Timeline for What are the pitfalls or challenges of converting clickstream/user action logs into test scripts?
Current License: CC BY-SA 3.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 7, 2017 at 17:41 | history | edited | Peter M. - stands for Monica | CC BY-SA 3.0 |
added 225 characters in body
|
| Nov 7, 2017 at 15:19 | vote | accept | c32hedge | ||
| Nov 2, 2017 at 20:35 | comment | added | c32hedge | Thanks! The last paragraph in the update is particularly helpful. | |
| Nov 2, 2017 at 20:28 | history | edited | Peter M. - stands for Monica | CC BY-SA 3.0 |
added 788 characters in body
|
| Nov 2, 2017 at 20:23 | comment | added | Peter M. - stands for Monica | I added a blurb about our experience with using record/replay vs PageObject tests. | |
| Nov 2, 2017 at 20:22 | history | edited | Peter M. - stands for Monica | CC BY-SA 3.0 |
added 788 characters in body
|
| Nov 2, 2017 at 19:43 | comment | added | c32hedge | "But they might provide a good starting place for editing to make the test more flexible." I'm well aware of the limitations of record/playback but this was more of a thought experiment of what else we could do if we were already implementing something (the detailed logging) that gave us most of the data we needed for a record/playback system. | |
| Nov 2, 2017 at 19:32 | comment | added | Peter M. - stands for Monica | @c32hedge - define "easily". Whole point was not editing/maintaining the recorded script, IIUC? Also, date was the trivial example. I have scripts which select one of the random objects of different types for different users. | |
| Nov 2, 2017 at 18:57 | comment | added | c32hedge | In your particular example, you could easily edit the generated script to replace the hardcoded date with your language-of-choice's function for getting today's date. But yes, there's still extra work involved to do that. | |
| Nov 2, 2017 at 18:45 | history | answered | Peter M. - stands for Monica | CC BY-SA 3.0 |