Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/hst/ghrs/documents/isrs/ghrs_isr093.pdf Дата изменения: Unknown Дата индексирования: Sun Mar 2 03:06:23 2014 Кодировка: Поисковые слова: http astrokuban.info astrokuban

Instrument Science Report GHRS 2014-093

Modifications to Final Reprocessed GHRS Headers before Ingest
D. Fraquelli (CSC/STScI), L. Gardner (STScI), A. Alexov (STScI), Daniel Durand (CADC) February 12, 2014

A

BSTRACT

The report describes the changes made to the FITS header keywords and the data files before the final reprocessed GHRS data were ingested into the archive

Introduction
In 2006 the GHRS dataset, which consists of 24,272 datasets, containing 298,965 FITS files, was reprocessed using the best available reference files. The dataset contains PODPS Data Quality (PDQ) and Observer Comment (OCX) files, which were produced manually by operations personnel. Both science and non-science (i.e., DARKs, etc.) exposures were reprocessed to produce as homogeneous a dataset as possible. See GHRS ISR-092 (Kamp et al., 2006) for details of the reprocessing. After reprocessing, the dataset was copied from CADC to ST-ECF and STScI. At STScI, the dataset was placed on disk for direct download, but was not ingested into the archive, nor was the catalog updated. Archive users who searched and downloaded GHRS data from STScI receive the original processed data, not the reprocessed version. To remedy the situation, we investigated the reprocessed dataset, determined what was needed to allow ingest and cataloging of the data, made the changes and then ingested and cataloged the data. What was done to the keywords and data is detailed below.

1


Data Set Definitions
We call the reader's attention to section 2.3, "Uncalibrated Data" from GHRS ISR-092.
"A total of 3949 datasets were left uncalibrated. The old archived version in DADS contained also only uncalibrated files for these datasets. Most of these datasets are target acquisition data that have either no coordinates, no start time, no exposure duration or no targetname specified. For these datasets, the uncalibrated files remained unchanged from the old processing archived in DADS."

Our charge was to ingest the final reprocessed data set into the archive. We did not pursue reprocessing any additional data (i.e, the 3949 datasets mentioned in GHRS ISR092). As they are not included in the ECF set, none of the details in this ISR apply to these files. We did spot check these datasets and confirmed that typically only the standard header packet (shf), unique data log (ulf) and trailer file (trl) are present in the archive. Although part of the science dataset for a typical exposure, these files do not contain data recognized as science data. In cases where a d0f file was present, the data were obtained for engineering purposes (i.e., during orbital verification) and/or with the SAF element in the light path. We did, however, retrieve the OCX and PDQ files for the datasets, made them FITS compliant and ingested them as part of the Augmented Set. The table below gives the number of datasets and files present in the Archive at the start of this work, in the ECF set and in the Augmented set. Original Archive Dataset Science Datasets OCX* Datasets PDQ* Datasets 24,272 10,735 24,010 Original Archive Files 264,220 10,735 24,010 298,965 ECF Dataset 20,323 8,570 20,137 49,030 ECF Files 245,997 8,570 20,137 274,704 Augmented Dataset 20,323 10,734 24,009 55,066 Augmented Files 245,997 10,734 24,009 280,740

Total (files 59,017 or datasets)

*The archive contains z2d40105x_ocx.fits and z2dn0101t_pdq.fits, both for DEFCAL's (an automatically scheduled, non-data producing observation whose sole purpose was to center the spectrum on the detector). These files are corrupted and cannot be read or fixed by FITS tools, so they were not included in the Augmented Set. Interested users may retrieve the data and use the linux/unix "more" command to view the content. Inspection of the above table shows a Science Dataset number difference between the Original Archive (hereafter called Archive) and the Augmented Datasets of 3949

2


datasets. We have confirmed these are the same 3949 datasets that were not recalibrated in the work described in GHRS ISR-092. The difference in the number of total datasets is 3951 because the Archive contains an additional PDQ and OCX dataset that the Augmented Dataset does not. See the footnote on the table. It was the Augmented Set that was verified, fixed and ingested into the archive.

Inspection and Correction of the GHRS Dataset
The reprocessed dataset consists of a set of FITS files for each exposure. We ran a FITS verifier, fverify, on these files to ensure they were valid FITS. A number of errors were received, most of which fell into one of seven types. Typical fverify errors are given in the Appendix. 1) File extensions: .c0f, .c1f, .c2f, .c3f, .c4f, .c5f, .cqf, .d0f, .q0f, .shf, .ulf, .x0f, .xqf FITS ASCII Table HDU has reserved Image WCS header keys which are not allowed in non-Image HDUs. 2) File extensions: .c0f, .c1f, .c2f, .c3f, .c4f, .c5f, .cqf, .d0f, q0f, .ocx, .ulf, .pdq, .shf, .trl, .x0f, .xqf, .ulf Data fill area invalid. 3) File extensions: .pdq, .ocx, .trl FITS ASCII Table HDU contains non-ASCII-text characters. 4) File extensions: .q0f, .d0f FITS ASCII Table HDU TDISP11 value `L1' is not allowed. 5) File extensions: .q0f, .d0f EQUINOX value = J2000 is not a floating point number. 6) File extensions: .c0f, .c1f, .c2f, .c3f, .c4f, .c5f, .cqf, .q0f, .d0f, .d1f DATE-OBS date string has illegal format. 7) File extensions: .pdq, .ocx, .trl FITS Warning (not error) about column names containing "-" characters. The following changes were made to bring the data into compliance with the current FITS standard. Following the order of the fverify issues above, our solutions to resolve each issue were: 1) For those files with a FITS BINARY/ASCII Table extension (HDU1) which contained a copy of the WCS header keywords, along with a few additional non-WCS keywords, we removed all Image-related WCS keywords. Note that the WCS keywords were also in the primary (Image) header, so they were redundant in Table extension. The following keywords were removed from the Table HDU: DATAMAX, DATAMIN, CTYPE1, CTYPE2, CRPIX1, CRPIX2, CRVAL1, CRVAL2, CD1_2, CD2_1, CD2_2, CD1_1. 2) The "Data fill area invalid" FITS error went away when we opened, fixed, and closed all the headers in all the files. The process of fixing the headers also fixed the FITS block sizes when the headers were written to disk, hence fixing this problem.

3


3) All non-ASCII-text characters were converted to spaces. We found three such characters throughout the log HDU extensions: unix bells, tabs and nulls. None of these are allowed in ASCII FITS Tables. 4) In some of the FITS ASCII Table HDUs, the header keyword TDISP11 value had to be reset from `L1' to `A1' to be FITS-compliant. The display format was for the column MIR_REVR in logical format; the logical display format (L1) is deprecated in FITS and was changed to character type format (A1). 5) The type and value of the EQUINOX keyword was changed from string to floating point, and from `J2000' to 2000 (float), to bring it into compliance with the current FITS standard. 6) The format of the DATE-OBS keyword was changed to comply with the FITS standard. The comment regarding the DATE-OBS format was similarly updated to `yyyy-mm-dd'. 7) We also cleaned up FITS Warning messages by deleting the "-" character from the column name: (PDQ-FILE Х PDQFILE; OCX-FILE Х OCXFILE; TRL-FILE Х TRAILERFILE) Nine OCX and PDQ files in the reprocessed data set were corrupted, incomplete and not readable by FITS tools. These files were also corrupted in the STScI and CADC archives, with the exception of one, z2sr0107t.pdq.fits. We determined that all relevant values in OCX and PDQ files are also present in the CAL data headers, and therefore the PDQ and OCX contain largely redundant information that could be deleted. The following is a list of files that were subsequently removed from the reprocessed set. z2co0407t.pdq.fits z2co040at.pdq.fits z2d40106x.ocx.fits z2d4010cx.ocx.fits z2d4010dx.ocx.fits z2d4010ex.ocx.fits z2d4010gx.ocx.fits z3d5030at.pdq.fits There were two duplicate sets of PDQ files: 1) z0l70601r.pdq.fits & z0l70601t.pdq.fits 2) z0l70602r.pdq.fits & z0l70602t.pdq.fits

4


We checked the "r" version against the "t" version headers and data content and found no differences other than those related to the file name itself. We decided to delete the "r" versions of these files. Therefore, the final processed FITS file total is 280,740 (280,750 original files minus 8 incomplete PDQ/OCX datafiles, minus two duplicate "r" PDQ files). This is the Augmented Set in the table above. We used python and shell scripts to make the changes listed above. Fverify was re-run after we fixed and cleaned up the data, to be sure no FITS errors remained. We confirmed that the FITS files were backwards compatible in GEIS format by running the IRAF task `strfits' on a number of test cases, to be sure no errors occurred. Files were randomly visually inspected in DS9 and FV. The HEASARC FITS diff tool (fdiff) was run on a number of test cases in order to be sure that the only differences present were those which were expected; we diff'ed the ECF versions vs our newly processed versions.

Keyword and Catalog Value Differences
The values of the FITS header keywords were compared to the existing values in the database (archive catalog). For the majority of the keywords, values matched between the headers and the database, or the differences were at the level of precision of the value or were a formatting difference. For those cases with a significant difference between the header and the database, an investigation was conducted to determine which should be used. In the table below we list 88 keywords where a difference was seen between the FITS files and the database, the number of datasets in which each difference was seen and the source (header or database) from which the value will be taken for this ingest. Header keyword Number of datasets Comment with differences ABSHFILE 3028 Header keyw ADC_CORR 107 Header keyw BCK_CORR 54 Header keyw BMD_CORR 16187 Header keyw CCR3 95 Header keyw CCR5 7663 Header keyw CCR6 692 Header keyw CCR7 7569 Header keyw CCR8 11191 Header keyw CCR9 10820 Header keyw CCRA 13731 Header keyw CCRB 15264 Header keyw

ord ord ord ord ord ord ord ord ord ord ord ord

deemed deemed deemed deemed deemed deemed deemed deemed deemed deemed deemed deemed

correct correct correct correct correct correct correct correct correct correct correct correct

5


CCRC CCRD CCRE DATE-OBS DECAPER1 DEC_V1 DIOFILE DIO_CORR DOPZER DQIFILE DQI_CORR ECH_CORR EXPEND EXPSTART EXP_CORR FLX_CORR GRATING GWC_CORR HEL_CORR IAC_CORR MAP_CORR MAXWAVE MDF_CORR MER_CORR MINWAVE MNF_CORR NETHFILE OPFORMAT PARENTID PA_V3 PHCFILE PHC_CORR PLY_CORR PPC_CORR PR_INV_L RA_APER1 RA_V1 SAAFILE SPEC_1 SPORDER TARGNAME UTC0 VIGHFILE VIG_CORR ZCST

9177 8830 16187 37 2278 9668 190 757 37 1397 2149 5347 36 37 2081 54 9 8880 54 3737 757 10764 1 124 10338 1 3027 1552 1097 139 11630 6402 28 2081 1099 2408 9671 312 11 107 1806 55 7139 54 7139

H H H H H H H H H H H H H H H H U H H H H H H H H H H U H H H H H H U H H H U H U H H H H

eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw se database eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw se database eader keyw eader keyw eader keyw eader keyw eader keyw eader keyw se database eader keyw eader keyw eader keyw se database eader keyw se database eader keyw eader keyw eader keyw eader keyw

ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed value ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed value ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed value ord deemed ord deemed ord deemed value ord deemed value ord deemed ord deemed ord deemed ord deemed

correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct

6


ZDEBTF ZDEBTR ZDETT1 ZDETT2 ZDRT ZET11 ZET31 ZFIAT ZFIBT ZFICT ZHDC1 ZHDC2 ZHVPST1 ZHVSPT2 ZLCOEF1 ZLCOEF2 ZMEBT1 ZMEBT2 ZOBBT ZOBT11 ZPABT1 ZPABT2 ZPCC1 ZPCC2 ZPCV1 ZPCV2 ZRIUTA ZRIUTB ZSCT1 ZSCT2 ZTST11

1771 1771 1771 1771 1771 1771 1771 1771 1771 1771 1771 1771 1771 1771 757 757 1771 1771 1771 1771 1771 1771 1771 1771 1771 1771 1771 1771 1771 1771 1771

Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Use database Use database Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw Header keyw

ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed value value ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed ord deemed

correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct correct

Those keywords whose values were used in calibration (all *CORR, CCR*, *TAB and *FILE keywords) and/or computed by the processing (CRVAL1, CRVAL2, MINWAVE, MAXWAVE) were deemed correct in the headers. UTC0 reports the base value of the spacecraft clock. The value is corrected from time to time, but is not changing constantly. It is used in the calculations for all other times reported in the FITS headers, including EXPSTART and EXPEND. Most cases, 45, where the database and the headers were different are confined to December 17, 1996, with the other 10 cases occuring on 4 others dates. We could find no problem reports addressing this issue. Because the other times are in agreement and the final processing would have the full set of UTC0 updates available during the processing, we elected to update the database with the UTC0, EXPSTART and EXPEND values from the headers.

7


The following 7 header keywords were updated with values from the database (as indicated in the table above): GRATING, OPFORMAT, PR_INV_L, SPEC_1, TARGNAME, ZLCOEF1 and ZLCOEF2. For GRATING and SPEC_1, a short term problem in the processing caused the grating name to be truncated in the headers of a small number of datasets. OPFORMAT had trailing blanks, which were removed. The ZLCOEF1 and ZLCOEF2 differences are for PHOTOSCAN exposures, where the database has actual values while the FITS headers have defaults. The actual values were retained. The TARGNAME database values were retained as they result from a long term cleanup effort. PR_INV_L suffered from inadequate handling of common features in some last names, such as apostrophes (i.e, O'Dell). Some early workarounds, while adequate for the database, did not play well with the FITS standard. We discovered the PR_INV_L differences after our initial reprocessing pass to fix the headers; therefore, we ran an additional pass on 1009 .shf.fits. These files have a slightly different processing date and FITS COMMENT block as opposed to the initial pass. Everything is documented in the headers. See below for more details. The FITSDATE header keyword was updated in all the files which contained this keyword; the value was updated to "16-DEC-2013" to reflect the fact that the FITS files were updated on this date, for all but 1009 headers. We updated 1009 .shf.fits FITS file headers with PR_INV_L corrections from the database once more in January 2014. Therefore, for those 1009 .shf.fits files, the FITSDATE is set to "3-JAN-2014" The TARDESCR, TARDESCR2, TARGCAT, TARCAT2, and TARGNAME keywords hold the target description, target categories, target keywords and target name as provided in the proposal. During the first ten or so years of the mission, instructions to the proposers on how to provide this information and the code that packaged the information into keywords that were passed to OPUS (the processing system) changed a number of times. A major effort was undertaken in 2005 to update the archive catalog (database) to bring the target descriptors, target categories and target keywords into agreement with the, by then, approved method of specifying the information. Rather than repeat this work, we elected to update TARGNAME in the header with the value from the database. We removed TARDESCR, TARDESCR2, TARGCAT and TARCAT2 as keywords and placed them in a series of comments in the header. Both the values from the header and the values from the database are contained in the comments add to the shf file of each dataset. An example is show here.
CO CO CO CO CO CO CO CO CO CO CO CO M M M M M M M M M M M M M M M M M M M M M M M M E E E E E E E E E E E E N N N N N N N N N N N N T T T T T T T T T T T T * * * * * * * * * * * * ------------------------------ STScI -------------------------------Header keywords TARGCAT, TARGCAT2, TARDESCR, TARDESC2 have been removed from the shf headers; their values and database values h a v e b e e n p r e s e r v e d in th is c o m m e n t b lo c k ( c r e a te d S e p te m b e r 2 0 1 3 ) . H N H N H N eader (old) keyword value TARGCAT: ULL eader (old) keyword value TARGCAT2: ULL eader (old) keyword value TARDESCR: ULL

8


CO CO CO CO CO CO CO CO CO CO CO CO

M M M M M M M M M M M M

M M M M M M M M M M M M

E E E E E E E E E E E E

N N N N N N N N N N N N

T T T T T T T T T T T T

* * * * * * * * * * * *

Header (old) keyword value TARDESC2: NULL D a ta b a s e (n e w ) k e y w o rd CALIBRATION D a ta b a s e (n e w ) k e y w o rd NULL Database (new) keyword CALIBRATION;DARK D a ta b a s e (n e w ) k e y w o rd NULL value TARGCAT: value TARGCAT2: value TARDESCR: value TARDESC2:

Finally, we documented our work in the headers. The following FITS COMMENT block was added to ALL GHRS files except the 1009 headers which had PR_INV_L header fixes (note, shf file has an additional COMMENT block, which is located after the following summary block, see above):
CO CO CO CO CO CO CO CO CO CO CO CO CO CO M M M M M M M M M M M M M M M M M M M M M M M M M M M M E E E E E E E E E E E E E E N N N N N N N N N N N N N N T T T T T T T T T T T T T T * * * * * * * * * * * * * * ------------------------------ STScI -------------------------------Details on the final header updates can be found in the GHRS ISR 93. (1) Headers were fixed to be FITS-compliant. ( 2 ) T h e f o llo w in g h e a d e r k e y s w e r e u p d a te d w ith v a lu e s f r o m th e d b : GRATING, OPFORMAT, SPEC_1, TARGNAME, ZLCOEF1 and ZLCOEF2 (3) Many reserved WCS keywords were removed from the FITS ASCII T a b le e x te n s io n ( H D U 1 ) b e c a u s e th e y a r e o n ly a llo w e d in I M A G E e x te n s io n s . (4) Non FITS compliant characters (bell, tabs, NULLs) were r e m o v e d f r o m H D U 1 in o c x , p d q , tr l f ile s . F in a l u p d a te s p e r f o r m e d a t S T S c I in D e c e m b e r 2 0 1 3 . ------------------------------ STScI --------------------------------

The following FITS COMMENT block was added to 1009 GHRS files headers which had PR_INV_L header fixes:
COM COM COM COM COM COM COM COM COM COM COM COM COM COM M M M M M M M M M M M M M M E E E E E E E E E E E E E E NT NT NT NT NT NT NT NT NT NT NT NT NT NT * * * * * * * * * * * * * * ------------------------------ STScI -------------------------------Details on the final header updates can be found in the GHRS ISR 93. (1) Headers were fixed to be FITS-compliant. ( 2 ) T h e f o llo w in g h e a d e r k e y s w e r e u p d a te d w ith v a lu e s f r o m th e d b : GRATING, OPFORMAT, PR_INV_L, SPEC_1, TARGNAME, ZLCOEF1, ZLCOEF2. (3) Many reserved WCS keywords were removed from the FITS ASCII Table extension (HDU1) because they are only allowed in IMAGE extensions. ( 4 ) N o n F I T S c o m p lia n t c h a r a c te r s ( b e ll, ta b s , N U L L s ) w e r e removed from HDU1 in ocx, pdq, trl files. F in a l u p d a te s p e r f o r m e d a t S T S c I in J a n u a r y 2 0 1 4 . ------------------------------ STScI --------------------------------

9


Ingest Changes
The database was populated from datasets produced by the production pipeline, which produces a standard dataset as follows, where the PDQ and OCX extensions were manually produced. Not every dataset has an OCX and/or PDQ. z2co0408c_ocx.fits z2co0408t_c0f.fits z2co0408t_c1f.fits z2co0408t_c2f.fits z2co0408t_c3f.fits z2co0408t_c4f.fits z2co0408t_cqf.fits z2co0408t_d0f.fits z2co0408t_pdq.fits z2co0408t_ulf.fits z2co0408t_q0f.fits z2co0408t_x0f.fits z2co0408t_shf.fits z2co0408t_xqf.fits z2co0408t_trl.fits

The final reprocessed datasets were produced using a standalone version of calhrs, the GHRS calibration software. When data are ingested, the database (archive catalog) is populated by placing the value of the headers keywords into specific tables. Because each dataset consists of a number of files, and some keywords may appear in the headers of more than one of these files, there is a defined order that indicates which file should be the source of the keyword value. The ingest software starts with the first file in the order, if the keyword is found, that value is placed in the database. If the keyword is not found, the next file is opened, and so on. The order of the files is fixed and the value in the database comes from the first instance of the keyword in the ordering. For pipeline processing the order was c0f, c1f, d0f, shf.

Acknowledgements
We thank Phil Hodge and Robert Jedrzejweski for very helpful discussion on the "nonASCII" character issues, and Mike Swam for his insight on the use of UTC0 in the science processing pipeline. We thank Charles Proffitt and Lou Strolger for helpful comments on the draft manuscript.

References
Kamp, I., Durand, D. and Micol, A. 2006, Instrument Science Report GHRS-92, "Final reprocessing of GHRS data."

Appendix
The following two examples list typical FITS verify errors as discussed in Section 2. 1) Example w.r.t. issues: EQUINOX non-floating point number, DATE-OBS illegal format, WCS Image keys not allowed in Table HDU, Data fill area invalid) File: z06a0101r.d0f.fits

10


2 Header-Data Units in this file. =================== HDU 1: Primary Array =================== *** Error: Keyword #223, DATE-OBS: (from CFITSIO error stack:) input date string has illegal format (ffs2dt): 9/05/90 *** Error: Keyword #217, EQUINOX: value = J2000 is not a floating point number. The value is entered as a string. 270 header keywords 32-bit floating point pixels, 2 axes (500 x 3), ==================== HDU 2: ASCII Table ==================== *** *** *** *** *** *** *** Error: Error: Error: Error: Error: Error: Error: Keyword Keyword Keyword Keyword Keyword Keyword checking #27, #23, #51, #15, #11, #19, data DATAMAX is not allowed in the Bin/ASCII table. DATAMIN is not allowed in the Bin/ASCII table. CTYPE1 is not allowed in the Bin/ASCII table. CRPIX1 is not allowed in the Bin/ASCII table. CRVAL1 is not allowed in the Bin/ASCII table. CD1_1 is not allowed in the Bin/ASCII table. fill: Data fill area invalid

110 header keywords Z06A0101R.D0H.TAB (25 columns x 3 rows) Col# Name (Units) Format 1 CRVAL1 D25.16 2 CRPIX1 E15.7 3 CD1_1 E15.7 4 DATAMIN E15.7 5 DATAMAX E15.7 6 RA_APER D25.16 7 DEC_APER D25.16 8 FILLCNT I11 9 ERRCNT I11 10 PKTTIME D25.16 11 CTYPE1 A8 12 OBSRPT I11 13 OBSINT I11 14 YDEF I11 15 XDEF I11

11


16 17 18 19 20 21 22 23 24 25

SAMPLE DELTAS LINE EXPOSURE BINID CARPOS ZSCOEF1 ZSCOEF2 ZSCOEF3 ZSCOEF4

E15.7 E15.7 E15.7 E15.7 I11 I11 E15.7 E15.7 E15.7 E15.7

++++++++++++++++++++++ Error Summary ++++++++++++++++++++++ HDU# Name (version) Type Warnings Errors 1 Primary Array 0 2 2 Z06A0101R.D0H.TAB ASCII Table 0 7 **** Verification found 0 warning(s) and 9 error(s). **** ///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////// ///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////// 2) Example w.r.t. issue: TDISP11 logical `L1' display format not allowed. File: z2dn0102t.d0f.fits 2 Header-Data Units in this file. ==================== HDU 2: ASCII Table ==================== *** Error: Keyword #75, TDISP11: Format L1 cannot be used for TFORM "A1". ///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////// ///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////// 3) Example w.r.t. issue: Warning regarding column names with `-` character, nonASCII-text characters. File: z06f0102c.ocx.fits 2 Header-Data Units in this file.

12


=================== HDU 1: Primary Array =================== 11 header keywords Null data array; NAXIS = 0 ==================== HDU 2: ASCII Table ==================== *** Warning: Column #1: Name "OCX-FILE" contains character '-' other than letters, digits, and "_". *** Error: row 24 data contains non-ASCII-text characters. *** Error: This ASCII table contains 1 non-ASCII-text characters *** Error: String in row #24, column #1 contains non-ASCII text. (Other rows may have errors). 13 header keywords z06f0102c.ocx (1 columns x 52 rows) Col# Name (Units) 1 OCX-FILE Format A132

++++++++++++++++++++++ Error Summary ++++++++++++++++++++++ HDU# Name (version) Type 1 Primary Array 0 2 z06f0102c.ocx ASCII Table Warnings Errors 0 1 3

**** Verification found 1 warning(s) and 3 error(s). **** ///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////

4) Specific error for 8 incomplete PDQ/OCX files:
archdevvm1> fverify z2sr0107t.pdq.fits FVERIFY V4.0.0 (CFITSIO V3.250) ------------------------------HEASARC conventions are being checked. File: z2sr0107t.pdq.fits *** Error: (from CFITSIO error stack:) ffopen could not interpret primary array header of file: z2sr0107t.pdq.fits **** Abort Verification: Fatal Error. ****

13


14