Search Results Weight By File Type?

We are crawling a series of folders using the NT CWS. Some folders contain primarily HTML content while others contain binaries (PDF, DOCs). The folders are split into "Web Content" and "Publications" at the root -- so they are already segregated by file type. Because the binary documents are relatively large and contain a lot of text (meeting minutes, etc), they seem to rise to the top of many search queries. Not sure if the search engine is simply counting the number of term matches or using some other algorithm for relevance. I've fiddled with the Search Results Manager - Banner Fields weights without much success.
Can anyone recommend a strategy that will weight our HTML documents higher than PDF/DOC files? I know through the Search API, we can restrict search to particular folders but that is different than actually affecting the weighting. Ideally, want to show any HTML docs and then show binary docs, below.

Bring up the search HUD, use the "+" drop down menu to add an "other metadata" criteria choose file name and type in the extension that you are looking for.
RB

Similar Messages

Search results not returning file names correctly

Have an onsite SharePoint Server 2010 Enterprise that contains a number of documents in a document library all containing the word "south" in their title. When a search is run with the keyword "south" all 3 files are shown in the results but only one displays
the correct title. The other 2 display alternate text (in this case the company name). The links all 3 results are correct, however the title of results is incorrect for 2 out of the 3.
I have duplicated these results using Office 365 and another on site SharePoint 2010 server.
I have duplicated these results converting the documents to DOCX format under Word 2007 and Word 2010.
Obviously, this is an issue with the files, however any ideas where to start looking to uncover where the SharePoint Server is pulling the information from?
In the end I need all 3 results to display the current name of the files it finds, not simply some random text with the files from 2 out of 3 of the results.
Thanks
Robert Crane
www.ciaops.com

Could it be because of this?
Pasted from here:
http://bpostutor.com/post/Hidden-SharePoint-2010-Feature-Changes-Document-Titles-in-Search-Results.aspx
Hidden SharePoint 2010 Feature Changes Document Titles in Search Results
SharePoint 2010 has a interesting feature which you may not know about. It's called Optimistic Title. It's part of the Office Search engine within SharePoint. What it does is determine a new, hopefully more relevant title for your documents
to be displayed in your search results based on document properties or the actual contents of the document (i.e. Text within the file). As you might expect this is closely tied to the Office document formats such as Word, PowerPoint, Excel, OneNote,
and Visio. Your end users may report that the titles that they see for search results differ greatly from the file name or the actual title of the document. This is particularly evident with PowerPoint files where the name of the first slide
is often used. The behavior is not entirely predictable. Different results can be expected from Office 2007 and Office 2010 created files and even those created in earlier versions of Microsoft Office.
If you want to change this functionality you need to actually go and edit the registry on your Search role server(s) within your SharePoint farm, restart the osearch14 service and then do a full crawl. The key you want to modify is the EnableOptimisticTitleOverride.
The default setting is 1. Change it to 0 to disable the feature.

Write diadem search results to text file

Is there a solution within DIAdem for exporting specific information from a search result into a text file? For example: I perform a search which returns all of the files, groups, and/or channels which I want to work with; now I would like to have, say, the file paths of all the results of my search exported to a text file. Is this possible? I could not find a direct way of doing this, but perhaps a script already exists for doing stuff like this?

Hi mrclary,
I did not have this code handy, but I was able to whip it up for you. Let me know if you have any questions.
Brad Turpin
DIAdem Product Support Engineer
National Instruments
Attachments:
Search Result Paths.zip ‏1 KB

When doing a Command+F search, can I EXCLUDE file types?

I need to find a logo called "5 Star". I know it is either a tiff or eps file.
I search brings 3000+ items.
Most of these are internet cache files, fonts and PDF files.
Can I EXCLUDE a file type when I do a command+F?
What happened to the "IS NOT" paramater in the find?
I see Name;
Is
Starts With
Ends With
Contains

Yes. If you use "Apple Command -f" , you can specify all sorts of search parameters (see OTHER) in an exclusive manner. You can only use the parameter choices which are provided, but for example, you can use "NAME + contains" "star" with KIND as "other" and specify "eps." or ".tif." Or use NAME + ends with ".eps."
I am not sure where the exclusion AND NOT feature went. Each list of possibilities is associated with the pre-canned options.

Search or sort by file type?

has anybody figured out how to search/sort by file type (.psd, .cr2, Etc.)?
I've been looking, but I can't find it.

Bring up the search HUD, use the "+" drop down menu to add an "other metadata" criteria choose file name and type in the extension that you are looking for.
RB

Spotlight doesn't find search results in .HTML file contents

When I use Spotlight to search, it doesn't find any .HTML file contents. If I rename an .HTML file to use a .TXT extension, the contents are found. Is there a way to enable Spotlight to find contents in .HTML files?

I remembered that used to be problem, so I just checked on my local version of my site, and Spotlight found the files on the basis of content without a problem. So I took a look at my installed mdimporters in /Library/Spotlight, to see what may have led to this happy turn of events. I thought perhaps the iWeb mdimporter might be doing it, but was unable to find a reference to HTML in its type declarations. I then took a look at in /System/Library/Spotlight and discovered that RichText.mdimporter has a declaration for public.html as a content type. You should have that mdimporter installed by default, but you might take a look to make sure it is there.
If the mdimporter is there, but the html files aren't being indexed, it could be that whatever program you are using to create them isn't giving them the appropriate content type. I code by hand using TextEdit, which saves them with the following metadata entry:
kMDItemContentType = "public.html"
kMDItemContentTypeTree = (
"public.html",
"public.text",
"public.data",
"public.item",
"public.content"
They are then evidently indexed without problem by the RichText.mdimporter. You might try opening one of your html docs in TextEdit (as plain text), resave, and see if Spotlight then picks up the content of the file. Of course, the content means the words that people see when they look at the page in a browser. If the content you are talking about is stuff that is inside an html tag, you are out of luck. I know of NO way to get Spotlight to find that. EasyFind will though, but it is a brute force search and will take awhile, even when restricted to a particular folder.
Francine
Francine
Schwieder

Navigation Bar Search results in jar:file:/// search

When typing in anything (intended to be a search, default search engine SHOULD be Google), my search is directed to a "File Not Found" Page with this as the search criteria: "jar:file:///C:/Program Files (x86)/Mozilla Firefox/omni.jar!/chrome/en-US/locale/browser-region/region.properties[SEARCH CRITERIA THAT I TYPED]" (without the brackets or quotes)
I have tried several times to change the keyword.url, but each time I close firefox or restart it, the search changes back to searching in the directory posted above.

I remembered that used to be problem, so I just checked on my local version of my site, and Spotlight found the files on the basis of content without a problem. So I took a look at my installed mdimporters in /Library/Spotlight, to see what may have led to this happy turn of events. I thought perhaps the iWeb mdimporter might be doing it, but was unable to find a reference to HTML in its type declarations. I then took a look at in /System/Library/Spotlight and discovered that RichText.mdimporter has a declaration for public.html as a content type. You should have that mdimporter installed by default, but you might take a look to make sure it is there.
If the mdimporter is there, but the html files aren't being indexed, it could be that whatever program you are using to create them isn't giving them the appropriate content type. I code by hand using TextEdit, which saves them with the following metadata entry:
kMDItemContentType = "public.html"
kMDItemContentTypeTree = (
"public.html",
"public.text",
"public.data",
"public.item",
"public.content"
They are then evidently indexed without problem by the RichText.mdimporter. You might try opening one of your html docs in TextEdit (as plain text), resave, and see if Spotlight then picks up the content of the file. Of course, the content means the words that people see when they look at the page in a browser. If the content you are talking about is stuff that is inside an html tag, you are out of luck. I know of NO way to get Spotlight to find that. EasyFind will though, but it is a brute force search and will take awhile, even when restricted to a particular folder.
Francine
Francine
Schwieder

Search / Replace on other Files Types

When do a mass search using DW CS3 and search on a directory,
the search ignore certain file extension such as .properties. How
can I tell the DW to search on ALL files or modify the files
extension list that the program searches on?

Read up on "Regular Expressions"
"indonaught" <[email protected]> wrote in
message
news:ff892l$cv3$[email protected]..
> When do a mass search using DW CS3 and search on a
directory, the search
> ignore
> certain file extension such as .properties. How can I
tell the DW to
> search on
> ALL files or modify the files extension list that the
program searches on?
>

Export results to different file type

I am very curious, if you can successfully export to csv format when choosing for query result in sql dev 2.1.
1. Worksheet
2. Enter a Query
3. Press the first green arrow to perform the query
4. Navigate to the results with the mouse and right click to have the possibility to export to csv.
In 1.5 I got loads of exportb option but in 2.1 I dont get any option. I upgraded to 2.1 because I needed to connect to a sql server as well as Oracle but since upgrading have had no joy in exporting the results from either, any advice?

So the context menu is still OK, you just don't get anything after that? You might be running into [this one|http://forums.oracle.com/forums/thread.jspa?threadID=873423].
As for SQL Server, 2.1 didn't allow for export, but 2.1.1 does.
K.

Word files appearing as search results

Hello,
I thought I'd come across a topic like this before, but I
can't seem to find one now. Here's my situation...
I have some MS Word files included in my RH project for users
to download. They are located in their own book -- let's call it
"Forms." I would like the "Forms" book to be the only place that
users find those files.
What I didn't realize was that links to download those files
would appear as search results. They sort of interrupt the other
search results, and the file names that I've given them were not
intended for clients. I can rename all of them, but I'd prefer to
remove them from the search results.
There must be a way to do this...right?
If it helps, I'm using RoboHelp HTML X5.0.2, and my Primary
Layout is WebHelp Pro.
Thanks! I've found a lot of help throughout these
forums.

Hello,
You have to use Archiving object - "FI_ACCRECV" to archive Customer Master Data (BP).
When general data for customer master records is archived, the system automatically checks whether the following conditions have been met:
- The deletion flag is set.
- The customer is not locked for deletion.
- Dependent data of the following types no longer exists in the system:
1. Business partner of the customer
2. Company code-specific data (FI data)
3. Sales data (SD data)
4. Transaction figures
5. Special general ledger transaction figures
6. Open items
7. Cleared items in the archive
8. Customer/Vendor links
For the archiving object all the required configurations should be maintained.
If you don't have archive server in place then do not take the option of storing (Just archive and delete the data).
-Thanks,
Ajay

How does SharePoint determine files are duplicates in search results?

In the search results, some files are grouped as duplicates (a hyperlink view duplicates appears under the search result).
How does SharePoint determines that 2 files are duplicates?
How does SharePoint determines the one that is shown in the search result (the 'main' file)?
Can we influence both?
Patrik | My Blog

I don't know if this helps, but I've been looking into the same problem that's come to light a few times during troubleshooting customised deployments of SharePoint recently. This is my understanding so far (paraphrased from http://blogs.technet.com/harikumh/archive/2008/11/14/some-interesting-facts-about-sharepoint-2007-search.aspx):
Document similarity or matching for the purposes of identifying duplicates is based only on a hash of the content of the document. None of the file properties are used in calculating the hash (i.e. things like filename, author, create and modify dates are not used). The SQL table MSSDuplicateHashes in the SSP’s search database holds all the 64bit hashes necessary to determine if one document is a near-duplicate of another against each indexed document. This table is read while doing a search to determine duplicates if removal of duplicates is enabled.
Steve

Yaourt doesn't show AUR searching results

yaourt doesn't show AUR searching results. When I type for instance:
$ yaourt splashy
I get:
1 archlinuxve/splashy 0.3.10-1 [installed]
A next-generation user-space boot splashing system for Linux systems
2 archlinuxve/splashy-themes 0.4-1 [installed]
Splashy Themes
google_ad_client="pub-3170555743375154";google_ad_width=468;google_ad_height=60;google_ad_format="468x60_as";google_color_border=
"ffffff";google_color_bg="ffffff";google_color_link="0771A6";google_color_url="99AACC";google_color_text="000000";//-->
</script> <script type="text/javascript" src="http://pagead2.googlesyndication.com/pagead/show_ads.js"></script></div><div
id="sub_nav"><ul><li><a href="http://archlinux.org/mailman/listinfo/aur-general">Discussion</a></li> <li><a href="http://bugs.archlinux.org/
index.php?tasks=all&project=2">Bugs</a></li> <li><a href="packages.php">Packages</a></li> <li><a href="account.php">Accounts</a></
li> <li><a href="index.php">AUR Home</a></li></ul></div></div><div id="lang_login_sub"><span id="lang_bar"><ul><li>Lang: </
li><li><a href="/packages.php?setlang=en" title="English">EN</a></li> <li><a href="/packages.php?setlang=pl" title="Polski">PL</a></li>
<li><a href="/packages.php?setlang=it" title="Italiano">IT</a></li> <li><a href="/packages.php?setlang=ca" title="Català">CA</a></li>
<li><a href="/packages.php?setlang=pt" title="Português">PT</a></li> <li><a href="/packages.php?setlang=es" title="Español">ES</a></li>
<li><a href="/packages.php?setlang=de" title="Deutsch">DE</a></li> <li><a href="/packages.php?setlang=ru" title="Русский">RU</a></li>
<li><a href="/packages.php?setlang=fr" title="Français">FR</a></li></ul>
==> Enter n° (separated by blanks, or a range) of packages to be installed
==> ----------------------------------------------
==>
It doesn't matter what I type after the command. I always get the header of AUR site as a result. How can i solve the problem? I have already done:
yaourt -Syu
and
pacman -Sc
few times.
The problem concerns 0.9-2 and 0.9.1-1 as well.

wain wrote:
ok your output is ok...
Please, try this:
wget -q "http://aur.archlinux.org/packages.php?setlang=en&do_Search=SeB=nd&L=2&C=0&PP=100&K=pacman" -O - | grep -A 2 "<a href='/packages.php?ID=" | sed -e "s/<\/span>.*$//" -e "s/^.*packages.php?ID=.*span class.*'>/aur\//" -e "s/^.*span class.*'>//" | grep -v " " | grep -v "^--"
it should be like that.
I tried that and I got:
google_ad_client="pub-3170555743375154";google_ad_width=468;google_ad_height=60;google_ad_format="468x60_as";google_color_border="ffffff";google_color_bg="ffffff";google_color_link="0771A6";google_color_url="99AACC";google_color_text="000000";//--></script> <script type="text/javascript" src="http://pagead2.googlesyndication.com/pagead/show_ads.js"></script></div><div id="sub_nav"><ul><li><a href="http://archlinux.org/mailman/listinfo/aur-general">Discussion</a></li> <li><a href="http://bugs.archlinux.org/index.php?tasks=all&project=2">Bugs</a></li> <li><a href="packages.php">Packages</a></li> <li><a href="account.php">Accounts</a></li> <li><a href="index.php">AUR Home</a></li></ul></div></div><div id="lang_login_sub"><span id="lang_bar"><ul><li>Lang: </li><li><a href="/packages.php?setlang=en" title="English">EN</a></li> <li><a href="/packages.php?setlang=pl" title="Polski">PL</a></li> <li><a href="/packages.php?setlang=it" title="Italiano">IT</a></li> <li><a href="/packages.php?setlang=ca" title="Català">CA</a></li> <li><a href="/packages.php?setlang=pt" title="Português">PT</a></li> <li><a href="/packages.php?setlang=es" title="Español">ES</a></li> <li><a href="/packages.php?setlang=de" title="Deutsch">DE</a></li> <li><a href="/packages.php?setlang=ru" title="Русский">RU</a></li> <li><a href="/packages.php?setlang=fr" title="Français">FR</a></li></ul>
instead of right output.
I have the same problem on my two diffrent mashines. I don't recall that I changed anything in config files. Btw, thanks for interest.

"File Type Codes" in OS-X

I've just learned that these codes were probably discontinued after OS-9. Here's why I care:
I have many QuickMail messages saved to various places on my hard disk. While I can still run QM, this surely will not be possible much longer. What I'd like to do is to change all those files to TEXT files, so I can always count on being able to read them. I thought I could use spot light to search for the QuickMail file type code then somehow change it to TEXT [or to that of Apple Mail, my present client], but that was OS 9 and other utilities, not X. Another problem is now I don't even know the QuickMail file type code.

Regarding QuickMail, AFAIK the format was not plain text (see this format description). If you have serious stuff to convert (rather than just a few messages), take a look at Email Alchemy. You may find additional options on the eMailman site.
Regarding type and creator codes, there's quite a bit of confusion about them. These are two 4-character codes used by Mac OS to determine what a file is (type) and what application created it (creator). They are metadata, and they are not "embedded" in the file; rather, they are maintained in a distinct database. Mac OS X still used them (alongside other methods), but, in Tiger, they were officially replaced by UTIs. They were still honoured by the OS until SL, when OS support for the creator code was dropped. (Sparking a fascinating controversy among Mac afficionados.) Both codes are still there, and you can see or change them with a suitable app (eg, Quick Change), but only the type code does anything useful -- if no other file metadata is available (such as filename extension), Mac OS X will use it.
Keep in mind that the type code is metadata -- changing it does not change anything in the file itself. For instance, you can change the type code from TEXT to JPEG, but that doesn't mean you've magically transmogrified plain text into a picture. You've merely told the OS to treat this file as a picture, which will cause an error.

Write the ldapsearch results into a file

We are using ldapsearch from the command line and want to write the search results into a file. How can we do this?

What OS are you using - If Unix / Linux then use the standard redirection > or >> facility.

SharePoint Search Results Not Showing all Results but in Refiners showing that Type of File is existing

Hi Team,
We have configured the Search Results and Search Refiner webpart in our Page. In few cases we are seeing that in Search Refiner we seeing the type of file is existing. But the respective file type files are not appearing in Search Results. When we click
on respective Search Refiner Type then the Search Result showing the respective value.
Could anyone help us on this.
Thanks, Dinesh
Dinesh Pulugundla, Microsoft Certified Technology Specialist.

Hello,
We are currently looking into this issue and will give you an update as soon as possible.
Thank you for your understanding and support.
Regards,
Forum Support
Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact
[email protected] .
Rebecca Tu
TechNet Community Support

Search Results Weight By File Type?

Similar Messages

Maybe you are looking for