Timeline for How do I really write tests without mocking/stubbing?
Current License: CC BY-SA 4.0
24 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 18, 2023 at 13:57 | review | Close votes | |||
| Jun 23, 2023 at 3:02 | |||||
| Jun 18, 2023 at 13:34 | history | protected | gnat | ||
| Jun 18, 2023 at 12:16 | answer | added | Kimball Robinson | timeline score: 0 | |
| Jul 31, 2020 at 1:52 | vote | accept | kibe | ||
| Jul 7, 2020 at 5:09 | history | edited | kibe | CC BY-SA 4.0 |
Grammar
|
| Jul 2, 2020 at 21:47 | comment | added | Adrian McCarthy | Possible duplicate of softwareengineering.stackexchange.com/questions/382087/… | |
| Jul 2, 2020 at 20:15 | comment | added | Jan Hudec | Integration tests, that use carefully prepared environment and no mocks, are the most important kind as those are the tests that actually validate your application. Unit tests have limited value for validation and are mainly a debugging tool, so don't worry skipping them when they become more work then they save by catching bugs early. | |
| Jul 2, 2020 at 13:49 | answer | added | Warbo | timeline score: 0 | |
| Jul 1, 2020 at 20:29 | comment | added | Davor Ždralo | @BernhardBarker - been there, done that, it was horrible. Mocks go obsolete as you change the classes that you're mocking, or you end up updating a bunch of unrelated tests every time you change something. Not very fun or refactoring friendly. | |
| Jul 1, 2020 at 13:34 | answer | added | Jared Smith | timeline score: 5 | |
| Jul 1, 2020 at 5:47 | comment | added | Bernhard Barker | Why does mocking make you afraid of refactoring? If you have good test coverage, you shouldn't have to worry about breaking things (in fact, I might argue mocks give you more robustness rather than less, as you can more easily cover more permutations of how different functions work). If you write modular code and classes with reasonable interfaces, you shouldn't have to change those interfaces much and the mocks should stay relatively constant, so refactoring shouldn't be much more work, if any. But yeah, it is probably more work to write tests initially. | |
| Jun 30, 2020 at 20:56 | comment | added | njzk2 | in a nutshell: by applying interface segregation principle, and providing implementations that behave the way your test require. (Caveat: not always possible, esp if you don't own the interfaces) | |
| Jun 30, 2020 at 18:00 | history | tweeted | twitter.com/StackSoftEng/status/1278025542986579969 | ||
| Jun 30, 2020 at 17:08 | comment | added | IMSoP | I am not convinced drawing a sharp line between "unit tests" and "integration tests" is helpful; I think it is much more helpful to a) think of a spectrum of "unit size"; and b) cover a variety of points on that scale. This is all somewhat orthogonal to how you handle collaborators which fall outside the system under test. | |
| Jun 30, 2020 at 15:49 | history | became hot network question | |||
| Jun 30, 2020 at 12:43 | answer | added | VoiceOfUnreason | timeline score: 24 | |
| Jun 30, 2020 at 10:48 | answer | added | Flater | timeline score: 65 | |
| Jun 30, 2020 at 7:03 | comment | added | kibe | @wasatz awesome, thank you! will look into it | |
| Jun 30, 2020 at 6:23 | answer | added | Euphoric | timeline score: 17 | |
| Jun 30, 2020 at 6:09 | answer | added | Kain0_0 | timeline score: 0 | |
| Jun 30, 2020 at 6:08 | comment | added | wasatz | I know posting links is frowned upon here, but this is too large for me to summarize here and I do think this is a very good, long, detailed blog post on the subject (with example code repo on github linked in the article): jamesshore.com/Blog/Testing-Without-Mocks.html | |
| Jun 30, 2020 at 5:54 | answer | added | casablanca | timeline score: 10 | |
| Jun 30, 2020 at 5:18 | review | Close votes | |||
| Jul 5, 2020 at 3:01 | |||||
| Jun 30, 2020 at 4:21 | history | asked | kibe | CC BY-SA 4.0 |