zzzFM file manager

Forum Forums General Software zzzFM file manager

  • This topic has 104 replies, 10 voices, and was last updated May 10-6:58 pm by Xecure.
Viewing 15 posts - 61 through 75 (of 105 total)
  • Author
    Posts
  • #58304
    Member
    XecureXecure

    I created a gitlab group, forked zzzfm, and built a i386 package. I have uploaded it here: https://gitlab.com/antix-contribs/zzzfm

    Please, test in 32 bits (you just need zzzfm-common_1.0.7-1_all and zzzfm_1.0.7-1_i386 .deb files (except if you want to test the gtk3 version, where instead you would download zzzfm-gtk3_1.0.7-1_i386 .deb file).

    If your testing works properly I will request to merge with skidoo’s main repo.

    #58309
    Member
    XecureXecure
    #58329
    Member
    Avatarcalciumsodium

    Hi @Xecure,

    I was able to successful install zzzfm-1.0.7-1 GTK2 on my antiX 19.1 32-bit system via terminal.

    But there is an interesting story.
    I first tried to install using gdebi. Gdebi was able to install zzzfm-common_1.0.7-1_all.deb. When I tried to install zzzfm_1.0.7-1_i386.deb, gdebi flashed a warning that it had to remove 7 dependencies and install 3 dependencies. So I clicked install. As it tried to remove the 7 dependencies, antiX crashed. When I rebooted, antiX crashed during the reboot saying some dependencies are missing. Basically, I had to reinstall my antiX OS.

    I wanted to see if this would happen on another antiX 19.1 32-bit system. Again, when I tried to install zzzfm_1.0.7-1_i386.deb, gdebi flashed a warning that it had to remove 7 dependencies and install 3 dependencies. This time, I did not proceed with the gdebi install. Instead, I installed by terminal using sudo apt install.

    I am not sure why this is happening on my 32-bit systems. Can someone reproduce this?
    Thanks

    • This reply was modified 1 week, 5 days ago by calciumsodium.
    • This reply was modified 1 week, 5 days ago by calciumsodium.
    #58333
    Member
    AvatarRobin

    Many thanks, Xecure!

    Can someone reproduce this?

    I’ll check as soon as possible, but I don’t have a free USB stick for duplicating antiX for testing this right now. So it’ll take some time.

    #58335
    Member
    XecureXecure

    Gdebi has failed me so many times I no longer trust it. I do the second method you described, using the terminal to install local .deb packages.
    Gdebi uses dpkg (I believe), which is no longer as smart as apt.

    Let us know if the zzzfm works as well on i386. We could also compare how much RAM it uses on that system.
    It is not 100% reliable, but t is a good tool for comparisons. I use the built in
    sudo ps_mem.py
    To see relative RAM use.

    #58336
    Member
    Avatarcalciumsodium

    These are my numbers using antiX 19.1 32-bit with all programs freshly and separately launched and then immediately sudo ps_mem.py

    For spacefm and zzzfm: 1 tab. Show devices, with internal drives, empty drives, mounted networks, and mounted others all checked.

    Private + Shared = RAM used Program

    26.4 MiB + 7.5 MiB = 33.8 MiB spacefm
    16.2 MiB + 4.3 MiB = 20.4 MiB zzzfm
    17.0 MiB + 3.4 MiB = 20.3 MiB pcmanfm
    42.4 MiB + 11.3 MiB = 53.7 MiB dolphin

    #58340
    Member
    Avatarskidoo

    View } Preferences } General(tab)
    [] Show thumbnails

    In my use, and during testing, I have this UNchecked.
    Someone might test whether memory usage is affected by the state of the ShowThumbnails setting.

    #58350
    Member
    Avatarskidoo

    @Robin
    >>> “copy name”/“copy path“/“copy parent”) the three similar commands produce unequally formatted output
    >>> Either all of them should get quoted, or none of them.

    copy_parent
    probably only exist as a menu item b/c we can optionally untick [hide] the pathbar for each of the 4 panels

    For the other 2, the comparative “inconsistency” in their behavior makes sense (to me, at least) when copy is performed while MULTIPLE rows are selected.

    If multiple fullpaths are copied, the likely paste destination is a script, or the commandline. So, they are necessarily conveyed all-on-a-single line (and are quote-enclosed to guard against the possibility of spaces occurring within filenames).

    If multiple names are copied, what is the likely paste destination? Apparently our ancestors agreed that the likely destination is a “list” of some sort, so… for the sake of convenience, the clipped content is conveyed as a multi-line, one-item-per-line snippet.
    .

    >>> SpaceFM always opens a new tab when clicking a device from left side menu, even if there exists an identical tab already. [..] it shouldn’t default to use new tabs again and again.

    Each time we left click on any of “Device” entries displayed in left-pane, the handler opens (displays) the Device’s toplevel content within a new tab added to the currently focused pane of the active xxxyyFM window. The handler does not know, does not care, how many panes are currently set visible nor how many concurrent xxxyyFM windows are open. Right click contextMenu does present a choice between “Open” and “Open in Tab”. Does one of these have more appeal? Which of these would be better suited for assignment as the left-click callback? (I have no firm opinion on this.) Maybe left click should “do nothing”, forcing the user to right-click and choose from various available actions? What text labels would be suitable for additional actions? Descriptive, but way too loooong: “open this device if not already open, and display it in pane4. Oh, and toggle pane4 visible if it isn’t already currently visible”. Workably short, but will defy translation: “OpenInP4andGoto”
    .

    >>> Display of some values in statusline at bottom of the window is confusing:
    There is no blank between unit identifier (e.g. G,M,K) and number of files.
    So it reads “23 G free / 24 G1 / 13 (12,3 K)”. Took some time to understand what “G1” (or “M26”, “G141”)
    in here actually means… Guessed some kind of secret code first.
    ^———- I have now added a few extra spaces of padding between the portions of secret code. FWIW, I’ve never seen “G141” ~~ apparently that is passed to gettext and back, and is rendered with localized formatting. English statusbar shows “13 G free / 29 G” followed by two or 3 spaces separating that portion from the next piece of “secret code”.
    .

    @anyone…
    >>> movement of the tablist. [..]the arrows right and left to its ends [..] The way it is, spaceFM will open and display every single tab on your way to the hidden end of the tablist, which is rather time consuming.

    Maybe the description will resonate with other testers. When I tested, using a 9-year-old PC, after opening 26 (or so)tabs… a “click an end arrow and hold down” maneuver zips end-to-end across the tabs in less than a second. None of the directories in my test contained 1000+ items, though, and thumbnail view was disabled… so maybe there’s a conditionally-triggered problem that I haven’t yet discovered.

    #58351
    Member
    Avatarcalciumsodium

    Regarding the Show Thumbnails vs no Show Thumbnails, I did not see much difference in memory usage using antix 19.1 32-bit. Here are the numbers,

    w/ Show Thumbnails UNchecked
    27.6 MiB + 7.0 MiB = 34.6 MiB spacefm

    w/ Show Thumbnails checked
    28.0 MiB + 7.0 MiB = 35.0 MiB spacefm

    w/ Show Thumbnails UNchecked
    17.5 MiB + 3.3 MiB = 20.8 MiB zzzfm

    w/ Show Thumbnails checked
    17.4 MiB + 3.2 MiB = 20.6 MiB zzzfm

    w/ Show thumbnail of files UNchecked
    16.9 MiB + 3.5 MiB = 20.4 MiB pcmanfm

    w/ Show thumbnail of files checked
    16.9 MiB + 3.5 MiB = 20.4 MiB pcmanfm

    #58352
    Member
    Avatarcalciumsodium

    with one tab
    17.4 MiB + 3.2 MiB = 20.6 MiB zzzfm

    with 26 tabs
    21.3 MiB + 4.6 MiB = 25.9 MiB zzzfm

    #58353
    Member
    Avatarcalciumsodium

    As mentioned in a previous post, I have a Samsung 60 Gb USB 3.0 Thumb drive formatted with exFAT system.

    When I insert that thumb drive into the USB port, zzzFM recognizes it and pops up on the left hand side of the screen, in this instance, as
    – sdc 60G 60G_USB

    I named the thumb drive 60G_USB.
    But when I click on it to open, I get this error message:

    Run: Mount /dev/sdc
    Status: Finished with error (exit status 1)

    FUSE exfat 1.3.0
    WARN: volume was not unmounted cleanly.
    fuse: mountpoint is not empty
    fuse: if you are sure this is safe, use the ‘nonempty’ mount option

    I cannot seem to mount it for some reason. I cannot seem to mount it on any of my antiX systems. I did an error check on it in Windows 10, and Windows 10 reported no error with it. It works great in a Windows environment. It has the speed of a USB 3.0. It is the newest thumb drive I have.

    Any suggestions on how to mount it? Is it because it was formatted with the exFAT system? I noticed that my other USB thumb drives, with are 2.0, were formatted with FAT32. They seem to mount without any problems.

    #58358
    Member
    XecureXecure

    But when I click on it to open, I get this error message:

    Run: Mount /dev/sdc
    Status: Finished with error (exit status 1)
    
    FUSE exfat 1.3.0
    WARN: volume was not unmounted cleanly.
    fuse: mountpoint is not empty
    fuse: if you are sure this is safe, use the ‘nonempty’ mount option

    try unmounting from terminal and then trying to mount again in zzzfm
    sudo umount /dev/sdc

    I remember I had some trouble in the past with some exfat or ntfs USB device, but I am not sure if adding the mounting options to /etc/fstab fixed it or what.
    Right now, I have no trouble with ntfs, but I no longer have a exfat drive to test.
    I also remember skidoo or dolphin_oracle pointing out how to add mounting instructions to spacefm Device handler, but I haven’t found it while searching the forum. Maybe I read it elsewhere.

    skidoo should be able to point us in the right direction.
    As it seems zzzfm workes well on i386 (ignoring the gdebi error), I will request to be added to skidoo’s git.

    #58362
    Member
    AvatarRobin

    @skidoo
    Thanks for looking into this.

    If multiple names are copied, what is the likely paste destination? Apparently our ancestors agreed that the likely destination is a “list” of some sort, so… for the sake of convenience, the clipped content is conveyed as a multi-line, one-item-per-line snippet.

    I agree that it can be sometimes useful to have the quotings settings like this, but when composing a terminal command from the copied content it would be desirable to have a more consistent behaviour. To make an all-purpose tool from it, my idea is: add modifiers to copy command these three copy commands, following a logic like e.g.:

    click mouse button on context menu item
    -- without modifier = react as it always has been, with different quotation.
    -- ALT + left mousebutton = copy all items single quoted
    -- CTRL + left mousebutton = copy all items double quoted
    -- CTRL + ALT + left mousebutton = copy all items unquoted

    This is a vague idea only, so if this is technically to complicated to realise, leave it simply as it was before, I’m fine with it (having memorised meanwhile in which of the three I’ll have to add or remove quotes after pasting).

    The handler does not know, does not care, how many panes are currently set visible nor how many concurrent xxxyyFM windows are open.

    This is something user can take care of by himself, I believe. If at all, it would make sense only to look for this within active pane I’d believe, but even this is not really needed. I was misled in remembering about the true reason of the problem I had in mind. What led me to the Idea of avoiding duplicates was another behaviour of spaceFM I didn’t mention in my first post, which is the true reason for many identical tabs opened: When SpaceFM is reopened (given, “save tabs” was set), all tabs connected to items not present while opening (like USB-devices, network folders…) will display users “home” folder instead. So it can happen you face 10 or more tabs with home-folders, in between others… What really was needed consequently was not to detect whether a tab with identical content is opened by user (he can care himself), but to close tabs simply instead of changing to home folder all of them when original content is not present anymore. There is no advantage of having them, since you can’t retrieve from all the identical home tabs what path they did originally contain (back button doesn’t work on these to bring back original content, even if missing device is mounted again meanwhile).

    Right click contextMenu does present a choice between “Open” and “Open in Tab”.

    That’s true, but you have always to find it from menu entries instead of simply clicking.
    Possible convenient solution for this could be again: Mouse button modifiers.

    left mouse button on item opens everything as before, in new tab.
    alt + left mouse button = open item in active tab.

    This way we wouldn’t need any additional menu entries, nor we have to decide which option has more appeal. Both of these would be only one click without searching a context menu and selecting fitting entry.

    >>> Display of some values in statusline at bottom of the window is confusing:
    There is no blank between unit identifier (e.g. G,M,K) and number of files.
    So it reads “23 G free / 24 G1 / 13 (12,3 K)”. Took some time to understand what “G1” (or “M26”, “G141”)
    in here actually means… Guessed some kind of secret code first.
    ^———- I have now added a few extra spaces of padding between the portions of secret code. FWIW, I’ve never seen “G141” ~~ apparently that is passed to gettext and back, and is rendered with localized formatting. English statusbar shows “13 G free / 29 G” followed by two or 3 spaces separating that portion from the next piece of “secret code”.
    .

    Well, from what you stated I take this is probably an error in translation files simply, I’ll search for it. I have to admit I have translated back the word “free” in the string from German “frei” before posting here to make it easier to identify the item I described in board language. So, if a single space between unit identifier and number of items is present when script program uses its original English strings, you don’t need to fix anything here in the script program. Btw, don’t understand what you mean with “never seen G141”? Right now my statusline reads “799 G frei / 1,8 T347 Objekte /media/home/sdc1-usb-TOSHIBA_External/Backup…”, so the secret code is T347 today 😉

    When I tested, using a 9-year-old PC, after opening 26 (or so)tabs… a “click an end arrow and hold down” maneuver zips end-to-end across the tabs in less than a second. None of the directories in my test contained 1000+ items, though, and thumbnail view was disabled… so maybe there’s a conditionally-triggered problem that I haven’t yet discovered.

    Conditions might be: 20 years old instead of only 9, fast but single core 32 bit processor instead of multi core 64 bit, and limited amount of ram, which probably forces real re-reading of directories (or redrawing of display-content for each window passed on the way, can’t determine) and/or possibly reading from slow USB or network devices. What is the true reason I can’t detemine. But with 12 tabs only you’ll need a minute to get from one end to the other, waiting for displaying each single tab, while Processor powers up to 100% (@ 1,7 GHz).

    So long
    Robin

    P.S.: Wasn’t able to free up a stick for testing zzzFM 32bit itself, will make up for it as soon as possible.

    • This reply was modified 1 week, 3 days ago by Robin. Reason: defined/corrected English wording
    #58374
    Member
    Avatarskidoo

    add modifiers to the copy command
    [..]
    if this is technically to complicated to realise

    Embrace the glory of DesignMode and custom commands, eh.
    It would be, like… a 1-liner command. It wouldn’t even require a fancy “script”.
    Pipe %F (or whatever detail), through sed command, to the xclip command

    pappy says: “You can lead a horse to water, but ya can’t brick ’em up

    To make an all-purpose tool from it, my idea is: add modifiers to the copy command…

    Each time you create a custom command, you have an opportunity to assign to it a keybind (aka hotkey aka keycombo aka accelerator).

    799 G frei / 1,8 T347 Objekte /media/ho..

    Seems like the person translating this took the liberty of inserting “Objekte”.
    shrug ~~ no harm, but minimial benefit is imparted by additionally displaying this static, generic, string (displayed regardless whether “file” or “folder”) and its presence will cause “ellipsized text” to be triggered sooner, when long pathnames are involved.

    so the secret code is T347 today 😉

    Thanks, made me chuckle.

    • This reply was modified 1 week, 4 days ago by skidoo.
    #58396
    Member
    Avatarolsztyn

    Just a question:
    So, considering zzzFM looks in many ways superior to SpaceFM, being a continuation of SpaceFM with serious drag bug fixed,
    what is holding antiX or developer (skidoo) from calling it SpaceFM 1.07 and antiX replacing the original SpaceFM with skidoo’s via update or upgrade? I think it makes perfect sense for antiX to permanently adopt this new version…

Viewing 15 posts - 61 through 75 (of 105 total)
  • You must be logged in to reply to this topic.