markup: handle netpbm encoded annotations

markup/pdf.py

@@ -2,7 +2,9 @@ import os
 import sys
 import subprocess
 import shutil
+import dumper
 
+from pdfminer.psparser import LIT
 from pdfminer.pdfparser import PDFParser
 from pdfminer.pdfdocument import PDFDocument
 from pdfminer.pdftypes import PDFObjRef, resolve1
@@ -59,8 +61,15 @@ def make_scribble(obj, pagenum, mediabox, workdir):
                  'rect': pdf_rect(rect, mediabox[3]),
                  'objid': im1.objid,
                  'image': path }
+    elif flter.name == 'FlateDecode':
+        path = export_netpbm(im1, workdir, pagenum)
+        return { 'page': pagenum,
+                 'rect': pdf_rect(rect, mediabox[3]),
+                 'objid': im1.objid,
+                 'image': path }
     else:
-        print('skipping non-jp2 image')
+        print('skipping unrecognized image')
+        # print(dumper.dump(im1))
         return None
 
 
@@ -85,6 +94,55 @@ def export_jp2(obj, workdir, pagenum):
     return png_path
 
 
+def export_netpbm(obj, workdir, pagenum):
+    oid = obj.objid
+    ensure_dir(workdir)
+
+    pbm_base = os.path.join(workdir, f"export-page{pagenum:03d}-obj{oid:05d}")
+    pbm_path = write_pbm(obj, pbm_base)
+
+    # stencil mask - use instead if present
+    smask = obj.attrs['SMask']
+    if smask:
+        print('extracting pbm mask')
+        mask = resolve1(smask)
+        mask_base = os.path.join(workdir, f"export-page{pagenum:03d}-obj{oid:05d}-mask")
+        mask_path = write_pbm(smask, mask_base)
+        pbm_path = mask_path
+
+    return pbm_path
+
+
+def write_pbm(obj, base_path):
+    obj = resolve1(obj)
+    color_space = resolve1(obj.attrs['ColorSpace'])
+
+    suffix = '.pgm' if color_space == LIT('DeviceGray') else '.ppm'
+    path = base_path + suffix
+
+    print('writing pbm: {}'.format(path))
+
+    data = obj.get_data()
+    with open(path, 'wb') as out:
+        if suffix == '.pgm':
+            out.write("P5\n".encode())
+        else:
+            out.write("P6\n".encode())
+
+        out.write("{} {}\n".format(obj.attrs['Width'], obj.attrs['Height']).encode())
+
+        if obj.attrs['BitsPerComponent'] == 8:
+            out.write("255\n".encode())
+        else:
+            out.write("65535\n".encode())
+
+        out.write(data)
+
+    set_file_perms(path)
+
+    return path
+
+
 def parse_pdf(fname, workdir, debug=0):
     PDFDocument.debug = debug
     PDFParser.debug = debug