FUNCTION eis_obs_structure, t0, t1, study_acr=study_acr, count=count, rast_acr=rast_acr, $ tl_id_input=tl_id_input, quiet=quiet, rast_id=rast_id, $ quick=quick ;+ ; NAME: ; EIS_OBS_STRUCTURE ; ; PURPOSE: ; Creates a structure containing information about all EIS ; rasters run during the specified time period T0 to T1. ; ; CATEGORY: ; Hinode/EIS; metadata ; ; INPUTS: ; T0: Start time of period in a standard format accepted by the ; ANYTIM routines. E.g., '1-feb-2011 00:00'. ; T1: End time of period in a standard format accepted by the ; ANYTIM routines. E.g., '2-feb-2011 00:00'. ; ; OPTIONAL INPUTS: ; Study_acr: Filters the output based on the study acronym. Can ; be an array, and string case does not matter. A ; partial match is OK, but the match must begin with ; the first character. ; Tl_id_input: This specifies a particular timeline ID (tl_id), ; and only results for this tl_id will be returned. ; Rast_acr: Filters the output on the raster acronym. ; Rast_id: Filters the output on the raster ID. ; ; KEYWORD PARAMETERS: ; QUIET: If set, then no informational messages will be printed ; to screen. ; QUICK: If set, then the routine does not query the Oslo catalog, ; which speeds up the routine but means the xcen_corr and ; ycen_corr tags are not filled. ; ; OPTIONAL OUTPUTS: ; Count: Number of rasters satisfying search criteria. ; ; OUTPUT: ; A structure array with various tags giving information about ; the rasters that were run. If no entries are found, then a ; value of -1 is returned. ; ; Please check notes below about xcen and ycen values. Generally ; you should use the xcen_corr and ycen_corr values. ; ; Some of the tags are: ; .nexp No. of exposures in an exposure set. ; .exp_times String containing the exposure times for the set. ; .nsteps No. of exposure sets. For a raster, this is the x-size ; of the raster. For sit-and-stare it is the number of ; repeats at the slit position. ; ; PROGRAMMING NOTES: ; XCEN_CORR, YCEN_CORR ; These are keywords stored at the OSDC ; (http://sdc.uio.no/search/form) that contain the best xcen and ; ycen values. That is, they are derived from the getxcen and ; getycen object methods, and then they are corrected for the ; AIA-EIS offset (eis_aia_offsets.pro). An internet connection ; is required to retrieve these values. If you find the values are ; NaNs, then this means the routine was not successful in ; retrieving the values. ; ; XCEN_PLAN, YCEN_PLAN ; Sometimes the xcen and ycen values for a raster have very ; large errors. In this case it is preferable to use the ; *planned* coordinates, and these are stored in the xcen_plan ; and ycen_plan tags. *However*, if there are raster repeats ; then xcen_plan and ycen_plan are the coordinates of the *first* ; raster in the sequence. ; ; RAST_ACR, STUDY_ACR ; If you're filtering the search based on the ; raster/study acronym, then it is much quicker to filter on ; the raster acronym due to how the databases are ; structured. Generally there's a one-to-one ; correspondence between the raster and study acronyms, but ; usually the acronyms are different. To find ; the raster acronym for a study, use the routine EIS_XSTUDY, ; find your study, and you should see the raster acronym ; towards the bottom of the widget. Note that you can search on ; the study and raster acronym together. ; ; Exposure times (NEXP, EXP_TIMES, NSTEPS) ; For a given slit position, EIS can take up to 8 exposures, and ; these exposures can have different exposure times. An "exposure ; set" denotes the set of exposures at the slit position. The tag ; NEXP gives the number of exposures in a set. Usually there is ; only one exposure (NEXP=1), but sometimes EIS may take short ; and long exposures to prevent saturation in strong lines ; (NEXP=2). ; ; The exposure times for the set are given in the tag EXP_TIMES. ; This is a string, with the exposure times separated by a space. ; For example, NEXP=2 and EXP_TIMES="5.0 50.0" means there are ; two exposures in the set of 5 s and 50 s duration. ; ; The number of exposure sets in a raster is given by NSTEPS. For ; a spatial raster, this is the x-size of the raster. For ; sit-and-stare observations, this is the number of repeated ; exposure sets. Thus if NSTEPS=30 and NEXP=2, then there were ; 60 exposures in total. ; ; EXAMPLES: ; IDL> s=eis_obs_structure('1-jan-2014','1-feb-2014') ; IDL> s=eis_obs_structure('1-jan-2014','1-jan-2015',study_acr='Atlas_30') ; ; The example below shows how different the runtime can be if ; rast_acr is used: ; IDL> s=eis_obs_structure('15-jan-2007','20-aug-2010',study_acr='pry_footpoints') ; IDL> s=eis_obs_structure('15-jan-2007','20-aug-2010',study_acr='pry_footpoints', $ ; rast_acr='cool_lines') ; The first one takes 1-2 minutes, the second takes a few seconds. ; ; MODIFICATION HISTORY: ; Version 1, 2-Apr-2012, Peter Young ; Version 2, 16-Jul-2012, Peter Young ; added obstitle, obs_dec, sci_obj and hop_id to the output ; structure. ; Version 3, 4-Jan-2013, Peter Young ; added STUDY_ACR optional input; fixed bug that led to ; incorrect rast_acr and rast_id tags; added COUNT= optional ; output. ; Version 4, 10-Apr-2014, Peter Young ; added some additional tags to the output structure. ; Version 5, 4-Aug-2014, Peter Young ; added search for tl_id. ; Version 6, 3-Apr-2015, Peter Young ; now filters out entries with bad date information. ; Version 7, 10-Nov-2015, Peter Young ; now remove any entries prior to 23-Nov-2006 13:14 as these ; have bad meta-data; also make sure date_end > date_obs. ; Version 8, 18-Nov-2015, Peter Young ; I've added the tags 'xcen_plan' and 'ycen_plan' which ; are the planned coordinates of the raster. ; Version 9, 19-Nov-2015, Peter Young ; Added 'offset_index' tag to output. ; Version 10, 13-Sep-2019, Peter Young ; Fixed problem with count output not working in some cases; ; updated header format. ; Version 11, 6-Dec-2019, Peter Young ; Fixed bug with rast_acr search - it is much faster now; ; moved the check on study_acr to improve speed; added some ; informational messages about speed. ; Version 12, 17-Jan-2020, Peter Young ; Removed engineering studies from output by checking ; rast_acr. ; Version 13, 01-Apr-2020, Peter Young ; Added 'atlas' tag to denote if raster is a spectral atlas. ; Version 14, 02-Apr-2020, Peter Young ; Fixed bug when /quiet set. ; Version 15, 05-Feb-2021, Peter Young ; The input rast_acr was being redefined in the code, ; so I've changed the internal parameter to ; rast_acr_out; the t0 and t1 parameters that were used for ; timing code have been changed to tt0 and tt1. ; Version 16, 08-Feb-2021, Peter Young ; Fixed bug whereby I was still computing the durations with ; t1-t0 instead ot tt1-tt0. ; Version 17, 04-Oct-2021, Peter Young ; Fixed directory problem that stopped routine working in ; Windows; removed the NRL thumbnail link since the site does ; not exist now. ; Version 18, 16-Dec-2021, Peter Young ; Added yws to output structure. ; Version 19, 03-Feb-2022, Peter Young ; Added rast_id optional input. ; Version 20, 23-Aug-2022, Peter Young ; Modified check on study_acr so that a partial match is allowed. ; Version 21, 25-Aug-2023, Peter Young ; Added the tags xcen_corr and ycen_corr to the output. These ; contain the xcen_corr and ycen_corr keywords from the OSDC, ; which contain the most accurate pointings for the rasters. ; Version 22, 28-Aug-2023, Peter Young ; The previous update didn't work if the tl_id option was used, ; so this has been fixed. ; Version 23, 12-Sep-2023, Peter Young ; Fixed bug in calculation of NSTEPS for sit-and-stare data; ; added extra information to header. ; Version 24, 13-Sep-2023, Peter Young ; Now handles case when xcen_corr and ycen_corr are both zero. ; Version 25, 14-Sep-2023, Peter Young ; Added /quick keyword. ;- IF n_params() LT 2 AND n_elements(rast_acr) EQ 0 AND n_elements(rast_id) EQ 0 AND n_elements(tl_id_input) EQ 0 THEN BEGIN print,'Use either' print,' IDL> obs=eis_obs_structure( t0, t1 [, study_acr=, count=, rast_acr=, rast_id= ])' print,'or' print,' IDL> obs=eis_obs_structure( rast_acr= [, t0, t1, study_acr=, count= ])' print,'or' print,' IDL> obs=eis_obs_structure( rast_id= [, t0, t1, study_acr=, count= ])' print,'or' print,' IDL> obs=eis_obs_structure( tl_id= )' count=0 return,-1 ENDIF IF n_elements(study_acr) NE 0 AND n_elements(rast_acr) EQ 0 THEN BEGIN print,'% EIS_OBS_STRUCTURE: *WARNING* generally it is much quicker to search on the raster acronym (RAST_ACR) rather' print,' than the study acronym. See the header for more information.' ENDIF ; ; The following give the paths to the EIS databases used by this ; routine. The databases are: ; ; dbexpt - The "as-run" raster database ; dbmain - The "as-run" study database ; dbrast - The raster technical database ; dbline - The line list technical database ; dbtlras - The "as-planned" raster database ; dbase_dir=concat_dir(getenv('SSW'),'hinode') dbase_dir=concat_dir(dbase_dir,'eis') dbase_dir=concat_dir(dbase_dir,'database') cat_dir=concat_dir(dbase_dir,'catalog') plan_dir=concat_dir(dbase_dir,'planning_db') tech_dir=concat_dir(plan_dir,'technical_db') tl_dir=concat_dir(plan_dir,'timeline_db') ; dbexpt=concat_dir(cat_dir,'eis_experiment') dbmain=concat_dir(cat_dir,'eis_main') dbrast=concat_dir(tech_dir,'eis_raster_db') dbline=concat_dir(tech_dir,'eis_linelist_db') dbtlras=concat_dir(tl_dir,'eis_science2_db') ; ; Construct time query ; IF n_elements(t0) NE 0 AND n_elements(t1) NE 0 THEN BEGIN t0_tai=anytim2tai(t0) t0_tai_str=trim(t0_tai,'(f15.3)') t1_tai=anytim2tai(t1) t1_tai_str=trim(t1_tai,'(f15.3)') ; query='date_obs > '+t0_tai_str+', date_end < '+t1_tai_str ENDIF ELSE BEGIN query='' ENDELSE ; ; Construct raster acronym query ; nra=n_elements(rast_acr) IF nra NE 0 THEN BEGIN IF nra EQ 1 THEN BEGIN r_query='rast_acr = '+rast_acr[0] ENDIF ELSE BEGIN r_query='rast_acr = [' FOR i=0,n_elements(rast_acr)-1 DO BEGIN r_query=r_query+trim(rast_acr[i]) IF i EQ n_elements(rast_acr)-1 THEN r_query=r_query+']' ELSE r_query=r_query+',' ENDFOR ENDELSE IF query EQ '' THEN query=r_query ELSE query=query+', '+r_query ENDIF ; ; Construct raster ID query ; nra=n_elements(rast_id) IF nra NE 0 THEN BEGIN IF nra EQ 1 THEN BEGIN r_query='rast_id = '+trim(rast_id[0]) ENDIF ELSE BEGIN r_query='rast_id = [' FOR i=0,n_elements(rast_id)-1 DO BEGIN r_query=r_query+trim(rast_id[i]) IF i EQ n_elements(rast_id)-1 THEN r_query=r_query+']' ELSE r_query=r_query+',' ENDFOR ENDELSE IF query EQ '' THEN query=r_query ELSE query=query+', '+r_query rast_id_save=rast_id ENDIF ; ; Timeline ID (tl_id) query. ; IF n_elements(tl_id_input) NE 0 THEN BEGIN t_query='tl_id = '+trim(tl_id_input) IF query EQ '' THEN query=t_query ELSE query=query+', '+t_query ENDIF IF query EQ '' THEN BEGIN print,'%EIS_OBS_STRUCTURE: either a time range T0,T1 should be specified, or RAST_ACR should be specified.' print,' Returning...' count=0 return, -1 ENDIF ; ; Extract information from the as-run EIS raster database ; IF NOT keyword_set(quiet) THEN print,'% EIS_OBS_STRUCTURE: querying as-run raster database...' tt0=systime(1) dbopen,dbexpt list=dbfind(query,/silent) dbext,list,'xcen,ycen,fovx,fovy,filename,tl_id,date_obs,date_end,rast_acr,rast_id,yws', $ xcen,ycen,fovx,fovy,filename,tl_id,date_obs,date_end,rast_acr_out,rast_id,yws dbclose tt1=systime(1) IF NOT keyword_set(quiet) THEN print,'% EIS_OBS_STRUCTURE: query took '+trim(string(format='(f15.1)',tt1-tt0))+' seconds.' ; ; Sort output arrays according to observation time ; k=sort(date_obs) date_obs=tai2utc(double(date_obs[k]),/ccsds) date_end=tai2utc(double(date_end[k]),/ccsds) xcen=xcen[k] ycen=ycen[k] fovx=fovx[k] fovy=fovy[k] filename=filename[k] tl_id=tl_id[k] rast_acr_out=rast_acr_out[k] rast_id=rast_id[k] yws=yws[k] ; ; Create the output observation structure ; n=n_elements(date_obs) str={ date_obs: '', date_end: '', xcen: 0., ycen: 0., xcen_plan: 0., ycen_plan: 0., $ xcen_corr: !values.f_nan, ycen_corr: !values.f_nan, $ fovx: 0., fovy: 0., $ filename: '', tl_id: 0l, stud_acr: '', rast_acr: '', link: '', rast_id: 0, $ wavelnth:'', wavemin: '', wavemax: '', obstitle: '', obs_dec: '', $ hop_id: 0, sci_obj: '', slit_width: 0, sit_stare: 0b, step_size: 0., $ slit_index: 0, nsteps: 0, nexp: 0, exp_times: '', offset_index: 0, $ atlas: 0b, yws: 0} struct=replicate(str,n) ; ; Populate the structure with items from the raster database. ; struct.date_obs=date_obs struct.date_end=date_end struct.xcen=xcen struct.ycen=ycen struct.fovx=fovx struct.fovy=fovy struct.filename=trim(str_replace(filename,'.gz','')) struct.tl_id=tl_id struct.rast_acr=trim(rast_acr_out) struct.rast_id=rast_id struct.yws=yws ; ; If the tl_id option was used, then t0 and t1 will not be defined, ; so set them here. (Needed for OSDC call.) ; IF n_elements(t0) EQ 0 THEN BEGIN t0=struct[0].date_obs t1=struct[n-1].date_end ENDIF ; ; 25-Aug-2023 ; The ycen values have become very inaccurate, so I now query the OSDC to ; return xcen_corr and ycen_corr values. ; IF NOT keyword_set(quick) THEN BEGIN ; ; By specifying the raster ID(s) in the OSDC search, it should make the ; search faster. ; rast_id=struct.rast_id ra_id=rast_id[uniq(rast_id,sort(rast_id))] nr=n_elements(ra_id) ra_id_str=trim(ra_id[0]) IF nr GT 1 THEN BEGIN FOR i=1,nr-1 DO ra_id_str=ra_id_str+','+trim(ra_id[i]) ENDIF ; t_osdc_0=systime(1) IF NOT keyword_set(quiet) THEN message,/info,/CONTINUE,'Querying OSDC to return xcen_corr and ycen_corr. This can be turned off with /QUICK.' o=obj_new('osdc',/quiet) o->show,'DATE_OBS,XCEN_CORR,YCEN_CORR' o->limit,0 o->condition,'INSTRUME: ,EIS' o->condition,'EPOCH_START: '+t0 o->condition,'EPOCH_END: '+t1 o->condition,'SS__L1LEAD: y' o->condition,'Gx: NONE' o->condition,'E__RAST_ID: '+ra_id_str o->search,osdc_data obj_destroy,o t_osdc_1=systime(1) IF NOT keyword_set(quiet) THEN message,/info,/CONTINUE,'OSDC query took '+trim(string(format='(f6.1)',t_osdc_1-t_osdc_0))+' seconds.' ENDIF ; ; Sometimes xcen_corr and ycen_corr come back as zero from the OSDC. This ; seems to be because the FITS files aren't available, usually because the ; files are new. I thus retain NaNs for these files. ; IF n_tags(osdc_data) NE 0 THEN BEGIN osdc_date_obs_tai=anytim2tai(osdc_data.date_obs) FOR i=0,n-1 DO BEGIN date_obs_tai=anytim2tai(struct[i].date_obs) getmin=min(abs(osdc_date_obs_tai-date_obs_tai),imin) IF getmin LE 1.0 THEN BEGIN osdc_xcen_corr=osdc_data[imin].xcen_corr osdc_ycen_corr=osdc_data[imin].ycen_corr IF osdc_xcen_corr NE 0. OR osdc_ycen_corr NE 0. THEN BEGIN struct[i].xcen_corr=osdc_xcen_corr struct[i].ycen_corr=osdc_ycen_corr ENDIF ENDIF ENDFOR ENDIF ; ; When no rasters are found, the code always seems to put an empty ; entry with a date corresponding to 1957 (!), so here I check for ; this and delete it. ; ; 10-Nov-2015: I've modified this to check for date_obs after ; 23-Nov-2006 13:13. This (I believe) is when the EIS planning tool ; started working, and so entries before this have bad meta-data. I ; also added a check for date_end > date_obs. There's one study ; in particular that fails this. ; ; 17-Jan-2020: I now check to make sure rast_acr is not 'NA' as this ; indicates an engineering study. (I think a negative rast_id is also a ; flag, but I'm not sure of this.) ; date_obs_tai=anytim2tai(date_obs) date_end_tai=anytim2tai(date_end) rast_acr_out=struct.rast_acr chck_tai=anytim2tai('23-nov-2006 13:13') k=where(date_obs_tai GE chck_tai AND date_end_tai GE date_obs_tai AND rast_acr_out NE 'NA',nk) IF nk EQ 0 THEN BEGIN IF NOT keyword_set(quiet) THEN print,'% EIS_OBS_STRUCTURE: no studies found. Returning...' count=0 return,-1 ENDIF ELSE BEGIN struct=struct[k] ENDELSE ; ; Create a link to the NRL thumbnail webpage. ; ; 04-Oct-2021, PRY: commented this out since NRL site no longer ; exists. ; ;struct.link='http://tcrb.nrl.navy.mil/~iuu/eis/thumbnails/rasters/'+strlowcase(trim(struct.rast_acr))+'/'+strlowcase(trim(struct.filename))+'.html' ; ; Extract the study acronym from the as-run study database. ; (note that ">" means "greater-or-equal-to" for the database software) ; IF NOT keyword_set(quiet) THEN print,'% EIS_OBS_STRUCTURE: querying as-run study database...' tt0=systime(1) dbopen,dbmain query='tl_id > '+trim(min(tl_id))+', tl_id < '+trim(max(tl_id)) list=dbfind(query,/silent) dbext,list,'stud_acr,tl_id,obstitle,obs_dec,sci_obj,jop_id,obs_num',stud_acr,st_tl_id, $ obstitle,obs_dec,sci_obj,hop_id,obs_num dbclose tt1=systime(1) IF NOT keyword_set(quiet) THEN print,'% EIS_OBS_STRUCTURE: query took '+trim(string(format='(f15.1)',tt1-tt0))+' seconds.' ; ; Extract the xcen and ycen values from the *as-planned* ; database. These go in the xcen_plan and ycen_plan tags of the output ; structure. ; Note that tl_id in the as-planned raster database is identified as ; 'sn_id'. ; IF NOT keyword_set(quiet) THEN print,'% EIS_OBS_STRUCTURE: querying as-planned raster database...' tt0=systime(1) dbopen,dbtlras query='sn_id > '+trim(min(tl_id))+', sn_id < '+trim(max(tl_id)) list=dbfind(query,/silent) dbext,list,'sn_id,x_point,y_point',tl_tl_id,xcen_plan,ycen_plan dbclose tt1=systime(1) IF NOT keyword_set(quiet) THEN print,'% EIS_OBS_STRUCTURE: query took '+trim(string(format='(f15.1)',tt1-tt0))+' seconds.' ; ; Populate the stud_acr tag of the observation structure. ; ; For obstitle, obs_dec and sci_obj, the string returned by the ; database software has 68 characters, with the last character being a ; "&". I don't think there's a fix for this (Ron's software ; returns the strings in this same format). I'm thus just ; truncating the string to 67 characters to remove the &. ; n=n_elements(stud_acr) FOR i=0,n-1 DO BEGIN k=where(struct.tl_id EQ st_tl_id[i],nk) IF nk NE 0 THEN BEGIN struct[k].stud_acr=trim(stud_acr[i]) struct[k].obstitle=trim(strmid(obstitle[i],0,67)) struct[k].obs_dec=trim(strmid(obs_dec[i],0,67)) struct[k].sci_obj=trim(strmid(sci_obj[i],0,67)) struct[k].hop_id=hop_id[i] struct[k].offset_index=obs_num[i] ENDIF ENDFOR ; ; The following loads up the tags 'xcen_plan' and 'ycen_plan' using ; information from the as-planned database (dbtlras). ; n=n_elements(tl_tl_id) FOR i=0,n-1 DO BEGIN k=where(struct.tl_id EQ tl_tl_id[i],nk) IF nk NE 0 THEN BEGIN struct[k].xcen_plan=xcen_plan[i] struct[k].ycen_plan=ycen_plan[i] ENDIF ENDFOR ; ; Perform a filter using the STUDY_ACR input. ; IF n_elements(study_acr) NE 0 THEN BEGIN nn=n_elements(study_acr) ind=-1 FOR i=0,nn-1 DO BEGIN ; ; I'm checking to see if the study acronym begins with the string contained in study_acr. ; A partial match is OK, but the match must begin with the first character. ; chck=strpos(strlowcase(struct.stud_acr),trim(strlowcase(study_acr[i]))) k=where(chck EQ 0,nk) ; k=where(trim(strlowcase(struct.stud_acr)) EQ trim(strlowcase(study_acr[i])),nk) IF nk NE 0 THEN ind=[ind,k] ENDFOR IF n_elements(ind) GT 1 THEN BEGIN ind=ind[1:*] struct=struct[ind] ENDIF ELSE BEGIN struct=-1 ENDELSE ENDIF ; IF n_tags(struct) EQ 0 THEN BEGIN count=0 IF n_elements(rast_id_save) NE 0 THEN rast_id=rast_id_save return,struct ENDIF ELSE BEGIN count=n_elements(struct) ENDELSE slit_width=[1,266,2,40] ; ; Extract information about the raster wavelength windows using the ; line list and raster definition databases. ; IF NOT keyword_set(quiet) THEN print,'% EIS_OBS_STRUCTURE: extracting line list and raster information...' tt0=systime(1) ; FOR i=min(struct.rast_id),max(struct.rast_id) DO BEGIN k=where(i EQ struct.rast_id,nk) IF nk GT 0 THEN BEGIN dbopen,dbrast query='id = '+trim(i) list=dbfind(query,/silent) dbext,list,'ll_id,n_windows,data_windows,rastertype,slit_index,scan_fm_nsteps,scan_fm_stepsize,nexp,exposures,sns_nexps', $ ll_id,n_windows,data_windows,rastertype,slit_index,scan_fm_nsteps, $ scan_fm_stepsize,nexp,exposures,sns_nexps n_windows=n_windows[0] dbclose,dbrast ; dbopen,dbline query='id = '+trim(ll_id) list=dbfind(query,/silent) dbext,list,'n_lines,wavelength',n_lines,wavelength n_lines=n_lines[0] dbclose,dbline ; IF n_lines GT 0 THEN BEGIN ; ; Check if raster is an atlas (4 windows, each of 1024 pix) ; IF n_lines EQ 4 AND min(data_windows[0:n_lines-1]) EQ 1024 AND max(data_windows[0:n_lines-1]) EQ 1024 THEN struct[k].atlas=1b ; wavemin=lonarr(n_lines) wavemax=lonarr(n_lines) FOR j=0,n_lines-1 DO BEGIN wavemin[j]=fix((float(wavelength[j])/100.-data_windows[j]/2.*0.0223)*100.) wavemax[j]=fix((float(wavelength[j])/100.+data_windows[j]/2.*0.0223)*100.) ENDFOR ; wave_str=trim(wavelength[0]) wavemin_str=trim(wavemin[0]) wavemax_str=trim(wavemax[0]) IF n_lines GT 1 THEN BEGIN FOR j=1,n_lines-1 DO BEGIN wave_str=wave_str+' '+trim(wavelength[j]) wavemin_str=wavemin_str+' '+trim(wavemin[j]) wavemax_str=wavemax_str+' '+trim(wavemax[j]) ENDFOR ENDIF struct[k].wavelnth=wave_str struct[k].wavemin=wavemin_str struct[k].wavemax=wavemax_str ENDIF ; struct[k].sit_stare=rastertype[0] struct[k].slit_width=slit_width[slit_index[0]] struct[k].slit_index=slit_index[0] ; ; For sit-and-stare observations, nsteps is the number of exposure sets and ; is obtained from sns_nexps ; IF rastertype[0] EQ 1 THEN BEGIN struct[k].nsteps=sns_nexps[0] ENDIF ELSE BEGIN struct[k].nsteps=scan_fm_nsteps[0] ENDELSE IF rastertype[0] NE 1 THEN struct[k].step_size=scan_fm_stepsize[0] nexp=nexp[0] struct[k].nexp=nexp exps=exposures[0:nexp-1] exp_str='' FOR ii=0,nexp-1 DO exp_str=exp_str+' '+trim(string(format='(f10.1)',float(exps[ii])/1000.)) exp_str=trim(exp_str) struct[k].exp_times=exp_str ENDIF ENDFOR tt1=systime(1) IF NOT keyword_set(quiet) THEN print,'% EIS_OBS_STRUCTURE: completed in '+trim(string(format='(f15.1)',tt1-tt0))+' seconds.' ; ; Restores the original input value to rast_id. ; IF n_elements(rast_id_save) NE 0 THEN rast_id=rast_id_save return,struct END