Delivered-To: igsmail@igscb.jpl.nasa.gov Message-ID: <40F7B236.2090002@gsi.go.jp> Date: Fri, 16 Jul 2004 19:47:18 +0900 From: Yuki Hatanaka User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; ja-JP; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) To: igsmail@igscb.jpl.nasa.gov Subject: [IGSMAIL-4975]: Update of RNXCMP for another bug fixing Content-Type: text/plain; charset=ISO-2022-JP Content-Transfer-Encoding: 7bit Sender: owner-igsmail Precedence: bulk ****************************************************************************** IGS Electronic Mail 16 Jul 03:47:33 PDT 2004 Message Number 4975 ****************************************************************************** Author: Yuki Hatanaka Dear colleagues, Another serious bugs are found in RNXCMP programs ver 2.4.1 and 2.4.1a. [The effect of the bug] Problems occure when event flag is 1 or 6. (A) When the event flag is set to 1: (A-1) RNXCMP ver. 2.4.1(a) ends with ERROR when more than 12 satellites are observed. (A-2) RNXCMP ver. 2.4.1(a) finish without error message, but the resultant Compact RINEX file is incorrect and can not be uncompressed by CRX2RNX (any version). (B) Both RNX2CRX and CRX2RNX (2.4.1, 2.4.1a and also all the past versions) doesn't work if event flag is set to 6 (cycle slip data) and more than 12 satellites are observed. The case (A-2) is especially serious since users can overlook unless he/she uncompresses the CRINEX file. [Release of bug-fixed version of programs] The bug mentioned in the above is fixed and the new version (2.4.2) of the programs is now avairable from ftp://terras.gsi.go.jp/software/RNXCMP_2.4.2/ Please update the programs to the newest version, and the versions 2.4.1 and 2.4.1a should not be userd any more. [A tool to recover the RINEX file from erroneous CRINEX] I recomend those users who already used RNX2CRX ver.2.4.1 or 2.4.1a to check if the files can be uncompressed by CRX2RNX. If the decompression is failed, compress the original RINEX file again but using the new program. For the case of problem (A-2), even if the original RINEX file is already discarded, there is a way to recover the RINEX from the erroneous CRINEX file created by 2.4.1 or 2.4.1a. I would provide a special program to do it: ftp://terras.gsi.go.jp/software/C2RFX242/ There are C-source code, DOS.exe, and binaries for several unix architectures. The usage is the same as CRX2RNX. (This is actually a special version of CRX2RNX.) Type C2RFX242 -h for showing help message. [Remarks on use of old versions of the programs] The following is the statistics of # of CRINEX files compressed by specific versions of RNX2CRX in IGS data archive: (DOY 191, 2004) 2.2beta 21 2.3beta 77 2.4.0 172 total 270 Surprisingly for me, 2.2beta (which was updated to 2.3beta in 1998) is still used!! As well as the bugs in 2.4.0 (and 2.3beta), which I mentioned in IGSmail #4967, 2.2beta containes some more bugs: - The compression stoped with error in the case the epoch flag (>1) is set and the "number of records follow" field is a blank (which means zero). I recomend users of old versions to update the programs to the newest ones if possible. If it's not possible, the users should be aware of the bugs and be careful so that their data is free from the effects of the reported bugs. Finally (I hope it's final), I apologize for the troubles you may have by the bugs and for inconvenience by the frequent update. Also I thank people who pointed out the problems. Best Regards, Yuki ------------------------------------------------- HATANAKA,Yuki, Geographical Survey Institute e-mail : hata@gsi-mc.go.jp -------------------------------------------------