Validation results

Release records#

• 1.8.4: Corpus benchmark and release comparison,
• 1.8.1: The standards-conformance run is preserved below.

The 1.8.4 benchmark reproduced every 1.8.3 file outcome and diagnostic detail across 46,944 normal inputs and 80 damaged inputs. These open/save benchmarks and the standards checks below measure different things.

Standards-conformance validation: KillerPDF 1.8.1#

Validation date: 2026-08-29

KillerPDF 1.8.1 rewrote 2,898 of 2,907 deliberately hostile conformance PDFs through The KillerPDF.Engine. The release candidate completed with zero rewrite failures, zero new veraPDF failures, and zero qpdf structural regressions.

Question under test: does saving a PDF through KillerPDF degrade its standards conformance? Every file in a 2,907-file public corpus was validated, resaved through KillerPDF's standard open/save pipeline, and validated again.

Result: Zero conformance regressions across every file KillerPDF saved. 74 files came out more conformant than they went in.

Tools#

Corpus#

2,907 PDFs from the public conformance suites: the veraPDF test corpus (PDF/A-1, PDF/A-2, PDF/A-4, PDF/UA-1, PDF/UA-2), the Isartor PDF/A-1b test suite, and the TWG test files. These are deliberately hostile files: most are constructed to violate exactly one clause of a standard, so any structural damage a resave introduces shows up as a new failed rule.

Method#

1. Validate the pristine corpus: verapdf --recurse --format json <corpus> > baseline.json
2. Resave every file through KillerPDF: KillerPDF.exe --batch-resave <corpus> <resaved> --log resave.csv
3. Validate the resaved tree the same way into after.json
4. Compare-VeraPDF.ps1 matches files by relative path and flags any file that fails a rule after the resave that it did not fail before
5. qpdf sweep: qpdf --check on original and resave of all 2,898 saved files; flag any file whose exit code worsened

veraPDF results#

The nine skips are four encrypted files that batch mode deliberately does not decrypt or strip, plus five parser-hostile files that do not provide enough reliable structure for a safe rewrite. Each one is a SKIP row in resave.csv. No file went missing for any other reason.

The 74 improvements are a side effect, not a goal: many corpus files carry deliberately malformed structure (bad trailers, broken xref, wrong stream lengths), and rewriting the file through a clean serializer repairs that class of defect.

qpdf structural sweep#

qpdf --check on the original and the resave of all 2,898 saved files:

No file's structural health got worse; 374 files with qpdf warnings came out clean.

Build under test#

KillerPDF 1.8 replaces the legacy document pipeline with The KillerPDF.Engine. This gate used the 1.8.1 Release candidate built from the current source tree.

Reproducing this run#

Everything needed ships in this folder or is a free download (veraPDF, qpdf, the public corpora). On a tree containing the corpus:

verapdf --recurse --format json C:\pdf-corpus > baseline.json
Start-Process -Wait KillerPDF.exe -ArgumentList '--batch-resave','C:\pdf-corpus','C:\pdf-corpus-resaved','--log','resave.csv'
verapdf --recurse --format json C:\pdf-corpus-resaved > after.json
.\Compare-VeraPDF.ps1 -Baseline baseline.json -After after.json `
    -BaselineRoot C:\pdf-corpus -AfterRoot C:\pdf-corpus-resaved `
    -ResaveLog resave.csv -CsvOut compare.csv
.\QpdfSweep.ps1 -Corpus C:\pdf-corpus -Resaved C:\pdf-corpus-resaved `
    -ResaveLog resave.csv -CsvOut qpdf-results.csv

The resave step must wait for KillerPDF to exit before the after-scan starts. The comparison uses resave.csv to distinguish the nine explicit skips from missing output. Any other missing file, newly failed rule, or new veraPDF parse failure is a regression.