From: Steven Baltakatei Sandoval <baltakatei@gmail.com>
Date: Tue, 14 Feb 2023 13:24:38 +0000 (+0000)
Subject: feat(user/bkfeh):Use unique names for sample files
X-Git-Tag: 0.8.1~11
X-Git-Url: https://zdv2.bktei.com/gitweb/BK-2020-03.git/commitdiff_plain/82fabd0214119cbc81c09edc2ba64418b77e00eb?ds=sidebyside;hp=-c

feat(user/bkfeh):Use unique names for sample files

- Note: Use both file order and file b2sum 32-bit hashes
---

82fabd0214119cbc81c09edc2ba64418b77e00eb
diff --git a/user/bkfeh b/user/bkfeh
index 0f8154a..911d51b 100755
--- a/user/bkfeh
+++ b/user/bkfeh
@@ -1,8 +1,8 @@
 #!/usr/bin/env bash
 # Desc: Wrapper for feh that accepts directory paths via posargs or stdin lines.
-# Version: 0.2.0
+# Version: 0.3.0
 # Ref/Attrib: [1] Tange, Ole. GNU Parallel with Bash Array. 2019-03-24. https://unix.stackexchange.com/a/508365/411854
-# Depends: GNU Parallel, GNU Bash v5.1.16, feh 3.6.3
+# Depends: GNU Parallel, GNU Bash v5.1.16, feh 3.6.3, GNU Coreutils 8.32 (b2sum)
 
 #===Declare local functions===
 yell() { echo "$0: $*" >&2; } # print script path and all args to stderr
@@ -153,7 +153,10 @@ displayMissing() {
     #==END Display errors==
 } # Display missing apps, files, dirs
 check_depends() {
-    if ! checkapp feh parallel; then displayMissing; die "FATAL:Missing apps."; fi;
+    if ! checkapp feh parallel bkshuf b2sum; then
+        displayMissing;
+        die "FATAL:Missing apps.";
+    fi;
     return 1;
 }; # check dependencies
 checkInt() {
@@ -249,7 +252,8 @@ save_sample() {
     # Input: arg1   list_paths        (list of files to take samples from)
     #        envvar BKFEH_SAMPLE_DIR  (environment variable set outside of this script)
     #        envvar BKFEH_SAMPLE_SIZE (space limit for sample dir files)
-    # Depends: yell(), GNU Parallel, GNU find, GNU Coreutils 8.32 (cut, find, du)
+    # Depends: GNU Parallel, GNU find, GNU Coreutils 8.32 (cut, find, du)
+    #   BK-2020-03: bkshuf (0.0.1), yell()
     local list_paths
     sample_count="100"; # max number of images to put in sample dir
     sample_max_space="10000000"; # max bytes to put in sample dir
@@ -263,33 +267,55 @@ save_sample() {
         sample_count="$BKFEH_SAMPLE_COUNT";
     fi;
     
-    if [[ ! -z "$1" ]]; then
+    if [[ -n "$1" ]]; then
         list_paths="$1"; # newline-delimited list of file paths to sample from
     else
         yell "ERROR:NO paths available to sample.";
     fi;
     
     if [[ -d "$BKFEH_SAMPLE_DIR" ]]; then
-        sample_dir="$BKFEH_SAMPLE_DIR";
+        #sample_dir="$BKFEH_SAMPLE_DIR";
         yell "STATUS:Environment variable BKFEH_SAMPLE_DIR set. Clearing and saving samples...";
         
         ## clear previous sample
-        count_samples="$(find $BKFEH_SAMPLE_DIR -maxdepth 1 -type f | wc -l)";
+        count_prev_samples="$(find "$BKFEH_SAMPLE_DIR" -maxdepth 1 -type f | wc -l)";
+        yell "STATUS:Deleting $count_prev_samples previous samples...";
         find "$BKFEH_SAMPLE_DIR" -maxdepth 1 -type f -exec rm '{}' \; ;
         
         ## save random sample
         yell "STATUS:Saving random sample of size $sample_count to $BKFEH_SAMPLE_DIR...";
-        list_paths_sample="$(echo "$list_paths" | shuf | head -n"$sample_count")";
+        list_paths_sample="$(echo "$list_paths" | bkshuf "$sample_count" | head -n"$sample_count")";
+        n_samp=0; # init sample file counter
+        sample_log="$BKFEH_SAMPLE_DIR"/paths.txt;
+        printf "%s,%s,%s\n" "n_samp" "file_hash" "file_path" >> "$sample_log";
         while read -r line; do
             if [[ -z "$line" ]]; then continue; fi;
             ### check size limit
             sample_act_space="$(du -bd1 "$BKFEH_SAMPLE_DIR" | cut -f1 )"; # actual used space
             cand_space="$(du -bd1 "$line" | cut -f1 )"; # size of candidate file to add
             sample_req_space="$((sample_act_space + cand_space))";
+
+            ### Customize file names
+            n_samp_w="$(printf "%s" "$sample_count" | wc -c)";
+            n_samp_fmt="%0""$n_samp_w""d";
+            n_samp_dd="$(printf "$n_samp_fmt" "$n_samp")"; # sample number fixed-width
+            file_path="$line";
+            #file_dir="$(dirname "$line")";
+            file_name="$(basename "$line")";
+            file_hash="$(b2sum -l32 "$line" | awk '{print $1}')"; # use file hash to avoid clobbering
+            file_ext="${file_name##*.}";
+            file_name="${file_name%.*}";
+            file_shortname="${file_name:0:32}";
+            file_name_new="$n_samp_dd"_"$file_hash".."$file_shortname"."$file_ext";
+            file_path_new="$BKFEH_SAMPLE_DIR"/"$file_name_new"            
             if [[ "$sample_req_space" -lt "$sample_max_space" ]]; then
                 #### add file to sample dir
-                cp -n "$line" "$BKFEH_SAMPLE_DIR" ;
+                must cp -n "$file_path" "$file_path_new";
+                #### note path in sample dir log
+                printf "%s,%s,%s\n" "$n_samp_dd" "$file_hash" "$file_path" \
+                       >> "$sample_log";
             fi;
+            ((n_samp++));
         done < <( echo "$list_paths_sample" );
     else
         yell "ERROR:Does not exist: $BKFEH_SAMPLE_DIR";