4.2.0

What's Changed

• fix: warning 'type qualifiers ignored on function return type' in hipo::bank::rowlist by @c-dilks in #54
• Update README.md by @gavalian in #56
• build: switch to Meson by @c-dilks in #58
• fix: finish meson porting and remove Makefiles and CMake files by @c-dilks in #59
• build: add option for warning (non-)suppression by @c-dilks in #62
• Issue 67 - Fix the char as short as int issue. by @mholtrop in #68
• fix: unknown warning option '-Wno-stringop-overflow' from clang++ by @c-dilks in #63
• feat: get the index of a bank in hipo::banklist, given its name by @c-dilks in #65

Full Changelog: 4.1.0...4.2.0

4.1.0

What's Changed

• CMake update to find liblz4 using pkg-config instead of cmake. by @mholtrop in #46
• fix: return nonzero exit code for failures by @c-dilks in #48
• fix: Makefile OS detection by @c-dilks in #52
• build: add libHipoDataFrame to pkg-config file by @c-dilks in #53
• refactor!: replace hipo::bank::iterator with hipo::bank::rowlist by @c-dilks in #50

Full Changelog: 4.0.1...4.1.0

release-4.0.1

Release-4.0.1
Changes include:

• implementation of composite bank structures used for holding decoded FADC and TDC banks when converting from EVIO
• added API for requesting bank list to be read at once from the event, simplifying the API without a need to refer to the dictionary (see examples in the readme file)
• added iterators for banks that allow linking one to many relations in the bank (see examples in README file)
• added a library for simple histograms, which allows filling histograms and displaying them on the terminal for debugging purposes while developing code (see README)
• a physics library is added (header only) that can be used to construct vectors and Lorentz vectors and do some physics.

release-v2.1

Release v2.1
Changes:

• The dc de-noising was removed from the distribution
• Improvements to RDataFrame interface
• Implementation of composite data format for online
• New interface to records for creating records without IndexArray, for quick online record passing

release-v2.0

Primary changes are done to extension module of DC de-noising, to comply with new standards of raw TDC/ADC order definitions. The de-noising code now runs three levels of de-noising labeling them order=0,40,50,60, where combination of all the orders is the un-denoised data.

release v1.9

Changes:

• Implemented de-noising for drift chambers. Runs multi-threaded, processes file, and replaces the DC::tdc bank with a new one with the same number of entries, but the noise hits are marked with order=10, leaving signal hits with order=0
• Implemented record (class hipo::dataframe) for fast record building for online usage, in this version of records there is no event index array (according to specifications), the events have to be parsed one by one in sequence. This class will be used in ET ring when converting EVIO to HIPO
• implanted simple data decoder for applying translation tables to raw data entries (currently works for TDCs, and the DC is the only implemented translation), to be used online for decoding the raw information into detector data.

version 1.8

Release 1.8 with a few improvements and fixes:

• fixed the bug in the event class where structures are not correctly added. (this bug exists only in 1.7)
• introduced no copy methods for reading banks/structures from records without the need of copying the entire event.
• updated benchmarking codes to include ROOT benchmarks and Java benchmarks. Implemented a script that can run all benchmarks (with the given HIPO file, it will convert it to ROOT and benchmark ROOT too).
• The functionality of writing tagged events is introduced in this version (previously only Java had the API). Users now can write events with different tags with hipo::writer class. (example in examples/readFileTags.cc)

release v1.7

Changes:

• event class modified to not write empty nodes/banks
• new class fusion is added to serve as an interface for extern "C" code
• implemented python interface with c-types allowing to read HIPO files from python
• examples in python added showing how to read banks and read entire columns into python arrays for NumPy

release 1.6

Release with new features/improvements in API for storing user-provided configurations in the HIPO file header.
The configurations are stored as key-value pairs, and the value can be JSON/XML strings that will be parsed
by user.
A new example file is added (examples/writeUserHeader.cc), it shows all API calls for writing the configuration
strings and for reading it from the file.

release 1.5 of hipo library

release 1.5 includes :

fixes in file header structure (removed typedef from struct definition)
reader::readDictionary now exits (halts the program) is it's called without opening the file
dictionary::getSchema exits (halts the program) is non exiting schema is requested. Users should first call dictionary::hasSchema then make decision if the bank does not exist.