bugfix in rich unpacker

This MR fixes a bug in RICH unpacker, that is shown by the following compiler warnings:

/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx: In function 'std::__cxx11::string mRichSupport::GetWordHexRepr(const uint8_t*)':
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx:69:32: warning: 'snprintf' output truncated before the last format character [-Wformat-truncation=]
   snprintf(cStr, buf_size - 1, "%02x%02x %02x%02x", byte[0], byte[1], byte[2], byte[3]);
                                ^~~~~~~~~~~~~~~~~~~
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx:69:11: note: 'snprintf' output 10 bytes into a destination of size 9
   snprintf(cStr, buf_size - 1, "%02x%02x %02x%02x", byte[0], byte[1], byte[2], byte[3]);
   ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx: In function 'std::__cxx11::string mRichSupport::GetBinaryRepresentation(size_t, const uint8_t*)':
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx:23:36: warning: 'snprintf' output truncated before the last format character [-Wformat-truncation=]
       snprintf(cStr, buf_size - 1, "%u", byte);
                                    ^~~~
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx:23:15: note: 'snprintf' output 2 bytes into a destination of size 1
       snprintf(cStr, buf_size - 1, "%u", byte);
       ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx: In function 'std::__cxx11::string mRichSupport::GetHexRepresentation(size_t, const uint8_t*)':
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx:47:34: warning: 'snprintf' output truncated before the last format character [-Wformat-truncation=]
     snprintf(cStr, buf_size - 1, "%02x", byte);
                                  ^~~~~~
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx:47:13: note: 'snprintf' output 3 bytes into a destination of size 2
     snprintf(cStr, buf_size - 1, "%02x", byte);
     ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx: In function 'std::__cxx11::string mRichSupport::GetWordHexReprInv(const uint8_t*)':
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx:91:32: warning: 'snprintf' output truncated before the last format character [-Wformat-truncation=]
   snprintf(cStr, buf_size - 1, "%02x%02x %02x%02x", byte[0], byte[1], byte[2], byte[3]);
                                ^~~~~~~~~~~~~~~~~~~
/home/cbmdock/cbmroot/fles/mcbm2018/unpacker/CbmMcbm2018UnpackerUtilRich2020.cxx:91:11: note: 'snprintf' output 10 bytes into a destination of size 9
   snprintf(cStr, buf_size - 1, "%02x%02x %02x%02x", byte[0], byte[1], byte[2], byte[3]);
   ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The code above assumes that the second snprintf() parameter, "maximal length", is the maximal number of characters one can put in the buffer (which is the buffer length minus 1). But snprintf() considers the second parameter as the full length of the buffer (including the last '\0').

Therefore in all the snprintf() calls above, all the messages are truncated.

I don't know how these messages are used, i.e., if my bugfix changes the results or if it only changes some log output.

added BugFix mCBM labels

requested review from @d.smith

assigned to @d.smith

changed the description

Dear @f.uhlig, @p.-a.loizeau,

you have been identified as code owner of at least one file which was changed with this merge request.

Please check the changes and approve them or request changes.

added CodeOwners label

Should be ok, this is legacy code from the mCBM beamtimes with the previous FPGA board.

The problem here was introduced when @f.uhlig replaced the original deprecated sprintf with the safer snprintf ~1 year ago in !1115 (merged) and this stayed since then as nobody maintained it.
Strange thing is that I though we fixed similar warnings in a follow up MR a few months afterward when we realized the range mistake, but the 3 files I re-checked today were not edited afterward. So maybe we will want to check all changes from !1115 (merged)

approved this merge request

Thanks @p.-a.loizeau.

@se.gorbunov Please merge (I never worked on the class changed in this MR).

added 2 commits

• a3d6cd98 - 1 commit from branch computing:master
• 5471d2cd - bugfix in rich unpacker

Compare with previous version

enabled an automatic merge when the pipeline for 5471d2cd succeeds

merged

added WarnFix label