b1b2e5b7727c19291acb605ad561dfbd225b0496
hiram
  Tue Jan 21 10:30:24 2020 -0800
cleaning up some errors for Ensembl v99 build no redmine

diff --git src/hg/utils/automation/findEnsFtpNames.sh src/hg/utils/automation/findEnsFtpNames.sh
index dac3122..467d53b 100755
--- src/hg/utils/automation/findEnsFtpNames.sh
+++ src/hg/utils/automation/findEnsFtpNames.sh
@@ -1,75 +1,85 @@
 #!/bin/sh
 
 # $Id: findEnsFtpNames.sh,v 1.1 2009/07/14 18:40:25 hiram Exp $
 
 VERSION=$1
 if [ "x${VERSION}y" = "xy" ]; then
     echo "usage: findEnsFtpNames.sh <ens version>"
     echo "where <ens version> is something like: 55"
     echo "this script will scan the ftp.ensembl.org site and extract"
     echo "the names from the files there that we need to create"
     echo "correspondence to UCSC database names"
     echo "when complete, look for result files:"
     echo "release.<ens version>.gtf.names"
     echo "release.<ens version>.MySQL.names"
     echo "release.<ens version>.fasta.names"
     echo "use those lists to edit EnsGeneAutomate.pm"
     exit 255
 fi
 
 echo "Scanning for GTF file names"
 
 echo "user anonymous hiram@soe
 cd pub/release-${VERSION}/gtf
 ls -lR
 bye" > ftp.rsp
 
 ftp -n -v -i ftp.ensembl.org < ftp.rsp > release.${VERSION}.gtf.ls-lR
 
 # the mus_musculus_ extra sequences are stuck at version 86
-egrep -v "CHECKSUMS|README" release.${VERSION}.gtf.ls-lR | awk '
+egrep -v "CHECKSUMS|README|Cyprinus_carpio_hebao_red|Cyprinus_carpio_german_mirror" release.${VERSION}.gtf.ls-lR | awk '
 {
 if (match($1,"^./[a-z0-9_]*:$")) {gsub(":$","",$1); printf "%s/", $1 }
 if (NF == 9) { if ((match($1,"^-rw")) && (match($NF,"'${VERSION}'.gtf.gz"))) {printf "%s\n", $NF} }
 if (NF == 9) { if ((match($1,"^-rw")) && (match($NF,"86.gtf.gz"))) {printf "%s\n", $NF} }
 }
 ' | sed -e "s#^./#'x' => '#; s#\$#',#" > release.${VERSION}.gtf.names
 
 echo "Scanning for MySQL table files"
 
 echo "user anonymous hiram@soe
 cd pub/release-${VERSION}/mysql
 ls -lR
 bye" > ftp.rsp
 
 ftp -i -n -v ftp.ensembl.org < ftp.rsp > release.${VERSION}.MySQL.ls-lR
 
 egrep "_core_${VERSION}.*:$" release.${VERSION}.MySQL.ls-lR \
   | sed -e 's/://;' | sed -e "s#^./#'x' => '#; s#\$#',#" \
      > release.${VERSION}.MySQL.names
 
 echo "Scanning for protein fasta files:"
 
 echo "user anonymous hiram@ucsc
 cd pub/release-${VERSION}/fasta
 ls -lR
 bye" > ftp.rsp
 
 ftp -i -n -v ftp.ensembl.org < ftp.rsp > release.${VERSION}.fasta.ls-lR
 
+
 awk '
-BEGIN{ D="notYet" }
+BEGIN{ D="notYet"; d="notyet" }
 {
   if (!match($1,"^drwx")) {
-    if (match($1,"^./[a-z_]*/pep:$")) {
+    if (match($1,"^./[0-9a-z_]*/pep:$")) {
         gsub(":$","",$1); D = $1;
+        d = tolower(D);
+        sub("./","", d);
+        sub("/pep","", d);
     }
-    if ((9 == NF) && match($1,"^-rw") && match($NF,"pep.all.fa")) {
+    if ((9 == NF) && match($1,"^-rw") && match($NF,"pep.all.fa.gz")) {
+        tl = tolower($NF)
+        if (index(tl, d) > 0) {
           printf "%s/%s\n", D, $NF
         }
     }
   }
+}
 ' release.${VERSION}.fasta.ls-lR \
 	| sed -e "s#^./#'x' => '#; s#\$#',#" > release.${VERSION}.fasta.names
 
+#         printf "%s/%s\t%s\t%s\n", D, $NF, d, tl
+#         printf "%s/%s\t%s\t%s\n", D, $NF, d, tl
+
 rm -f ftp.rsp