User:AutoFormat/Headers

Header control table for AutoFormat
Each table line is a header recognized by the bot code.


 * Header is the canonical format
 * level is the minimum level it should occur, POS headers for example are 3, even though they occur at 4 when in a numbered etymology section
 * NS flags headers that are non-standard but recognized, the presence or absence of NS here is not policy.
 * POS flags headers of POS sections, even if non-standard (e.g. Transitive verb)
 * variants and errors list forms and misspellings that should be corrected to this header. It is not necessary to list variants that differ only by +s or -s at the end. These must all be in lower case, separated by commas, the bot matches to lower case.
 * variants by +s or -s are recognized for standard headers, not those with NS (for example, we have Idiom and Idioms, Idioms must have NS set, and be listed after Idiom!)
 * the order of these tables doesn't matter to the bot s/w (but see previous comment), it is roughly the common POS first, the other L3 headers, then the standard L4 headers, with some logical grouping
 * note that the lines the bot is to read must start with exactly "| " and have all 5 columns separated by exactly "||"

The bot will do a fuzzy match to all variants if there is no exact match in the table, so most one and two off misspellings will be corrected. For example: "Derivated terms" will match "derivate terms" and be corrected to "Derived terms".

Language specific or non-English POS
and some other non-standard cruft, like expression which has to be here because expressions is

other L3 things
"Etymology n" for any n 1 to 24 gets special treatment in the code, introducing a section nested one level deeper.

level 4/3 headers
These are headers that must be at L4 (Synonyms, Translations) and headers that are used at L4 or at L3 (even though shown in WT:ELE only at L4). These levels are in the case without numbered etymologies, level adjustments for that are in the code.

WT:ELE list of other L4 headers: Antonyms, Hypernyms, Hyponyms, Meronyms, Holonyms, Troponyms, Coordinate terms, See also.

Other headers
Of varying status, but recognized, rather than automatically flagged with rfc.

Examples are Homophones, proposed for Pronunciation sections; Devanagari spellings, used a lot, but should be fixed (probably by some other bot) to Alternative forms.

Mostly these should be "level 3" and NS, so that other checks are disabled.

Some might be POS, if POS processing is still a good idea (probably not).

Prenoun and Preverb are used for Ojibwe, need standardizing.

Classifier is used in Thai, probably should be Counter

Verb xx for xx = I to XI are used in Arabic, need to be changed to Verb with the form indicated on the inflection line, as in all other languages.