<html dir="ltr"> <head> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> <style id="owaParaStyle" type="text/css">P {margin-top:0;margin-bottom:0;}</style> </head> <body ocsi="0" fpstyle="1"> <div style="direction: ltr;font-family: Tahoma;color: #000000;font-size: 10pt;">Dear Rivet, YODA authors, I am using Rivet and YODA. I create many histograms (on the order of 400) and creating the .pdf plots for the .dat aabout 10 minutes. As I fine tune the YMin and YMax ranges, I want to be able to test it easily. In short, I want to be able to plot only one or a few histograms instead of them all. I find two potential solutions, but I have questions for both. 1. rivet-mkhtml ${RIVET_OUTPUT_FILE} -c ${RIVET_ANALYSIS}.plot -m counter* I see in rivet-mkhtml that -m and -M allow for a regular expression of the histogram. However, this does not work for me. The documentation says -m PATHPATTERNS, --match=PATHPATTERNS only write out histograms from analyses whose name matches any of these regexes -M PATHUNPATTERNS, --unmatch=PATHUNPATTERNS Exclude histograms whose $path/$name string matches these regexes There is an inconsistency between the two options, one says name, the other says $path/$name I tried all options but could not find one that works. My plots get saved in plots/RivetHbbBJets and there .dot files such as counter_El.dat h1D_3_ResponsePtCalMuNuMu.dat I would like to keep only those with "counter_" in the name for example, or remove those with "counter_" in the name. 2. I found in a Rivet tutorial online by Andy Buckley (which I found very useful, by the way) that I can replace the rivet-mkhtml command with two commands: # here make all the .dat files (it works) rivet-cmphistos ${RIVET_OUTPUT_FILE} -o plots --hier-out # here create only the histograms that I want (it works, so regular expressions work here) make-plots ./plots/RivetHbbBJets/*counter*.dat --pdfpng # but then we want one more step, that of making the index.html that has all the plots in one webpage. How is that done in a new command as well? I tried this again, but it did not work. rivet-mkhtml ${RIVET_OUTPUT_FILE} -c ${RIVET_ANALYSIS}.plot -m counter* It only creates an index.html in plots, but not in plots/RivetHbbBJets I create a similar script myself in bash, but it would be useful if it would provided in a simple form (presumably in python for consistency) by Rivet/Yoda. #!/bin/bash if [ $# -ne 1 ]; then cat <<EOF Usage: $0 Folder Usage: $0 afs/phas.gla.ac.uk/user/a/abuzatu/public_ppe/RivetHbbBJets EOF exit 1 fi FOLDER=$1 INDEX_FILE="index.html" # go to the folder that has the files pushd ${FOLDER} rm -f ${INDEX_FILE} for FILE in `ls -1 *.pdf` do STEM=${FILE%.*} echo " ${STEM} " >> ${INDEX_FILE} echo " <a href=\"./${STEM}.pdf\"> <img src=\"./${STEM}.png\" alt=\"Mountain View\"> </a> " >> ${INDEX_FILE} done # return to folde r when we run from popd exit The way to run everything in step 2: #RIVET_INPUT_FILE="${RIVET_INPUT_FILE_PATH}/ZHvvbb_1000000.hepmc" #RIVET_INPUT_FILE="${RIVET_INPUT_FILE_PATH}/ZHvvbb_3000000.nobias.hepmc" RIVET_INPUT_FILE="${RIVET_INPUT_FILE_PATH}/ZHvvbb_3000000.ptbias.hepmc" RIVET_OUTPUT_FILE="./yoda/ZHvvbb.yoda" RIVET_ANALYSIS="RivetHbbBJets" RIVET_PLOTS_FOLDER="/afs/phas.gla.ac.uk/user/a/abuzatu/public_html/${RIVET_ANALYSIS}" RIVET_RUN="rivet --pwd -a ${RIVET_ANALYSIS} ${RIVET_INPUT_FILE} -H ${RIVET_OUTPUT_FILE} -n 100 --histo-interval=10000 " #-l Rivet.Analysis=DEBUG -v " RIVET_MKHTML="rivet-mkhtml ${RIVET_OUTPUT_FILE} -c ${RIVET_ANALYSIS}.plot --mc-errs -M \"./plots/RivetHbbBJets/counter_lepton.dat\"' " # -n 8 -v RIVET_CMPHISTOS="rivet-cmphistos ${RIVET_OUTPUT_FILE} -o plots --hier-out" MAKE_PLOTS="make-plots ./plots/${RIVET_ANALYSIS}/counter*.dat --pdfpng -n 4" MAKE_HTML="./make_html.sh ${RIVET_PLOTS_FOLDER}" `echo ${RIVET_CMPHISTOS}` `echo ${MAKE_PLOTS}` rm -rf ${RIVET_PLOTS_FOLDER} mkdir -p ${RIVET_PLOTS_FOLDER} mv ./plots/${RIVET_ANALYSIS}/*.png ${RIVET_PLOTS_FOLDER}/. mv ./plots/${RIVET_ANALYSIS}/*.pdf ${RIVET_PLOTS_FOLDER}/. `echo ${MAKE_HTML}` Thank you, Adrian </div> </body> </html>