Skip to content

fix: ft pg-lift --bam passes the output BAM positionally to pg-pansn - #153

Open
mtcicero26 wants to merge 1 commit into
mainfrom
fix/pg-lift-bam-output
Open

mtcicero26 wants to merge 1 commit into
mainfrom
fix/pg-lift-bam-output

Conversation

@mtcicero26

Copy link
Copy Markdown
Collaborator

ft pg-lift --bam fails on every input: the final stage runs ft pg-pansn - --strip --out <OUT> …, but pg-pansn takes its output BAM positionally, so clap stops with unexpected argument '--out' found. BED mode is unaffected (pg-inject does take --out).

  • The final BAM stage now passes the output BAM positionally.
  • The ft stage arguments are built by first_stage_args / final_stage_args instead of inline vectors.
  • New unit test pipeline_stage_args_parse parses every stage (BAM and BED mode) with ft's own Cli. It fails on the old arguments with the original error and passes with the fix.

Tested end to end with vg 1.77.0 on a small Minigraph-Cactus graph (GRCh38 + HG002 Q100 maternal/paternal), lifting FiberHMM-called Fiber-seq and scDAF-seq BAMs from HG002 to GRCh38. Other things we saw after vg inject | vg surject (supplementary flags, stale MD/NM, st on strand flips, BED12 blocks) are in a separate issue.

🤖 Generated with Claude Code

pg-pansn takes its output BAM as a positional argument, but pg-lift's
final BAM stage passed --out, so every BAM-mode lift stopped with
"unexpected argument '--out'". The ft stage arguments are now built by
first_stage_args/final_stage_args, and a unit test parses every stage
with ft's own CLI.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant