CIRS REVIEW Documents, labels and formats in general. Text should be 80 characters wide with ^M at end. PDS can provide IDL software to make the change. I agree with Diane Conner about the double quotes and pair of single quotes in vanilla examples. It appears that either will work when running the software. DATA/NAV_DATA/AIRM.FMT, CIFM.FMT, ISPM.FMT, FIFM.FMT, OBS.FMT, SDAV.FMT, SPM.FMT, and likely others in the other subdirectories. Have blank lines beyond end of text. DATA/NAV_DATA/POI.FMT and other formats. Suggest including UNIT in column descriptions, i.e. distance "KM", angles "DEGREE" or "MILLIRADIANS", wavenumbers "CM-1", power "W CM-2 SR-1". Ex. OBJECT = COLUMN NAME = SPACECRAFT_TO_IMAGE_POINT_ZPD DATA_TYPE = PC_REAL START_BYTE = 265 BYTES = 72 ITEMS = 9 ITEM_BYTES = 8 UNIT = "KM" END_OBJECT = COLUMN DATA/UNCALIBR/HSK.FMT, IHSK.FMT, DATA/NAV_DATA/HSK.FMT, IHSK.FMT No column descriptions at all. Column descriptions are also missing from other formats. SOFTWARE/DOC/cirs_vanilla_examples.pdf Example 2 and 3 request spectra. What are the units of the retrieved spectra? Flux units are not listed in the examples. How does request the wavenumber using vanilla? DOCUMENT/DATASIS.PDF and other *.PDF documents. Text in the PDF document is of poor quality on screen, but graphics are fine. Is there a way to improve appearance? Printer output is fine. VOLSIS.PDF states that vanilla-guide.* are in the DOCUMENT/ directory. They are in the SOFTWARE/DOC/ directory. Using Netscape: file:/home/macbeth1/lhuber/CIRS/SOFTWARE/DOC/vanilla-guide.html#Figure1 Figure 1 (Image1.gif) and Figure 2 (Image2.gif) are not displayed nor present in any directory. SOFTWARE/DOC/cirs_calib_soft.pdf: Section 2.5.1 Temperature Maps is empty. Next section is 2.6 Automatic archiving, which also looks incomplete. DATASIS.PDF Geometric Definitions: Any mention of latitude should be prefaced by planetographic or planetocentric so that each definition can stand alone. It is not clear that the geometric data provided explicitely provides the orientation of the spectrograph slit as projected onto the planet. The body_pole_clock_angle specifies the direction of the target body's rotation axis in an image. See: http://pdsproto.jpl.nasa.gov/DDColStdVal/newdd/Top.CFM With some work, this angle could be derived from the Q points. Check that all angles are labeled either degrees or milliradians. Example: SOLAR_PHASE is not listed as degrees. BODY_ANGULAR_SEMIDIAMETER & PRIMARY_ANGULAR_SEMIDIAMETER "The angular diameter of the body as seen by the spacecraft, in milliradians." Is this the equatorial or polar angular semidiameter? Should "diameter" in the definition be replaced with "semidiameter", e.g. radius? CALIB/ directory Only CALINFO.TXT & .LBL in directory, no other files. CALINFO.TXT states that AIFM.FMT, AIFMyymmddhh.DAT and AIFMyymmddhh.VAR should be in the directory. AIFM.FMT is in DATA/NAV_DATA/. SOFTWARE in general The source code and MAKEFILEs are written for use at the GSFC. It includes many references to /usr/local/bin/ for both software and data. It also includes several rsync to GSFC machines. See below for details. The PDS archive users will likely want to establish their own directory structure. Would it be possible to rewrite/compile the software to be run from a directory other than /usr/local/bin/? We do appreciate that Vanilla can be run from any directory as long as it is included in the unix path. PDS recommends that the software be modified to allow the users to define their own directories. One approach is to have a single file where the user can input the location of the CIRS and Vanilla software, perl, cc/gcc, SPICE kernels and data tables. I.e. SOFTWARE/SRC/CIRS_SPE/cirs_spec.c contains references to /data/tables/cruise/current_calib/apod_spectra. A data and software installation guide is needed. The guide should explain the directory structure for both data and software, how the user customizes the software for use on their own machine, the use of the FIXEM.SH to change the software file names, what PERL and C versions were used incase there is a problem with backwards compatibility, and how to compile the software using makefile. CIRS Spectral Utility Programs Requires that the executables be in /usr/local/bin/vanilla/ and data in /data/.... For our system, /usr/local/bin/ owner is root, group other and does not provide us write permission. Until the directory is issue is resolved, it is not possible to run the utility programs on most PDS user's machines. SOFTWARE/SRC/CIRS_CAL/MAKEFILE.CMD get:, put:, get_test:, put_test: all rsync to cirsrss.gsfc.nasa.gov for geometry information(?). The rsyncs must be replaced. SOFTWARE/SRC/CIRS_GEO/VMAKELAB.PL Contains line: open(C,"ckbrief $_ /data/naif/misc/SCLK.ker ... No SCLK.ker present in dataset at PDS. Is SCLK.ker necessary? If it should be provided for the archive? How should the reference to /data/naif/misc/SCLK.ker be changed for a generic install. SOFTWARE/SRC/CIRS_GEO/CIRS_MAK.PL & CIRS_PRE Beneath set up dirs, several directories (/data/tables/cruise, ..., /data/fmt/) are refered to that are not in the PDS data. Are these directories need to use the archive, or just used to create the archive? SOFTWARE/SRC/CIRS_BIU/MAKEFILE.CMD NoiseDetector.o is mixed case. Should be all lower case? Typo $(CC) -o $(BIN)/cirs_biu cirs_biu.o ifgm_rw.o noisedetector.o\ should read $(CC) -o $(BIN)/cirs_biu cirs_biu.o ifgm_rw.o noisedetector.o \ A blank space is needed between ".o" and the "\". SOFTWARE/BIN/SOLARIS/cirs_avifm, cirs_avifm2, cirs_avifm_grid, cirs_geom, cirs_ifm No source code found for these executables. Similarly for BIN/LINUX/. Source code is required. SOFTWARE/SRC/CIRS_SPE/cirs_spec.c and cirs_do_fcal.lp Refer to /data/tables/cruise/current_calib/(un)apod_spectra. The directory is not present in the archive. Should the directory be part of the archive, or is it generated by the software? SOFTWARE in general. There are many references with a leading "/", including to directories not included in archive. Need to delete leading "/", or change to "./" on paths related to the archiving (i.e. /data/, /bin) in mulitple procedures to make the paths relative rather than absolute. Locate using grep -i '"/' */*.* and grep -i "'/" */*.* from SOFTWARE/SCR/. Running cirs_spec generates the error: Unable to stat /data/tables/cruise/current_calib/apod_spectra, bailing out Bad path: /data Unable to load dataset: /data/tables/cruise/current_calib/apod_spectra This is directly related to the leading "/" issue. SOFTWARE/CIRS_PRO/CIRS_MAK.PL (and many others) /data/fmt/ is refered to. No such directory exists in the archive. Presumably, the format should be either DATA/NAV_DATA/, DATA/UNCALIBR/, or DATA/APODSPEC/. Should the .FMT files be in one of the above directories, or should a data/fmt/ directory be added? Either way, leading "/" on path should be deleted. SOFTWARE/DOC/cirs_calib_soft Section 1.2 Important Warning and Disclaimer warns the users that the software was specifically tailored and coded to run on local data systems. This is unacceptable to PDS. Such an extensive, and important software package must be written to allow SOLARIS and LINUX users to operate the software on their systems. Mark Vincent PDS Atmospheres Node