Timeline for Follow a pipe using less?
Current License: CC BY-SA 3.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 21, 2017 at 11:44 | comment | added | mik | @PiotrDobrogost yes, it would block waiting for new data or EOF, but my point is that the screen does not need to be updated during this waiting (just because there is no new data) | |
| Jan 19, 2017 at 13:36 | comment | added | Piotr Dobrogost | @mik And here, we are interested in the no data case. Earlier you wrote *After pressing F or G on pipe input less not only does a blocking read, but does it in a loop waiting for EOF. * My point is that even if there was no waiting for EOF less would block on read after it had read all data from pipe. | |
| Jan 19, 2017 at 13:11 | comment | added | mik | @PiotrDobrogost with blocking read less wouldn't be able to update screen, if there is no data; when some data appear, blocking read will return it, and less would be able to update screen without a separate thread | |
| Jan 19, 2017 at 10:10 | comment | added | Piotr Dobrogost | @mik If less would update screen in that loop, there would not be a problem. Blocking read has little to do with this issue. If the read were blocking then less wouldn't be able to update screen unless the read was in a different thread, would it? | |
| Jun 27, 2016 at 9:21 | comment | added | mik | @Flow this is not a hang the question was about, but merely a wait which is expected in this situation - a wait for more data in the file when EOF is reached (which will not happen for a closed pipe, by the way), or for an interrupt to exit the follow mode | |
| Jun 23, 2016 at 12:03 | comment | added | Flow | If the input is a pipe, less hangs until the pipe is closed on the other side. That's not my observation. For example echo foo |less +F results in less displaying Waiting for data... (interrupt to abort), so it does not hang until the pipe is closed, but until an interrupt is received. | |
| Feb 12, 2015 at 22:46 | comment | added | mik | If less would update screen in that loop, there would not be a problem. Blocking read has little to do with this issue. | |
| Feb 12, 2015 at 22:29 | comment | added | mik | After pressing F or G on pipe input less not only does a blocking read, but does it in a loop waiting for EOF. And an EOF on a pipe happens only if its other side is closed. | |
| Dec 11, 2014 at 11:24 | comment | added | Piotr Dobrogost | If the input is a pipe, less hangs until the pipe is closed on the other side. That's kind of misleading statement. What happens is that less calls read in a blocking mode, waiting for new data or or pipe's closure. | |
| Oct 29, 2014 at 19:38 | vote | accept | haelix | ||
| Jun 27, 2014 at 8:37 | history | edited | mik | CC BY-SA 3.0 | minor grammatic improvements |
| Jun 26, 2014 at 9:06 | history | edited | jasonwryan | CC BY-SA 3.0 | Grammar and formatting |
| S Jun 26, 2014 at 8:21 | review | Late answers | |||
| Jun 26, 2014 at 9:06 | |||||
| S Jun 26, 2014 at 8:21 | review | First posts | |||
| Jun 26, 2014 at 8:27 | |||||
| Jun 26, 2014 at 8:07 | history | edited | mik | CC BY-SA 3.0 | added 6 characters in body |
| Jun 26, 2014 at 8:02 | history | answered | mik | CC BY-SA 3.0 |