Query language
Type words or phrases in the query form, then click the button to perform the query, you can search for any word or phrase in the Web site (for example, the "Executive Query" button in the query table example). This section will introduce the following topics:
Logic and similar operators: Show how to perform more accurate queries by inserting logic and similar operators.
Wildcard: Help you find a page that contains words similar to the word gives.
Free text query: Explain how to express a query based on the phrase, not to use precise words.
Vector Space Query: Explain how to get query results matching words and phrase lists.
Attribute value query: Tell you how to query the properties value of the file.
Sample: Example of various queries is given.
Attribute name list: List and instructions that can be used for querying properties.
Search Generate a list of files that contain a word or phrase (regardless of the word or phrase anywhere in the file). The following list gives the rules for the query:
Continuous words are treated as a phrase, and they must appear in the same order in the matching document.
The query is not case sensitive, so you can type inquiry with uppercase or lowercase.
You can search for any words, but not included in the exception list (for English, including A, AN, AND, AS, and other words), they will be ignored in the query.
The words in the exception list will be treated as a placeholder in the phrase to perform similar queries. For example, if you query "Word for Windows", the result will give "Word for Windows" and "Word and Windows" because for is a virtual word that appears in the exception list.
The punctuation symbol will be ignored at the time of search, such as the number (.), Colon (:), semicolon (;) and comma (,).
To use special treatment characters in the query, such as &, |, ^, #, @, $, (,), please check with query (").
To search for words or phrases containing quotation marks, please use quotation to enclose the entire short screen and use two quotes to enclose the words enclosed in quotes. For example, "World-Wide Web OR" "Web" "will search for World-Wide Web or" Web ".
You can insert logic operators (AND, OR and NOT) and similar operators (Near) to specify additional search information.
Wild characters (*) can match words with the prefixed prefix. Query ESC * You can match the entry "ESC", "Escape", and more.
Free text query may not specify queries as query syntax.
You can specify a vector space query.
You can execute an attribute value query for ActiveX "! (OLE) and file properties.
Logic and similar operators
Logic and similar operators can create more precise queries.
Search target
Example
result
Two entries are included in the same page
Access And Basic- or -access & Basic
Contains pages of words "Access" and "Basic".
One of the same page contains one of the two entries
CGI or isapi- or -cgi | isapi
Contains pages with words "CGI" or "Isapi".
Contains the first entry, but does not contain the second entry
Access and not basic- or -access &! Basic
Contains words "Access", but does not include a page of the word "Basic".
Page that does not match a property value
Not @Size = 100- or -! @size = 100
Size is not a 100-byte page.
The same page contains two entry and close to each other.
Excel Near Project- or -excel ~ Project
A page containing words "Excel" and "Project" that are close to each other.
prompt:
You can add parental arc in the query expression, and the part of the brackets in the expression is performed first than the other parts of the query. Using dual quotes (") can indicate that the logic or Near operator keyword needs to be ignored in the query. For example," Abbott and Costello "will match the page containing the phrase, not the page that matches the logical expression. As an operator Word and the word is a virtual word.
The Near operator is similar to the And operator. Near also returns a page containing two words. However, the NEAR and AND operators differ from the NEAR to see if the words are close. That is, the level of the page containing a relatively close to the search word will be greater than or equal to the level of the page remotely separated by the word. If the search for words are more than 50 words, this page will be set to zero
In the content query, the NOT operator can only be used after the AND operator, which only uses a page that matches the front content constraint. For attribute values queries, the NOT operator may not be used with the And operator.
The priority of the AND operator is higher than that. For example, the first three queries below, but the fourth different: a and b or cc or a and bc or (a and b) (c or a) and b
Note that in all languages, symbols (&, |,!, ~) Are the same as English keywords and NEAR (INDEX Server support). If the browser is set to the following six languages, the localized keyword can also be used.
Language
Keyword
German
Und, Oder, Nicht, NAH
French
ET, OU, SANS, PRES
Spanish
Y, O, NO, CERCA
Dutch
En, of, NIET, NABIJ
Swedish language
OCH, Eller, INTE, N 腞 A
Italian
E, O, NO, Vicino
Note that the Near operator can only be used for words or phrases.
Wildcard
Wildcards can help you find pages that contain words similar to your words.
Search target
Example
result
Word with the same prefix
Comput *
A page containing a word with "Comput", such as "Computer", "computing", etc.
Words based on the same word
fly **
Contains a page based on the same word "fly", such as "flying", "flown", "flew", etc.
Free text query
In the free text query, the query engine finds the page with the best matching of words or phrases. This kind of query is matched, not a precise word. In the free text query, logic, similar, and wildcard operators will be ignored. Free text query is prefixed in $ contents.
Search target
Example
result
Match the file file
$ Contents How Do I Print in Microsoft Excel?
Refer to the page of Printing and Microsoft Excel.
Vector space query
Query engine supports vector space query. Vector query returns a page of matching words and phrase a list, each page specifies the level of the page and query matching.
Search target
Example
result
Page containing specified words
Light, BULB
Contains files with the most matching words with search words
Page containing weighted prefix, words and phrases
Invent *, Light [50], BULB [10], "Light Bulb" [400]
Contains files, words "Light", "BULB", and phrase "light bulbs" with "invent," prefixed (entry)
The components in the vector query are separated by commas.
Components in vector queries can be weighted with [Weight] syntax.
Vector query returned page does not have to match all the entry of the query.
When the result is sorted by level, the vector query is best.
Attribute value query
With an attribute value query, you can find a file that contains attribute values that match the standard. The attributes that can be used to query include basic information of the file (such as file name, file size) and ActiveX property (included in the Document Summary), and the ActiveX property is created by the ActiveX application and stored in the file. Here are two types of properties queries:
Relationship property query includes "AT" character (@), attribute name, relational operator, and attribute value. For example, you want to find a file that exceeds a megabyte file, you can perform query @size> 1000000.
Regular expression queries include the regular expression of the number symbol (#), attribute name, and attribute value. For example, you want to find all video (.avi) files, you can perform Query #FileName * .avi. Regular expressions do not match the specific properties content (#Contents) and all (#all). The attributes that cannot be retrieved in the query cannot be used for # query, including HTML META properties that are not stored in the property cache.
This section contains the following topics:
Attribute name
Relational operator
Attribute value
Attribute name
The attribute name begins with "AT" (@) or number symbol (#), @ for relational query, # for regular expression query.
If the property name is not specified, it will be assumed to be @Contents.
The properties available in all files include:
Attribute name
Description
All
Match words, phrases, and any properties
Contents
File words and phrases
Filename
Name of the file
Size
File size
Write
The last revision time of the file
The ActiveX property value can also be used for queries, and most of the Web site files created by the ActiveX application can be queried by the following attributes:
Attribute name
Description
DOCTITLE
Title of the document
DOCSUBJECT
Document theme
Docauthor
Document author
Dockeywords
Keywords in the document
DocuComments
Note in the document
Attribute Name Complete list, see the list of property names later.
Relational operator
Relational operator is used in relational properties queries.
Search target
Example
result
Attribute values related to fixed values
@Size <100 @Size <= 100 @Size = 100 @Size! = 100 @Size> = 100 @Size> 100
Size and query matching file
Have all bit-specific attribute values
@attrib ^ a 0x820
Compressed file with archive properties
Have some bit more than possible attribute values
@attrib ^ s 0x20
File with archive properties
Attribute value
Search target
Example
result
Specified value
@Docauthor = Bill Barnes
File created by "Bill Barnes"
Value started with a prefix
#Docauthor george *
Author name is a file that heads with "george"
File with any extension of extension
#filename *. | (Exe |, DLL |, SYS |)
Files with .exe, .dll or .sys extension
Modified file after a date
@Write> 96/2/14 10:00:00
Documents modified after 10:00 GMT on February 14, 1996
Modified file after the relative date
@Write> -1d2h
Modified files within 26 hours
Match a vector vector
@vectorprop = {10, 15, 20}
The vector value is {10, 15, 20} ActiveX documentation
Each value is a vector matching vector
@vectorprop> ^ a 15
Each value of the vector value is greater than 15 ActiveX documentation
At least one value matches the standard match
@vectorprop = ^ s 15
There is at least one value greater than 15 in the vector value.
When using a regular expression query, make sure to bring (#) characters before the attribute value, otherwise bring "AT" characters. A equals (=) relationship operator is defined as a regular expression query.
File name (#filename) is the only property that supports the regular expression of the wildcard in the left side of the text.
Date and time values are YYYY / MM / DD HH: MM: SS or YYYY-MM-DD HH: MM: SS format. The first two and the whole time of the year can be ignored. If you ignore the first two digits of the year, less than or equal to 29 will be interpreted as the twenty-first century, greater than or equal to 30, will be interpreted as a twentieth century. All dates and times are GMT (GMT).
Relative to the date and time of the current time can be represented by minus or integer unit and time unit. The time unit is interpreted as: (Y) year, (m) month, (w) week, (d), (h) hours, (n) minute and (s) seconds. As an option, you can specify a triple millisecond value after the time expression. For example, 1997/12/8 10: 10: 03: 452
The currency value is x.y format. Here, X is an integer value of the amount, Y is a small value. The unit does not assume the value.
The logical value is: (t) or (true) corresponds to true, (f) or (false) corresponds to FALSE.
The vector (vt_vector) is represented as: the left bulk ({) begins, following the comma-separated value list, ends with the right bulk (}).
The single-value expression is represented by the vector as: a relational operator, and then (^ a) (corresponding to all) or (^ s) (corresponding to some).
The digital value can be a decimal or hexadecimal (add 0X in front).
The Contents property value does not support the relational operator. If the relationship operator is specified, no results will be found. For example, @ Contents Microsoft will look for documents containing Microsoft, but @ contents = Microsoft will not find anything.
Regular expression
The regular expression in the properties query is defined as follows:
In addition to the asterisk (*), a comma (.), Any character outside the question mark (?) And vertical line (|) is default to match yourself.
Regular expressions can be enclosed in quotation marks ("), if there is a space () or right bracket ()), you must be enclosed with quotation marks.
The role of characters *,., And?, Like they in Windows, the asterisk matching any character, the number of the junctions match (.) Or string, the question mark matches any single character.
Character | is a transduction character, after the character, the character has a special meaning:
(Start a group, there must be followed by it).
) End a group, the front must be (relative.
[Start a character class, there must be] (do not translate) relative.
{Matching of the start count, there must be} back.
} The matching of the end count must be {opposite.
Separate the OR clause.
* Match zero or multiple appearances of front expressions.
• Match zero or once in front of the previous expression.
Match one or more appearances of the previous expression.
Other characters, including |, match yourself.
The following characters between square brackets ([]) have special meanings:
^ Does anything except the following class. It must be the first character.
] Matching]. Only after ^, it will otherwise end the role of the class.
- Range operator. The front and back can be a normal character.
Other match yourself (start or end).
The following syntax is applied between braces ({}):
| {m |} Accurately match the M times of the previous expression. (0 | {m, N |} Match the appearance of the front expression from m to N times, including M and N. (0 To match *,., And?, Please enclose with parentheses (for example, | [*] Sample will match "* sample"). Query example Example result @Size> 1000000 Greater than a megabyte page @Write> 95/12/23 Modified page after this date Apple Tree Contains a page with phrase "apple trees" Apple Tree Equally @Contents Apple Tree Equally Microsoft and @Size> 1000000 Contains words "Microsoft" and greater than greater than one megabyte page "Microsoft and @SIZE> 1000000" Contains pages with specified phrases (different from above) #filename * .avi Video file (because the query contains the regular expression, use the # prefix) @attrib ^ s 32 Page with archive properties @docauthor = john smith Page created by the author $ Contents why is the sky blue? Match the query page @Size <100 & #filename * .gif GIF file greater than 100 bytes Attribute name list These properties are available in queries, and additional properties should be viewed by the web server. Friendly name type of data Attributes A_href DBTYPE_WSTR | DBTYPE_BYREF HTML HREF's text, this property name is created by Microsoft? Site Server, with the index server property name HTMLHREF. You can query, but you can't retrieve it. ACCESS VT_FileTime The last access time of the file. All (Not applicable) Search for each property of the string. You can query, but you can't retrieve it. Allocsize DBTYPE_I8 Disk size for file allocation. Attribact DBTYPE_UI4 File properties, in the Win32 SDK documentation. ClassID DBTYPE_GUID Object class identity, such as WordPerfect, Word, etc. CHARACTERIZATION DBTYPE_WSTR | DBTYPE_BYREF Description or summary of documents, used by Index Server. Contents (Not applicable) The main content of the document. You can query, but you can't retrieve it. Create VT_FileTime Create time of the file. Directory DBTYPE_WSTR | DBTYPE_BYREF The physical path of the file does not contain the file name. Docappname DBTYPE_WSTR | DBTYPE_BYREF The name of the application created the file. Docauthor DBTYPE_WSTR | DBTYPE_BYREF Document author. DocBytecount DBTYPE_14 The number of bytes of the document. Doccategory DBTYPE_STR | DBTYPE_BYREF Categories of documents, such as memo, plan or note. Docucharcount DBTYPE_I4 The number of characters of the document. DocuComments DBTYPE_WSTR | DBTYPE_BYREF Notes on the documentation. Doccompany DBTYPE_STR | DBTYPE_BYREF The company name of the document writing. DocucreatedTM VT_FileTime Document creation time. DOCEDITIME VT_FileTime Edit all the time used by the document. DochiddenCount DBTYPE_14 The number of hidden slides in the Microsoft® PowerPoint document. Dockeywords DBTYPE_WSTR | DBTYPE_BYREF Document keyword. DOCLASTAUTHOR DBTYPE_WSTR | DBTYPE_BYREF Recently edited the user of the document. DOCLASTPRINTED VT_FileTime The most recent print time of the document. Doclastsavedtm VT_FileTime The most recent save time of the document. DoculineCount DBTYPE_14 The number of rows containing the document. Docmanager DBTYPE_STR | DBTYPE_BYREF The name of the manager of the document author. DOCNOTECUNT DBTYPE_14 The number of pages with comments in the PowerPoint document. DocPageCount DBTYPE_I4 The number of pages of the document. Docparacone DBTYPE_14 The number of graphics of the document. DocPartTitles DBTYPE_STR | DBTYPE_VECTOR The name of the document section. For example, in Excel, some headings are the name of the electronic worksheet; in PowerPoint is a slide title; in Word for Windows, it is the name of each sub-document in the main document. DocuPresentationTarget DBTYPE_STR | DBTYPE_BYREF PowerPoint presentation target format (35mm, printers, videos, etc.). DOCREVNUMBER DBTYPE_WSTR | DBTYPE_BYREF The current version of the document. ā ?? 僐 ?????? 㑩 ??? painted ?? DOCSLIDECOUNT DBTYPE_14 The number of slideshows of the PowerPoint document. DOCSUBJECT DBTYPE_WSTR | DBTYPE_BYREF The subject of the document. DOCTEMPLATE DBTYPE_WSTR | DBTYPE_BYREF Document template. DOCTITLE DBTYPE_WSTR | DBTYPE_BYREF Title of the document Docwordcount DBTYPE_I4 The number of words of the document. FileIndex DBTYPE_I8 The unique identifier of the file. Filename DBTYPE_WSTR | DBTYPE_BYREF file name. Hitcount DBTYPE_I4 The number of hits in the file (word matching query). HTMLHREF DBTYPE_WSTR | DBTYPE_BYREF HTML HREF text. You can query, but you can't retrieve it. HtmlHeading1 DBTYPE_WSTR | DBTYPE_BYREF The text in the HTML document is H1 text. You can query, but you can't retrieve it. HTMLHEADING2 DBTYPE_WSTR | DBTYPE_BYREF The text of the HTML document is H2. You can query, but you can't retrieve it. HtmlHeading3 DBTYPE_WSTR | DBTYPE_BYREF The text in the HTML document is H3 text. You can query, but you can't retrieve it. HtmlHeading4 DBTYPE_WSTR | DBTYPE_BYREF The text in the HTML document is H4. You can query, but you can't retrieve it. HTMLHEADING5 DBTYPE_WSTR | DBTYPE_BYREF The text in the HTML document is H5 text. Can be inquired, ā ?? 僐 ????? 㑩? 疀 ?? but cannot retrieve it. HtmlHeading6 DBTYPE_WSTR | DBTYPE_BYREF The text in the HTML document is H6. You can query, but you can't retrieve it. IMG_ALT DBTYPE_WSTR | DBTYPE_BYREF Tag candidate text. You can query, but you can't retrieve it. Path DBTYPE_WSTR | DBTYPE_BYREF The physical path of the file contains the file name. Rank DBTYPE_I4 The level of the line, the range from 0 to 1000, the larger the number, the more match. RankVector DBTYPE_I4 | DBTYPE_VECTOR Vector set of independent components of the vector. Shortfilename DBTYPE_WSTR | DBTYPE_BYREF Short (8.3) file name. Size DBTYPE_I8 The file size is byte. USN DBTYPE_I8 Update serial number, only for NTFS drivers. VPath DBTYPE_WSTR | DBTYPE_BYREF Point to the full virtual path of the file, including the file name. If there are multiple possible paths, one of the most conformable queries will be selected. Workid DBTYPE_I4 The INTERNAL ID of the file, INDEX Server is used. Write VT_FileTime Recently writes the time of the file. Define new attribute names To define the properties in the previous list, you must list them in the [Names] section of the .idq file. To use these attributes defined in the .IDQ file in a list of restrictions, sort methods, or as the search, use the following format: [Names] # Not in the standard list in the standard list = guid ["name" | PropID] In the grammatical, "name" is the property name (below the example "Sales"), and the PropID is a hexadecimal attribute ID. Note that you have to put the friendly name with quotation marks, but the attribute ID does not use quotation marks. For example, it is assumed that the HTML Meta tag is defined as the attribute name so that someone can search, the attribute to be defined is Sales. Define Sales properties Add the following line below [Names] in .IDQ file: Metadescription (DBTYPE_WSTR) = D1B5D3F0-C0B3-11CF-9A92-00A0C908DBF1 "SALES" GUID number from the metatagclsid parameter from the registry, this parameter is below: HKEY_LOCAL_MACHINE / SYSTEM / CURRENTCONTROLSET / Control / HTMLFILTER / METATAGCLSID Then, in the HTML file, you want to display a mark, define the META instructions. For example, suppose you want to search all files containing the sales plan: In File1.htm: In File2.htm: In File3.htm: Note Make sure that the Meta Name tag is added to the file start and the HTML tag. You can now search all files about the sales plan, please send the following query: @Metadescription Projections This query returns all files that contain the word Projections in the Content field of the Meta tag. In this example, File1.htm and File2.htm will be returned. However, if you want to search for sales, for example, in 1997, please send the following query: @metadescription 1997 The file3.htm will be returned.
