Jump to content

wearenotamused

Experienced Members
  • Posts

    30
  • Joined

  • Last visited

Everything posted by wearenotamused

  1. The asterisk wildcard character seems not to be supported in the domain of a whitelisted cookie in the Mac version, while it is in the Windows version. In the Windows version, while it isn't possible to add a domain containing a wildcard to the list of cookies to be kept (the right side of Options > Cookies) using the GUI, such a pattern can be added to the list in the registry and it will subsequently appear in the GUI. CCleaner itself comes preconfigured with a "*.piriform.com" whitelist entry. In the Mac version (at least as of 1.07.236), the situation doesn't seem to be analogous. The app doesn't come with any "*.piriform.com" whitelist entry, and any such entry added to the CookiesToKeep dictionary in ~/Library/Preferences/com.piriform.ccleaner.plist manually doesn't appear in the list in the GUI. I haven't tested whether such an exception still operates, despite not appearing in the GUI, but I doubt it. I'm curious why this difference exists between the Windows and Mac versions. Is it, for example, due to the API used to read the plist file? (I wouldn't think Apple would tolerate such a restriction since plists are used everywhere in OS X.) Any Piriformer who sees this: I'd much appreciate whatever brief explanation you might give.
  2. I can't reproduce what you're describing. Right-clicking (i.e. with two fingers) on both App names and individual entries underneath them is giving me the same context menu I get in Windows. Nothing seems off vertically or horizontally about the registered targets of my clicks. I get the same context menu content for Safari > Internet Cache as I do for System > Unneeded Languages, so I'm assuming you're say it's the context menu for the latter based on the menu's position on the screen, not its content. I'm using the same versions of OS X and CCleaner that you are, but on a non-Retina MacBook Air. Are you seeing strange behavior in other apps too? If so, I'd think hardware or . If not, perhaps CCleaner doesn't support the higher DPI of Retina screens properly (e.g. miscalculating click positions, though I'd think Cocoa/the-OS would handle that)? Could someone else with a Retina try to reproduce this?
  3. In the detailed results of an analysis, the Description and Size column labels depress (when clicked) as if clicking them is supposed to do something, but it doesn't. In contrast, the column labels in the Tools > Uninstall table depress in the same way and do serve to sort the table. I'm still too new to Mac to recognize whether, e.g., those are standard Cocoa-provided tables (and column headers) and the depression behavior is enabled by default or can't be disabled, or whatever. In other words I don't know whether to take the depression behavior as a sign that the developers intend for those columns to be sortable, in which case this would be a bug report. (I'm guessing not.) If they currently don't intend for them to be sortable, consider this a request that they do.
  4. Please eliminate the maximum window width constraint. It reduces the amount of info that can be visible at one time, and for what? See attached screenshots for examples of the situation. The CCleaner window will not allow itself to be widened to make more of the truncated fields visible. (I does let me reduce its width.)
  5. What do you think of this idea? You should be able to get what you're looking for out of two presets.
  6. It would be nice to be able to save a "selections state" as a preset that could be loaded at any time. By "selections state" I mean the set of the states of all check-boxes (on both tabs--[OS] and Applications), excluding the ones that haven't been changed from their default, the exact same checkbox state information currently stored by CCleaner in the registry (on Windows) for persistence between sessions. In principle you should be able to have as many presets as you have disk space for, but obviously in practice people would only use a few, so a max of at least 10 probably wouldn't be unreasonable (in my opinion). Ideally, users would be able to load (and save) them without going into Preferences. Consider putting a small button in the empty space to the right of the [OS] ("Mac OS X" in my case) and Applications tab labels on the Cleaner pane, that dropped down a menu containing a list of existing presets and, separated from them by a horizonal divider, a "Save as preset" that would open a prompt for a name. Existing presets would be loaded by simply clicking their name. (I've seen this kind of preset interface in at least a few other softwares, but the only one I can think of at the moment is the print options dialog for a high-end printer. ) This would address the needs of people seeking multiple "levels" of cleaning (see this topic, for example), but be more flexible and (I'd think) easier to implement--it would require fewer changes to the UI anyway. If stored in files (presumably, ini's on Windows) such presets could double as a way for users to backup their selections or transfer them between machines, at least within platform. As the number of applications and options supported by CCleaner (or winapi2.ini) grows, so would the benefit of not having to manually recreate a selection pattern. Thoughts?
  7. If what you describe happened to me, my first thought would be that the dmg was corrupted during download. Given the relatively small size of the dmg, I'd just redownload and try again. If after doing that you get the same behavior, here's my first question in helping you troubleshoot: By default the system's disk mounting component (haven't learned the names of all the OS X components yet) validates each dmg before it tries to mount it, and shows a little dialog with a progress bar as it's doing it. When you say "nothing happens" does that include that little dialog or did you mean nothing after that? FYI: A single-click on something in Spotlight's list is equivalent to double-clicking that thing in Finder, so a second click won't doing anything for you and one day under the right circumstances might have an effect you'll consider undesirable. On a related note, I've found Spotlight doesn't "do" right-clicks either. It'll treat it like a left-click.
  8. From what browser? I likely won't be able to solve your problem, but I know anyone investigating it would want to know the answer to that. Same here.
  9. I wish there were available for display on the File list tab a column for Mean fragment size, as this quantity would give an indication of the frequency of disk seeks necessary in reading a given file sequentially, but unlike other possible metrics that would give an even better idea of that (e.g. median), it's trivial to calculate based on data Defraggler already has: it's simply Size / Fragments. Defragmenting the files with the lowest mean fragment size first would give more performance-improvement "bang" for time-spent "buck" than prioritizing those with the highest number of fragments. I understand any concern there might be about having too many columns, which is why I say "available for display". I don't mind if it's not shown by default, as long as it can be shown by (I suggest) right-clicking on any of the column headers (as in Windows Explorer's details view).
  10. When specifying registry locations in custom ccleaner entries (whether in winapp1.ini or winapp2.ini), it would be useful to be able to use patterns like we can in FileKeyN specs. Feel free to think of this as two severable requests, which I'll list in order of importance to me: Support patterns in the value mask. (Note that this would make it possible to empty a key without deleting the key itself, behavior that is not currently possible but which might be necessary for certain applications and/or security environments.) Support patterns in the key name Thanks.
  11. I understand the general desire to keep the number of entries to a minimum for multiple reasons, but would someone mind sharing the reasoning behind removing separations of entirely different types of data, like cache and logs? I pick those because they exemplify replaceable and irreplaceable data, which, depending on a user's reason for using CCleaner, they may well want to treat differently on an app-by-app basis.
  12. Could switches be added to df.exe for getting various fragmentation-related statistics pertaining to drives, files, and folders: fragmentation level of a drive or folder (w/ or w/o excluded files--indicated by another switch), number of fragments making up a file or folder entry, etc? Ideally the output would be a single line containing a single number for easy handling by scripts. Among other uses, this would allow requested behaviors like this to be achieved with a few lines of batch or bash (thank God for cygwin) and a scheduled task, with limitless customization potential. Thank you for Defraggler (and df)!
  13. You'd also want to take into account which disks are on the same disk controller. The max throughput of a controller is total, not per-disk.
  14. Dare I even bring up native command queuing and variable spindle speed?
  15. Couldn't agree more. The current behavior is rather puzzling. Doesn't including excluded files in the measurement make it an objectively worse indicator of whether it's time to defragment? Any advantage to this approach over using process priority? (I can only think of one: if total CPU usage was the concern rather than the performance of other processes.)
  16. That's the Tools property sheet for drives. Defraggler already replaces Windows's defrag with respect to that "Defragment Now..." button. To see what he's talking about, Computer Management, right-click My Computer and choose Manage. See "Disk Defragmenter" under "Storage" in the tree.
  17. The registry exists for a reason: performance (of the system, not necessarily of CCleaner). This idea is well-intentioned but would force all CCleaner installations--portable or not--to be poorly behaved Windows apps. Doing that in the name of making portable use absolutely seemless is a crappy trade-off. I think your earlier idea is much better: This one would be simple for both developers and users, and do so without forcing all installations into being poorly behaved. In the history of Windows, the registry was a replacement for the use of .ini files, so there shouldn't be any reason a non-portable installation would need to have a ccleaner.ini around. It might as well be the portable installation flag that the .exe looks for as soon as it starts. And a "Run at startup" option should really be disabled and ghosted in a portable installation of any program.
  18. Mmm this might be getting a bit out of CCleaner's scope. Cool, I hadn't thought about looking for a ready-made tool to do that. Thanks.
  19. Totally! I mean, after all, how's Silverlight supposed to compete with Flash if CCleaner doesn't delete its cookies?
  20. What about this more parsimonious (in my opinion) alternative: Store screen-normalized position (and size?) instead. That way, it can be closed in the lower right-hand corner on a JumboTron, and next open in the lower right-hand corner on a 640x480 dinosaur. If you consider size of elements within the window and bring upper and lower bounds into it, there's lots of clever stuff you can do. Even without that, though, normalized position should ensure some part of the title bar is always on screen. At the very least, if the position in the INI is off-screen under the current display settings, disregard it and open at 0,0 instead. Next step multiple monitors?
  21. We're talking about a non-default option--"IF you want the current behavior, don't go out of your way to change it." As for "preserving trash", keep in mind the entire point of the Recycle Bin is to allow you to confirm something is trash (by not missing it) before it's truly deleted. If you'd rather not have that insurance, yet the presence of a check box in the preferences adds complexity to your deletions, then "you're doing it wrong".
  22. I can actually see some logic in at least giving the user this option. Deleting to the Recycle Bin would fit with the whole Windows convention, after all. You could always make a "Skip Recycle Bin" preference enabled by default. I have to disagree with this part, though. If the user minds the new deletions mixing with previous ones, they can empty it manually just as they would do before some manual deletions. If they went out of their way to turn off the Recycle Bin skipping in the preferences, they should be able to figure out that they won't free the space until they empty it. Then again, I suppose you could give them a reminder message before exiting (but only if it applied to them). Kind of an odd place to put stuff you want to keep, no? Something you might consider: Save yourself the 'shift'-press by disabling the recycle bin altogether (how-to) and making yourself a 'pre-delete' folder instead. A fringe benefit of this is that, should you ever have to recover that file system, those files wouldn't be named, e.g., De09.txt. I'm sure the Recycle Bin's capacity is in the registry. If it runs out of space, the user could be shown the same prompt they'd get if they tried to delete something that was too big to fit... a message asking if they want to permanently delete or cancel. Plus, if the user Analyzed first, CCleaner would know how much stuff it was going to delete, and could even prompt them about it as soon as they hit Run. Something like "This will exceed your Recycle Bin's capacity. Permanently delete instead?" They could click Continue, or Cancel, empty it, and try again. They might just want to leave the 'Skip Recycle Bin option' enabled then! Same way you differentiate stuff from different manual deletion sprees: sort on the Date Deleted column. Really? Isn't there a Shell function you can pass the filename to? As a matter of fact, if you passed an array of all the filenames found in the Analysis, the Windows shell itself might take care of the potential insufficient space issue above, just as if the user had multi-selected those files in an Explorer window and press Delete. (I must say this Shell integration would be right in line with the oh-so-sexy real-time updating space usage quotes in Defraggler! ) The Recycle Bin is perfect for this. If something breaks, you can just open it, sort by Date Deleted, select the big streak of deletions (or even just targeted ones if you know what was 'missed' by something), right-click and select Restore. Voila; everything goes back where it came from. (If a new Chrome history file, for example, had been created since the deletion, Windows would ask if you wanted to overwrite it.) You could be indiscriminate in the restoration because you could just run CCleaner again (only more conservatively this time). I'm not a fan of bloat either, but the problem, in my opinion, is more often in the organization and presentation of the options than in the fact that they're offered. At any rate, removal of cruft is CCleaner's game. This would be a single check box allowing the user to pick the sense of "removal" they wanted: the more absolute sense or the more Windowsy. I myself would keep the default and skip the recycle bin (since I know the stuff CCleaner removes--log files, hotfix uninstallers, history, MRU, etc--would only be missed by a user), but I think it's a worthy idea for an option to help users with traumatic memories of overly ambitious system cleaners.
  23. Can't something be clean yet fragmented? Any write to disk can potentially cause fragmentation. Should virtually every program defrag after itself? If they're on an NTFS partition, Defraggler can defrag open registry hives just like any other file. If they're on FAT, try pagedefrag.
  24. I could be wrong, but I'd think this might require CC to keep (and Piriform to keep up-to-date) more knowledge about how various apps store their history than is currently required to just clear it all. For example, if an app's history is kept in a dedicated file, clearing the whole history could be as simple as deleting that file. To clear only part would mean having to alter that file in a particular way, which requires knowing how the file is structured internally, something that could even change from version to version of the app in question. It'd be nice if apps exposed an interface (through a command-line option or something) to abstract their implementation for the rest of the world!
  25. I would hope CCleaner would use environment variables, registry keys, config files, etc to abstract as many locations as possible (Windows and third-party apps), not just Temp. Is it pretty good about all the others as far as people know--temp just got missed? As my prof would say, "respect the abstraction!"
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.