v6

[nikos]

I think we discussed "custom properties" before, right? The problem is that they wouldn't be fast-searchable. The generic comments property is for this purpose and it is indexed for many file types (=fast search)

I saw that other FMs are embedding this Everything tool, I might consider about it but wouldn't know how to go about it. Isn't everything just for filenames? no file content?

as for long path support, it is available in xplorer2 for 10 years or so

[otlaolap]

Here are a few thoughts triggered by jimspoon's comment above.

I use two other programs all the time while I am using X2: Everything, as mentioned by jimspoon, because it finds things instantaneously and selectively and can take console command lines to create lists; and File Locator Pro, because it is more powerful and selective than X2's search. Both programs provide great utility to me, and the three together work very well. From time to time I have used Bulk Rename Utility because it too is much more powerful than X2's built-in renamer. (I know you are trying to provide something simpler and less complex.) don't know how well trying to incorporate features from these programs would pay off, though, because in some way my mind enters a different domanin when I am doing data searches (FL Pro) and general disk organizing and maintaining and working (X2), and tying all of that into one domain might be, well, mentally tangling. Also the level of expertise in these other programs is extensive, and would be hard for a single person to approach. (FL Pro for example uses eight threads at 100%, maxing out my CPU, while searching for data across files, and that alone is hard.)

A very useful feature to me would be more direct access to what I call "jug" files (.jug filetype). These are simple lists of files with fully-qualified file names. When I open a .jug file, X2 now gives me (via AutoHotkey hacking) an equivalent scrap with the jug-file contents. Thrre is probably no difference between a .cida and a .jug, but because of the AutoHotkey hackery the .jug file is treated much more as a normal file: you open it, you have a scrap that works; you add or remove files from it; you save it; you edit it elsewhere or generate it elsewhere (and file lists are produced easily in lots of other programs) and open it; in all cases you have again a scrap that works. Native mode access to file lists, more visible or mentally accessible in X2, might be good. It seems to me this has been requested before, possibly by kilmataed or frizzlefry or others in that pantheon.

And I second the desire for multiple programmable columns. I really have had to work hard overloading the single available column when trying to deal with several different desirable attributes. And you may remember I had to go to hackery on the comments field to add additional attributes so that I could see them. The only way I found to do this was (automated) binary editing on the extended attributes of the file. Everything else requires keyboard interactions and the whole thing I wanted was something automatic (for example a count of the lines in the file: I have lots of jug files (.cida files) on disk and I wanted to sort them, inside X2, by the number of lines (i.e. files) they contain, so that I could deal with them from fewest to most or vice-versa).

Finally, it would be nice if you were not forced into changing the base controls used inside X2 or my whole house-of-hackery falls down.

[Tuxman]

Isn't everything just for filenames? no file content?

It is just for filenames. But it’s pretty damn fast at it.

[nikos]

File Locator Pro, because it is more powerful and selective than X2's search....
I have used Bulk Rename Utility because it too is much more powerful than X2's built-in renamer.

so what do you find in these programs that xplorer2 cannot do? sometimes people are not aware that features exist in xplorer2
if there are any useful features in them, please state them explicitly and I'll see what can be done

FLP uses all your CPUs because after all these years it is still choo-choo search! Why don't these guys use an index?

you know that the single programmable column can be easily "reprogrammed" using the saved programs, right? I will try to add another one for v6

as for your JUG files, scraps can already open text files with contents, if there's one path per line. You can also paste the text and it will create the contents

[jimspoon]

Re long path support -

To test this I created a folder path: "C:\0--------1---------2---------3---------4---------5\0--------1---------2---------3---------4---------5\0--------1---------2---------3---------4---------5\0--------1---------2---------3---------4---------5". Each folder name is 50 characters long. The total path, if you count the "C:" and the backslashes, is 206 characters (actually I think 207 when you count a terminating NUL). If I try to nest another folder of the same name in the last nested folder, the total would reach 258 including C:, the 5 backslashes, and the terminating NUL. And I get this message:



If I do the same operation in Directory Opus or Total Commander, the new folder is created without an error message.

Since Windows 10 version 1607, Windows can be enabled to accept paths up to 32767 path length (still, no component of the path may be longer than 256):

https://learn.microsoft.com/en-us/windo ... s=registry

I don't think I'd ever need a path length of thousands of characters, but I frequently run into the 256 (260?) limit.

[Gandolf]

... Why don't these guys use an index? ...

Because indexes are of no use if you use removable drives. Drive H:\, for example, can be any of about 10 possible drives, depending which drive I plug in. Indexing in that situation does not work. It gets totally confused. Everything goes completely balmy if it finds a drive that it has already indexed, but now is (partially) different. Some files may have the same name, but are not the same file. x²'s search does not work for the same reason (if indexing is checked).

FileLocator Pro (and Agent Ransack) have no problem in this situation - I use them many, many times a day. They might be slower, but they show what is on the drive, not a mixture of what was on the last drive plugged in, and what is on the current drive.

I search for a file by name, containing a specified text, in every sub-directory of a given directory (123456.txt | Fred Bloggs | C:\Users\Gandolf\2024\January\). I then get a list I can export, to the clipboard, then paste it into an editor and edit appropriately. x² can do the search, but I have to enter the directory name every time (I'm not necessarily focused on it). Also, I haven't found a way of exporting the file names and the contents. The CSV is just a list of the files, that's not what I want. It's no quicker than FileLocator, and possibly slightly slower.

There's my rant for to-day!! Now I feel better

[jimspoon]

The latest version of Everything, 1.5a, does support content indexing and searching:

Content Indexing - voidtools forum - https://www.voidtools.com/forum/viewtop ... =12&t=9793

Not only that, you can index, search, and sort by all the properties (both Windows Property System properties and "Everything" properties):

Properties - voidtools forum - https://www.voidtools.com/forum/viewtop ... =12&t=9788

While Everything 1.5a is described as an "alpha" version, in fact it has been completely solid for me, and I've been using it for years. If you're still using version 1.4, you're really missing out on a lot of capabilities.

[jimspoon]

Nikos - another wish which perhaps is not a moonshot.

There are some settings which (I think) affect every pane and every tab in the file pane. Sometimes I would prefer that I could have a different setting for different tabs, and that the tab would remember its setting.

I'm thinking that Tools > Options > Window > Folders Sorted As is one of these settings.

[otlaolap]

FLP uses all your CPUs because after all these years it is still choo-choo search! Why don't these guys use an index?

you know that the single programmable column can be easily "reprogrammed" using the saved programs, right? I will try to add another one for v6

as for your JUG files, scraps can already open text files with contents, if there's one path per line. You can also paste the text and it will create the contents

FLP does provide indexes and use them if you want them. Sometimes I do. Often, though, when I search, I want a byte-for-byte search of the current content of every file. That is why I use FLP. I do not trust indexes in general that update themsevlves (windows and its programs do crash) for when I am looking for exactly who is using a given character string. And a search with 8 CPUs sure beats a search with 1.

In one use of the single programmable column, I compute a color for a file based on multiple attributes about that file. Does the file have one of a number of companion files of a particular types? More than one? Does it differ from a specific other companion file? And so on. With that color I use color coding to color the name of the file as it is displayed by X2. This was a lot of work. It is always active for this application (which uses its own X2 profile). I cannot switch it out, in this application, for another programmable without wiping out the color coding. And I cannot switch it out without keyboard interactions.

File lists (jugs) are prepared by many other programs. It is nice to simply click on them and have them open as a scrap.

[nikos]

There are some settings which (I think) affect every pane and every tab in the file pane. Sometimes I would prefer that I could have a different setting for different tabs, and that the tab would remember its setting.

TOOLS > ADVANCED OPTIONS, search for "Each folder tab can have different view settings"

It is nice to simply click on them and have them open as a scrap.

I thought you could open anything with a scrap, but apparently it only checks for TXT at present. So it would load a TXT with paths but not JUG. Is this a standard extension?

[jimspoon]

Different view settings for each tab - thanks for pointing that out.

[pj]

A few years back, during a discussion about click-and-drag, a suggestion to add checkboxes to a file list was made to use the boxes to simplify selection of multiple files and avoid the pitfalls of ctrl and ctrl-shift selections.

I'd like to raise that suggestion again. Perhaps as a toolbar option on a per tab/pane basis instead of something that's always on.

Many times I've been selecting files in a very long list, forgot to hold the ctrl or accidentally clicked the mouse and lost the selection and all the work that went into building the list to that point. The checkbox would eliminate that. I use that with my many emails on YAHOO email and it's very convenient.

-----------------------------------------------
PJ in FL

[Gandolf]

Does "Sticky selection" not do what you want?

[IHipschman]

Some menu options such as actions/change attributes don't seem to be available in a browse flat window. There are times when this would have been helpful.

[nikos]

change attributes is in scrap actions menu? Make sure you don't have plain menus enabled