The Voynichese project (VP) defines a simple syntax for querying words in the Voynich Manuscript. While this syntax is only used internally, by the VP's query processor, it's important to be aware of how it operates in order to effectively query the manuscript.
Note that, unlike most query languages, Voynichese queries are evaluated at the word level. As such, word delimiters like whitespace and punctuation are not allowed.
Voynichese queries may use the following characters:
a,c,d,e,f,g,h,i,k,l,m,n,o,p,q,r,s,t,v,x,y,z,*,^,$
The characters a-z each match the corresponding EVA character.
The wildcard character "*" matches one or more EVA characters. Note that the wildcard may also be represented as a dash "-", for example when used in an URL.
The "^" character matches the start of a word.
The "$" character matches the end of a word.
For example, the query ^daiin$ will exactly match the EVA word daiin, whereas the query daiin (excluding the ^ and $ symbols) will match any EVA word containing daiin, such as chodaiindy.
See the QueryResults overview for information about quantities resulting from Voynichese queries.
When interacting with the Voynichese query user interface, queries are implicitly wrapped between "^" and "$" whenever the Exact option is selected.
The following screenshot shows a query for daiin with the Exact option selected - highlighted in red - which internally evaluates to ^daiin$.
Queries containing the wildcard character "*" match a variable number of words. For example the query ^da*n$ will match dain, daiin and daiiin in addition to daiiiolkaiin. In these cases the wildcard character matches i, ii, iii and iiiolkaii respectively.
Voynichese queries may contain any number of wildcards. For example, the query ^q*e*y$ will match qokeedy, qopeedy, qoeey and so on.
Note that multiple consecutive wildcard characters are collapsed into a single one - for example the query ^q***y$ is equivalent to ^q*y$.
A combination of the wildcard character "*" with the Exact option may be used to query words by suffix or prefix.
For example, the query ^*edy$ matches all words ending with edy. Note that without the Exact option, this query becomes *edy which will also match words like chedyty. It is therefore important to use the Exact option when performing a prefix or suffix query.
An example of a prefix query is ^qo*$, which matches words like qokeedy and qokol.
It's possible to query the manuscript to analyze the occurrence of a given character, using non-exact queries.
For example, the simple query o will match any word containing the character o one or more times. Note that the exact version of this query, ^o$, will only match the word o. As such, character queries should not use the Exact option.