Monitor (tail -f) prompts to reload file
-
I have been seeing this same issue. I tried the suggestion of downloading the zip and running in portable mode, but still seeing prompts to reload the file. Sometimes it does update the file even though it prompts, and sometimes I have to acknowledge the prompt to get it to reload. I did occasionally see this behaviour in the past but it was very rare. I noticed that so far I have only been able re-create this issue when monitoring files on the network. When I monitor a file on my local computer I have not yet seen the prompt (although I rarely do this).
-
this issue is verified as reproducible.
@SinghRajenM has committed a fix that solves all testing scenarios so far.
the fix is currently in the queue and might get merged, if no further issues or concerns arise.best regards.
-
I had the same problem with a 32-bit version(v.7). I installed this solved my problem 64 v.7.6.4
-
@melle-sterk I cant edit my prev submission. but I do still have the problem after installing both version. it only looked like it dindt for a second. So ignore my prev message
-
Still seems to be an issue with v7.7
-
welcome to the notepad++ community, @Paul-Renold
yes, you are correct.
a fix has been submitted to address this issue, but it has not been merged into 7.7 for issue strategic reasons, due to the major changes within this release.
many thanks and best regards.
-
I installed the V7.8 32-bit (built time Oct 7 2019 - 00:58:16) and continue not working.
Now, there is not prompt, but don’t update the changes. -
Improving my last answer.
The functionality “tail -f” seams working when data changes and stream is closed, but sometimes not works if the file stream remains opened.I tested it, redirecting java application output to one file (some_java_application > out.log).
The Notepad++ only updated at the begin (when file was cleaned) and at the end (when application ends, and file stream was closed).
If I disable the “tail -f”, the prompt comes again as expected, when focus in Notepad++ with files changes, even when stream remains opened (that means the Notepad++ have capability to detect changes on files with stream open, but not when “tail -f” enabled).
I’ll comment again if I reproduce this with a simple application, but bellow I reproduce partially with native ms-dos applications.After that, I tried to reproduce the problem only using windows base applications, and that works better, but not perfect.
Executing the bellow command on the console:
ping -t 0.0.0.0 > test_tail_notepad_plus_plus.txtThe Notepad++ with “tail -f” enablled only updates when:
- At the begin, when file was cleared;
- Receiving focus on Notepad++;
- Press F5 on Explorer with file folder opened (native file Explorer of windows);
There are missing updates on acceptable time for tail functionality, even Notepad++ don’t have focus (for example refresh rates of 1 second).
I’m running Notepad++ V7.8 32bit on windows 10 64bit.Thanks
Manuel Romeiro -
@Manuel-Romeiro it seems these all are windows limitation. It is explained here - https://github.com/notepad-plus-plus/notepad-plus-plus/pull/5591#issuecomment-487341747
Refreshing every second could be an option. But it will keep utilizing CPU even if there is no change as well because the monitoring thread has to do polling every second which is against the design of change notification.
-
For reference to all -
From MSDN
FILE_NOTIFY_CHANGE_SIZE
0x00000008Any file-size change in the watched directory or subtree causes a change notification wait operation to return. The operating system detects a change in file size only when the file is written to the disk. For operating systems that use extensive caching, detection occurs only when the cache is sufficiently flushed.
FILE_NOTIFY_CHANGE_LAST_WRITE
0x00000010Any change to the last write-time of files in the watched directory or subtree causes a change notification wait operation to return. The operating system detects a change to the last write-time only when the file is written to the disk. For operating systems that use extensive caching, detection occurs only when the cache is sufficiently flushed.
-
@SinghRajenM Yes, it’s exactly that issue.
I’ll continue using the functionality knowing that limitation.
Thanks