[Basespace_Fetch] Extending grep -E to project ID track (#981)

* addition of grep -E to project_id search and new search params

* making regex more robust

* docs

* strengthening separator check

* condense to elif

* doc clarifications

* update description

* rm unnecessary continue

* separator selection to previous functionality

---------

Co-authored-by: xonq <konkelzach@tuta.io>

Author: awh082834

docs/workflows/data_import/basespace_fetch.md

@@ -145,6 +145,9 @@ _If you already have a command-line environment available_, you can skip ahead t
 !!! warning "Sample_Name _and_ Sample_ID"
     If the Sample_Name and Sample_ID in the BaseSpace sample sheet are different, set the `basespace_sample_id` input attribute to "`this.basespace_sample_id"`.
 
+!!! warning "Nested Samplenames"
+    Erroneous matches may occur when a samplename is nested within another samplename in the same batch, separated by an underscore or space. This will occur when no lane suffix is present within the Basespace dataset. For example, "sample1" will retrieve "sample1_1", however, "sample1_L1" will NOT retrieve "sample1_1_L1". This should be taken into consideration when using certain naming conventions. 
+
 /// html | div[class="searchable-table"]
 
 {{ render_tsv_table("docs/assets/tables/all_inputs.tsv", input_table=True, filters={"Workflow": "BaseSpace_Fetch"}, columns=["Terra Task Name", "Variable", "Type", "Description", "Default Value", "Terra Status"], sort_by=[("Terra Status", True), "Terra Task Name", "Variable"]) }}

tasks/utilities/data_import/task_basespace_cli.wdl

@@ -39,9 +39,14 @@ task fetch_bs {
     #Grab BaseSpace Run_ID from given BaseSpace Run Name
     run_id=$(${bs_command} list run --retry | grep "~{basespace_collection_id}" | awk -F "|" '{ print $3 }' | awk '{$1=$1;print}' )
     echo "run_id: ${run_id}" 
+
+    # NOTE: substring matching will occur when the data table does not append a suffix; 
+    # e.g. where "sample1" will retrieve "sample1_1", however, "sample1_L1" will NOT retrieve "sample1_1_L1"
+    # This cannot be resolved without explicitly knowing the suffix prior to parsing. Noted in documentation
+
     if [[ ! -z "${run_id}" ]]; then 
       #Grab BaseSpace Dataset ID from dataset lists within given run 
-      dataset_id_array=($(${bs_command} list dataset --retry --input-run=${run_id} | grep -E "${dataset_name}([_]|$)" | awk -F "|" '{ print $3 }' )) 
+      dataset_id_array=($(${bs_command} list dataset --retry --input-run=${run_id} | grep -E "${dataset_name}(_| )[^_|^ ]* *\|[^\|]*\|[^\|]*\|[^\|]*\|" | awk -F "|" '{ print $3 }' )) 
       echo "dataset_id: ${dataset_id_array[*]}"
     else 
       #Try Grabbing BaseSpace Dataset ID from project name
@@ -50,7 +55,7 @@ task fetch_bs {
       echo "project_id: ${project_id}" 
       if [[ ! -z "${project_id}" ]]; then 
         echo "project_id identified via Basespace, now searching for dataset_id within project_id ${project_id}..."
-        dataset_id_array=($(${bs_command} list dataset --retry --project-id=${project_id} | grep "${dataset_name}" | awk -F "|" '{ print $3 }' )) 
+        dataset_id_array=($(${bs_command} list dataset --retry --project-id=${project_id} | grep -E "${dataset_name}(_| )[^_|^ ]* *\|[^\|]*\|[^\|]*\|[^\|]*\|" | awk -F "|" '{ print $3 }' )) 
         echo "dataset_id: ${dataset_id_array[*]}"
       else       
         echo "No run or project id found associated with input basespace_collection_id: ~{basespace_collection_id}" >&2
@@ -72,21 +77,23 @@ task fetch_bs {
     # setting a new bash variable to use for renaming during concatenation of FASTQs
     for elm in ./dataset_${dataset_id}/*.fastq.gz; do
       echo "Checking Basespace file: $elm"
-      if [[ $(echo "$elm" | cut -d '/' -f 3) =~ [-] && $sample_identifier =~ [_] ]]; then
-        echo "Basespace sample name for $(echo "$elm" | cut -d '/' -f 3) contains dashes, input sample identifier $sample_identifier contains underscores, renaming identifier..."
-        SAMPLENAME_RENAMED=$(echo $sample_identifier | sed 's|_|-|g' | sed 's|\.|-|g')
+      filename=$(basename "$elm")
+
+      if [[ "$filename" =~ [-] && "$sample_identifier" =~ [_] ]]; then
+        echo "Basespace sample name for $filename contains dashes, input sample identifier $sample_identifier contains underscores, renaming identifier..."
+        SAMPLENAME_RENAMED=$(echo "$sample_identifier" | sed 's|_|-|g' | sed 's|\.|-|g')
       fi
-      if [[ $(echo "$elm" | cut -d '/' -f 3) =~ [_] && $sample_identifier =~ [-] ]]; then
-        echo "Basespace sample name for $(echo "$elm" | cut -d '/' -f 3) contains underscores, input sample identifier $sample_identifier contains dashes, renaming identifier..."
-        SAMPLENAME_RENAMED=$(echo $sample_identifier | sed 's|-|_|g')
+      if [[ "$filename" =~ [_] && "$sample_identifier" =~ [-] ]]; then
+        echo "Basespace sample name for $filename contains underscores, input sample identifier $sample_identifier contains dashes, renaming identifier..."
+        SAMPLENAME_RENAMED=$(echo "$sample_identifier" | sed 's|-|_|g')
       fi
-      if [[ ($(echo "$elm" | cut -d '/' -f 3) =~ [_] && $sample_identifier =~ [_]) || ($(echo "$elm" | cut -d '/' -f 3) =~ [-] && $sample_identifier =~ [-]) ]]; then
-        echo "Both Basespace sample name and input sample identifier for $(echo "$elm" | cut -d '/' -f 3) contain matching separators..."
-        SAMPLENAME_RENAMED=$sample_identifier
+      if [[ ("$filename" =~ [_] && "$sample_identifier" =~ [_]) || ("$filename" =~ [-] && "$sample_identifier" =~ [-]) ]]; then
+        echo "Both Basespace sample name and input sample identifier for $filename contain matching separators..."
+        SAMPLENAME_RENAMED="$sample_identifier"
       fi
-      if [[ ! ($(echo "$elm" | cut -d '/' -f 3) =~ [_]) || ! ($(echo "$elm" | cut -d '/' -f 3) =~ [-]) ]]; then
+      if [[ ! ("$filename" =~ [_]) || ! ("$filename" =~ [-]) ]]; then
         echo "Filename doesn't use underscore or hyphen separators, using input sample identifier as-is"
-        SAMPLENAME_RENAMED=$sample_identifier
+        SAMPLENAME_RENAMED="$sample_identifier"
       fi
     done