Print full ocrmypdf output on failure; sanitize control characters in warnings.log

_pipeline/bookio.py

@@ -30,7 +30,7 @@ def write_warnings(book_dir, stage, entries):
     if os.path.exists(path):
         with open(path, encoding="utf-8") as f:
             keep = [ln for ln in f.read().splitlines() if ln and not ln.startswith(stage + " |")]
-    one = lambda t: " / ".join(str(t).split("\n")).strip()     # one entry = one line
+    one = lambda t: " / ".join(re.sub(r"[\x00-\x08\x0b-\x1f]", " ", str(t).replace("\r", "\n")).split("\n")).strip()     # one entry = one line
     lines = keep + ["%s | %s | %s | %s" % (stage, k, one(w), one(m)) for k, w, m in entries]
     with open(path, "w", encoding="utf-8") as f:
         f.write("\n".join(lines) + ("\n" if lines else ""))

_pipeline/ingest.py

@@ -108,5 +108,6 @@ def ocr_if_needed(cfg, book_dir, book):
     src = os.path.join(book_dir, book["source_pdf"])
     r = subprocess.run(["ocrmypdf", "--skip-text", "-l", cfg["ocr"]["language"], src, out], capture_output=True, text=True)
     if r.returncode != 0:
+        print("ocrmypdf failed, exit %d. stdout: %r stderr: %r" % (r.returncode, r.stdout[-1500:], r.stderr[-3000:]))
         return [("ocr-failed", where, "ocrmypdf exit %d: %s" % (r.returncode, (r.stderr.strip()[:700] + " ... " + r.stderr.strip()[-500:]) if len(r.stderr) > 1250 else r.stderr.strip()))]
     return [("ocr", where, "OCR applied to a scratch copy; original untouched")]