Recent Posts

Pages: [1] 2 3 ... 10
1
SlickEdit User Macros / F1 API Help for cppreference
« Last post by Dennis on July 16, 2024, 07:12:41 PM »
Kind of rough, but works for major cases.

Tools > Options > Import/Export Options > Import...
2
I frequently spin up a new instance of SlickEdit for searching, slicing, dicing, and trying to understand various log files.  Since I don't want these directories polluting my workspace history, I usually start "vs" with the "-sc" option and point it to a default configuration directory, like "vs -sc /tmp/defcon1"

It would be very useful to have a shortcut "-sc" option for named configuration directory, such as "vs -sname logconfig", which would place the configuration in a directory such as "$HOME/.slickedit/logconfig/28.0.2/..."

I can create a shell alias to work around this, but I think this would be generally useful.
3
General Programming / Re: Windows handle leaks in SlickEdit versions after v25
« Last post by Graeme on July 12, 2024, 04:56:16 AM »
Can you try closing your files one at a time and see if the handle leak stops.  If the problem is caused by one specific file or file type, you could maybe send it to slickedit support.  It might be a Windows 7 problem, since no-one else seems to be getting this.
4
How did the consumer react when it happened?  If the client turned off the power, it would be a catastrophic hardware failure, and (nearly) all bets are off.  NTFS journaling should preserve the directory structure, as well as the file allocation tables.  However, pending file content may not have been committed (even if it was committed by the OS and the hard drive returned success, the hard drive firmware may have deferred the actual write for performance reasons, causing it to be lost -- yes, hard drives do this these days to increase throughput numbers).

It is hard to tell what happens to the operating system once the kernel handle space gets exhausted. The file system may or may not lose data. However, this is irrelevant to this bug. The bug is caused by something in the SlickEdit code not doing the proper FileClose() on all open handles. As a result of that, after some time, not only the editor itself becomes unusable but the entire operating system as well. Even if the file system does not lose any data (which is not the case), it is still a very serious bug.
5
General Programming / Re: Windows handle leaks in SlickEdit versions after v25
« Last post by repayzippy on July 12, 2024, 02:49:45 AM »
It appears that SlickEdit v25 is the last version that does not leak handles
on Windows. I have tried v26 and v27 - both of them leak handles
at the rate of hundreds per second even if the editor is completely idle.
After several hours of being open, the editor eventually eats ups all the kernel
space making it impossible for other processes to open any new handles.
At this point the entire system becomes unusable and dies.

The attached screenshot shows the handle count for v25 with about 30 files
open (a mix of Python, Kotlin and TypeScript code). The Task Manager shows
the handle count of a few hundred. More importantly, it does not change at all
when the editor is idle. If I open exactly the same project with exactly
the same set of files in v26 or v27, the Task Manager shows the handle counter
growing rapidly at a constant rate. The rate is very high - hundreds new handles
are open every second.

This indicates that there is a handle leak somewhere in the editor. Most likely,
some background thread is constantly opening new files but does not close them,
leaking the kernel memory.

Just in case, this is on Windows 7 32-bit, although I do not think it matters.
How did the consumer react when it happened?  If the client turned off the power, it would be a catastrophic hardware failure, and (nearly) all bets are off.  NTFS journaling should preserve the directory structure, as well as the file allocation tables.  However, pending file content may not have been committed (even if it was committed by the OS and the hard drive returned success, the hard drive firmware may have deferred the actual write for performance reasons, causing it to be lost -- yes, hard drives do this these days to increase throughput numbers).
6
SlickEdit® / Re: Slick-C Stack in Version Control > History Diff
« Last post by Clark on July 11, 2024, 08:53:31 PM »
A hot fix is available for this.

Download the hot fix here:

https://www.slickedit.com/hotfixes

Then load the hot fix: Help>Product Updates>Load Hot Fix...
7
SlickEdit® / Re: Slick-C Stack in Version Control > History Diff
« Last post by Clark on July 11, 2024, 08:09:39 PM »
Reproduced. We will look into this
8
SlickEdit® / Slick-C Stack in Version Control > History Diff
« Last post by Dennis on July 11, 2024, 06:13:35 PM »
To reproduce the issue:

1) Open a file that is controlled under Git
2) Version Control > History Diff for ...
3) Right click in the editor control


Code: [Select]
Stack trace written to file: /tmp/vsstack.XXXXXX
 Control -ctlfile1-readonly referenced but does not exist
diffedit.ex 4632 diffedit:edit_window_rbutton_up()  p_window_id: 337  p_object: OI_EDITOR  p_name: _ctlfile1  form_name: _history_diff_form
diffedit.ex 11218 diffedit:_EditControlEventHandler()  p_window_id: 337  p_object: OI_EDITOR  p_name: _ctlfile1  form_name: _history_diff_form
diffedit.ex 12808 _diff_form._ctlfile1.range_first_nonchar_key-all_range_last_nonchar_key()  p_window_id: 337  p_object: OI_EDITOR  p_name: _ctlfile1  form_name: _history_diff_form
dlgeditv.ex 7514 show(-modal -xy  _history_diff_form,@24462,eeddced489df362b22fafcbfa1b31aa7a96e1403,0)  p_window_id: 459  p_object: OI_MINIHTML  p_name: ctlminihtml1  form_name: _svc_query_output_form
svc.ex 17573 svc_history_diff(/home/XXXXXXXX.local/XXXXXX/.../XXXXXXX.cpp,,eeddced489df362b22fafcbfa1b31aa7a96e1403,1)
  p_window_id: 530
  p_object: OI_EDITOR
  p_name:
svcquery.ex 3603 _svc_query_output_form.ctlminihtml1.on_change(32,histdiff:eeddced489df362b22fafcbfa1b31aa7a96e1403)  p_window_id: 530  p_object: OI_EDITOR  p_name:
controls.ex 1444 _ul2_minihtm.lbutton_down()  p_window_id: 530  p_object: OI_EDITOR  p_name:

There is no "_ctlfile1_readonly" control is on the history diff form.  The right click handler should check if it's there before using it.
9
SlickEdit® / Re: Is SlickEdit out of business?
« Last post by Clark on July 11, 2024, 01:34:38 PM »
Looks like your email went to spam. Support will answer your email.
10
SlickEdit® / Is SlickEdit out of business?
« Last post by SJA on July 11, 2024, 04:15:04 AM »
I didn't get a reminder to renew my support contract in May.  They don't answer their phone or return messages.  They don't respond to emails.
Pages: [1] 2 3 ... 10