00902944f6
Added code to match capturing groups and store into a Group (used to be MatchIndex)
2024-12-09 01:28:18 -05:00
80ea262064
Updated test-case structs to reflect the name of the new type
2024-12-09 01:06:18 -05:00
f5eb9c8218
Defined postfixNodes for LPAREN and RPAREN
2024-12-09 01:05:47 -05:00
20fbd20994
Added helper function to expand a slice to a given length
2024-12-09 01:05:26 -05:00
11f7f1d746
Added fields to state, to determine capturing group information. 0th group refers to entire match
2024-12-09 01:05:01 -05:00
822d1f319f
Added initial support for capturing groups
2024-12-09 01:04:31 -05:00
745fab9639
Clone lookaroundNFA when cloning a state; use compiled regex for
...
lookarounds instead of compiling a new one
2024-11-27 12:15:30 -05:00
34e9aedbd6
Compile lookaround regex to avoid compiling each time we want to use it
2024-11-27 12:15:01 -05:00
6208f32710
Added support for numeric ranges: <5-38> will match all numbers between 5 and 38, inclusive on both ends. Also print line number on which matches occur, if we are in printing (and single line) mode
2024-11-27 11:48:04 -05:00
cbd6ea136b
If the NFA starts with an assertion, make sure it's true before doing anything else. Also, check for last-state _lookaround_ rather than just last state, before breaking (instead of aborting) when the assertion fails
2024-11-27 11:46:38 -05:00
eb6a044ecf
Added angle brackets to list of special characters (which need to be escaped to be used literally
2024-11-27 11:45:27 -05:00
393769f152
Accounted for last character being a newline when checking for EOS (we can be at the second-last character if the last one is a newline
2024-11-27 11:44:39 -05:00
e36310b32d
Added function (and helper functions) to generate a regex that matches all numbers in a range
2024-11-27 11:43:57 -05:00
298285e44c
Added more test cases
2024-11-27 11:43:34 -05:00
0de3a94ce3
Fixed bug with lookaheads: f(?=f) would not match anything in 'ffa', because of the 'a' at the end of the string. Fixed by checking if there are other last states when an assertion fails, rather than immediately aborting
2024-11-24 15:04:51 -05:00
fe1136c54c
Fixed bug with parentheses in lookaround regex; fixed bug with reading last line of test string (if it doesn't end in a newline)
2024-11-24 15:02:58 -05:00
25c333bea4
Added function to determine if a state is a lookaround
2024-11-24 15:01:06 -05:00
74c177324b
Added more test cases
2024-11-24 15:00:47 -05:00
7916629c4d
Added substitute flag - substitute matched text with given text
2024-11-23 10:12:22 -05:00
ee02e7575e
Added function to generate all case variations of a rune
2024-11-23 09:26:27 -05:00
c87a4b7136
Added case-insensitve flag
2024-11-23 09:26:11 -05:00
924e2a8dbc
Added some AI-generated test cases (llama3.1:405b)
2024-11-22 00:12:41 -05:00
21b2d5a2a9
Added lookaround-related fields to postfixNode struct
2024-11-22 00:12:22 -05:00
77d19cd84e
Added lookaround-related fields to State struct, added lookaround support to checkAssertion()
2024-11-22 00:11:51 -05:00
051a8551f3
Match zero-length match at end of string, even if the start node is an assertion (end of string, lookarounds, etc.)
2024-11-22 00:10:58 -05:00
11c0a0552f
Added support for lokarounds; parsing and adding nodes for different lookarounds
2024-11-22 00:10:15 -05:00
c807d6664e
Changed generation of characters for non-whitespace, non-digit and non-word characters - it's basically an inverted character class now
2024-11-20 10:39:24 -05:00
d986999001
Added more tests
2024-11-20 10:38:57 -05:00
ea64ddc88a
Removed unnecessary duplication of assertion checking
2024-11-20 10:38:41 -05:00
1ba871d618
Removed dotChars() function, moved notDotChars() setting to main()
2024-11-20 10:38:22 -05:00
1a1a8f4f9c
Moved flag-checking after flag.Parse()
2024-11-20 10:37:33 -05:00
e882f41400
Added fields to denote all the characters that an 'allChars' postfixNode _shouldn't_ represent (useful for inverting character classes)
2024-11-20 09:41:37 -05:00
b3ee1fe5e8
Convert an inverting character class into an 'allChars' node, with the characters marked as exceptions
2024-11-20 09:40:40 -05:00
708a9e1303
Added field to denote all characters which an 'allChars' node _shouldn't_ match (useful for invertinc character classes
2024-11-20 09:39:24 -05:00
c694c47be7
Added flag to print match indices, and to enable multi-line mode
2024-11-20 01:06:23 -05:00
2569f52552
Wrote toString function for MatchIndex
2024-11-20 01:04:31 -05:00
160b2f9215
Added newline character as an escaped node
2024-11-20 01:04:01 -05:00
992c5a9300
Replaced isAlphaNum() with isNormalChar(), which returns true if the character isn't special (also returns true for unicode characters, which the previous function didn't
2024-11-20 00:24:43 -05:00
1e0502c6aa
Added unicode tests
2024-11-20 00:23:57 -05:00
c56d81a335
Added unicode support to dot metacharacter - it now matches _any_ unicode character (almost)
2024-11-18 16:44:43 -05:00
8a1f1dc621
Added unicode support
...
Replaced strings with rune-slices, which capture unicode codepoints more
accurately.
2024-11-18 10:41:50 -05:00
805766a5ba
Added support for -l : only print lines with at least one match (or with exactly 0 matches, if -v is enabled
2024-11-18 10:02:34 -05:00
dcd712dceb
Added support for -o flag: only print matching content
2024-11-18 09:36:16 -05:00
f2b8812b05
Added support for -v flag, to invert which values are printed in color. Also got rid of unecessary 'else' clause
2024-11-17 22:19:55 -05:00
11641596fa
Read multiple lines from stdin and apply regex to each one; Convert the array of matchIndex structs into a flat array of indices; speeds up process of checking if we have to print a character in color
2024-11-17 21:49:11 -05:00
b55b80ec6c
Updated TODO
...
I didn't like the existing capturing group implementation, so I moved
that to a separate branch. This branch does not (at the moment) any code
relating to capturing groups.
2024-11-17 21:29:18 -05:00
137ea3c746
Made findAllMatchesHelper non-recursive, added pruneIndices (improved performance) and more changes
...
I made findAllMatchesHelper a non-recursive function. It now only
returns the first match it finds in the string (so I should probably
rename it).
These indices are collected by findAllMatches and pruned (to
remove overlaps). The overlap function has also been rewritten, to make
it (I believe) less than O(n^2). I also used the uniq_arr type to make
checking for uniqueness O(1) instaed of O(n) (as it was with
unique_append()). This has resulted in massive performance gains.
There's been a lot of changes here, and I probably haven't documented
all of them.
2024-11-07 16:16:50 -05:00
9201ed49bd
Changed type from matchIndex to MatchIndex
2024-11-07 16:12:21 -05:00
9a073aa514
Added node types for left and right parentheses
2024-11-07 15:55:37 -05:00
7d265495f5
Got rid of list for uniq_arr (O(n) deletion) and instead have separate method to create list (O(n) list creation)
2024-11-07 15:55:13 -05:00
