<!-- list-format.mldoc -->
<!-- Entities.sgml entry
<!ENTITY ListFormat SDATA "list-format-sig.sml">
-->
<!DOCTYPE ML-DOC SYSTEM>
<COPYRIGHT OWNER="Bell Labs, Lucent Technologies" YEAR=1998>
<VERSION VERID="1.0" YEAR=1998 MONTH=6 DAY=11>
<TITLE>The ListFormat structure</TITLE>
<INTERFACE>
<HEAD>The <CD/ListFormat/ structure</HEAD>
<SEEALSO>
<STRREF TOPID/Format/
<STRREF TOPID DOCUMENT=SML-BASIS-DOC/List/
<STRREF TOPID DOCUMENT=SML-BASIS-DOC/StringCvt/
</SEEALSO>
<PP>
The <STRREF NOLINK/ListFormat/ structure provides simple utilities
for converting between lists and their string representations.
<STRUCTURE STRID="ListFormat">
<SIGBODY SIGID="LIST_FORMAT" FILE=LIST-FORMAT>
<SPEC>
<VAL>fmt<TY>{init : string, sep : string, final : string, fmt : 'a -> string} -> 'a list -> string
<COMMENT>
<PROTOTY>
fmt {<ARG/init/, <ARG/sep/, <ARG/final/, <ARG/fmt/}
</PROTOTY>
takes an initial string (<ARG/init/), a separator (<ARG/sep/),
a terminating string (<ARG/final/), and an item formating
function (<ARG/fmt/), and returns a list formatting function.
The list <CD/[a, b, ..., c]/ gets formated as
<CODE>
init ^ (fmt a) ^ sep ^ (fmt b) ^ sep ^ ... ^ sep ^ (fmt c) ^ final
</CODE>
<SPEC>
<VAL>listToString<TY>('a -> string) -> 'a list -> string
<COMMENT>
<PROTOTY>
listToString <ARG/f/ <ARG/l/
</PROTOTY>
formats a list in SML style. Equivalent to:
<CODE>
fmt {init="[", sep=",", final="]", fmt=f} l
</CODE>
<SPEC>
<VAL>scan<TY>{init : string, sep : string, final : string, scan : (char,'b) StringCvt.reader -> ('a,'b) StringCvt.reader} -> (char,'b) StringCvt.reader -> ('a list,'b) StringCvt.reader
<COMMENT>
<PROTOTY>
scan {<ARG/init/, <ARG/sep/, <ARG/final/, <ARG/scan/} <ARG/re/
</PROTOTY>
scans a list of the specified format from a stream of characters. If
successful, it returns the list and the remainder of the stream;
otherwise, it returns <CONREF STRID="Option" DOCUMENT=SML-BASIS-DOC/NONE/.
<PP>
The format of the list is specified by an expected initial string
<ARG/init/, a list item separator <ARG/sep/, an expected
terminating string <ARG/final/, and a function <ARG/scan/ for scanning
a list item. Whitespace is ignored.
<PP>
When scanning a string, <CD/scan/ checks for the separator before
the terminator, thus if the separator is a prefix of the terminator,
the function will not work.
</STRUCTURE>
</INTERFACE>
