Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Removed protocol ('http') from links in documentationv1.1.0 | Roberto Ierusalimschy | 2023-06-26 | 3 | -14/+14 |
| | |||||
* | Missing files in 'pack' script | Roberto Ierusalimschy | 2023-06-23 | 1 | -2/+2 |
| | |||||
* | Added a README file | Roberto Ierusalimschy | 2023-06-23 | 3 | -4/+11 |
| | |||||
* | Copyright notice in module 're' | Roberto Ierusalimschy | 2023-06-23 | 2 | -34/+10 |
| | |||||
* | Fold capture deprecated | Roberto Ierusalimschy | 2023-06-23 | 1 | -52/+16 |
| | |||||
* | Improved documentation for accumulator captures | Roberto Ierusalimschy | 2023-06-22 | 3 | -59/+75 |
| | |||||
* | Accumulator pattern added to the 're' module | Roberto Ierusalimschy | 2023-06-22 | 2 | -8/+15 |
| | |||||
* | Detail in 're.lua' | Roberto Ierusalimschy | 2023-06-19 | 1 | -2/+2 |
| | | | | Distinction between 'm' and 'mm'. | ||||
* | Some fixes in vibibility check for back captures | Roberto Ierusalimschy | 2023-06-19 | 6 | -46/+97 |
| | |||||
* | Full captures can contain nested captures | Roberto Ierusalimschy | 2023-06-15 | 5 | -97/+173 |
| | | | | | | Nested captures can be recognized because they start (and end) inside the character range of the full capture. This optimization can remove a lot of 'close' captures from the capture logs. | ||||
* | Captures point to string positions using indices | Roberto Ierusalimschy | 2023-06-07 | 6 | -30/+60 |
| | | | | | That uses 4 bytes (uint) instead of 8 (char*); the size of the structure 'Capture' reduces from 16 to 8 bytes in 64-bit machines. | ||||
* | First implementation for the accumulator capture | Roberto Ierusalimschy | 2023-06-06 | 8 | -22/+131 |
| | |||||
* | Code size stored in code itself | Roberto Ierusalimschy | 2023-05-29 | 7 | -28/+66 |
| | | | | | | | Most patterns do not have code, as they are not directly used for a match; they are created only to compose larger patterns. So, we shouldn't waste space to store the size of their code, as a NULL pointer already indicates that the size is zero. | ||||
* | New type name 'uint' (unsigned int) | Roberto Ierusalimschy | 2023-04-28 | 5 | -17/+19 |
| | |||||
* | Small optimization in size of charset trees | Roberto Ierusalimschy | 2023-04-27 | 3 | -29/+35 |
| | | | | Got a byte that was wasted for padding to be used in the bitmap. | ||||
* | New macros 'fillset' and 'clearset' | Roberto Ierusalimschy | 2023-04-27 | 6 | -11/+22 |
| | |||||
* | Compact charsets used in trees, too. | Roberto Ierusalimschy | 2023-04-27 | 9 | -120/+178 |
| | |||||
* | New module 'lpcset' | Roberto Ierusalimschy | 2023-04-26 | 4 | -79/+104 |
| | | | | For code related to compact sets. | ||||
* | Instruction array grows by factor of 1.5 | Roberto Ierusalimschy | 2023-04-25 | 1 | -9/+18 |
| | |||||
* | Fixing several bugs left in previous commit | Roberto Ierusalimschy | 2023-04-25 | 3 | -22/+47 |
| | |||||
* | Smaller encoding for charsets in code | Roberto Ierusalimschy | 2023-04-24 | 3 | -75/+133 |
| | |||||
* | Towards a smaller encoding for charsets in code | Roberto Ierusalimschy | 2023-04-23 | 6 | -28/+69 |
| | |||||
* | Field Instruction.key put inside a union | Roberto Ierusalimschy | 2023-04-21 | 5 | -26/+28 |
| | | | | So that we can get its space for other uses, if needed. | ||||
* | Small extra test | Roberto Ierusalimschy | 2023-04-11 | 1 | -2/+6 |
| | |||||
* | Documentation | Roberto Ierusalimschy | 2023-04-10 | 15 | -134/+15 |
| | | | | | Removed '$Id' from all files + updated copyright year + other changes in comments and documentation | ||||
* | Capture-list grow factor changed to 1.5 | Roberto Ierusalimschy | 2022-12-09 | 1 | -2/+2 |
| | |||||
* | Clearer error message for 'pattern / nil' | Roberto Ierusalimschy | 2022-08-18 | 1 | -1/+3 |
| | |||||
* | Details | Roberto Ierusalimschy | 2022-07-06 | 3 | -5/+6 |
| | |||||
* | Better computation of size limit in 'growcap' | Roberto Ierusalimschy | 2022-02-22 | 1 | -4/+13 |
| | | | | | | Old code mixed size in elements (counted with 'int's) with size in bytes (division by the size of an element). Moreover, when it could not double the size, it grew it one by one, which is undoable performance-wise. | ||||
* | Bug: IBackCommit must remove dynamic captures, too | Roberto Ierusalimschy | 2019-07-19 | 3 | -1/+14 |
| | | | | | | Like a fail, a IBackCommit instruction must remove any dynamic capture made inside an 'and' pattern. (The added test for this problem needs assertions on to detect the bug.) | ||||
* | Added small note about LuaRocks in the Web page | Roberto Ierusalimschy | 2019-05-28 | 1 | -0/+7 |
| | |||||
* | 'lpeg.version' changed to a string | Roberto Ierusalimschy | 2019-04-22 | 4 | -14/+11 |
| | |||||
* | Implementation of UTF-8 ranges | Roberto Ierusalimschy | 2019-04-17 | 9 | -20/+222 |
| | | | | | New constructor 'lpeg.utfR(from, to)' creates a pattern that matches UTF-8 byte sequences representing code points in the range [from, to]. | ||||
* | Removed 'unsigned char' limit on number of rules in grammars | Roberto Ierusalimschy | 2019-04-14 | 7 | -48/+75 |
| | | | | | | Added a new tree-type node 'TXInfo', which follows 'TRule' nodes, to store extra information about a node. (In this case, the rule number, with an 'unsigned short' field.) | ||||
* | Some details before release 1.0.2v1.0.2 | Roberto Ierusalimschy | 2019-03-11 | 7 | -20/+107 |
| | | | | | | | - changed release number - small corrections in 'pack' - removed "about" section in .html files - added files 'HISTORY' and 'lpeg-128.gif' | ||||
* | Fixed bug when resizing capture list | Roberto Ierusalimschy | 2019-02-27 | 2 | -42/+51 |
| | | | | | | | | | | Fixed the bug reported in http://lua-users.org/lists/lua-l/2018-11/msg00080.html The field 's' of the open group of captures inside a run-time capture was being reused by the open group of the results of the run-time capture, even though that open-group entry was being removed from the list and then added again. | ||||
* | Avoid stack overflow when handling nested captures | Roberto Ierusalimschy | 2019-02-20 | 5 | -20/+50 |
| | | | | | | The C code uses recursion to handle nested captures, so a too deep nesting could create a stack overflow. The fix limits the handling of nested captures to 'MAXRECLEVEL' (default is 200 levels). | ||||
* | Removed extra information from RCS keyword strings | Roberto Ierusalimschy | 2019-02-20 | 15 | -15/+15 |
| | | | | | Version numbers and dates from RCS keyword strings removed from all source files; keeps only the file name. | ||||
* | Fist version of LPeg on GIT | Roberto Ierusalimschy | 2019-02-20 | 18 | -0/+7899 |
LPeg repository is being moved to git. Past versions won't be moved; they are still available in RCS. |