Tabular: prepend header

Tool

toolshed.g2.bx.psu.edu/repos/bgruening/text_processing/tp_awk_tool. The survey split awk into operation-named recipe pages; this page covers the recurring BEGIN { print "header" } idiom, not awk in general.

tp_text_file_with_recurring_lines plus concatenation can build constant prefixes, but IWC reaches for awk more often for this header-prepend operation.

When to reach for it

Use this when a tabular or text-like output needs a fixed first line inside the workflow. Common shapes include prepending a header and passing rows unchanged, prepending a header while skipping an old first row, and prepending a header while normalizing a few fields.

Do not use this for header stripping alone; that is usually Remove beginning1. Do not use it for collection header dedupe; use tabular-concatenate-collection-to-table with one_header: true.

Parameters

• code: awk program. Corpus uses both quoted one-liners and multiline |- blocks.
• infile: connected input.
• variables: usually [] in the cited examples.

Version pins vary (9.3+galaxy1, 9.5+galaxy0, 9.5+galaxy2, 9.5+galaxy3) with stable parameter shape. Prefer the newest/current pin already used in the workflow; do not change pins solely for cleanup.

Idiomatic shapes

Prepend a simple header, pass rows unchanged:

tool_id: toolshed.g2.bx.psu.edu/repos/bgruening/text_processing/tp_awk_tool/9.3+galaxy1
tool_state:
  code: 'BEGIN{print "Metric\tAlternate"}; {print}; '
  infile: { __class__: ConnectedValue }
  variables: []

Anchored by the VGP purge-duplicates IWC exemplar.

Prepend a header, skip the old first row, and normalize rows:

tool_id: toolshed.g2.bx.psu.edu/repos/bgruening/text_processing/tp_awk_tool/9.5+galaxy2
tool_state:
  code: |-
    BEGIN {OFS="\t"; print "genome\tcompleteness\tcontamination"}
    NR > 1 {
        if ($1 !~ /\.fasta$/)
            $1 = $1 ".fasta"
        print $1, $2, $3
    }
  infile: { __class__: ConnectedValue }
  variables: []

Anchored by the MAGs generation IWC exemplar.

Pitfalls

• Duplicate headers. If input already has a header, use NR > 1; plain {print} preserves the old header under the new one.
• Set OFS when printing fields. print $1, $2, $3 uses OFS; without BEGIN {OFS="\t"}, awk emits spaces.
• Quoted and multiline code both occur. Use one-line quoted strings only for simple pass-through. Use multiline |- for NR > 1, conditionals, or field rewriting.
• Do not name this awk. The page is operation-named per the survey decision. Sibling awk recipes get their own operation pages.
• Header injection is not collection header dedupe. Collection-wide header dedupe belongs to tabular-concatenate-collection-to-table.

Legacy alternative

tp_text_file_with_recurring_lines plus tp_cat can prepend constant text, but it is not the lead recommendation for this corpus pattern. Use it only when a separate reusable text file is already part of the workflow design.

See also

• iwc-tabular-operations-survey — §2g, §5, and §7 awk split decision.
• tabular-concatenate-collection-to-table — collection header dedupe via one_header.
• tabular-compute-new-column — simple per-row expressions without awk.
• tabular-filter-by-regex — line-level regex filtering without header injection.