier the first value was never used, as it was first multiplied by 1.7	2015-09-16 05:22:07 +02:00
Ruben Bridgewater	cd5cfb4a8e	Emit an error when connection permanently goes down ... Closes #724 and #615	2015-09-16 05:22:07 +02:00
Ruben Bridgewater	403bfb0200	Merge pull request #835 from fintura/remove-stuff ... Remove send_command safety checks. Fixes #629	2015-09-16 05:21:18 +02:00
Ruben Bridgewater	eea9d2af76	Merge pull request #847 from fintura/check-password ... Check that the password is from type string	2015-09-16 05:20:43 +02:00
Ruben Bridgewater	30f227b7b1	Merge pull request #846 from fintura/parser ... Improve js parser	2015-09-16 05:19:50 +02:00
Ruben Bridgewater	aab1fc8f4f	Merge pull request #838 from fintura/consistent-commands ... Consistent commands arguments. All commands are from now on behaving the same no matter if they are on multi or no and they all take an array as either the first or second argument. Fixes #686 #369 #422 #390 and Closes #634	2015-09-16 05:17:22 +02:00
Ruben Bridgewater	005e869d83	Remove send_command safety checks ... This checks are only important for users who use send_command directly instead of using the convience method. As the readme clearly stats how send_command should work and any user would have run into errors if misused, these checks can be removed. If any user might misuse the function anyway, it is very likely that another error will be thrown because of that Fix #629 and insert tests	2015-09-16 05:10:39 +02:00
Ruben Bridgewater	013831c00f	Improve js parser	2015-09-15 23:18:26 +02:00
Ruben Bridgewater	c269b7539c	Check that the password is from type string	2015-09-15 22:35:25 +02:00
Ruben Bridgewater	a0c92b0756	Add a lot more tests to indicate that the everything is consistent ... Add a test for #390 More special cases	2015-09-15 04:14:40 +02:00
Ruben Bridgewater	c522ca1264	Fix inconsistent command argument handling ... Earlier multi.command and client.command diverged a lot in the way they accepted arguments. This is now consistent This will also fix some bugs like using multi.hmset with arrays	2015-09-15 04:00:02 +02:00
Ruben Bridgewater	e24f056b2d	Merge pull request #841 from fintura/command ... Add .command_used to errors returned by the parser. Fix authentication failure being emitted instead of returned by a callback if present. Fixes #427	2015-09-15 03:57:00 +02:00
Ruben Bridgewater	ebbb0146b9	Fix auth emitting the error even though a callback is present ... Fix auth manipulating the returned error And this is also removing some dead code	2015-09-15 03:50:47 +02:00
Ruben Bridgewater	1a06cfb6ec	Add .command_used to errors thrown by the parser	2015-09-15 03:50:46 +02:00
Ruben Bridgewater	e4bd9bf0c5	Merge pull request #839 from fintura/statements ... Small stuff like: let v8 inline instead of doing in manually	2015-09-15 03:44:47 +02:00
Ruben Bridgewater	6ea70271d6	v8 is going to inline a couple of things that were necessary a few years ago	2015-09-15 03:39:33 +02:00
Ruben Bridgewater	4bdcf8a598	Fix regression: version detection in tests did not work properly anymore	2015-09-14 23:53:52 +02:00
Ruben Bridgewater	785af29385	Merge pull request #842 from fintura/speedup ... Small speedup: do not call data.toString() on debug messages	2015-09-14 23:13:19 +02:00
Ruben Bridgewater	4a8da6dce5	Merge pull request #823 from fintura/debug ... Refactor debug mode not to break old code and to work with NODE_DEBUG too	2015-09-14 23:12:50 +02:00
Ruben Bridgewater	4f79370887	Merge pull request #816 from fintura/callback ... Remove try callback and emit errors if no callback is present. Fixes #456, #591, #522 and #755	2015-09-14 23:11:26 +02:00
Ruben Bridgewater	4f0443cdd4	Emit errors instead of throwing them ... Thrown errors might kill the users app. By emitting the errors the user is able to catch all errors in one place without the app going down	2015-09-14 21:28:40 +02:00
Ruben Bridgewater	c6ae7832a3	Remove try callbacks and emit an error in case of no callback has been provided	2015-09-14 21:28:40 +02:00
Ruben Bridgewater	1b261fcaad	Revert "added test for multi.hmset's array handling" ... This reverts commit bb221adec4. The test is broken and only passes because there's a bug with multi hmset using arrays. See #686 and #838 for more details	2015-09-14 21:23:47 +02:00
Benjamin E. Coe	bf2ccd9580	Merge pull request #837 from NodeRedis/windows-tests ... retrofit tests so that they work on Windows	2015-09-14 08:43:12 -07:00
Ruben Bridgewater	656706a263	Reduce the reauthenticate timeout to 333ms ... It should not take all that long to load everything into redis	2015-09-14 15:46:17 +02:00
Benjamin Coe	bb221adec4	added test for multi.hmset's array handling	2015-09-14 00:09:38 -07:00
Benjamin Coe	be708906fa	add test for providing auth after client is created	2015-09-13 22:21:03 -07:00
Benjamin Coe	68936c5eb2	based on code-review added back check for error	2015-09-13 19:17:52 -07:00
Ruben Bridgewater	e6aa3826f7	Small speedup: do not call data.toString() on debug messages	2015-09-13 23:51:36 +02:00