Just to mention, for people using Intune to deploy notepad++. if you activate the “managed installers” option, any apps not signed wil lbe signed with a valid certificate and therefore users don’t have popup for the “unknown publisher”. I think you can do the same with SCCM. Will not help regular users, but jsut in case.

@scampsd

Although the registry is actually the Windows standard, I would personally avoid it and rather use the plugin config directory. Can be determined via NPPM_GETPLUGINSCONFIGDIR.
However, I would create a subdirectory with the plugin name and add a json, xml, toml … file there.

I figure I’d better document the solution(s) for others that may tread this path. A summary of some of the steps needed (of course built on top of Mark’s work!):

Ensure you register your modeless dialog with NPP (NPPM_MODELESSDIALOG, in Main.cs we call NppFormHelper.RegisterFormIfModeless(Form, bool)) The WPF System.Windows.Controls.UserControl should be in an System.Windows.Forms.Integration.ElementHost, which is in a System.Windows.Forms.Form. I’ve subclassed the ElementHost to handle WM_SHOWWINDOW to update the Visible property, and to trigger a (slightly delayed) refresh of the visuals when we get a WM_SIZE. There I invalidate the visuals and update layouts. There may be a way to skip this, but it’s working, so I’m not trying to fix it. In the WPF UserControl, when it’s Loaded, we have to get the presentation source (as a HwndSource) and add a hook to respond to the WM_GETDLGCODE and tell the source that we DLGC_WANTCHARS and DLGC_WANTARROWS, and maybe we also need DLGC_HASSETSEL.

I’ve pulled most of it into a method called MakeModelessDialog. And now for a short recording of it working!

2025-06-28 Typing in Notepad++ modeless WPF dialog works.gif

@rdipardo said in 8.8.2 - coming soon??:

edition in a user-local path

I already checked such possibility (for N++ installer, not for portable eds) and have the non-admin user-installations on my TODO, but so low spare time… (#15182)

@rdipardo said in 8.8.2 - coming soon??:

can still have context menu entries — scoped to HKEY_CURRENT_USER by default

Very good. I think the NppShell project should take this into account somehow and allow the HKCR registrations when needed/requested. This HKLM default is(?) IMO the reason for a possible regsvr32 zombie process when trying to register the NppShell manually & unelevated (#15182)

@Coises said in 8.8.2 - coming soon??:

I don’t think I’ve ever used a package manager.

But OTOH you also said that for some SW (Firefox) you allow (welcome(?)) the autoupdates. Such package managers do just that - they remove the burden from the user to constantly check and update their software (even for a portable non-admin user-installation), but at the same time they leave them free to enter this process (if they wanna) and therefore it is not completely hopeless like with the locked stores like WindowsStore or GooglePlay. One example - with the WinGet you can go simply back in the app-versions if you need to (as you do with your “Software\Old”), this is of course impossible with e.g. the GooglePlay (damn hell to them for it!).

SW-updating “world” just changed and not everything is so bad. So e.g. the WinGet (or more user friendly WinGetUI) way is ok for me (I am not using it for N++, I do it old-school as you do, I admit :-) ). Some relevant links:

https://community.notepad-plus-plus.org/topic/24884/winget-real-autoupdater-finally
https://github.com/microsoft/winget-pkgs/tree/master/manifests/n/Notepad%2B%2B/Notepad%2B%2B

Here is a funny scenario:

type “menu” in the Search textbox; select Plugins -> NppMenuSearch -> Menu Search… press Enter.
As the result, the Search textbox does not have a focus!

Well, I know this is a strange use-case: call itself, but anyway :)

Hello, @alan-kilborn, and All,

Alan, you’ve just understood all my stuff quite correctly and even more regarding your last example with $+{Test} and $+{foo}, whose I did not think of !

Best Regards,

guy038

@Coises said in Config Files Need Updating, Too:

For what it’s worth, as the author of Controlled Auto-indent, I can guarantee it was the culprit. It could crash under other circumstances, too, having nothing to do with ConfigUpdater; I just hadn’t hit one of them, and if anyone else did before you, they didn’t report it.

I appreciate that, and I’ll go ahead and download it and test it against the old and new versions of ConfigUpdater. I’ve already downloaded, but not installed the new version of CU yet.

Update: It works fine under the old CU and the new CU. I tried it with CAI enabled and not enabled, and the CU ran through everything, so the crash condition from CAI appears to be gone. Glad I could help find that edge case, unknowingly. :-)

Replace was where I was having the issue and it brings up some challenges that Find does not - basically, you would need a shortcut for “Ignore and Find Next” and one for “Replace and Find Next” and you’d have to remember those shortcuts.

@Coises; TBH, now that I’ve learned to resize the screen. I’m okay with it. I think most users have multiple monitors these days and use the find/replace dialog on the second monitor.

@Coises and @mkupper - I wasn’t aware of the caret button to move the dialog to two lines. I think that and an OPTION to dock the find dialog in the lower-right corner so I don’t have to re-size the screen would be perfect for what I want.

Again - none of this addresses the context issue - which isn’t a primary concern of mine, but I still think it should be accommodated, if possible.

@Alexander-Anisimov said in Notepad++ lags when text files are on FTP:

You may be right. In my understanding, WebDAV is another way of connecting protocol but still is FTP, although different.
May be I am wrong.

You are wrong. :-)

WebDAV is a protocol that is similar to HTTP that web browsers use with web servers. As WebDAV looks like HTTP on the network it works with NAT, firewalls, web proxies, etc. Accessing a WebDAV server is just like accessing a web server. Many web browsers have built-in support for WebDAV.

That said, if there are bandwidth or latency issues accessing a particular WebDAV server then all applications, including Notepad++ will be impacted.

WebDAV itself is not slow or laggy.

If I was stuck with needing to edit files that are on a slow or laggy server then I probably would create script that first copies the file to the local hard drive, runs Notepad++ on the file, and then keeps an eye on both the local and remote copies of the file. If the local copy changes then I push it out to the remote server. If it changes on the remote side then pop-up a warning. A tricky aspect is that Notepad++ allows people to exit Notepad++ and then resume editing. Thus I can’t depend on Notepad++.exe going away as a signal that I no longer need to monitor and synchronize that file but also need to watch Notepad++'s session.xml file.

For those who weren’t watching:

ozone10 made a better fix for the tab-control: PR #16673 has been accepted ozone10 also fixed my “secondarily” report about the listbox scrollbar from above, which I reported in issue #16674, fixed in PR #16676 (accepted) progressbars also don’t go to darkmode, and ozone10 has proposed PR#16687, which hasn’t been accepted yet, but hopefully will be.

So at least 2 of those 3 (and hopefully the third) will be in v8.8.2

Meanwhile, CollectionInterface v1.1 has been released – the new version will be in PluginsAdmin for v8.8.2, or you can install it from the assets in the release at https://github.com/pryrt/NppPlugin-CollectionInterface/releases/tag/v1.1.0 :

Adds DarkMode support (with N++ v8.5.4 or newer; in earlier versions, dialog will be LightMode still) [I do my own subclassing on the TabControl for v8.5.4-v8.8.1; it will use the builtin starting in v8.8.2] Allow selecting multiple files for bulk download (#6) Add label to progress bar so “100% [DONE]” is more clear (#9) Fix crash if firewall blocks download (#13) and make error handling better in other edge cases as well

@guy038 said:

the site’s ability to highlight any sub-section of the regex in green is really awesome

https://jmrware.com/articles/2010/dynregexhl/DynamicRegexHighlighter.html

Hello, @ottiditto, @pbarney, @peterjones and All,

A tiny piece of information : The & character is not a regex character, so the following syntax is sufficient :

FIND (?-s)^((.*? layer=")(25)(".*?>>NAME.*\R))

REPLACE ${1}${2}21${4}

Best Regards,

guy038

@Kunal-Mehta

The problem with all these different “cloud” drives is that they don’t use a standard protocol, which makes it difficult to support multiple of them.

@PeterJones said in in correct version reporting:

some insight as to why the Windows program/features install version isn’t getting updated for you.

@eli-rosen said in in correct version reporting:

user for almost 20 years,

The following will be only my guess.

Since @eli-rosen uses the N++ installers for so long, maybe he previously used there also a v7.9.1 x64 one, but then he has started to use that Automox (sorry, IDK it) and it can do whatever stuff, so maybe it’s simply forgetting/blocking to update the corresponding Registry item.

It shouldn’t be a remnant from a previously installed 32-bit v7.9.1. In such a case there should be a separate uninstall-item visible (with a corresponding Registry hive HKLM\SOFTWARE\WOW6432Node\Microsoft\Windows\CurrentVersion\Uninstall\Notepad++\)

@Alan-Kilborn said in Some functions not shown in function list:

There’s also a sample macro called “Trim Trailing Whitespace and Save” or something close to that, that conveniently does the action every time you save (well, every time you run the macro, which, if assigned to your favorite save-keycombo, e.g. Ctrl+s, will do it).

I used Ctrl+S remapped to Macros > Trim Trailing and Save for years before starting to use editorconfig – and actually I still use that mapping, because I don’t have a .editorconfig file at the root directory of all my drives, so there are still times when the editorconfig plugin doesn’t know to trim trailing for me. Since I almost exclusively use the keystroke vs the toolbar or menu command for my Save, I (almost) always get the trailing removed when I save.

@Alexander-Anisimov said in Copying file itself into clipboard in Notepad++:

how can I bind a hotkey something like Ctrl+Shift+C to it?

That’s explained in the FAQ. Search that page for “shortcut” (step 4 of the instructions) if you didn’t notice it.

@mkupper said in Question(s):

I verified that at least for my current setup that the zoom level seems to have no affect on the printed results.

Same here. Perhaps I was unclear.

If I leave the zoom at default and adjust the font size for Default Style to be large enough for me to read easily on my monitor, I would set a font size of 14 or 16 points with Liberation Mono. (14 is bit smaller than I like, 16 is too big, 15 isn’t an option.) Those are too large for printing, though.

If I set Default Style to 12 points, printing is reasonable. If I then zoom in 3 steps, I’m just about where I want to be to read easily on the screen — without affecting printing.

The question I was answering was, “Why not just change your default font size?” Because that does affect printing (and, though I hadn’t remembered until I tried just now, also isn’t as fine-grained as zoom).

@Vitalii-Dovgan said in Announcement: NPPM_GETOPENFILENAMES and related are being deprecated in v8.8.2:

My last question was rather related to NPPM_GETFULLCURRENTPATH, NPPM_GETCURRENTDIRECTORY and the other messages from the same family. These messages:

are internally limited by MAX_PATH; do not provide ability to request/retrieve the length required for the returned path/name.

See the first part of this comment and the first part of the reply by the maintainer of Notepad++.

The MAX_PATH limitation is present in the file handling routines of Notepad++, not just the interface for these messages. As I read that reply, Don has rejected (for now, at least) consideration that the MAX_PATH constraint might someday be removed. If you know MAX_PATH is enough, there’s no need to request the actual length ahead of time.

I agree that there is some inconsistency in what is considered appropriate for deprecation, but I return to the observation that it will always be possible to screw up calling a C interface, and there were and are compelling reasons to keep the plugin interface as a pure C interface. (If we could pass a std::wstring or a std::vector<std::wstring> the whole problem would go away.) It’s a judgement call what is “too error-prone.” Not worth all that much debate.

@Alan-Kilborn Hehe, “for in much wisdom is much grief”.