refactor checking file extensions

dlibrary/dlibrary.py

@@ -329,8 +329,17 @@ def link_ordered_files(ordering, dest, start_index):
         link_path = dest / f'{idx:04d}{ext}'
         link_path.symlink_to(relpath(src_path, dest))
 
+def check_extension(path, exts):
+    return path.suffix.lower() in exts
+
+def is_pdf(path):
+    check_extension(path, ['.pdf'])
+
+def is_image(path):
+    check_extension(path, IMAGE_FILE_EXTENSIONS)
+
 def ignoreable(path):
-    return path.name in IGNOREABLE_FILES or path.suffix.lower() in IGNOREABLE_EXTENSIONS
+    return path.name in IGNOREABLE_FILES or check_extension(path, IGNOREABLE_EXTENSIONS)
 
 def ls_ignore(directory):
     return [
@@ -338,9 +347,12 @@ def ls_ignore(directory):
         if not ignoreable(path)
     ]
 
-def descendant_files_ignore(directory):
+def descendant_files_ignore(path):
+    if path.is_file():
+        return [path]
+
     result = []
-    for item in ls_ignore(directory):
+    for item in ls_ignore(path):
         if item.is_dir():
             result.extend(descendant_files_ignore(item))
         else:
@@ -389,7 +401,7 @@ def collate(args):
     collation_staging_area.rmdir()
     con.close()
 
-def collate_split_regex(srcs, dest, start_index, earlier=None, later=None):
+def try_collate_split_regex(srcs, dest, start_index, earlier=None, later=None):
     early_srcs = []
     middle_srcs = []
     late_srcs = []
@@ -443,11 +455,56 @@ def median(items):
     items.sort()
     return items[len(items) // 2]
 
+def superior_or_equal(a, b):
+    return len(a) >= len(b) and all(a[i] >= b[i] for i in range(len(b)))
+
+def try_collate_images_vs_pdf(srcs, dest, start_index):
+    pdfs = [src for src in srcs if 'pdf' in src.name.lower()]
+    if len(pdfs) != 1:
+        return False
+    outer_pdf = pdfs[0]
+
+    inner_pdfs = [f for f in descendant_files_ignore(outer_pdf) if is_pdf(f)]
+    if len(inner_pdfs) != 1:
+        return False
+    inner_pdf = inner_pdfs[0]
+
+    non_pdf_srcs = [src for src in srcs if src != outer_pdf]
+    images = []
+    non_images = []
+    descendant_files = [f for src in non_pdf_srcs for f in descendant_files_ignore(src)]
+    for f in descendant_files:
+        if is_image(f):
+            images.append(f)
+        else:
+            non_images.append(f)
+            break
+
+    if len(non_images) != 0 or len(images) == 0:
+        return False
+
+    pdf_sizes = pdf_image_sizes(inner_pdf)
+    standalone_sizes = [standalone_image_size(f) for f in images]
+    if abs(len(pdf_sizes) - len(standalone_sizes)) > 2:
+        return False
+
+    median_pdf_size = median(pdf_sizes)
+    median_standalone_size = median(standalone_sizes)
+    if not (median_pdf_size and median_standalone_size):
+        return False
+
+    if superior_or_equal(median_standalone_size, median_pdf_size):
+        return collate_from_paths(non_pdf_srcs, dest, start_index)
+    elif superior_or_equal(median_pdf_size, median_standalone_size):
+        return collate_from_paths([outer_pdf], dest, start_index)
+    else:
+        return False
+
 def collate_from_paths(srcs, dest, start_index):
     if len(srcs) == 1 and srcs[0].is_dir():
         return collate_from_paths(ls_ignore(srcs[0]), dest, start_index)
 
-    if len(srcs) == 1 and srcs[0].suffix.lower() == '.pdf':
+    if len(srcs) == 1 and is_pdf(srcs[0]):
         print(f'Extracting images from {srcs[0]}')
         return link_pdf(srcs[0], dest, start_index)
 
@@ -462,19 +519,19 @@ def collate_from_paths(srcs, dest, start_index):
             if len(descendant_files_ignore(lo_res_dir)) == len(descendant_files_ignore(hi_res_dir)):
                 return collate_from_paths([hi_res_dir], dest, start_index)
 
-    textless_split = collate_split_regex(srcs, dest, start_index, later=TEXTLESS_REGEX)
+    textless_split = try_collate_split_regex(srcs, dest, start_index, later=TEXTLESS_REGEX)
     if textless_split != False:
         return textless_split
 
-    epilogue_split = collate_split_regex(srcs, dest, start_index, later=EPILOGUE_REGEX)
+    epilogue_split = try_collate_split_regex(srcs, dest, start_index, later=EPILOGUE_REGEX)
     if epilogue_split != False:
         return epilogue_split
 
-    cover_split = collate_split_regex(srcs, dest, start_index, earlier=COVER_REGEX)
+    cover_split = try_collate_split_regex(srcs, dest, start_index, earlier=COVER_REGEX)
     if cover_split != False:
         return cover_split
 
-    if all(src.is_file() and src.suffix.lower() in IMAGE_FILE_EXTENSIONS for src in srcs):
+    if all(src.is_file() and is_image(src) for src in srcs):
         ordering = complete_prefix_number_ordering(srcs)
         if ordering:
             print(f'Symlinking image files: {ordering[0]}...')
@@ -483,33 +540,9 @@ def collate_from_paths(srcs, dest, start_index):
         else:
             return None
 
-    pdfs = [src for src in srcs if src.is_file() and src.suffix.lower() == '.pdf']
-    if len(pdfs) == 1:
-        pdf = pdfs[0]
-        images = []
-        non_images = []
-        descendant_files = [
-            src for src in srcs if src != pdf and src.is_file()
-        ] + [
-            f for src in srcs if src.is_dir() for f in descendant_files_ignore(src)
-        ]
-        for f in descendant_files:
-            if f.suffix.lower() in IMAGE_FILE_EXTENSIONS:
-                images.append(f)
-            else:
-                non_images.append(f)
-                break
-        if len(non_images) == 0 and len(images) > 0:
-            pdf_sizes = pdf_image_sizes(pdf)
-            standalone_sizes = [standalone_image_size(f) for f in images]
-            if abs(len(pdf_sizes) - len(standalone_sizes)) <= 2:
-                median_pdf_size = median(pdf_sizes)
-                median_standalone_size = median(standalone_sizes)
-                if median_pdf_size and median_standalone_size:
-                    if median_standalone_size[0] >= median_pdf_size[0] and median_standalone_size[1] >= median_pdf_size[1]:
-                        return collate_from_paths([src for src in srcs if src != pdf], dest, start_index)
-                    if median_pdf_size[0] >= median_standalone_size[0] and median_pdf_size[1] >= median_standalone_size[1]:
-                        return collate_from_paths([pdf], dest, start_index)
+    images_vs_pdf = try_collate_images_vs_pdf(srcs, dest, start_index)
+    if images_vs_pdf != False:
+        return images_vs_pdf
 
     return None
 
@@ -534,17 +567,17 @@ def manual_collate(args):
     index = 0
     for path in args.paths:
         if path.is_dir():
-            entries = [p for p in path.iterdir() if p.suffix.lower() in IMAGE_FILE_EXTENSIONS]
+            entries = [p for p in path.iterdir() if p.is_file() and is_image(p)]
             ordering = complete_prefix_number_ordering(entries)
             if ordering is None:
                 ordering = entries
                 ordering.sort()
             link_ordered_files(ordering, collation_dir, index)
             index += len(ordering)
-        elif path.suffix.lower() in IMAGE_FILE_EXTENSIONS:
+        elif is_image(path):
             link_ordered_files([path], collation_dir, index)
             index += 1
-        elif path.suffix.lower() == ".pdf":
+        elif is_pdf(path):
             pdf_page_count = link_pdf(path, collation_dir, index)
             if pdf_page_count is None:
                 return
@@ -563,10 +596,10 @@ def analyze(args):
 
     for f in files:
         print(f'{relpath(f, extract_dir)}', end='')
-        if f.suffix.lower() in IMAGE_FILE_EXTENSIONS:
+        if is_image(f):
             size = standalone_image_size(f)
             print(f'\t{fmt_size(size)}')
-        elif f.suffix.lower() == '.pdf':
+        elif is_pdf(f):
             sizes = pdf_image_sizes(f)
             if len(sizes) == 0:
                 print(f'\tContains no images')