Movatterモバイル変換


[0]ホーム

URL:


Jump to content
WiktionaryThe Free Dictionary
Search

Template:affix

From Wiktionary, the free dictionary

prefix-base-suffix


The followingdocumentation is located atTemplate:affix/documentation.[edit]
Useful links:subpage listlinksredirectstransclusionserrors (parser/module) •sandbox
Lua logo
Shortcut:
{{af}}

This template displays themorphemes (parts) that compose a word and categorizes its affixes, used inetymology sections. Despite the name, this template may be used for compounds similar to{{compound}}.

Parameters

|1=
The language code. SeeWiktionary:Languages.
|2=,|3= etc.
The parts of the terms to show. Either:
  1. At least one of the provided parts must begin or end with a hyphen (prefix, suffix, interfix);
  2. or there must be two or more parts (it may be a compound, with or without prefixes, suffixes, interfixes).
A term can begin with a language code followed by a colon. This overrides the default provided by|1=. When this is provided, the language name is displayed before the part, and a derivational category is also added to the entry. This is used for the occasional term which was directly affixed at the time of borrowing, and for which no native base form existed at the time, such asvexillology.
|pos=
The part of speech to include in the category names, in plural form. Defaults toterms. This is separate from (and conceptually very different from) the|posN= parameters, which specify clarifying parts of speech or similar descriptive text to display in parentheses after individual terms, next to the gloss.
|lit=
The literal meaning of the overall term. This is separate from the|litN= parameters, which specify literal meanings of individual parts.
|sort=
The sort key to use when placing the entry in categories. This is rarely needed, as the default generated sort key is usually correct.
|sc=
The script code. SeeWiktionary:Scripts. This is rarely needed as the script can generally be automatically detected. This is separate from the|scN= parameters, which specify script codes for individual parts in languages distinct from the overall language specified using|1=.
|nocat=1
Do not add categories to the entry.
|type=
Type of compound, e.g.bahuvrihi oralliterative. See#Compound subtypes below. If this parameter is given, extra text is generated before the listed parts (e.g.Bahuvrīhi compound of FOO + BAR), and the page is placed in an additional category (e.g.Category:Hindi bahuvrihi compounds).
|nocap=1
Do not capitalize the first letter of the extra text generated by|type=.
|notext=1
When|type= is used, do not generate the extra text that is normally placed before the listed parts. The page will still be placed in the additional type-specific category unless|nocat=1 is given.

The following parameters are available for each matching part in the numbered/positional parameters. TheN must be replaced by the corresponding part index. For example, for the second part (positional parameter 3), use 2 forN. Most of these parameters correspond directly to the equivalent parameters in the standard{{l}} (and{{m}}) template.

|altN=
The alternative display form. Equivalent to the third parameter of{{l}}.
|tN=
A gloss to show after the part. Equivalent to the fourth parameter of{{l}}.The parameter|glossN= is deprecated for this purpose.
|trN=
The transliteration, as in{{l}}.
|tsN=
The transcription, in languages where the transliteration and transcription differ markedly, as in{{l}}.
|gN=
The gender (multiple comma-separated genders may be specified for a given part).
|idN=
The sense id, as in{{l}}. As distinguished by mere linking templates, the affix template, categorizing terms by affix, also adds the sense id to the category name to distinguish various purposes of the same affix shape, hence the sense id in the dictionary entry for the affix should have a reasonable name in respect to this categorization. Changes the category name fromlanguage name termsaffixed withaffix tolanguage name termsaffixed withaffix (value of id parameter).
|posN=
The part of speech label to show after the part, as in{{l}}. If this parameter containspatronym(ic), the entry will be added to alanguage's patronymic category. If this parameter containsdiminutive and the|pos= parameter is defined, the entry will be added to a diminutive part-of-speech category. For instance, the parameter values|pos=noun and|pos2=diminutive suffix will add an entry to the language'sdiminutive nouns category.
|litN=
The literal meaning to show after the part, as in{{l}}.
|typeN=
Override the affix type, e.g.|type2=prefix.Not the same as the compound type specified using|type=.
|qN=
Qualifier text to display before the part, in parens and normally italicized.
|qqN=
Qualifier text to display after the part, in parens and normally italicized.
|lN=
Comma-separated label(s) to display before the part, as in{{lb}}.
|llN=
Comma-separated label(s) to display after the part, as in{{lb}}.
|refN=
Reference(s) to display after the part, as in{{IPA}}.
|langN=
The language code to use for this particular part. This overrides the default provided by|1=. When this is provided, the language name is displayed before the part, and a derivational category is also added to the entry. This is used for the occasional term which was directly affixed at the time of borrowing, and for which no native base form existed at the time, such asvexillology.
|scN=
The script code to use for this particular part. This overrides the default provided by|sc=.

Examples

Basic examples:

{{af|en|complete|-ness}}complete +‎-ness
{{af|en|pre-|date}}pre- +‎date
{{af|en|acro-|-nym}}acro- +‎-nym
{{af|en|pest|-i-|-cide}}pest +‎-i- +‎-cide
{{af|en|volley|ball}}volley +‎ball
{{af|en|small|en- -en}}small +‎en- -en

In foreign languages

1. For Portugueseaviãozão(large plane), which can be viewed as being composed ofavião(plane), an interfix-z- and an augmentative suffix-ão, use the following:

{{af|pt|avião<t:plane>|-z-|-ão<pos:augmentative suffix>}}
which produces
avião(plane) +‎-z- +‎-ão(augmentative suffix)

This can equivalently be written using separate parameters (the older way of doing things) like this:

{{af|pt|avião|t1=plane|-z-|-ão|pos3=augmentative suffix}}


2. The Portuguese medical termacromia(achromia (lack of pigmentation)) can be viewed as being composed directly from Ancient Greekχρῶμα(khrôma,color) using the Portugueseprivative (i.e. expressing "lack of") suffixa- and the abstract noun suffix-ia. Write this as follows:

{{af|pt|a-<id:privative>|grc:χρῶμα<t:color>|-ia}}
which produces
a- +‎Ancient Greekχρῶμα(khrôma,color) +‎-ia

In this case, there are multiple Portuguese prefixes spelleda-, with different origins and meanings. To identify the privative prefix specifically, we use the<id:...> modifier (see{{senseid}}).


3. The Italian relational adjectivelitoraneo(coastal,littoral) was constructed directly from the Latinlītus(shore) based on its stemlītor- (as found for example in the genitive singularlītoris). Write this as follows:

{{af|it|la:[[lītus|lītoris]]<t:shore>|-aneo}}
which produces
Latinlītoris(shore) +‎-aneo

The use of the bracketed expression allows us to display the accusative singular form while linking to the lemma form (the nominative singular). This could also be implemented using the<alt:...> modifier.


4. The Portuguese toponymNova Escócia(Nova Scotia) is composed of two words, literally meaning "New Scotland". This can be written as follows:

{{af|pt|nova<pos:feminine of{{m|pt|novo|t=new}}>|Escócia<t:Scotland>}}
which produces
nova(feminine ofnovo(new)) +‎Escócia(Scotland)

Here, we use the<pos:...> ("part of speech") modifier to display the grammatical derivation of the wordnova. This example shows that arbitrary template code can be embedded inside an inline modifier.


5. For the Japanese term(kona,powder), the following is an actual etymology:

Appears to be{{af|ja|sort=こな|<tr:ko><t:powder><ng:see above>|<alt:な><tr:na><pos:suffixing element><ng:meaning unclear>}}.
which produces
Appears to be(ko,powder,see above) +‎(na,suffixing element,meaning unclear).

The use of the<alt:...> modifier with an empty term serves to prevent the term from being linked. The|sort= parameter is needed in general with any Japanese templates that generate categories, because the sort key for this language cannot be automatically generated. (It is based on the pronunciation of the term, and many Japanese terms have multiple pronunciations.)


6. An example of using<type:...>, for Tagalogsumasaibabaw:

From{{af|tl|sa-|ibabaw|-um-<type:infix>}} with initial{{rdp|tl|nocap=1}}.
which produces
Fromsa- +‎ibabaw +‎-um- with initialreduplication.

Inline modifiers

All of the per-term parameters described above can alternatively be specified asinline modifiers, using a syntax likeдетекти́в<tr:dɛtɛktív><t:detective> to specify modifiers such as transliterations, glosses and qualifiers. In this example, for the Russian termдетекти́в(dɛtɛktív,detective), the manual transliterationdɛtɛktív and gloss "detective" are given. Specifically, the following modifiers are recognized; see the above documentation for more information on the exact meaning of each modifier.

  • t orgloss: gloss
  • alt: alternative display text
  • tr: transliteration
  • ts: transcription, for languages where the transliteration and pronunciation are markedly different
  • q: qualifier, e.g.high-register; this appears before the term, parenthesized and italicized
  • qq: qualifier, e.g.high-register; this appears after the term, parenthesized and italicized
  • l: comma-separated label(s), e.g.rare; these appear before the term, parenthesized, italicized and appropriately linked
  • ll: comma-separated label(s), e.g.rare; these appear after the term, parenthesized, italicized and appropriately linked
  • ref: reference(s), as in{{IPA}}
  • lit: literal meaning
  • pos: part of speech
  • type: affix type (not compound/etymology type)
  • g: comma-separated list of gender/number specifications
  • id: sense ID; see{{senseid}}
  • lang: language code
  • sc: script code

Examples

The following:

From {{af|pt|Patrícia|t1=Patricia|pos1=given name|-inha|pos2=diminutive suffix}}.

can be equivalently written as follows using inline modifiers:

From {{af|pt|Patrícia<t:Patricia><pos:given name>|-inha<pos:diminutive suffix>}}.

Both produce the following:

FromPatrícia(Patricia,given name) +‎-inha(diminutive suffix).

Note how the use of inline modifiers frees you from having to make sure the numbers of indexed parameters like|t1=,|pos2= correctly line up with the terms they are referring to.

Categories

The template will automatically categorize the entry based on the types of affixes it sees.

  • If a part begins with a hyphen, it is a suffix.
  • If a part ends with a hyphen, it's a prefix.
  • If a part has a hyphen at both ends, it's an interfix.

This behavior can be overridden using the inline modifier<type:TYPE> (or just<TYPE> for short). This should be used, for example, for Proto-Indo-European roots, which are conventionally written with a trailing hyphen but aren't prefixes, and for certain Korean suffixes such as-고-(-go-) that are written with a hypen on both ends. The recognized affix types are

prefix orpreprefix
suffix orsufsuffix
infix orininfix
interfix orinterinterfix
circumfix orcircumcircumfix
non-affix ornaf orrootnon-affix

What is considered a hyphen depends on the specified language, so it can also support language-specific types of hyphen. In particular, the Hebrew script has its own script-specific hyphen character (which has the correct right-to-left behavior). The Arabic script does not traditionally use hyphens, though some languages have elected to use the kashida (ـ) for this purpose.

Some languages (e.g. most East Asian languages) don't use hyphens with affixes. For these languages, you should still add a regular hyphen (-) to the affix as appropriate, but the term will be displayed and linked without a hyphen.

Compound subtypes

The|type= parameter can be used to specify the subtype of compound when{{affix}} is used to specify the parts of a compound. For example, if|type=bahuvrihi is given (or|type=bahu or|type=bv), extra textBahuvrīhi compound of is generated before the listed parts (where the wordbahuvrīhi links appropriately to the appendix, with an explanation of what this term means). In addition, the page is placed in a type-specific category, e.g.Category:Hindi bahuvrihi compounds if the language in question is Hindi (codehi).

The parameter|nocap=1 can be used to disable capitalization of the first letter of the extra text generated, and|notext=1 can be used to suppress the extra text entirely. Using|type= along with|notext= is still useful because of the extra type-specific category that the page is placed in. It's also possible to use|type= without any parts, e.g.{{affix|hi|type=bahuvrihi}}. This generates just the extra text (without the wordof after it), and places the page in the type-specific category.

The following types are currently supported:

TypeAlias ofDisplay(click on link for explanation)Category
aboradapted borrowingadapted borrowinglang adapted borrowings fromsource
adapadapted borrowingadapted borrowinglang adapted borrowings fromsource
adapted borrowingadapted borrowinglang adapted borrowings fromsource
allitalliterativealliterative compoundlang alliterative compounds
alliterativealliterative compoundlang alliterative compounds
antantonymousantonymous compoundlang antonymous compounds
antonymousantonymous compoundlang antonymous compounds
bahubahuvrihibahuvrīhi compoundlang bahuvrihi compounds
bahuvrihibahuvrīhi compoundlang bahuvrihi compounds
bvbahuvrihibahuvrīhi compoundlang bahuvrihi compounds
coordcoordinativecoordinative compoundlang coordinative compounds
coordinativecoordinative compoundlang coordinative compounds
descdescriptivedescriptive compoundlang descriptive compounds
descriptivedescriptive compoundlang descriptive compounds
detdeterminativedeterminative compoundlang determinative compounds
determinativedeterminative compoundlang determinative compounds
dvadvandvadvandva compoundlang dvandva compounds
dvandvadvandva compoundlang dvandva compounds
dvidvigudvigu compoundlang dvigu compounds
dvigudvigu compoundlang dvigu compounds
endoendocentricendocentric compoundlang endocentric compounds
endocentricendocentric compoundlang endocentric compounds
exoexocentricexocentric compoundlang exocentric compounds
exocentricexocentric compoundlang exocentric compounds
iz1izafet Iizafet I compoundlang izafet I compounds
iz2izafet IIizafet II compoundlang izafet II compounds
iz3izafet IIIizafet III compoundlang izafet III compounds
izafet Iizafet I compoundlang izafet I compounds
izafet IIizafet II compoundlang izafet II compounds
izafet IIIizafet III compoundlang izafet III compounds
karmakarmadharayakarmadhāraya compoundlang karmadharaya compounds
karmadharayakarmadhāraya compoundlang karmadharaya compounds
kdkarmadharayakarmadhāraya compoundlang karmadharaya compounds
kenkenningkenninglang kennings
kenningkenninglang kennings
rhyrhymingrhyming compoundlang rhyming compounds
rhymingrhyming compoundlang rhyming compounds
synsynonymoussynonymous compoundlang synonymous compounds
synonymoussynonymous compoundlang synonymous compounds
tattatpurusatatpuruṣa compoundlang tatpurusa compounds
tatpurusatatpuruṣa compoundlang tatpurusa compounds
tptatpurusatatpuruṣa compoundlang tatpurusa compounds

Limitations

  • If an entry both begins and ends with a hyphen, it is considered an interfix, not an infix. There is currently no support for infixes as such.

See also

Template Data

This is theTemplateData documentation for this template used byVisualEditor and other tools.

TemplateData for affix

No description.

Template parameters[Edit template data]

This template prefers inline formatting of parameters.

ParameterDescriptionTypeStatus
language code1lang

the 2- or 3-letter language code for the current section

Example
en
Stringrequired
first part2

the first part of the affixed word

Example
neuro
Page namesuggested
second part3

the second part of the affixed word

Example
genic
Page namesuggested
script codesc

no description

Stringoptional
first part of speechpos1

no description

Stringoptional
second part of speechpos2

no description

Stringoptional
first part alternate textalt1

Text to show instead of the first part for the link to its entry

Stringoptional
second part alternate textalt2

Text to show instead of the second part for the link to its entry

Stringoptional
first glosst1

brief gloss of the first part

Stringoptional
second glosst2

brief gloss of the second part

Stringoptional
first part transliterationtr1

Transliteration of the first part (for non-Latin scripts)

Stringoptional
second part transliterationtr2

Transliteration of the second part (for non-Latin scripts)

Stringoptional
Retrieved from "https://en.wiktionary.org/w/index.php?title=Template:affix&oldid=75235160"
Categories:
Hidden category:

[8]ページ先頭

©2009-2025 Movatter.jp