summaryrefslogtreecommitdiff
path: root/hiredis.c
AgeCommit message (Collapse)Author
2012-11-06Try again later for EINTR errors (see issue #99)Fabien MARTY
2012-08-21On stack buffer to read replies set to 16k (was 2k).antirez
It was verified experimentally that this value, on Linux kernels, provides better performances compared to the 2k value. However larger values apparently don't produce any noticeable effect on performances.
2012-08-21Max depth of multi-bulk reply moved from 2 to 7.antirez
Hiredis can handle multi bulk replies with a fixed (hardcoded) level of nesting. This should be changed in the future in order to avoid hardcoded limits. As a quick fix this commit moves the max nesting from 2 to 7, so that there are no problems when processing replies from the SLOWLOG command, from Redis Sentinel, or generated by Redis Lua Scripts (that are allowed to generate replies with any level of nesting).
2012-08-21Configurable reader max idle buffer size.antirez
Hiredis used to free unused redisReader buffers bigger than 16k. Now this limit is configurable (see the documentation updated by this commit) in order to allow working with big payloads without incurring to speed penalty.
2011-12-20DRY list of integer formatsChip Salzenberg
2011-12-02increase read depth by one, would allow ruby redis client to read from slowlogmrb
2011-11-28Fix minor inconsistencies in comment.John Zwinck
2011-07-09Abort on invalid formatPieter Noordhuis
There is no way we can guess the width of the argument when we cannot infer its type from the format specifier.
2011-07-09Use correct type when calling va_arg in formatterPieter Noordhuis
2011-06-19Revert "Add function to retrieve formatted reply"Pieter Noordhuis
This reverts commit 77540aa31694aa1e14d41f60a0452e49a2fed86a. The change in buffer strategy is too large to put in a minor release. It is put in a separate branch in the meantime, so it can be refined and released together with a minor version bump.
2011-06-09Fix for issue #45Pieter Noordhuis
2011-05-05Add function to retrieve formatted replyPieter Noordhuis
This is done by only truncating the read buffer once a full reply has been read. The buffer is no longer truncated halfway through reading a reply. In addition: pass offset/length of protocol and content via the read tasks.
2011-04-21Update licensePieter Noordhuis
2011-04-21Let the async API segfault on OOM for nowPieter Noordhuis
2011-04-21Move chrtos to hiredis.c (not used elsewhere)Pieter Noordhuis
2011-04-21Handle OOM when doing blocking requestsPieter Noordhuis
2011-04-21More safetyPieter Noordhuis
2011-04-21Make command formatters gracefully abort when out of memoryPieter Noordhuis
2011-04-21More OOM checks in the protocol readerPieter Noordhuis
2011-04-21Include fmacros.h to make sure strerror_p is defined on LinuxPieter Noordhuis
2011-04-21Rename *ReplyReader* to *Reader*Pieter Noordhuis
2011-04-21Use static buffer for error string on contextPieter Noordhuis
2011-04-21Create protocol reader when creating contextPieter Noordhuis
2011-04-21Reply object functions can be set by the user after creating the readerPieter Noordhuis
2011-04-21Privdata can be set directly on the structPieter Noordhuis
2011-04-21Remove useless castsPieter Noordhuis
2011-04-21Return when memory cannot be allocatedPieter Noordhuis
2011-04-21Remove accessor functions now the redisReader struct is publicPieter Noordhuis
2011-04-21Make redisReader struct publicPieter Noordhuis
2011-04-21Test if there is a freeObject functionPieter Noordhuis
2011-04-20Use pre-allocated buffer for error strings in readerPieter Noordhuis
2011-04-20First pass at making the protocol reader properly handle OOMPieter Noordhuis
2011-04-03Lazily destroy protocol reader bufferPieter Noordhuis
2011-04-03Individual reply object functions may be NULLPieter Noordhuis
2011-03-06Fix formatCommand to work with all empty interpolationsPieter Noordhuis
2011-02-04Use select(2) for enforce a timeout on blocking connect(2)Pieter Noordhuis
2011-01-27Properly reset state after protocol error in reply readerPieter Noordhuis
2011-01-27redisReplyReaderFeed should take a const char*Pieter Noordhuis
2011-01-27Avoid warnings with -Wstrict-prototypesPieter Noordhuis
2011-01-19Let sds.c figure out where the range stopsPieter Noordhuis
2011-01-19Discard consumed part of buffer less oftenPieter Noordhuis
2011-01-07Return error on socket timeout for a blocking contextPieter Noordhuis
2010-12-28Add function to explicitly free an async contextPieter Noordhuis
2010-12-16Add myself to license in some filesPieter Noordhuis
2010-12-02Add (nearly) full printf support by delegating to vsprintfPieter Noordhuis
2010-12-01Wait with setting CONNECTED until there is an fdPieter Noordhuis
2010-11-26Use r->len in another spotPieter Noordhuis
2010-11-26Use custom strollPieter Noordhuis
2010-11-26Reply objects don't need to be calloc'edPieter Noordhuis
2010-11-26Use explicit buffer length field in the reply readerPieter Noordhuis