extend-catalog.py
1#!/usr/bin/env python3
2"""Generate the reviewed adapters' contracts, examples and runnable input fixtures."""
3import argparse,json,yaml
4from pathlib import Path
5ROOT=Path(__file__).resolve().parent
6specs=json.loads((ROOT/'extra-specs.json').read_text())
7parser=argparse.ArgumentParser(description=__doc__)
8parser.add_argument('operators',nargs='*',help='Regenerate selected operators, or all when omitted.')
9selected=parser.parse_args().operators or list(specs)
10assert not set(selected)-specs.keys(),'Unknown Galaxy operator'
11DESCRIPTIONS={
12'start':'First line number.','increment':'Positive increment between numbered lines.','width':'Minimum number-label width.','style':'Line-number alignment: ln left, rn right or rz zero-filled.',
13'column':'One-based selected field, or zero for a whole-line key where allowed.','numeric':'Compare keys as finite decimal/scientific numbers.','reverse':'Use descending key order.','ignore_case':'Compare without case differences.','header':'Header line count, or new header label for label-column operators.',
14'count':'Prefix each emitted group by its occurrence count.','mode':'Emit every group, repeated groups, or singleton groups.','separator':'Literal separator between values in the selected field.',
15'find':'Nonempty literal search text.','replacement':'Literal replacement text.','first_only':'Replace only the first match per row or selected field.','skip_header':'Keep the first input row unchanged.',
16'pattern':'Literal text or IUPAC DNA motif, as described by this operator.','invert':'Keep lines that do not contain the pattern.',
17'names':'Unique header names to select, remove or reorder.','discard':'Discard columns not explicitly selected.','text':'Literal line content.','position':'Insert the line as a header or footer.','times':'Number of generated repetitions.',
18'keep':'Keep only the named columns instead of removing them.','strip_prefix':'Characters stripped from the beginning of the first header name when matching.',
19'key_column':'One-based column to keep first, or zero to sort every header.','label':'Explicit input name to append or prepend.','prepend':'Place the added label column first.','widths':'Positive fixed-width character counts for each field.','pad_short':'Pad short fixed-width rows with spaces.',
20'keep_prefix':'Include the SOLiD primer nucleotide in the output.','mate':'One-based mate number, 1 or 2.','min_coverage':'Inclusive minimum accepted pileup depth.',
21'sequence_column':'One-based chromosome/sequence-name column.','position_column':'One-based input position column.','base_column':'One-based reference-base column.','coverage_column':'One-based pileup coverage column.',
22'reference_column':'One-based reference-allele column, or zero for N.','sample_column':'One-based VCF sample column, or zero for all samples.','prefix_chr':'Add chr to sequence names without that prefix.',
23'alphabet':'DNA or RNA alphabet for complementing.','to_stop':'Stop translation before the first stop codon.','cds':'Validate a complete standard-code CDS and omit its terminal stop.',
24'strand':'Search forward, reverse-complement, or both strands.','genome_size':'Estimated genome size for NG50, or zero to omit NG50.','motifs':'Literal DNA motifs, each written as sequence or name=sequence.','windows':'Number of adjacent windows considered on each side of the indel.'}
25sequence={'reverse_complement','translate','codon_count','base_distribution','find_subsequences','fasta_stats','colorspace_to_fasta'}
26genomic={'split_paired_reads','pileup_interval','divide_pg_snp','vcf_to_pgsnp','motif_flank_counts'}
27for name in selected:
28 s=specs[name]
29 docs={}
30 for key,prop in s['parameters']['properties'].items():
31 desc=prop.get("description",DESCRIPTIONS.get(key,key.replace("_"," ").capitalize()+"."));prop['description']=desc
32 docs[key]={'description':desc,'help':desc+' '+s['scope'],'example':s['params'].get(key,prop['default'])}
33 source='context:/galaxy/extended/'+name+'.txt'
34 job={'name':s['description'],'steps':[{'id':'read','operator':'item_source@1.0','inputs':{},'params':{'uri':source,'view':'bytes'}},{'id':'process','operator':s['name']+'@'+s['version'],'inputs':{'source':'read'},'params':s['params']}]}
35 extra_inputs=s['fixture'].get('inputs',{})
36 for port,value in extra_inputs.items():
37 read_id='read_'+port
38 job['steps'].insert(1,{'id':read_id,'operator':'item_source@1.0','inputs':{},'params':{'uri':'context:/galaxy/extended/'+name+'-'+port+'.txt','view':'bytes'}})
39 job['steps'][-1]['inputs'][port]=read_id
40 (ROOT/'data/extended'/f'{name}-{port}.txt').write_text(value)
41 job['steps'].append({'id':'client_sink','operator':'client_sink@1.0','inputs':{'result':'process'}})
42 s['inputs']={port:{'representations':['bytes'],'formats':[]} for port in ['source',*extra_inputs]}
43 s['documentation']={'inputs':{port:'UTF-8 bytes stream. '+s['scope'] for port in s['inputs']},'outputs':{'result':s['description']+' '+s['scope']},'parameters':docs,'example_job':yaml.safe_dump(job,sort_keys=False,allow_unicode=True)}
44 s['icon']={'path':'M4 3h12l4 4v14H4zM8 11h8M8 15h8m-8 4h5'}
45 data=ROOT/'data/extended'/f'{name}.txt';data.parent.mkdir(parents=True,exist_ok=True);data.write_text(s['fixture']['data'])
46 jobpath=ROOT/'jobs-extended'/f'{name}.yaml';jobpath.parent.mkdir(parents=True,exist_ok=True);jobpath.write_text(s['documentation']['example_job'])
47(ROOT/'extra-specs.json').write_text(json.dumps(specs,indent=2)+'\n')
48classes='\n'.join('object '+v['entrypoint'].split('.')[-1]+' extends '+v.get('implementation',('ExtendedSequence' if k in sequence else 'ExtendedGenomic' if k in genomic else 'ExtendedText'))+'("'+k+'")' for k,v in specs.items())
49schemas=[json.dumps({k:{'parameters':s['parameters']} for k,s in list(specs.items())[i:i+15]},separators=(',',':')) for i in range(0,len(specs),15)]
50registry='package xstream.galaxy\n// Generated by extend-catalog.py.\nobject ExtendedContract { val specs = new com.fasterxml.jackson.databind.ObjectMapper().createObjectNode(); private val mapper=new com.fasterxml.jackson.databind.ObjectMapper();\n'+''.join('specs.setAll[com.fasterxml.jackson.databind.JsonNode](mapper.readTree("""'+chunk+'""").asInstanceOf[com.fasterxml.jackson.databind.node.ObjectNode]);\n' for chunk in schemas)+'}\n'
51(ROOT/'src/main/scala/xstream/galaxy/ExtendedContract.scala').write_text(registry+classes+'\n')