@Todd-Marshall said in **Title: Proposal to Add `Ctrl + ]` Shortcut for Matching Braces**:

I explored this issue with DuckDuckGo AI and asked it to summarize in the form of a submission here

LLM/AI/ChatGPT-generated content is not allowed here, as described in the Read here before posting post that you obviously didn’t read.

I let this post through without immediately banning you, because you were obviously desparately trying to keep it on-topic, which is more than most LLM-posters do. But no more AI posts, or you will be banned.

At the closing brace }, Ctrl + B moves to before the opening brace {, without highlighting it.

That is not what I see. If I have the care just before or just after the closing brace, both opening and closing braces are highlighted. And when I do Ctrl+B to go to the opening brace, the caret is just before the initial brace, and both braces are still highlighted.

So I don’t think there needs to be a feature request.

You shouldn’t believe everything AI tells you.

If AI was telling you that the Ctrl+B brace jumping doesn’t work right, it was hallucinating. If you were claiming in your AI prompt that brace-jumping doesn’t work right, you were mistaken. And the nonsense you plugged into the AI became the nonsense claim in this “feature request” If you believe I am mistaken, you will need to provide plenty of evidence, because it works right for me in any situation I can come up with.

I think you either mis-described the situation to the AI in your prompt, or you let the AI make up any problem that it could propose a solution to, even if that problem doesn’t really exist. Don’t use a tool that you don’t understand, and don’t expect to be able to input garbage in an LLM prompt or the underlying training and get anything but garbage out: an LLM trained on lies will spew lies, and an LLM given a prompt with bad information can repeat that bad information back out.

@PeterJones said in All of a sudden cannot type anything in Notepad++ 8.8.8:

I don’t know how that one plugin works

I just grabbed a copy on a portable N++ to experiment. So yes, when you first install NppVim, it starts Notepad++ in “Vim” mode. So, just in Vim, in order to add text, you need to use the a (append) or i (insert) or o (add new line) or the capital versions thereof. NppVim hijacks the status bar, and will list --NORMAL-- when it’s in the mode that accepts the hjkl navigation keys (and similar) or --INSERT-- when it’s in the insert/append mode. It’s only when you are in --INSERT-- mode that the Notepad++ INS/OVR INS-key toggle has any meaning, and it’s only when you are --INSERT-- mode that you can type text in the editor.

760ee28d-7f24-4943-93fb-9133407a104d-image.png
vs
4f3fa4eb-ab7b-469e-a49e-ad4508a5411d-image.png

and now, just a few seconds of using Vim-style editing commands, just to prove that’s how it works, my brain has already reverted to vi keys, and I tried to edit this post with vi keystrokes. so thanks for that 😜

update:
If you don’t want to have to use the Vim keys with that plugin, you can use Plugins > NppVim > Toggle Vim Mode to toggle off that plugin’s features temporarily.
2de62f18-5e03-45b2-a632-4a0eb28bf875-image.png

So, as of now, my working theory is that @Salam-Elias had the NppVim plugin installed without understanding the implications, and the problems with typing, mentioned in the original post, were 100% due to user error and that plugin.

@ioc2e3:

I’m fairly sure the only one of those you can edit in Notepad++ is json.

You’ll have to look up the rest yourself. In some cases there might be more than one way the same file extension is used, so you’ll have to figure out which program wrote the files. (For example, .aff can be a file used as part of a spell-checking dictionary or an Advanced Forensic Format file.)

Even for JSON, you usually shouldn’t just edit a JSON file in Notepad++ unless you know what you are doing. JSON files are highly structured and any given JSON file is normally maintained by a specific program or set of programs for specific purposes. Usually there is a correct way to change the data stored there through a user interface provided by the application, rather than diving in yourself.

It might seem like a pain in the butt, but you need to work out which programs created each of the files that concern you and research those programs to figure out how to examine and/or change these files. If Notepad++ can open them at all, it will probably give you nothing but the ability to mess them up.

@Kaptain-Biwouak,

1/ last updates about Perl language don’t allow anymore to detect perl subroutines with the menu function (It was working well with Notepad++ 8.8.1).

By “the menu function”, I am assuming you mean the View > Function List.

But when you say it doesn’t “detect perl subroutines”, you will have to be more specific, because Function List does work for Perl files in v8.8.8 – though it is known to have efficiency issues.

b60af6c5-90e6-4a07-b34d-a45c6524db4c-image.png

the version of the Perl FunctionList parser found here will hopefully be in the next version of Notepad++ (presumably v8.8.9), so if you put the new XML in %AppData%\Notepad++\functionList\perl.xml (you might have to create the functionList subdirectory) and restart Notepad++, it will use the more efficient parser.

After doing that, if it still doesn’t show your subroutine, please click the </> button in the reply, and paste a small Perl program that shows one or two subroutines that don’t show up in the FunctionList.

2/ on-going modifications usually shown with * on the tab bar are no more accurate; Several tabs have now this status even if one file is currently being modified.

I only see an * on the title bar if a tab that has unsaved edits is active. If I switch to a tab that has no edits, the * is not shown on the title bar. Can you give a minimal series of steps to reproduce the problem? because I cannot get it to be wrong about the *.

So, I’ve checked the session.xml to see how many files and in which encoding are.
https://i.imgur.com/H10X5Sh.png

What I’ve discovered here are inconsistent results.
While I expected all files to be in UTF-8 due to my settings, here I found:

Some (most) UTF-8 files with encoding “-1” in session.xml Some UTF-8 files with encoding “10007” in session.xml Some “Cyrillic -> Macintosh” files with encoding “10007” in session.xml Some “Cyrillic -> Windows 1251” files with encoding “1251” in session.xml

Somehow, N++ for itself has decided and selected alle these different encodings. What I need is to have all my files always in UTF-8, and now I’m thinking maybe to mass-convert all the file to UTF-8 somehow…

@b-squared said in Can't install any plugin - hash mismatch:

Hi,

I get the following error when trying to install a plugin from the Plugin Admin tool. I’ve tried several including Compare, JSON Viewer, 3P, HEX-Editor, JSON Tools.

Your Debug Info shows that you are using 64-bit Notepad++ version 8.8.8; Plugins Admin in that version should be offering Preview HTML 1.3.3.3. From your screenshot, it looks like it’s trying to load a 32-bit plugin with version 1.3.2.0.

I’m thinking that somehow your copy of the plugins list is not getting updated properly. Just to be sure… is this a regular installed version or a portable version, and exactly where are you getting the download?

Any way to install plugins manually, or will I just run into the same issue?

You can install plugins manually and there isn’t a hash check when you do that. Before we go that way, though, we need to figure out why your installation copy of the plugins list is messed up — I have to think it might not be the only thing that’s messed up.

My last message was wrong. Please excuse me.

@andwan0

Your issue has been fixed for the N++ v8.8.6+.

@Hyung-jun-Chang While not exactly what you asked for, if you press Ctrl+Alt+F it will fold the row where your marker is.
in your picture example, to fold number 3 you can press row 949 and then Ctrl+Alt+F. This should also work with rows 585, 586, 587, 867, 947, 948.
This should be a workaround instead of the “-” at the final row of the fold.

@Bruno-Guillemette said in NPP RSS Feed broken?:

I’ve looked into this and we’re being blocked by them sometimes. They’d have to whitelist our IP range, which is: 8.29.198.0/24 and that would fix it.

This Forum is a Community of Notepad++ users. If you want to contact the developer, you’d have to either create an issue in the main GitHub repo (which isn’t really where website errors belong, but sometimes people post them there) or email Don directly (his contact information is on the Notepad++ website). If you would like him to “whitelist” an IP range to be able to see the RSS feed, you’d have to contact him, not us.

My guess is there isn’t any setting on the Notepad++ website itself which is directly blocking Feedly. My guess is that Feedly isn’t passing the CloudFlare captcha, so CloudFlare (which inserts itself between user requests and the underlying website) is the one that’s blocking Feedly, not any setting on the Notepad++ website. (And honestly, the fact that the employee in the email you quoted is not knowledgable enough to have tried accessing the RSS feed directly, and seeing that a CloudFlare are-you-a-human prompt comes up, makes me wonder about the quality of their support.) However, I seem to remember that Don was able to make it so that the updater-XML URL was “unblocked” for CloudFlare, so maybe he’d be able to do the same for the RSS URL… but I don’t know.

@Bjorgen-Eatinger said in The Nightmare %> Issue:

This issue only occurs in a case like this:

Sadly, you did not give nearly enough for us to be able to investigate your problem any further. I put in exactly that line (and only that line) into bjorgen.asp on my local IIS, and when I go to that page, it gives me a server error:
5728353b-eb82-47ee-baa1-0082457cc706-image.png

… And I confirmed that other pages, like the one I showed above, are still working on my local server.

From what I understand, the ' as a comment is VBSCRIPT syntax, not ASP-specific syntax. Because I thought everything outside of <% ... %> was just interpreted as ASP (and my quick web searches seem to confirm that conclusion on my part). In which case, putting the ' before a <% ... %> line is not “commenting out” anything, it’s just putting a single-quote in the static HTML portion of the code. Moreover, you aren’t supposed to nest <% ... %> inside each other, as far as my searches tell me. So your single line is really confusing… because if it’s outside of a containing <% ... %>, then I don’t see why you think that the ' is commenting anything. And if it’s inside a containing <% ... %>, I don’t see why you have another <% ... %> nested inside. This is why a complete example is needed for us to be able to see what you mean. It doesn’t have to be huge – my example document from earlier (with the . ' what goeth here in the rendering) is only about a dozen lines.

And per the Microsoft documentation linked in the Lexilla issue,

<% i = i +1 'This statement increments i. (This script will work.) %>

… is valid / complete syntax: the ' comment only comments out the text portion – the %> closer tag does not get hidden by the comment inside the <% ... %>.

Taking that line, and putting it into zufuliu.asp with a few other lines,

Text before <% i = i +1 'This statement increments i. (This script will work.) %> text after <br>i = <%= i %>

… to actually render something in the webpage, I see:
d8c908a2-4250-41f8-85a7-07f073fe642a-image.png
And Notepad++ is highlighting that as expected: the i = i + 1 is code, the 'This statement increments i. (This script will work.) is a comment, and the %> still closes the code, and everything after still runs.

Moreover, if I use a ' to “comment out” the <% ... %> line, it doesn’t actually make that code not run. It just interprets the ' as part of the HTML, not as part of the VBSCRIPT.
2a2540ba-16a9-4beb-957d-da2b7353ae7e-image.png

So your single line of ' <% Sub ApplyPasswordReset(oConnGlobal, postData, ByRef result). %> does nothing to show us where the bug might be, or what’s going wrong.

To be able to prove whether there’s a bug in Lexilla, a bug in Notepad++, or just user error, you need to give a complete known-good ASP code that shows the problem, that we can open with IIS to see that the code works as expected (ie, to show it’s valid ASP that you are trying to edit), that we can open with Notepad++ to see that it highlights “wrong”, and opens with SciTE to show that it highlights “right” or “wrong” depending on what you say (the last time, I believe I understood you to say that it was rendering right in SciTE),

If you want help, you need to give all the information from the 3 simple steps I put in my last post and reiterated here. Please understand, I am not saying this to be mean: the Lexilla people do not believe there is a bug in their library (the issue has been closed as PEBKAC). Nothing you have shown us gives evidence of a bug in Notepad++. If you don’t provide exactly what I described above, your report will be diagnosed to be user error, and nothing will ever happen.

At this point, it’s your decision as to what happens next.

@PeterJones nice I will try and upload this theme as well.

@jeffs555-s ,

With more tabs than will fit on screen, most programs put the left scroll arrow on the left side of the tab bar and the right scroll arrow on the right side of the tab bar.

I am not sure I’d agree with the word “most”.

Microsoft’s new Windows 11 Notepad app (the one that handles multiple tabs) does. But Chrome just keeps shrinking the tab shape, and after a few dozen, I stopped trying to see if I ever got it to show scroll controls OpenOffice/LibreOffice Calc (and my memory of Excel) put both the ◀ go-left and ▶ go-right and add-sheet buttons on the lower left Google Sheets webapp puts both the < and > on the lower right As far as I can tell, Visual Studio doesn’t give such buttons; instead, they just have the ▼ pulldown, similar to the one on Notepad++'s menu bar. Mozilla Thunderbird (email client), like Chrome, seems to just shrink the tabs; again, I could never get it to show scroll controls.

So I tried a variety of apps, from different categories of tabbed-like interfaces, and only one of them puts the left-scroll on the left and the right-scroll on the right. In my quick experiment, “most” put them immediately next to each other, wherever they go. But really, it seems there’s no global consistency on such scroll controls. That’s not to say that there aren’t a gazillion other apps that do split them between left and right sides of the Tab Bar – just that it apparently depends on one’s set of applications that they use.

I can understand wanting it as a feature, if it’s what you are used to. It isn’t currently an option in Notepad++ (though it could be requested, as described in our feature request FAQ); and, as far as I know, there aren’t any plugins that currently move the Tab-Bar-scroll-left button to the left side of the Tab Bar (and I don’t know who would win the war between a plugin and the main app as far as trying to position that button, so I don’t know if a plugin could enforce a different location for it)

@bvklaveren said in sha512 certificate cannot be validated:

I have seen the instructions, but installing the self-signed certificate on all computers in the organization is not a practical solution.

Okay. Then ignore the self-signed, since there’s a commercially-signed alternative that can give you confidence that the binary is being distributed by the one who it claims to be.

It only takes one certificate to give the confidence, but it also means that you can ignore any other certificate on the same file, because having two certificates (one self-signed that you choose to distrust and the other commerically-signed which you choose to trust) gives you exactly the same amount of confidence that the binary is valid as does having just the commercially-signed certificate with none others. Whereas for someone who does trust the self-signed, they can choose to trust that self-signed and ignore the commercially-signed as being tainted by money.

Also, why would you include a self-signed certificate if you also have “real” one that can be validated without extra steps?

The “why’s and wherefore’s” are all available in the public, in various discussions on this forum (search for certificate) and the relevant section of the User Manual and in the various announcement pages in the official website. But I’ll sum it up, from my outside perspective:

Because for months, there wasn’t a commercial certificate available, because the corporations behind the certificates are set up to make things difficult to open source projects. And once he did get the new globalsign cert, he didn’t feel it was worth it to undo all the effort he had put into creating the self-signed certificate.

Lastly, it is not in the msi, so why is it in the exe?

apples and oranges?

The MSI is a container, which can be signed, and is signed separately from any EXE or DLL that might be contined therein. So the developer apparently chose to only sign that MSI with the commercially-derived – probably because his system was already setup to sign the EXE with both (because it was setup to sign with just the self-signed, then he added the globalsign cert on top of that once it was available), whereas the MSI didn’t come until a month or two after the globalsign was available, so he probably didn’t think it was worth the extra effort to figure out how to sign the MSI with the self-signed as well.

@mpheath ,

Thanks again for you reply! I plan to test your BufferRestore script after my next “break” in my current project(s). After looking closer at your post and web page, this looks to be a good option for me to use until NP++ has a built in tool…

I do own a rack of servers and plan to spin up a couple extra VMs once I release my own platform/environment for my apps (within the next couple months). The challenge with one of the (main) current platforms which I’m using is, while I pay annually to be a developer for them, their licensing does not allow me to run on more than 1 PC at a time, so it requires me to log out/in each time I switch between PCs or VMs. I can’t maintain the variable “state” if forced to log out.

There are many additional issues with Windows Rebooting… When the current test environments get “killed”, all the opened files are now locked and all the variables are flushed. So it requires unlocking the files, and recreating the previous workflow to recreate the variable “state”. Using a log file doesn’t work well in this environment, because it typically goes back beyond the last save… and could only be applied if the saved file was in the exact state as when the log file was started… not sure if that makes sense?

@PeterJones @guy038 @W-H @Ava-Thompson and everyone

Thanks a lot for checking out NppVim and for all the support. Really appreciate the feedback and the nice words.

Quick updates:

Repo here: https://github.com/h-jangra/NppVim I’m still actively working on the plugin, adding more motions, fixing bugs and making things feel closer to real Vim. More text objects, better repeat for t/f etc all in progress.

I’m glad to hear interest in supporting the project. I’ll set up GitHub Sponsors soon and will update this thread.

Thanks again for trying it out and giving feedback, it motivates me to keep improving it.
Please feel free to open issues on GitHub for any bugs or features you want prioritized.

I seem to recall that there was a bug in the main GitHub repo about this kind of issue, where it was extremely slow to move tabs when a lot of tabs were open.

I also seem to recall that this issue was solved, probably in the last year. I’d recommend that @Wilson-Azambuja-Porto-Alegre-RS install the most recent Notepad++ version and see if that fixes the problem.

@Dan-McGovern ,

-openFoldersAsWorkspace works just fine with -settingsDir and -titleAdd and other similar options, in my experiements.

But, as far as I can tell, once -openFoldersAsWorkspace is specified anywhere on the command line, all path arguments are treated under the rule, “if it’s a folder, put it in the FolderAsWorkspace panel, and don’t open any file-paths directly”.

That seems like the intent, because I don’t think it was considered that notepad++.exe myfile.txt -openFoldersAsWorkspace SomeDirectory\ would be a normal use-case (ie, when adding the feature, didn’t think that you’d want to open a file from one location, plus a folder-as-workspace from another, in the same command-line)

If you don’t have the -multiInst argument, and if you don’t have the multi-instance preference set, then you could break it up into two calls of notepad++.exe

"%np%" -settingsDir="..." -openFoldersAsWorkspace "%workspace%" "%np%" -settingsDir="..." "%file%"

In that incantation, the oFAW has to come first, because of the known-and-never-solved issue that oFAW doesn’t work on an already-running instance of Notepad++.

… But since I seem to remember that you do have the multi-instance preference set, that’s not likely to work for you. In which case, until you do a feature request, and it was accepted and implemented and released, I don’t know that you’ll get exactly what you want.

I’ll update the manual to better explain current behavior with the oFAW option.