Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 1 | #/bin/bash |
| 2 | |
| 3 | EURECO="/scratch/project_2010889/eureco" |
Harald Lüngen | 197aa20 | 2024-09-04 17:42:31 +0300 | [diff] [blame] | 4 | TEST="/scratch/project_2010889/TEST" |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 5 | |
| 6 | CORPUS="$EURECO/klk-fi-v2-vrt" |
| 7 | TEIDTD="$EURECO/tei/tei_all.dtd" |
| 8 | |
| 9 | DAT="$CORPUS/data" # input vrt files of klk-fi by year |
Harald Lüngen | 197aa20 | 2024-09-04 17:42:31 +0300 | [diff] [blame] | 10 | XML="$CORPUS/XML" # output zipped vrtxml files of selection of newspapers |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 11 | |
| 12 | |
Harald Lüngen | cb223bd | 2024-09-19 10:52:09 +0300 | [diff] [blame] | 13 | TEI="$CORPUS/TEI" # output zipped tei files of the selection |
| 14 | #TEI="$TEST/TEI" # output zipped tei files of the selection |
| 15 | |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 16 | TEIFORMAT="tei"; |
| 17 | #TEIFORMAT="i5"; |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 18 | |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 19 | MASK=""; |
| 20 | #MASK="-m"; |
| 21 | |
| 22 | PERL_BAD=0; # to supress warnings about locale |
| 23 | |
Harald Lüngen | 197aa20 | 2024-09-04 17:42:31 +0300 | [diff] [blame] | 24 | mkdir -p $TEST |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 25 | mkdir -p $XML |
| 26 | mkdir -p $TEI |
| 27 | |
| 28 | |
| 29 | |
| 30 | # Multiple threads for a loop in bash: |
| 31 | |
| 32 | ## for stuff in things |
| 33 | ## do |
| 34 | ## ( something |
| 35 | ## with |
| 36 | ## stuff ) & |
| 37 | ## done |
| 38 | ## wait # for all the something with stuff |
| 39 | |
| 40 | |
| 41 | |
| 42 | # Generate corpus files by year and source in vrt |
| 43 | |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 44 | #for VRT in $DAT/*.vrt # loop to be applied to yearly files |
| 45 | for VRT in "$DAT/klk_fi_v2_2021.vrt" |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 46 | do # ( |
| 47 | |
| 48 | BASENAME=`basename $VRT .vrt` |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 49 | YEAR=`echo $BASENAME | gawk 'BEGIN {FS="_"} {print $4}'` |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 50 | |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 51 | ##TMP echo "Generating proper XML files from $VRT in $XML/$YEAR by source..." |
| 52 | ##TMP gawk -v OUTDIR="$XML" -v YEAR="$YEAR" -f vrt2xml.awk $VRT |
| 53 | |
Harald Lüngen | cb223bd | 2024-09-19 10:52:09 +0300 | [diff] [blame] | 54 | |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 55 | |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 56 | #-------------------------------------------------------- |
Harald Lüngen | 6feedd2 | 2024-08-29 13:33:59 +0300 | [diff] [blame] | 57 | # Checking Wellformedness of the XML and generating TEI |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 58 | #-------------------------------------------------------- |
| 59 | |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 60 | ##TMP for s in $XML/*/ |
| 61 | for s in $XML/Suomen_Kuvalehti/ |
| 62 | do # ( # threading |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 63 | SOURCE=`basename $s` |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 64 | x=$s/$SOURCE$YEAR.xml |
| 65 | |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 66 | #--------------------------------------------------------------- |
| 67 | # Some existence checks and removal of empty files beforehand: |
| 68 | #--------------------------------------------------------------- |
| 69 | |
| 70 | if [[ ! -f $x ]] ; then |
| 71 | echo "Warning: klk2eureco.sh: File $x does not exist; skipping"; |
| 72 | break; |
| 73 | fi |
| 74 | |
| 75 | if [[ -z `grep -l "<text " $x` ]] ; then |
| 76 | #echo "Warning: klk2eureco.sh: File $x does not contain any <text>s, skipping" |
| 77 | echo "Warning: klk2eureco.sh: File $x does not contain any <text>s,removing" |
| 78 | rm -fv $x; |
| 79 | break; |
| 80 | fi |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 81 | |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 82 | #--------------------------------- |
| 83 | # checking well-formedness of $x |
| 84 | #--------------------------------- |
| 85 | |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 86 | echo " checking wellformedness of $x" |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 87 | xmllint --stream --noout $x |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 88 | if [[ $R != "" ]] ; then |
| 89 | echo "Error: xmllint error with error return code $R" >&2; |
| 90 | break; |
| 91 | fi |
| 92 | |
| 93 | |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 94 | |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 95 | #-------------------- |
| 96 | # calling vrt2tei.pl |
| 97 | #-------------------- |
| 98 | |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 99 | BASENAME=`basename $x .xml` |
| 100 | mkdir -p $TEI/$SOURCE |
Harald Lüngen | cb223bd | 2024-09-19 10:52:09 +0300 | [diff] [blame] | 101 | t="$TEI/$SOURCE/$BASENAME.$TEIFORMAT.xml" |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 102 | |
Harald Lüngen | 197aa20 | 2024-09-04 17:42:31 +0300 | [diff] [blame] | 103 | echo " generating $t using vrt2tei.pl, and prettifying..." |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 104 | perl vrt2tei.pl -t $TEIFORMAT $MASK $x | |
Harald Lüngen | 7aac973 | 2024-09-17 09:07:25 +0300 | [diff] [blame] | 105 | xml_pp | # xml_pp works but takes ages; |
| 106 | gawk -f rearrange-idsDoc.awk > $t # rearranging the start and end tag of idsDoc |
| 107 | # xmllint- format geht out of memory und --stream machen funktioniert nicht |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 108 | ls -l $t |
Harald Lüngen | 6feedd2 | 2024-08-29 13:33:59 +0300 | [diff] [blame] | 109 | |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 110 | # if $t is empty, remove it: |
| 111 | if [ ! -s $i ]; then |
| 112 | rm -f $i |
| 113 | fi |
| 114 | |
| 115 | # validating $t: |
Harald Lüngen | cb223bd | 2024-09-19 10:52:09 +0300 | [diff] [blame] | 116 | if [ "$TEIFORMAT" == "i5" ] ; then |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 117 | echo " validating $t..." |
Harald Lüngen | cb223bd | 2024-09-19 10:52:09 +0300 | [diff] [blame] | 118 | xmllint --stream --noout --valid $t |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 119 | else |
| 120 | echo " validating $t..." |
| 121 | xmllint --stream --noout --dtdvalid $TEIDTD $t # scheint so zu funktioneren - nicht kombinieren mit --format! |
Harald Lüngen | cb223bd | 2024-09-19 10:52:09 +0300 | [diff] [blame] | 122 | fi |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 123 | |
Harald Lüngen | 34c0561 | 2025-02-04 15:08:08 +0200 | [diff] [blame^] | 124 | ###TMP # echo " zipping $x..." |
| 125 | ###TMP # gzip -f $x |
| 126 | ###TMP |
| 127 | ###TMP # echo " zipping $t..." |
| 128 | ###TMP # gzip -f $t |
| 129 | |
| 130 | echo # ) & |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 131 | done |
Harald Lüngen | d488d5c | 2024-09-25 09:01:51 +0300 | [diff] [blame] | 132 | wait # ) & # wait seems to have to be there for it to terminate |
Harald Lüngen | 5bebb0c | 2024-08-27 16:44:34 +0300 | [diff] [blame] | 133 | |
| 134 | done |
| 135 | # wait |
| 136 | |
| 137 | |
| 138 | |