• Introduction
• Labels
• Match whole fields / any part
• Wildcard characters
• Record
• Start at record 1 / continue
• Accented and other non-standard characters
• Miscellaneous

Each box on the search form stands for a field in each record of the database, and each record relates to one line in the selected texts. The whole database contains 162,125 records. The bands across the form divide it up as follows:

• basic text data
• rhyme and rhyme words
• phonemic patterns, including alliteration and assonance
• accentual rhythm, syllable divisions, and word accents

Your run a search by clicking the Results page button at the bottom of the Search form. This retrieves up to 30,000 records at a time. You can also look at the database entry for one text line at a time by clicking on the reference numbers in the Text files.

An empty box matches anything in that field: running a search with all the boxes empty simply returns the first 30,000 lines in the database. In the drop-down boxes you can select entries; in the others you can enter search strings (sequences of characters). You can use as many boxes as you like. Click on the labels next to each box to go to the relevant information page for each section.

Filling in more than one box returns all (and only) the records which match all the entries in all the boxes filled in (in technical terms the searches are 'and', not 'or', queries). For instance selecting 3a rima in the Metre box and 4 accents in the Even box and entering ITA in the Rhyme box produces all lines in terza rima with 4 accents on even syllables and a rhyme in ITA..

You browse through the search results using the navigation buttons at the bottom of the Results page. This can be slow if your search has returned a large number of records. Narrow the search down as much as you can to speed things up. Opening one of the Text files and clicking on the reference numbers you want is the quickest way of browsing line by line through the entries for a particular text .

If you want to make a search that returns more than 30,000 lines, see Start at record 1 / continue below. Note also that

• searches are not case-sensitive: A and a produce the same result;
• a blank space counts as a character.

See also the information below about Accented and other non-standard characters, about Wildcard characters, and about Match whole fields / any part.

The labels next to each box in the Search Form or the Results Page are all links. Click on them to go to the relevant point in an information page for each section. You can also click on the list of divisions in the section above to go to the related information page. The present page contains general information about searching.

The default for searches is to match the search string exactly with any part of the field in question. Thus ILE in the Rhyme box on the Search form will return not only rhymes in ILE but those in ABILE, etc., as well. If you want to match whole fields, for instance in this case to return only the rhymes in ILE, click the button next to Match whole fields at the bottom of the Search form. Click the button next to any part to return to matching any part of fields.

When searches are started by clicking on the reference numbers in the text files, the search engine only returns a single line. To continue browsing through the database entries for the text you have to click on further reference numbers in the text file you are interested in. The alternative is to enter a Record number in the search form. See Record below.

The search engine always matches the string in the search box exactly with the data in the search field, unless you use wildcard characters. These are the key to making full use of the database. You can put wildcards anywhere in your search string and you can put in as many as you like. The wildcards are

• _ (underscore) representing any single character or blank
• % representing any number, including zero, of any characters or blanks
• [ ] enclosing any number of alternative characters: for instance [aeiou] matches any one of the five vowels

Note also that [0-9] will match any single number and [a-z] any single letter, uppercase or lowercase.

Putting % at both ends of your search string when you match whole fields has exactly the same effect as clicking the button to match any part of fields.

Some examples in the Rhyme box when you match whole fields:

• ILE will produce all and only the rhymes in ILE
• %ILE will not only produce rhymes in ILE but also rhymes in IBILE etc.
• _LE will produce all rhymes in ALE, ELE, ILE, OLE and ULE
• _ will produce all rhymes consisting in one character (in practice one accented vowel)
• __ (double underscore) will produce all rhymes consisting in two characters
• [aeiou][aeiou] will produce all rhymes consisting in two vowels
• %[bcdfglmnprstvxz] will produce all the rhymes ending in a consonant (not many)

For other examples, see the instructions for the different boxes by clicking on the relevant label.

As said above, each line of verse in the database is part of a single database record. The numbers at the bottom of the Results page show the number of lines or records that have been returned by a search: [1/30000] means that the record in question is first out of a total of 30,000. The > button  at the bottom takes you to the next record in the search results, the < button to the previous one, >| to the very last in the sequence, and |< back to the first one. These numbers and buttons do not appear when only one line is returned by a search.

The Record box at the bottom right-hand corner of the Results page gives the unique number of the record in question, which follows the normal order of lines and the chronological order of texts. If you wish, you can enter a specific number in the equivalent Record box in the Search form, in order to run searches starting from the record number in question. You can use this as a way of browsing line by line through database entries for a text from a particular point onwards; however it is probably more convenient to do this by browsing the relevant text file and clicking on the reference numbers. See Browse Texts.

For reasons of its own, when the database displays the results of a search it sometimes starts with a record other than the first in text sequence that the search has returned: e.g. the first record you see could appear as [14/30000]. Simply press the <| button if you want to go back to the first record returned.

For technical reasons the search engine will also not return more than 30,000 records at a time: see the following section if you want to run searches returning more than 30,000 records.

The default for searches is to start at record 1. You may want to run a search returning more than 30,000 records in order to conduct broad statistical surveys of the whole corpus; for instance, you may want to determine the distribution of accented syllables or accented  vowels in rhyme. You can see how many records a search has returned from the numbers displayed at the bottom of the Results page (see preceding section).

When a search of this kind returns 30,000 records the first time you run it and you want to continue, click on the  >| button at the bottom of the Results page to go to the last record in the search results; the number of that record is displayed next to the Record label in the bottom right-hand corner.

Then click the button for continue next to the button for Start at record 1 at the bottom of the Search form. Run the search again, and it will start from the record after the last one in the previous search and return up to the next 30,000 lines. If you need to search the next 30,000 after that, go to the last record in the current set of search results, and so on from there.

The search engine makes no distinction between the accented and unaccented forms of vowels. Entering i in a search box, for instance, will match not only i, but also í, ì, î, and ï (and their uppercase equivalents). Accented characters will not be recognized if you enter them as such in the search form, so no instructions are given as to how to do this.

The only other non-standard characters in the database, in the sense of characters not in the UK keyboard, are the quotation marks « and ». You can search for these (only in the Line box) by entering respectively < and >, which are not otherwise used in the database.

Some of the search boxes may be filled in in yellow, if you have the Google Autofill function switched on. This has no significance for the database programme.