Versione italiana  

Retrieve All

The "Retrieve All" button gives a table with data on all people in our database in the chosen categories.

Search

Rather than downloading the entire table, in may be better to search people by name (once again, in the chosen categories). This can be done by typing an input search string in the search field and clicking on the "Search" button. The input string shall be a space-separated sequence of subwords that are expected to occur in the name (first name and surname) of the person to be looked up. How the search is performed, and when a name matches an input string is approximately explained by the following rules. The precise rules, in terms of (unix) regular expressions, are given below.

• The string "str1 str2 … strn" (i.e., the words str1, str2, …, strn separated by spaces) matches the person Name Surname if the strings str1, str2, …, strn occur in the same order in the string "NameSurname". For instance the input string "ius pe pea" would find Giuseppe Peano (if he were in our department). Moreover:
• the search is case-insensitive, as the example just given shows.
• Non-ASCII characters, like accented letters, are handled as follows:
  • If a person's name contains accented letters, then it will be found even if the input string has the corresponding unaccented letter in place of it, or also the same (unaccented) letter followed by either tick (characters ' or `). For instance, any of the the following string will find our librarian Stefania Castanò: "castano", "castano'", "Castanò".
  • As a matter of fact, every non-ASCII character matches any character. Thus also "Cästanì" will return Castanò.

Search results. Every name found is preceded by a category tag with the following meaning:

academic staff	PhD student
support staff	others

Detailed Search Rules: The input string I is treated as a regular expression to be matched against the string N taken from our people database in case-insensitive mode. The string N is obtained by juxtaposition of the strings in the database fields containing name and surname. It is important to note that these string are formatted as HTML code. Before the matching is tested the following changes are applied.

• Every space character in I is replaced by ".*"; moreover ".*"; is also prepended to I.
• The characters "'" and "`" possibly contained in I are replaced by "'?" and "`?" respectively.
• Every character with ASCII code greater than 127 in I is replaced by ".".
• In N, some HTML entities (basically, those occurring in our databases) are replaced by the corresponding ASCII characters. In particular, this is what happens for accented letters (and german Umlaute), that are replaced by the corresponding unaccented letters: for any vowel V each of the substrings "&Vgrave;", "&Vacute;", "&Vuml;" occurring in I is replaced by V.

The above implies that regular expressions may be successfully used in the search string. For instance, the string "[dt][^a]" matches all persons whose name contains a "d" or a "t" not immediately followed by "a".