Guido van Rossum | 6c6b78d | 1991-12-30 16:03:05 +0000 | [diff] [blame] | 1 | # These bits are passed to regex.set_syntax() to choose among |
| 2 | # alternative regexp syntaxes. |
| 3 | |
| 4 | # 1 means plain parentheses serve as grouping, and backslash |
| 5 | # parentheses are needed for literal searching. |
| 6 | # 0 means backslash-parentheses are grouping, and plain parentheses |
| 7 | # are for literal searching. |
| 8 | RE_NO_BK_PARENS = 1 |
| 9 | |
| 10 | # 1 means plain | serves as the "or"-operator, and \| is a literal. |
| 11 | # 0 means \| serves as the "or"-operator, and | is a literal. |
| 12 | RE_NO_BK_VBAR = 2 |
| 13 | |
| 14 | # 0 means plain + or ? serves as an operator, and \+, \? are literals. |
| 15 | # 1 means \+, \? are operators and plain +, ? are literals. |
| 16 | RE_BK_PLUS_QM = 4 |
| 17 | |
| 18 | # 1 means | binds tighter than ^ or $. |
| 19 | # 0 means the contrary. |
| 20 | RE_TIGHT_VBAR = 8 |
| 21 | |
| 22 | # 1 means treat \n as an _OR operator |
| 23 | # 0 means treat it as a normal character |
| 24 | RE_NEWLINE_OR = 16 |
| 25 | |
| 26 | # 0 means that a special characters (such as *, ^, and $) always have |
| 27 | # their special meaning regardless of the surrounding context. |
| 28 | # 1 means that special characters may act as normal characters in some |
| 29 | # contexts. Specifically, this applies to: |
| 30 | # ^ - only special at the beginning, or after ( or | |
| 31 | # $ - only special at the end, or before ) or | |
| 32 | # *, +, ? - only special when not after the beginning, (, or | |
| 33 | RE_CONTEXT_INDEP_OPS = 32 |
| 34 | |
| 35 | # Now define combinations of bits for the standard possibilities. |
| 36 | RE_SYNTAX_AWK = (RE_NO_BK_PARENS | RE_NO_BK_VBAR | RE_CONTEXT_INDEP_OPS) |
| 37 | RE_SYNTAX_EGREP = (RE_SYNTAX_AWK | RE_NEWLINE_OR) |
| 38 | RE_SYNTAX_GREP = (RE_BK_PLUS_QM | RE_NEWLINE_OR) |
| 39 | RE_SYNTAX_EMACS = 0 |
| 40 | |
| 41 | # (Python's obsolete "regexp" module used a syntax similar to awk.) |