ADL (Automatic Directory Listing) Search
By Ray Blake (Ray_99) with additional valuable input from The Great Wizard.
We're an odd lot in these hubs, aren't we? We're completists, all desperate to find that last episode of the series, or the special episode that's thought to be lost. Or maybe we love a show from which we have just one episode. No one else seems to have heard of it and we'd kill for more episodes.
Given our funn 151r1715b y ways, we can't help but regard the search tool in DC++ as an annoyingly blunt instrument, defying the use of Boolean operands, for instance (it just treats them as extra search words!) and returning a limited number of hits from any given user. You also get hundreds of unwelcome hits for every useful one because the function does not insist on all terms being found (Ask The Great Wizard how he feels about Rupert Holmes, as opposed to Sherlock!) [TGW note: I'm not crazy about Oliver Wendell Holmes or that well-endowed fellow, either.] Plus, you can't schedule searches to test automatically for new hits every week, say.
ADL (Automatic Directory Listing) Search isn't a panacea, but it's a huge leap forward in usability. Although available from version .24, ADL Search is best implemented in the latest versions, v.242 and v.25. When introduced, it was completely undocumented. Even now, the 'what's this' button on the ADL screen gives misleading information, as we'll see. So, why am I such an enthusiastic user of the new feature? Let's walk through what it does and how to use it and you'll find out along the way.
What ADL search does is to act like a search engine on all the file lists you download. It can store multiple search terms and will scan each file list before it arrives on your screen, presenting any 'hits' in a separate folder at or near the top of the share called "<<<ADLSearch>>>" (in v.242 and v.25 it actually bolds this folder name too). It then goes on to show you all of the rest of the list too, so you lose nothing at all.
To access the function, look for its icon on the toolbar. You'll find it just to the right of the old Search icon, and it looks like a magnifying glass in front of a manila file. Alternatively, you can access it from the 'file' menu. Either way, it will open up a screen with a lot of white space waiting for you to set up your searches. The headings should be:
Search string | Source Type | Destination | Active | Min Size | Max Size
(NB: If you're using v.25, you won't see the 'Active' column, but will instead see a checkbox to the extreme left of the search string once you've actually entered one.)
To enter an item, hit the 'Add' button and complete the simple form. Here's some help on the fields:
Search String: What you enter here will be used as the search term. Note that it will look for full string matches only. For instance, "Sherlock Holmes" will only hit on both words, together, and not on "Rupert Holmes", nor on just "Holmes". If you think that's restricting, just remember you can set up lots of these searches individually. For example, you can set up separate searches on name variations like "Vonnegut" and "Vonegut."
Source Type: The pull-down list gives you a choice of Filename, Directory or Full Path. To be on the safe side, go for Full Path, which will return a hit whether the search string is found in the folder name, the filename or both.
Min/Max Filesize and Size Type: Should be self-explanatory. The only use I can think of for this in an OTR context is in making sure you get a high enough bitrate. But, like me, you'll probably always leave these fields empty.
Destination Directory: Defaults to ADSL Search and I always leave it there. You could, though, specify different directory names for different search terms if you build a long list of them so that the hits come in pre-organised for you (you sad person!) [TGW note: Ray: I am one of these "sad people." I search for about a dozen different terms and I LIKE them organized into folders..<G>]
Finally, you'll see a check box inviting you to check it to activate the search or empty it to inactivate it. This allows you temporarily to suspend searching on a string without removing the term permanently. Try as I might, I can't think of a scenario in which you'd want to do this, but it's there if you need to.
Hit the OK button, and you'll have an entry in the list. Repeat this process for each of the other search terms that interest you. Any of these can be removed or edited in the conventional way by highlighting the line and clicking the edit(v.242)/properties(v.25) or remove button (right-clicking worked in v.24, but not in v.242, for some reason. It is working again in v.25)
[TGW note: Be
creative and thorough. Remember the limitations of the engine. It does not know
that "LOTR" is "Lord of the Rings," or that "
Now, don't spend time wondering what the 'Move Up' and 'Move Down' buttons mean. I'll explain in a moment. Instead, click the 'What's this?' button and read the text.
The third paragraph deserves a mention, because it is misleading. It says:
"Note also that the order of the searches is important. The top item is matched first, then the second, etc. If a search is matched, the ones below will not be tried. You can alter the order with 'Move Up' and 'Move Down', which will move the current selection."
To me, this sounds like it checks the entire list term by term. Once it finds a search term, all subsequent ones are ignored. In other words, if you had "Sherlock Holmes" as Term number one and "Blue Carbuncle" as Term number 2, then a file with the name "SH - Blue Carbuncle" would not get a hit if there was any hit at all on "Sherlock Holmes". It's not me, is it? That IS how it reads, right?
Well, relax, because that's not what it means, and the Blue Carbuncle hit would be returned in our example no matter what else was in the file list. What it means is that each individual file is tried with all the search terms in order, and as soon as one hits, it stops trying the others ON THAT FILE and moves to the next file to start again. The only circumstances I can see that being any kind of a worry are if you set up a separate 'Destination Directory' for each search term and expect the same file to turn up in more than one. It won't.
Now, back to those 'Move Up' and 'Move Down' buttons, which I thought were about as useful to the average hub user as a tissue paper oven glove. The Wiz has educated me, though. The search order is important in that the terms you expect to find most frequently should be listed first. This speeds up the search markedly. It's a simple principle of database design and query structure. Since a search target is no longer reviewed after a single hit, the search engine will proceed more quickly if common terms are listed first.
Hang on now, we're nearly there. Just one last function to cover and we're done. When you download a share list with ADL Search terms in place, you'll expect to see the "<<<ADLSearch>>>" folder now and again. When you open it and see a file that you want, you can set it to download in the normal way, just as though it really were in a folder on the user's drive called "<<<ADLSearch>>>". However, if you want to know what other files he has sitting alongside that one, right-click on it. You'll see an option called 'Go to directory' which will take you to the point in the file list where the file actually sits. Neat, huh?
I've saved one of the best features of ADL until last; unlike regular searching, ADL use will NEVER get you kicked for 'spam searching' however much you do it. That's because all the searching activity happens on your PC only, once the file list is on board.
Well, that's it. I hope you get as much use out of using the new ADL as I have, and make sure you pass on the knowledge. I'll keep this article in my share, near the top with a descriptive folder name, in case you want to refer others to it.
|
