| | 970 | == CLARIN FCS-QL Grammar Specification #fcsQLEBNF |
| | 971 | The version of the CLARIN FCS-QL is tied to the FCS Core version starting with version 2.0. |
| | 972 | |
| | 973 | === FCS-QL EBNF === |
| | 974 | {{{#!comment |
| | 975 | Please keep the EBNF nicely formatted. Thanks! |
| | 976 | }}} |
| | 977 | {{{ |
| | 978 | [1] query ::= main-query within-part? |
| | 979 | |
| | 980 | [2.11] main-query ::= simple-query |
| | 981 | | simple-query "|" main-query /* or */ |
| | 982 | | simple-query main-query /* sequence */ |
| | 983 | | simple-query quantifier /* quatification */ |
| | 984 | |
| | 985 | [3.11] simple-query ::= '(' main_query ')' /* grouping */ |
| | 986 | | implicit-query |
| | 987 | | segment-query |
| | 988 | |
| | 989 | [4] implicit-query ::= flagged-regexp |
| | 990 | |
| | 991 | [5] segment-query ::= "[" expression? "]" |
| | 992 | |
| | 993 | [6] within-part ::= simple-within-part |
| | 994 | |
| | 995 | [7] simple-within-part ::= "within" simple-within-scope |
| | 996 | |
| | 997 | [8] simple-within-scope ::= "sentence" |
| | 998 | | "s" |
| | 999 | | "utterance" |
| | 1000 | | "u" |
| | 1001 | | "paragraph" |
| | 1002 | | "p" |
| | 1003 | | "turn" |
| | 1004 | | "t" |
| | 1005 | | "text" |
| | 1006 | | "session" |
| | 1007 | |
| | 1008 | [11.11] expression ::= basic-expression |
| | 1009 | | expression "|" expression /* or */ |
| | 1010 | | expression "&" expression /* and */ |
| | 1011 | |
| | 1012 | [12.11] basic-expression ::= '(' expression ')' /* grouping */ |
| | 1013 | | "!" expression /* not */ |
| | 1014 | | attribute operator flagged-regexp |
| | 1015 | |
| | 1016 | [13] operator ::= "=" /* equals */ |
| | 1017 | | "!=" /* non-equals */ |
| | 1018 | |
| | 1019 | [14] quantifier ::= "+" /* one-or-more */ |
| | 1020 | | "*" /* zero-or-more */ |
| | 1021 | | "?" /* zero-or-one */ |
| | 1022 | | "{" integer "}" /* exactly n-times */ |
| | 1023 | | "{" integer? "," integer "}" /* at most */ |
| | 1024 | | "{" integer "," integer? "}" /* min-max */ |
| | 1025 | |
| | 1026 | [15] flagged-regexp ::= regexp |
| | 1027 | | regexp "/" regexp-flag+ |
| | 1028 | |
| | 1029 | [16] regexp-flag ::= "i" /* case-insensitive; Poliqarp/Perl compat */ |
| | 1030 | | "I" /* case-sensitive; Poliqarp compat */ |
| | 1031 | | "c" /* case-insensitive, CQP compat */ |
| | 1032 | | "C" /* case-sensitive */ |
| | 1033 | | "l" /* literal matching, CQP compat*/ |
| | 1034 | | "d" /* diacritic agnostic matching, CQP compat */ |
| | 1035 | |
| | 1036 | [17] regexp ::= quoted-string |
| | 1037 | |
| | 1038 | [18] attribute ::= simple-attribute |
| | 1039 | | qualified-attribute |
| | 1040 | |
| | 1041 | [19] simple-attribute ::= identifier |
| | 1042 | |
| | 1043 | [20] qualified-attribute ::= identifier ":" identifier |
| | 1044 | |
| | 1045 | [21.11] identifier ::= identifier-first-char identifier-char* |
| | 1046 | |
| | 1047 | [21.12] identifier-first-char ::= [a-zA-Z] |
| | 1048 | |
| | 1049 | [22] identifier-char ::= [a-zA-Z0-9\-] |
| | 1050 | |
| | 1051 | [24] integer ::= [0-9]+ |
| | 1052 | |
| | 1053 | [26] quoted-string ::= "'" (char | ws)* "'" /* single-quotes */ |
| | 1054 | | """ (char | ws)* """ /* double-quotes */ |
| | 1055 | |
| | 1056 | [27] char ::= <any unicode codepoint excluding whitespace codepoints> |
| | 1057 | | "\" escaped-char |
| | 1058 | |
| | 1059 | [28] ws ::= <any whitespace codepoint> |
| | 1060 | |
| | 1061 | [29] escaped-char ::= "\" /* backslash (\) */ |
| | 1062 | | "'" /* single quote (') */ |
| | 1063 | | """ /* double quote (") */ |
| | 1064 | | "n" /* generic newline, i.e "\n", "\r", etc */ |
| | 1065 | | "t" /* character tabulation (U+0009) */ |
| | 1066 | | "x" hex hex /* Unicode codepoint with hex value hh */ |
| | 1067 | | "u" hex hex hex hex /* Unicode codepoint with hex value hhhh */ |
| | 1068 | | "U" hex hex hex hex hex hex hex hex /* Unicode codepoint with hex value hhhhhhhh */ |
| | 1069 | |
| | 1070 | [30] hex ::= [0-9a-fA-F] |
| | 1071 | }}} |
| | 1072 | === Notes === |
| | 1073 | * based on Poliqarp with inspiration from others |
| | 1074 | * "attribute": the annotation layer to be used, e.g. "word", "lemma", "pos" or qualified "pos:stts" the supported values for this construct are beyond the grammar and are defined in supplementary documents |
| | 1075 | * "simple-within-scope": possible values for scope |
| | 1076 | * "sentence", "s", "utterance", "u": denote a matching scope of something like a sentence or utterance. provides compatibility with FCS 1.0 ("Generic Hits", "Each hit SHOULD be presented within the context of a complete sentence.") |
| | 1077 | * "paragraph" | "p" | "turn" | "t": denote the next larger unit, e.g. something like a paragraph |
| | 1078 | * "article" | "session": something like a whole document |
| | 1079 | * {{{[27]}}} and {{{[28]}}} "any $SOMETING codepoint" are a pain to get easily done in at least ANTLR and JavaCC. Especially in combination with {{{[29]}}} :/ |
| | 1080 | * regex are not defined/guarded by this grammar :/ |
| | 1081 | * non-continuous rule numbers are currently intended; we've already removed some. Rules will be renumbered, when grammar is fixed. |
| | 1082 | * Integrated Peter B's suggestion {{{[2v2]}}} and {{{[3v2]}}} together with {{{[11v2]}}} and {{{[12v2]}}} for resolving structural ambiguity eventhough antlr handles this perfectly fine. |
| | 1083 | * Changed "identifier" {{{[21]}}} to only be allowed to start with a letter e.g. not digits and - (hyphen) to more resemble XML names. |
| | 1084 | |
