22 lines
2.0 KiB
Python
22 lines
2.0 KiB
Python
from pathlib import Path
|
|
import json,hashlib,zipfile,xml.etree.ElementTree as ET
|
|
out=Path(__file__).resolve().parent;inv=json.loads((out/'INVENTARIO-ACERVO.json').read_text('utf-8'));dest=out/'FONTES-LEVANTADAS.json';d=json.loads(dest.read_text('utf-8'))
|
|
names=['melhorar e transformar o Portal do Investidor atual.docx','upgrade do portal de investimentos v6 programacao.docx']
|
|
paths=[Path(x.get('path')or x.get('absolute_path')or Path(inv['corpus_inventory_reused']['root'])/x.get('relative_path',''))for x in inv['corpus_files']]
|
|
for p in paths:
|
|
if p.name not in names or str(p)in[x['path']for x in d['sources']]:continue
|
|
if not p.is_file():continue
|
|
with zipfile.ZipFile(p)as z:
|
|
ns={'w':'http://schemas.openxmlformats.org/wordprocessingml/2006/main'};tree=ET.fromstring(z.read('word/document.xml'))
|
|
paras=[''.join(n.text or ''for n in para.findall('.//w:t',ns))for para in tree.findall('.//w:p',ns)]
|
|
embeds=[{'entry':i.filename,'bytes':i.file_size,'CRC32':format(i.CRC,'08x')}for i in z.infolist()if not i.is_dir()and(i.filename.startswith('word/media/')or i.filename.startswith('word/embeddings/'))]
|
|
d['sources'].append({'path':str(p),'available':True,'bytes':p.stat().st_size,'sha256':hashlib.sha256(p.read_bytes()).hexdigest(),'suffix':'.docx','paragraphs':paras,'embedded':embeds,'selection':'Exact source named by existing park requirements; no master relationship inferred'})
|
|
if p.name==names[0]:
|
|
found=set()
|
|
for i,s in enumerate(paras):
|
|
if any(k in s.casefold()for k in ['ficha deve ser preenchida por parque','além dos campos comuns, cada parque','ficha operacional','infraestrutura comum','modelo de gestão','lotes internos']):
|
|
found.update(range(max(0,i-2),min(len(paras),i+18)))
|
|
print(json.dumps({'name':p.name,'sha256':hashlib.sha256(p.read_bytes()).hexdigest(),'selected_paragraphs':[{str(i+1):paras[i]}for i in sorted(found)]},ensure_ascii=True))
|
|
dest.write_text(json.dumps(d,ensure_ascii=False,indent=2)+'\n','utf-8')
|
|
print(json.dumps({'sources_total':len(d['sources'])}))
|