From the administration panel, new calculations can be added. Initially, only the name is required. When created, the calculation will contain suitable default values.

For wiki pages:

Thus, visitors can provide feedback like:

• Did this page help you solve the issue?
• Was this page easy to understand?

Introduced in Tiki5, the advanced rating feature allows for more control over the aggregation of scores.

Rating methods are defined globally and will be used for all supported objects. They are defined through the Advanced Rating administration panel (tiki-admin.php?page=rating). Multiple methods can be created. If a method contains type-specific calculations, it will be ignored when performing the calculation.

Features currently supporting sorting through advanced rating:

• Articles
• Wiki
• Comments

Note that the sort mode to use when needing to sort by advanced rating is either adv_rating_xx_asc or adv_rating_xx_desc, where xx is the ratingConfigId.

Feature request: Can we make this take the name of the config instead of the ratingConfigId as well?

By default, each calculated value is kept for 1 hour (3600 seconds). This limit does not apply when recalculating on vote, but is used for every other technique to avoid recalculating the same scores over and over again.

The calculation is defined as a small piece of code, similar to functional languages, which is very close to mathematical representations. Creating custom formulas is expected to require some mathematical skills. However, this documentation should provide examples for most frequent cases.

The editor in the administration panel performs extensive validation and will make it impossible to save the formula unless it can be evaluated. Checks are performed for:

• Syntax errors
• Unknown functions
• Missing arguments
• Invalid argument values
• Unknown input variables

(rating-average (object type object-id))

It can be altered to limit the vote consideration to a limited time span, 30 days for example.

(rating-average
    (object type object-id)
    (range (mul 3600 24 30))
)

In the language, spaces do not matter. Only the parenthesis indicate structure. rating-average is a function that fetches the ratings for a given object. type and object-id are standard variables fed when calculating a rating. object and range are configuration options of the function.

mul is a mathematical function. (mul 3600 24 30) is equivalent to 3600*24*30.

The functions can be combined in various ways. For example, we could calculate a score that considers the votes from the past month, but gives extra emphasis on the recent ones.

(add
    (rating-average
        (object type object-id)
        (range (mul 3600 24 30)))
    (rating-average
        (object type object-id)
        (range (mul 3600 24 7)))
)

Even though the votes are 1-5, the final score can be on an entirely different scale. The language is also extensible if the calculation needs to be combined with other factors or weight. See Rating Language.

All available options are documented in the following section.

Any comment block is stripped from the formula at parse-time

(mul
  1
  2
  (comment Simple enough?))
    -> 2

Performs a simple multiplication accepting multiple input values.

(mul 3 4)
   -> 12
(mul (mul 3 4) 5)
   -> 60
(mul 3 4 5)
   -> 60
(mul 4 0.5)
   -> 2

Performs a simple division accepting multiple input values.

(div 3 4)
   -> 0.75
(div (mul 3 10) 5)
   -> 6
(div 30 5 3)
   -> 2
(div 4 0.5)
   -> 8

Performs a simple sum accepting multiple input

(add 3 4)
   -> 7
(add (add 3 4) 5)
   -> 12
(add 3 4 5)
   -> 12
(add 4 0.5)
   -> 4.5

Performs a simple substraction accepting multiple input

(sub 3 4)
   -> -1
(sub (sub 3 4) 5)
   -> -6
(sub 3 4 5)
   -> -12
(add 4 0.5)
   -> 3.5

Rounds to a specific number of digits (new in Tiki12)

(round 4.556234342234 2)
   -> 4.56
(round 4.556234342234)
   -> 5

Returns the first non-empty value from the list.

(coalesce 3 4)
   -> -3
(coalesce (sub 3 3) 5)
   -> 5
(coalesce 0 0 (str) -10)
   -> -10
(coalesce 0 0 0 0 0)
   -> 0

Generates a static string when needed and the processor attempts to process the string as a variable. Any arguments will be concatenated using spaces.

Note: The quoted string syntax was included in Tiki13.

(str hello-world)
   -> "hello-world"
(str hello world)
   -> "hello world"
(str 
    hello
    world
    foobar)
   -> "hello world foobar"
(str (mul 2 3) "= 6")
   -> "6 = 6"

Concatenates a string of text. (new in Tiki12)

Note: The quoted string syntax was included in Tiki13.

(concat 
    (str $)
    1234
)
   -> "$1234"
(concat 
    14
    (str %)
)
   -> "14%"
(concat 14 "%")
   -> "14%"

Generates a map (or dictionary).

(map
  (key1 1)
  (key2 2)
  (key3 (str value3))
)
 -> {"key1": 1, "key2": 2, "key3": "value3"}

Compares multiple values.

(equals 2 (add 1 1) (sub 4 2))
 -> 1 (equivalent of 2 == 1+1 && 2 == 4-2)
(equals (add 1 1) 3)
 -> 0

Conditionally evaluates a branch.

(if (equals 2 2) 42 -1)
 -> 42
(if (equals 2 1) 42 -1)
 -> -1

Ensures all elements evaluate to true.

(and 3 2 1 2 3)
 -> 1
(and 2 3 0 2)
 -> 0

Ensures that at least one element evaluates to true. Elements are evauated sequentially until a false element is found. Others are left unevaluated.

(or 3 2 1 2 3)
 -> 1
(or 2 3 0 2)
 -> 1
(or 0 0)
 -> 0

Generates a hash based on multiple values. Used primarily to generate aggregate hashes in the PluginActivityStream. Note that because it is a hash, the exact value coming out does not matter. Only that given the same parameter, it will produce the same value.

(hash 1)
 -> [sha1("1")]
(hash 1 2 3 4)
 -> [sha1("1/2/3/4")]
(hash 1 2 (map (a 3) (b 4)))
 -> [sha1("1/2/3/4")]

Calculates the average of multiple values. All entries in the list will be flattened if arrays are present.

(avg 1 2 3)
-> 2
... given list contains [1, 2, 3]
(avg list)
-> 2

Produces a multi-dimensional array out of a text string. Each line is expected to be an independent value, each line will be split by a separator into the specified keys.

... given str contains a list of 3 comma-separated values
(split-list
    (content str)
    (separator ,)
    (keys a b c))
-> [{a: 1, b: 2, c: 3}, {a: 2, b: 3, c: 4}]

For a list of value pairs, such as the output of split-list, evaluates a formula for each set of values, returns the list of results.

Within the formula, variables coming from the list will be used first. Fallback will be on the other variables available in the execution context.

... given items contains [{a: 1, b: 2, c: 3}, {a: 2, b: 3, c: 4}]
(for-each
    (list items)
    (formula (mul a b c)))
-> [6, 24]

... given items contains [{a: 1, b: 2, c: 3}, {a: 2, b: 3, c: 4}]
... and d contains 10
(for-each
    (list items)
    (formula (mul c d)))
-> [30, 40]

등급 함수는 등급 기록 테이블에서 점수를 계산합니다. 사이트에 수행된 각 등급은 데이터베이스 내에 보관되며 사용자 지정 등급을 계산하기 위해서 사용될 수 있습니다. 다양한 선택사항이 문서의 품질 향상을 지원하거나 피드 수집기로 들어오는 데이터의 순위를 매기기 위하는 등의 사이트 상의 중요성을 반영하기 위하여 점수를 계산하도록 적용됩니다.

• object (개체), 필수이며 항상 이 내용상 (개체 유형 개체ID).
• range (범위), to limit how long votes are considered. Argument is provided as a number of seconds.
• ignore (무시), with anonymous as an argument to only consider votes from registered users.
• keep (유지), to only consider one vote per visitor. Unless the option is present, all of the votes are taken into account. The option can be either latest or oldest to indicate which one to keep.
• revote (재투표) can be specified if keep is specified. Indicates the time period required between votes. For example, users could be allowed to vote more than once per day, but only their latest vote each day would be considered, if revote is set to mul(24 3600). If the user voted yesterday as well as today, both votes will be counted.

계산에 포함될 수 있도록 기사에서 정보를 추출. 첫 번째 인자는 언제나 항상 'article' (기사)가 되어야함 . 그 외 다른 값이 되는 경우, 연산은 평가된 개체에 대하여 건너뛰어지게 되어, 이를 공식 유형-특화시켜 버립니다.

사용가능한 속성들:

• rating (등급), 기사에 첨부된 정적 등급
• view-count
• age-second
• age-hour
• age-day
• age-week
• age-month

(article-info type object-id rating)

(article-info (str article) 42 age-month)

보통의 개체 속성에서 정보를 추출함.

(attribute
    (object type object-id)
    (property tiki.proposal.accept)
)
    -> [value for page in a rating calculation]
(attribute
    (object (str wiki page) 14)
    (property tiki.proposal.accept)
    (default 0)
)
    -> [value for page id 14]

트래커 항목에서 정보를 추출함. 필드 값은 숫자로 자동으로 변환됨. 값이 발견되지 않거나 적용 불가능 한 경우 0 이 제공됨.

(tracker-field 
  (object type object-id) 
  (field priority)
)
    -> [value contained in the tracker item field with permanent name ''priority'']

개체상에 존재하는 모든 나열된 범주의 점수에 1 을 부여함.

(category-present
  (object type object-id)
  (list 3 4)
)
    -> [0, 1 or 2 - Depending on how many of categories 3 or 4 are on the object]

통합 검색 (unified search) 이 사용될 때, 재계산은 재색인 동안 이루어지도록 구성될 수 있습니다, 고로 이 스크립트가 필요 없어집니다.

refresh_all();

• 등급
• Rating Revamp

고급+등급 | 고급등급