# `Accrue.PDF.ChromicPDF`
[🔗](https://github.com/szTheory/accrue/blob/accrue-v0.3.0/lib/accrue/pdf/chromic_pdf.ex#L1)

ChromicPDF adapter for `Accrue.PDF` (D-32, D-33).

## CRITICAL: Accrue does NOT start ChromicPDF (D-33, Pitfall #4)

This adapter ONLY calls into a ChromicPDF instance that the **host
application** has started in its own supervision tree. Neither this
module nor `Accrue.Application` (Plan 06) adds ChromicPDF to any
supervisor — doing so would fight with the host's config and make the
Chrome pool impossible to tune per-environment.

### Recommended host wiring

    # Dev / test
    children = [
      {ChromicPDF, on_demand: true}
    ]

    # Prod
    children = [
      {ChromicPDF, session_pool: [size: 5]}
    ]

Consult the `ChromicPDF` README for the full option set.

## Opts translation (RESEARCH Summary point 5)

The `Accrue.PDF` facade accepts `:header_html` / `:footer_html` (readable
English names). This adapter translates them to the `:header` / `:footer`
keys that `ChromicPDF.Template.source_and_options/1` actually wants. If
`:archival` is `true`, we route to `ChromicPDF.print_to_pdfa/1` instead
of `print_to_pdf/1`.

---

*Consult [api-reference.md](api-reference.md) for complete listing*