A Missing(?) attribute?

[johngalt]

Hi, Nikos,

Been a long time since I've had to post anything - X2 is just that good.

I'm using 2.2.0.2 ULT [Unicode] x64 11/14/2012 Licensed to me (using my real name) w/ ref code X2UL.1-BFFECB0D.  I've found a bit of a dilemma.

Problem:  X2 won't display non-Windows file attributes (in my case, T for Temporary - apparently a Linux tag).

Background:  I swap back and forth between the 2 OSs and was extracting some music from several of my discs to finally upload to Google Music in order to stream when I want to on my Android devices.  I found several tracks could not be uploaded to Google Music for whatever reason - in X2 everything looked normal.  However, when I started inspecting the files a little closer, I found that some of them had a T attribute - shown only in the file Properties dialog on the Details tab.

I searched for this problem online, and sure enough, the T can be set by Linux OSs, as well as other programs (say, Live Messenger, files downloaded from a friend on Linux, etc.)  I found a utility that was able to strip the T attribute relatively easily (Nir Sofer has some great utilities) but it led me to think - why didn't X2 show it in the first place?

Assumption:  I'm guessing that X2, being designed for Windows, only shows Windows default attributes, correct?  However, for whatever reason, NTFS stores non Windows Attributes as well.

So, is this even remotely likely to make it into a future build?  I'm doubting it, but I'll still make the RFF here.

Thanks again for your great product.  Loyal user now for...a very long time!  :D

[Kilmatead]

However, for whatever reason, NTFS stores non Windows Attributes as well.

For what it's worth, FILE_ATTRIBUTE_TEMPORARY is actually an official NTFS attribute - just one that isn't expected to exist long enough to be practical to the end user:

A file that is being used for temporary storage. File systems avoid writing data back to mass storage if sufficient cache memory is available, because typically, an application deletes a temporary file after the handle is closed. In that scenario, the system can entirely avoid writing the data. Otherwise, the data is written after the handle is closed.

Considering Nikos recently (v 2.2.0.1) enabled the -I- (Indexed) and -P- (Sparse) attributes in the [S]tock columns, one never knows.

However, in the mean time if you select the column "Attributes" (as opposed to "[S] Attributes") via <Alt+K>, the 'T' attribute will be visible to you in x2.



For those interested in the difference between "regular" and [S] columns, the [S] stands for Stock, which means native to x2, thus the data is extracted slightly faster than the other columns.  This can come into play when you have folders with thousands of objects in them and you're trying to apply complex colour-coding, searching, or filtering rules, the results will be processed "just that little bit" faster.

[Brig]

I was playing around with the Attributes column, [S] vs. otherwise, and I noticed this odd thing. I have Infobars set up under my two panes with these columns: Type [S], Attributes, Selection information [S], Infotip [S]. The weird thing is, Infotip works only on my hard drive--it goes blank when looking at a network folder. Was this always the case and I'm just really slow on the uptake? Does anyone else see this?

Thanks folks.

[x2, latest beta; Windows 7 Pro, SP1, 32 bit]

[nikos]

one of the things the last beta did was turn off the infotips for network folders. If you don't mind the delay, just untick the advanced option "Faster network access..."

[Brig]

Ah, that did it. Thanks. I didn't notice any particular delay either.

[johngalt]

However, for whatever reason, NTFS stores non Windows Attributes as well.

For what it's worth, FILE_ATTRIBUTE_TEMPORARY is actually an official NTFS attribute - just one that isn't expected to exist long enough to be practical to the end user:

A file that is being used for temporary storage. File systems avoid writing data back to mass storage if sufficient cache memory is available, because typically, an application deletes a temporary file after the handle is closed. In that scenario, the system can entirely avoid writing the data. Otherwise, the data is written after the handle is closed.

Considering Nikos recently (v 2.2.0.1) enabled the -I- (Indexed) and -P- (Sparse) attributes in the [S]tock columns, one never knows.

However, in the mean time if you select the column "Attributes" (as opposed to "[S] Attributes") via <Alt+K>, the 'T' attribute will be visible to you in x2.



For those interested in the difference between "regular" and [S] columns, the [S] stands for Stock, which means native to x2, thus the data is extracted slightly faster than the other columns.  This can come into play when you have folders with thousands of objects in them and you're trying to apply complex colour-coding, searching, or filtering rules, the results will be processed "just that little bit" faster.

I should have known there was already a method to do this.  I've been using X2 for so long I completely miss the little things these days lol.

Thanks!

[Kilmatead]

Nikos, while we're on the subject of applied attributes, I know when you added the -I- flag, you did it back-arse-wards, so that technically it's only displayed when items are "NOT" indexed (which is rather annoying for those of us who disable indexing).  Except the odd thing is it shows up on all objects on USB drives, but only selectively on C:\ drive objects.

Doing a search of the C: drive for the -I- attribute tells me there are actually 5,100 objects so denoted (meaning, by your logic, they are not indexed).  Curiously, the indexing service tells me there are actually only 72 items on the drive which are indexed, which means of the 187,116 objects on the C: drive (what, you don't flatten yours on a regular basis?), there should be 187,044 objects displaying the -I- attribute, not 5,100.

So, by your standards are there are actually 182,016 objects which qualify as indexed, but they are not searchable via WDS, and the indexing service itself says they don't exist.

So (before you tell me to go crawl back under the rock I crawled out from ), my question is: What good is the back-arse-wards -I- attribute if it's not particularly accurate in the first place?  Hmm?

(And yes, I do sit around all day dreaming this shite up just to pester you. )

[nikos]

this I is totally useless. It doesn't have to do anything with windows desktop search indexing. I don't know what index they are referring to!

[Kilmatead]

Actually, to be fair, it is related to Windows Indexing (I did a few tests when you first started bothering with WDS), but I imagine that what happens is that the Indexing service only goes around applying the attribute at the time the object is actually indexed, it never actually "un-indexes" the attribute because the service doesn't process files it doesn't index (to turn it off, you simply ban it from scanning certain areas of the system).  And when it wants to re-index the drive, it simply zeroes the existing cache index, and starts over - never seeing the "old" files, which are thusly left incorrectly flagged.

So, yeah - rather useless as a reliable indicator.  And you inverting the bloody thing didn't help matters much!

[Kilmatead]

...this -I- is totally useless. It doesn't have to do anything with windows desktop search indexing. I don't know what index they are referring to!

To answer this, we have to go back 159 years to the history of the Crimean War and the incompetence of some of the individuals who were in it and precisely why Into the valley of Death rode the six hundred.

The Light Brigade was commanded by Lord Cardigan who, by all accounts was a bit of an old mug, as the Irish would say (I'll be playing his part in this little stage-play).  His immediate superior was the evil Lord Lucan (and trust me, to the Irish, this guy epitomises Evil) who was in overall command of the cavalry, and of particular importance, the Heavy Brigade.  Our fearless leader Nikos Bozinis will be portraying this part, for reasons as will become obvious.

Not to recount the whole story, but basically a certain Lord Raglan sent down an order from on high for reconnaissance to be done on some Russian artillery, and Lord Lucan dispatched the intrepid Cardigan to do the investigating, as the Light Brigade (the hint is in the name) were a lightweight cavalry unit which relied more on speed and manoeuvrability than having actual weaponry (above the stock sabres) or decent armour - and thus were ideal to just run around and "have a look at what the Cossack lads were up to", as they say.  

Now there is some question as to the accuracy with which the orders were relayed, and a suspicious undercurrent of Cardigan being Lucan's somewhat hated brother-in-law, so exactly why what occurred isn't known, but effectively the Light Brigade was ordered to charge (not merely reconnoitre) under heavy enemy fire the battalions and artillery at the other end of the valley - crucially, under the impression that the Heavy Brigade would be following along after to supply much needed support after the Light Brigade cut its swathe.

So, Cardigan (that would be me) goes rushing in spending a quiet evening investigating the -I- attribute and what it might have to do with object indexing states... making the honest assumption that Lucan (that would be Nikos) had not only explained its addition clearly in the first place, but would indeed be following up with support in case I ran into trouble figuring this stuff out.

Now, remember what I said about incompetence and evil?  As it turns out, Lord Lucan dispatched Cardigan on a fool's errand, and more importantly, never even clarified what the Heavy Brigade was going to do once it entered the valley, never mind the reckless sallying done in the name of MSDN by the brave Cardigan.

Had Lord Lucan actually read the documentation of the FILE_ATTRIBUTE_NOT_CONTENT_INDEXED attribute, he would have realised that this is not actually an attribute of descriptive status (the reconnoitring I was under the impression I was doing), but rather a property directive explicitly commanding that "The file or directory is not to be indexed by the content indexing service", in other words, it's the inverted state of "Allow this file to have its contents indexed in addition to file" checkbox that's available when you right-click on an object to check its Properties, and you select "Advanced":



Lord Lucan will share Lord Cardigan's notation that the option explicitly uses the term "Allow" (which means actual status indeterminate), it does not state it as a positive (or a negative) indicator, just that it's a directive which "may allow" the indexing service to look you in the eye before it skewers you on the pikes of the Cossack battalions!

So, while I'm out there taking artillery fire...

Flash'd all their sabres bare,
Flash'd as they turn'd in air
Sabring the gunners there,
Charging an army, while
All the world wonder'd.

...I was doing so with the foolhardy notion that the evil Lord Nikos Lucan actually had the Heavy Brigade not merely out of the redoubt, but indeed providing actual assistance on the battlefield...

Plunged in the battery-smoke
Right thro' the line they broke;
Cossack and Russian
Reel'd from the sabre-stroke
Shatter'd and sunder'd.

My optimism grew with every confident swash of sword, and buckling of the - well, buckles, I guess - that I would see the banners and flags of compatriots in arms coming to my rescue... but...

Then they rode back, but not,
Not the six hundred.

See that?  Not the six hundred!  Those were my lads out there.    What the heck was Lord Lucan doing?  Taking a long leisurely lunch on his accolades?  Indeed, as always from the sources of both ignoble and literary flames, my ire doth produce something inexplicable yet strangely stylish in a Cardigan vs. Anorak sort of way.  :roll:

The charge of the Light Brigade continues to be studied by modern military historians and students as an example of what can go wrong when accurate military intelligence is lacking and orders are unclear.

(Now, in closing, if that doesn't give Fred more than a chuckle, my Sunday will have been in vain, and my free-association metaphors dead in the wind.  And perhaps next time Nikos will actually read the documentation before he adds an attribute "just because it has something to do with WDS" and he thought that would be "cool" to have that sort of convenient tie-in! )

[drac]

Kilmatead,

As usual, and interesting, entertaining, educational commentary on a technical issue.  And while I enjoyed reading your post, when I was done, I was not really sure what Nikos (Lucan) had done.  I understand it is related to the I attribute but I am not clear what the issue is and why it is so severe as to warrant the time and effort you put into your literary dissertation.  So for dolts (I might as well beat you to it) like me who are not as intimately familiar with the workings of X2, could you please include a “Cliff Notes” version of your commentary.

[Kilmatead]

I am not clear what the issue is and why it is so severe as to warrant the time and effort you put into your literary dissertation.

This is just what I do - I enjoy taking little things and blowing them all out of proportion and (when I'm lucky, like today) blowing the dust back in Nikos' face with spiteful (spitful?) glee.    Besides, considering that the 19th-century was obviously so superior to anything humans have done since, it behoves us to remember that having satellite technology in our pockets hasn't made any of us "better people".  Indeed, there's a case to be made that the human perception of progress is actually a retrograde illusion, and the resulting entropy has us by the neck with our feet dangling as it is now ("The flocke goeth to wrecke and vtterly perisheth" - Ephraim Udall).  Literary dissertations always warrant time and effort no matter how insignificant their source events - that's what makes them fun - you never know where you'll end up.  :D  (Besides, it's Sunday - what else are they for than castigating man's aspirations?)

As befits the lighthearted nature of the post, the hint is in the original quotation of Nikos':

It doesn't have to do anything with windows desktop search indexing. I don't know what index they are referring to!

Essentially it means (as I said above, once you wash away the literary drivel) that Nikos rather neglected to read the documentation pertaining to FILE_ATTRIBUTE_NOT_CONTENT_INDEXED before he added it to x2's [S]tock columns (under the long-shot belief that it might prove useful to someone).  He added this at the same time as the WDS integration, no doubt thinking himself clever and (probably!) rather hoping no one would notice his rather complete lack of research on the matter.

Is this important?  Not really.  Considering the few minutes it took me to create the meaningless utility it inspired, there's not much to it - except to say that it would be nice if these little wild-goose chases he sends me on at least started with a decent murder mystery or something, and not some flunky attribute that clutters up the display.  If I were a paranoid chap, I'd find it metaphorically suggestive that glancing at attribute columns these days makes me self-conscious about all the 'eyes' that are suspiciously watching me desecrate their demesnes.

And before anyone gets upset about me saying "the 19th-century was obviously so superior to anything humans have done since", I'll note that the 15th-century was obviously superior to the 19th, the 12th superior to the 15th (if for nothing else, tonal-scale music), and then probably the 3rd or 4th centuries qualify for accolades as well, depending on how you nitpick, and whether you think St. Augustine was a just a delusionist prat or one of the first to document that a wayward youth need not necessarily lead to wrack and ruin (unless you're really lucky ), and in-so-doing completely transfigure our western perception of the "self" through biography, and all the writings that followed.

[nikos]

without spending time to reproduce your hypothesis, I grant you that you may have found the true meaning of this I attribute. Can you explain to me why these two files in the same folder have it different?

Code: Select all

Name	Size	Modified	Attributes
trialpay_wall.htm	11.0 kB	22/07/2010 17:25:42	---A-----I-
testalert.htm	886 B	28/05/2009 17:24:20	---A-------

sure enough, WDS cannot find text in the one called I, but why did it index the one HTM and not the other?

[Kilmatead]

...but why did it index the one HTM and not the other?

...personally I don't think the windows indexing service is very robust or intelligent - it does have a strange habit of skipping things - hence the proliferation of advice from people about forcefully rebuilding the index completely fresh from time to time if you rely upon it.  If these .htm's had sibling folders, it would be interesting to see if they're treated similarly or not...

(And when you start asking me for help, the world must truly be coming to an end!  Is that the 3rd horseman of apocalypse I hear trotting by on his way to Bangladesh?  One more horseman to go...)

[nikos]

you studied the subject far and longer than I did, so you are the expert
another failing of this "I" is that files that are in non indexed folders they don't have it. The parent folder itself doesn't have it to show whether it is indexed or not... which is the reason why I gave up on it originally, before you started your dissertation on the subject