Metadata never fails a drop: deterministic title/author detection (metadata, first-page text, OCR of scanned first pages, file name), sources recorded in book.json; decrypt scratch copy before OCR

_pipeline/tests/helpers.py

@@ -0,0 +1,26 @@
+"""Small PDF builder for the metadata and fetch tests (needs reportlab, which the CI unit-test step installs)."""
+from reportlab.lib.pagesizes import letter
+from reportlab.pdfgen import canvas
+
+
+def make_pdf(path, pages, title=None, author=None):
+    """pages: list of lists of text lines; an empty list = a page with only a picture-like box (no text layer)."""
+    c = canvas.Canvas(path, pagesize=letter)
+    if title:
+        c.setTitle(title)
+    if author:
+        c.setAuthor(author)
+    for lines in pages:
+        y = 700
+        if not lines:
+            c.rect(100, 100, 300, 400, fill=1)
+        for ln in lines:
+            if ln == "":
+                y -= 30
+                continue
+            c.setFont("Helvetica", 18)
+            c.drawString(100, y, ln)
+            y -= 24
+        c.showPage()
+    c.save()
+    return path