summaryrefslogtreecommitdiffstats
path: root/grammar
Commit message (Collapse)AuthorAgeFilesLines
* bugfix: comments inside objects (e.g. action()) were not properly handledRainer Gerhards2012-10-041-0/+1
|
* Implemented different grammar for pri filtersRainer Gerhards2012-09-211-1/+1
| | | | | | | | The method introduced by the previous commit caused issues in v7 as it was too generic. Most importantly, it permitted simple words (like "stop") to (invalidly) be detected as pri filters. Now, the grammar is close to the initial one, and each pri filter must at least have a comma or a period inside it, which does not conflict with simple words.
* bugfix: some valid legacy PRI filters were flagged as errornousRainer Gerhards2012-09-211-1/+1
| | | | | | closes: http://bugzilla.adiscon.com/show_bug.cgi?id=358 This happend to filters of the style "local0,local1.*", where the multiple facilities were comma-separated.
* Fix length calculation for script string constantsRainer Gerhards2012-09-141-2/+2
| | | | | | This could lead to various problems, like if-filters not working. Note this is a regression from yesterdays escape fix, so there exist no released version with this problem.
* Fixing string unescaping (new code from this morning's commit)Rainer Gerhards2012-09-131-1/+1
|
* bugfix: missing support for escape sequences in RainerScriptRainer Gerhards2012-09-133-2/+148
| | | | | | | Only \' was supported. Now the usual set is supported. Note that v5 used \x as escape where x was any character (e.g. "\n" meant "n" and NOT LF). This also means there is some incompatibility to v5 for well-know sequences. Better break it now than later.
* bugfix: expression-based filters with AND/OR could segfaultRainer Gerhards2012-06-111-2/+4
| | | | | | | | due to a problem with boolean shortcut operations. From the user's perspective, the segfault is almost non-deterministic (it occurs when a shortcut is used). Thanks to Lars Peterson for providing the initial bug report and his support in solving it.
* regression fix: last memleak bugfix caused abort in some situationsRainer Gerhards2012-06-051-1/+3
| | | | | The problem was introduced by the fix I did roughly an hour ago. Obviously, no released version is affected.
* restructured cfsysline handling a bitRainer Gerhards2012-06-051-3/+2
| | | | so that we have cleaner code
* cosmetic: removed autoconf warning messagesRainer Gerhards2012-06-041-4/+4
|
* bugfix: number conversion in expression evaluation was not always correctRainer Gerhards2012-04-231-2/+4
| | | | | was invalid when the property was no string. Currently all props are strings, so this bug never could hit in practice.
* conf file: bufgix: legacy parsing of some filters did not work correctlyRainer Gerhards2012-02-012-2/+1
|
* added support for v6 config system to omfwdRainer Gerhards2011-10-111-2/+2
|
* bugfix: facility local<x> was not correctly interpreted in legacy filtersRainer Gerhards2011-09-281-1/+1
| | | | | Was only accepted if it was the first PRI in a multi-filter PRI. Thanks to forum user Mark for bringing this to our attention.
* bugfix: config parser did not support properties with dashes in themRainer Gerhards2011-09-151-1/+1
| | | | ...inside property-based filters. Thanks to Gerrit Seré for reporting this.
* milestone: queue object now has a param handler for new conf interfaceRainer Gerhards2011-07-223-3/+48
| | | | ... and action queue defs use this new interface (but not yet the main queues)
* need to handle legacy-legacy omusrmsg format stricterRainer Gerhards2011-07-221-1/+1
| | | | otherwise, the grammar for if-constructs was broken
* milestone: all syntaxes for new config handler implementedRainer Gerhards2011-07-222-1/+131
| | | | | | so far, mostly compile-time tested. Real testing happens as side-activity when implementing other parts of the new config system
* omusrmsg now supports the new config systemRainer Gerhards2011-07-211-1/+1
|
* bugfix: abort in omfile (in brand-new code)Rainer Gerhards2011-07-213-1/+7
|
* fixing minor memory leaksRainer Gerhards2011-07-212-1/+46
|
* milestone: done plumbing to call plugin create action instance entry pointRainer Gerhards2011-07-201-2/+3
|
* milestone: size syntax implementedRainer Gerhards2011-07-191-0/+48
|
* milestone: glbl obj parameters settable via new conf interfaceRainer Gerhards2011-07-192-0/+3
|
* milestone: two syntaxes for get conf param implementedRainer Gerhards2011-07-192-4/+76
|
* milestone: first steps at global() conf obj implementationRainer Gerhards2011-07-193-0/+184
| | | | | also, the foundation for accessing conf file params has been laid. Still more work to do...
* removed some printf's left in lexerRainer Gerhards2011-07-151-4/+5
| | | | | may cause some mild confusion if someone actually uses the not-yet-officially-supported construct
* better check for config errorsRainer Gerhards2011-07-154-23/+63
|
* renamed exprret type to var in spite of more generic useRainer Gerhards2011-07-152-49/+55
|
* preparing for 6.3.3v6.3.3Rainer Gerhards2011-07-131-0/+1
|
* forgotten to support the discard action ("~")Rainer Gerhards2011-07-131-0/+1
|
* doc: v6 compatiblity doc addedRainer Gerhards2011-07-131-3/+4
|
* lexer bug: *.somewhat PRIs not detectedRainer Gerhards2011-07-131-1/+1
| | | | regression from last update
* removed emergency config, do error reporting on fatal config file errorRainer Gerhards2011-07-122-4/+5
|
* fixed lexer: facilities local0-local7 were not correctly handledRainer Gerhards2011-07-111-1/+5
|
* cleaup & emergency config system reactivatedRainer Gerhards2011-07-092-0/+24
|
* milestone/[WORKS AGAIN!]: looks like the new conf format is integratedRainer Gerhards2011-07-093-1/+54
| | | | finally completed $IncludeConfig processing.
* milestone/[PARTWORK]: made sure all legacy selectors workRainer Gerhards2011-07-092-1/+3
| | | | | ... except for the not guaranteed $outchannel construct, which we could not fix without syntax change. Shouldn't hurt too many.
* checking number of parameters provided to functionRainer Gerhards2011-07-091-2/+32
|
* milestone/[PARTWORK]: implemented RainerScript functionsRainer Gerhards2011-07-093-7/+129
|
* optimized function representationRainer Gerhards2011-07-082-9/+31
|
* milestone/[PARTWORK]: integrated all variable types (msg/sys/cee)Rainer Gerhards2011-07-081-3/+21
|
* milestone/[PARTWORK]: implemented comparison operationsRainer Gerhards2011-07-082-33/+222
|
* milestone/[PARTWORK]: implemented "CONTAINS"Rainer Gerhards2011-07-081-7/+19
|
* milestone/[PARTWORK]: obtaining msg vars integrated, "==" works for stringsRainer Gerhards2011-07-074-20/+61
|
* milestone/[PARTWORK]: integrted script filter, but var access is missingRainer Gerhards2011-07-077-12/+85
|
* milestone/[NONWORKING]: first integration of new parser, rules are not yet ↵Rainer Gerhards2011-07-063-9/+10
| | | | handled
* milestone: abstracted parser interface ...Rainer Gerhards2011-07-063-9/+44
| | | | | so that both testdriver and other callers (rsyslog!) can use it without changing the parser (this simplifies development).
* milestone: some plumbing for rsyslog/stand alone parsingRainer Gerhards2011-07-055-48/+83
|
* milestone: improved build systemRainer Gerhards2011-07-055-31/+79
| | | | | ... still had quite some glitches, as usual. This time it hopefully works under all circumstances (well, let's hope for "usual cir..." ;)).
generated by cgit v1.2.3 (git 2.25.1) at 2025-07-16 15:33:03 +0000